Encyclopedia of Electrochemical Power Sources (7-Volume Set) [2 ed.] 0323960227, 9780323960229

Table of contents :
Volume 1
Volume 2
Volume 3
Volume 4
Volume 5
Volume 6
Volume 7

Citation preview

ENCYCLOPEDIA OF ELECTROCHEMICAL POWER SOURCES SECOND EDITION

This page intentionally left blank

ENCYCLOPEDIA OF ELECTROCHEMICAL POWER SOURCES SECOND EDITION EDITOR-IN-CHIEF Jürgen Garche Ulm University, Ulm, Germany VOLUME 1

SECTION EDITORS Yuichi Aihara Nanoscale Organisation and Dynamics Group, School of Science and Health Western Sydney University, Penrith, NSW, Australia; Nissan Motor Corporation, Yokohama, Japan

Klaus Brandt Lithium Battery Consultant, Wiesbaden, Germany

Sylvain Brimaud Center for Solar Energy and Hydrogen Research (ZSW), Ulm, Germany

Angelika Heinzel University of Duisburg-Essen (emerita), Duisburg, Germany

Ludwig Jörissen Center for Solar Energy and Hydrogen Research (ZSW), Ulm, Germany

Peter Kurzweil Technical University of Applied Sciences (OTH), Amberg-Weiden, Germany

Joachim Scholta Center for Solar Energy and Hydrogen Research (ZSW), Ulm, Germany

Elsevier Science Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands 125 London Wall, London, EC2Y 5AS, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge MA 02139, United States Copyright © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. Publisher’s note: Elsevier takes a neutral position with respect to territorial disputes or jurisdictional claims in its published content, including in maps and institutional affiliations. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers may always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-323-96022-9 For information on all publications visit our website at http://store.elsevier.com

Publisher: Oliver Walter Acquisitions Editor: Blerina Osmanaj Content Project Manager: Snehil Sharma Associate Content Project Manager: Kiruthigadevi N Designer: Matthew Limbert

EDITORIAL BOARD Editor-in-Chief Jürgen Garche, Ulm University, Ulm, Germany Jürgen Garche graduated in chemistry at the Dresden University of Technology (DTU) in Germany in 1967. He was awarded his Ph.D. in theoretical electrochemistry in 1970 and his habilitation in applied electrochemistry in 1980 from the same university. He worked at the DTU in the Electrochemical Power Sources Group for many years in different projects, mainly related to conventional batteries, before he moved 1991 to the Centre for Solar Energy and Hydrogen Research (ZSW) in Ulm, where he was, until 2004, the head of the Electrochemical Energy Storage and Energy Conversion Division. He was professor of Electrochemistry at Ulm University and guest professor at Shandong University—China, 2005, Sapienza University of Rome—Italy, 2009, 2013, 2016, and 2023, TUM-CREATE—Singapore, 2014–2016, Dalian Institute of Chemical Physics—China, 2016, CNR Institute for Advanced Energy Technologies, Messina— Italy, 2019. After he retired from the ZSW, he founded in 2004 the consulting firm Fuel Cell and Battery Consulting (FCBAT). Since 2015, he is a senior professor at Ulm University. He has published more than 300 papers, 10 patents, and 11 books, among others as editor-in-chief of the first edition of Encyclopedia of Electrochemical Power Sources. He is listed in “World’s most Influential Scientific Minds” by Thomas Reuters (2014) and in the book “Profiles of 93 Influential Electrochemists” (2015).

Section Editors Yuichi Aihara, Nanoscale Organisation and Dynamics Group, School of Science and Health, Western Sydney University, Penrith, NSW, Australia; Nissan Motor Corporation, Yokohama, Japan Yuichi Aihara received his Ph.D. from Mie University in Japan. He is an electrochemist/battery engineer and since 2022 has been the deputy general manager of the Advanced Material and Processing Laboratory, Nissan Research Centre, Nissan Motor Co. Ltd., in Kanagawa, Japan. He started his research career in the battery industry in 1991 at Yuasa, Japan, and conducted research on solid polymer electrolytes and ion conduction mechanisms in various media. In 2003, he joined Samsung Research Institute Japan and from 2003 to 2010, he investigated fuel cells by performing research and development on intermediate temperature proton-conducting membranes. From 2010 to 2020, he was the principal engineer and later director of Battery R&D; he developed sulfide-based solid-state batteries. During 2020–2022, he was the manager of Solid-State Battery Development at Mercedes Benz AG, Stuttgart. In addition, he undertook

v

vi

Editorial board

postdoctoral studies at the University of Rome during 2000, where he studied solid polymer electrolyte cell under the guidance of Professor B. Scrosati. Also, he is a long-time collaborator of the Nanoscale Group, Western Sydney University, Australia, where he has held the position of an adjunct professor position since 2021. He has published more than 70 papers related to battery technology and ion conduction mechanisms and has an h-index of 36 (Scopus). He has also filed more than 120 patents.

Klaus Brandt, Lithium Battery Consultant, Wiesbaden, Germany Klaus Brandt received his Ph.D. in Physics from the Technical University of Munich. He has spent his career in the battery industry in Canada, the USA, and Germany. He conducted and managed research and development in battery start-ups, large battery companies, and chemical companies in the field of materials for lithium batteries, cell development, and battery design. He was a cofounder of Moli Energy, now E-One Moli Energy, the first company to produce rechargeable lithium metal anode batteries. Whereas the focus of his work was initially on batteries to power consumer devices, he later focused on large batteries to power, for example, electric vehicles. He is now a consultant to the lithium battery industry and to the automotive industry. He has published more than 30 scientific papers and has been awarded more than 20 patents.

Sylvain Brimaud, Center for Solar Energy and Hydrogen Research (ZSW), Ulm, Germany Sylvain Brimaud graduated in Chemistry at the University of Poitiers in France in 2005. He received his Ph.D. for research works on the physical and electrochemical properties of Pt nanoparticles from the same university in 2008. He then moved in 2009 to Ulm University (Germany) as post-doc, where he pursued research works on the basic understanding of electrocatalytic processes on noble metal electrodes. Since 2018, he has worked as a scientist and team leader at the Zentrum für Sonnenenergie- und Wasserstoff-Forschung BadenWürttemberg (Ulm, Germany). His current research works focus on the development of functional electrodes for electrochemical power sources such as hydrogen fuel cells, zinc-air batteries, nickel-zinc and zinc-manganese batteries, as well as on the corrosion of stainless steel, together in cooperation with academic and industrial partners. He has published more than 35 scientific papers.

Angelika Heinzel, University of Duisburg-Essen (emerita), Duisburg, Germany Angelika Heinzel received her Ph.D. degree in physical chemistry from the University of Oldenburg and subsequently worked at the Fraunhofer Institute for Solar Energy Systems, where she has been Head of the “Energy Technology” department since 1997. She held the Chair for Energy Technology in the Department of Mechanical and Process Engineering at the University of Duisburg-Essen (2001–2021). At the same time, she established the institute ZBT, Center for Fuel Cell Technology Duisburg and was its managing director. Her research has mainly been dedicated to hydrogen and fuel cell technology.

Editorial board

vii

Ludwig Jörissen, Center for Solar Energy and Hydrogen Research (ZSW), Ulm, Germany Ludwig Jörissen studied chemistry at the University of Ulm from where he graduated in physical chemistry. Subsequently he was awarded a doctorate in science. After a postdoctoral fellowship at the institute of physical and chemical research in Wako Shi, Japan, he joined ZSW in 1990. He is currently the head of the department of “Fuel Cell Fundamentals.”

Peter Kurzweil, Technical University of Applied Sciences (OTH), Amberg-Weiden, Germany Peter Kurzweil graduated in chemistry from the Technical University of Munich, Germany, in 1987. He received his PhD in 1990 in the field of impedance spectroscopy and technical electrochemistry. He joined Dornier GmbH, Friedrichshafen (Daimler Group) in the Applied Research/Energy Technology division, where he worked on electrolysis, supercapacitors, and fuel cells for automotive and space applications, including as project manager. Since 1997, he has been professor of chemistry, electrochemistry, and instrumental analytics at the Technical University of Applied Sciences (OTH) Amberg-Weiden, Germany. He has published around 60 scientific papers, patents, textbooks, and book chapters, including contributions to the previous edition of this encyclopedia.

Joachim Scholta, Center for Solar Energy and Hydrogen Research (ZSW), Ulm, Germany Joachim Scholta graduated in physics at the University of Münster. He was awarded a Ph.D. for investigations on materials and operating techniques for Phosphoric Acid Fuel Cells (PAFC) at the TU Darmstadt in 1993. He continued his research at the Center for Solar Energy and Hydrogen Research (ZSW) in the field of low and high temperature polymer membrane fuel cells as well as on fuel cell systems. Since 2009, he is responsible for the group of stack development at the ZSW. He has published more than 50 papers, patents, and book chapters on low and high temperature PEMFC, including investigations on flow field design, water transport, visualization techniques, and related topics.

This page intentionally left blank

FOREWORD Without water, food, and energy, life could not exist. Life on earth will not continue until we build a sustainable world, one that is cleaner and healthier for our children and grandchildren. In the case of energy, this dictates that we use predominantly non-carbon dioxide-producing sources, such as wind, solar, hydro, and even nuclear. A number of these are intermittent, so they require storage of the energy generated. This storage can range from seasonal, typically served by very large pumped hydro facilities, to the much shorter duration systems typically served today by large batteries. In 2024, two of these in California exceeded 3 GWh each and serve to store solar energy, which today is in excess at midday. As the Nobel committee stated: “They have laid the foundation of a wireless, fossil fuel-free society, and are of the greatest benefit to humankind.” Now, we as scientists and engineers need to reduce our talking and take more action. We need to recruit new students into the field, so we can progress faster, and we need to do this in a sustainable and modern manner. This new Encyclopedia of Electrochemical Power Sources will serve as the to-have bible in the electrochemical storage field. It will help in bridging the gap between scientists and engineers and help join the multiple fields/meetings in the electrochemistry area to stop the segregation of related fields such as batteries, supercaps, electrolyzers, fuel cells, hydrogen generation, and storage. We need to pull together to achieve our goal of a healthier, more sustainable society. Today’s battery systems are not exactly sustainable. It takes between 40 and 80 kWh of energy to build a 1 kWh battery. Components can travel tens of thousands of miles from the mine to the finished product. Manufacturing uses toxic chemicals, such as NMP, in the making of the electrodes. Manufacturing waste is excessive. Cells only achieve a maximum of 25% of their theoretical energy density. The manufacturing formation process takes a huge CAPEX and weeks of time. Recycling, when it occurs, is energy-intensive and often uses brute-force methods rather than trying to directly upgrade the components. Batteries are a likely source of PFAS. None of the above are conducive to a sustainable system. A typical circular economy for a battery or any other electrochemical storage system is shown in the figure. Electrochemistry plays a key role in many of the component steps.

The circular economy for batteries (Courtesy of Olga Petrova, Binghamton University)

ix

x

Foreword

The excessive energy used in manufacturing demands that we as electrochemists must come up with improved materials, materials that can be sourced locally, and cleaner materials processing including the mining (e.g., electroreduction rather than carbothermal). We must work closely with our colleagues so that our electrochemistry links smoothly with the control systems and has the ability to be readily recycled, maybe using electrochemical methods. It goes without saying that all energy used in the manufacturing chain must be clean energy. The need to source locally may well mean that the active materials in the cell will be different on different continents, so electrochemists need to consider that need. We are already seeing a move away from cobalt containing oxide cathodes to iron phosphate materials, whose components are readily available throughout the world. However, the latter store much less energy so we need to find means to modify them, for example, using manganese or vanadium. Fundamental electrochemistry is needed to solve this challenge, also to increase the ionic and electronic conductivities of the electrode materials so that we can use much thicker electrodes, thereby eliminating maybe half of the current collectors and separators. Higher conductivities will also allow us to have fast charging for electric vehicles, a major barrier to broader adoption. We need to better understand electroplating of metals, so we can safely use lithium or sodium metal as the anode. This could easily alone increase the cell energy density by 50%. Researchers must improve today’s electrochemical systems and search for new ideas. This research must extend all the way from fundamental studies that are the cornerstones of innovation to manufacturing. It must also include safer systems. Obviously we need to work with the systems engineers so that there is less energy demand on the energy source—better power systems, lower chip energy demand (remember the days when a laptop computer generated so much heat that it could not be placed on your lap), better aerodynamics of vehicles, even better regenerative breaking in vehicles, better operation under extreme temperature conditions (climate is one reason why electric vehicles are so much more popular in California than in the frigid areas of the world). Lithium batteries are now more than 50 years old and are beginning to dominate portable as well as stationary storage. It is a major technological science in the 21st century. It enabled the communications industry with, for example, the iPhone; there is no longer any need for fixed wired telecommunication systems. It enables the transportation system to go electric. It is enabling renewable energy on a large scale and at the same time stabilizing the electric grid. This is a great time to bring the 2nd edition of Encyclopedia of Electrochemical Power Sources to all those students, scientists, and engineers who want to launch and/or switch their careers toward electrochemical energy storage and help generate a sustainable society for the generations following us. M. Stanley Whittingham, Distinguished Professor, Chemistry and Materials, Binghamton University (SUNY), Binghamton, New York, USA; Nobel Prize Laureate in Chemistry, 2019 Akira Yoshino, Honorary Fellow, Asahi Kasei Corporation, Tokyo, Japan; Nobel Prize Laureate in Chemistry, 2019

PREFACE Electricity has long been known as a natural phenomenon. Today, electricity is an ideal bulk energy vector in almost every respect and is already crucial for life in the twenty-first century. In the future, electricity will become even more important than it is today. However, this requires that electricity is produced without CO2 emissions, primarily from renewable sources, and that it can be stored in large quantities to compensate for fluctuations in electricity production from renewable sources. An important technology for storing electricity is electrochemical energy sources, which store electricity in the form of chemical energy and can convert it into electricity on demand via an electrochemical pathway. The battery, as the most important representative of electrochemical energy sources, dates back to the first practical demonstration of a Cu-Zn cell by Alessandro Volta in 1800, although it was not yet rechargeable. The first practically used secondary or rechargeable battery, the lead–acid battery, was invented by Gaston Planté in 1859. The carbon-zinc cell developed by Georges Leclanché in the 1860s was a practical primary or nonrechargeable battery that is still widely used today as the lead–acid battery as well. Later alkaline batteries (1900 Ni-Cd, Ni-Fe, Ni-Zn, 1970 Ni-MH), mid-1960s Na-S/Na-NiCl2 batteries, and mid-1970s redox flow batteries were developed. The first widespread applications were batteries relatively early for flashlights at the turn of the 19th and 20th centuries for powering automobiles, as well as from about 1920 for starting light ignition (SLI) of automobiles. Later, batteries were used in the industrial sector (storage, emergency). As the performance of both primary and secondary batteries improved, their use in the 3C (camera, computing, and communications) market expanded. Until about the mid-1980s, battery development was technology driven. However, as global environmental and resource problems increased, development became more demand- and government-driven. National and international battery development programs were launched, in particular, to drive the development of electric vehicles and stationary bulk electricity storage. The Battery Energy Storage Systems (BESS) were developed not only for the storage of renewable energy but also for various grid-balancing tasks to decentralize the energy supply and make it more efficient. The introduction of the high-energy rechargeable Li battery (early 1990s) significantly accelerated the realization of these programs, especially its use in new applications such as electric vehicles and stationary energy storage. While the global production capacity of rechargeable batteries was only 5 GWh in 1990 and 10 GWh in 2000, it was already 30 GWh in 2010 (one year after the publication of the first edition of the Encyclopedia of Electrochemical Power Sources). With the further increase in LIB performance and, above all, the reduction in battery costs (2000–approx. 1000 $ kWh−1, 2010–approx. 400 $ kWh−1, and 2020–approx. 150 $ kWh−1), Li battery production has increased exponentially to approx. 1 TWh, where production capacity has even reached 2.6 TWh in 2023 (according to Bloomberg NEF, 2024). Forecasts for the future vary widely. Elon Musk (Tesla), for example, estimates that the world will need 300 TWh of battery cell production to fully transition to a sustainable world. It should also be mentioned that this development was also subject to environmental and resource requirements from around 2010 and led to the new EU Battery Regulation in Europe in 2023, which, among other things, relates to the environmentally friendly (low carbon) production, use, and disposal of batteries. In this context, recycling and the circular economy will become increasingly important. The high social recognition of the development of batteries, especially the lithium-ion battery, was also expressed in the award of the 2019 Nobel Prize in Chemistry to John B. Goodenough, M. Stanley Whittingham, and Akira Yoshino. As a special tribute to the second edition of the Encyclopedia, M. Stanley Whittingham and Akira Yoshino have written the foreword to the Encyclopedia. Professor Goodenough, who died in 2023 at the age of 100, contributed a chapter about Li-ion batteries to the Encyclopedia.

xi

xii

Preface

In addition to batteries, hydrogen and fuel cell technologies can help solve global environmental and resource problems. The fuel cell was first demonstrated by Christian Friedrich Schönbein and William Robert Grove in 1838/39, years before the discovery of the lead-acid battery. However, there were no significant practical applications until the mid-1990s, apart from the US niche application in the space program (mid-1960s). It was only with the start of Daimler’s vehicle development in 1994 that fuel cells attracted industrial interest. For the time being, the use of FC for electric vehicles is more favored for trucks/buses than for cars. Large electricity storage is possible with electrolyzer/fuel cell systems; here, high-temperature fuel cells are generally preferred due to their higher efficiency. The production capacity for fuel cells in 2009 was 120 MW; in 2023 the production capacity, however, amounted to 12 GW. The number of FC cars was approximately 400 in 2009 and amounted in 2022 to 20.700. The first edition of the Encyclopedia of Electrochemical Power Sources covered the development of electrochemical power sources up to 2009 and represents a unique, comprehensive, and interdisciplinary reference that provides a complete description of the basic design and function of the various types of electrochemical power sources, as well as the materials and components used in their design and manufacture. Coverage extends beyond the electrochemical power sources themselves to include important ancillary topics such as hydrogen generation and storage systems, safety, and recycling. It also includes information on important analytical techniques used in the research and development of this field of science and technology. A large part of the Encyclopedia was devoted to the applications of electrochemical energy sources in portable, stationary, and transportation applications. The information provided also highlighted the environmental and economic impacts of electrochemical energy sources. The first edition was very successful with more than 202,871 article downloads (till 2024). This is the third highest usage in the chemistry and chemical engineering Major Reference Work (MRW) portfolio. The Encyclopedia has become a standard reference in the energy-related, applied electrochemistry community for students, researchers, scientists, and engineers in the fields of electrochemistry, materials science, sustainable energy, and power engineering, as well as for government and environmental agencies, industrialists, and the armed forces. The second edition of the Encyclopedia includes an update of the state-of-the-art chapters of the first edition and new chapters reflecting new developments over the last 15 years. These significant developments in advanced materials, better cell/battery designs, and cost-reduced production processes in this time have been reflected in the increase in global cell production capacity from 30 GWh to 2.6 TWh and the reduction in battery cost from 400 $ kWh−1 to 150 $ kWh−1 or even below. Unfortunately, the editorial team for the first edition of the Encyclopedia, with the exception of Jürgen Garche, was no longer available for the second edition due to age and health reasons. However, as this team created the basic content for the Encyclopedia, many thanks are due to Chris K. Dyer, Jürgen Garche, Patrick T. Moseley, Zempachi Ogumi, David A. J. Rand, and Bruno Scrosati for their great work. Jürgen Garche, editor-in-chief of the first edition, has taken over as editor-in-chief of the second edition as well and has formed a new team with Yuichi Aihara (Japan), Klaus Brandt (Germany), Sylvain Brimaud (France/Germany), Angelika Heinzel (Germany), Peter Kurzweil (Germany), Ludwig Jörissen (Germany), and Joachim Scholta (Germany). This new and relatively young editorial team consists mainly of German colleagues. However, this does not detract from the internationality of the authorship. More than 600 authors and coauthors from universities, national laboratories, industrial companies, government agencies, and consulting organizations in more than 28 countries have contributed to the second edition of the Encyclopedia, resulting in a truly global view of the state of the art. The second edition is not arranged alphabetically by sections, and within the sections, not alphabetically by chapter titles as in the first edition. Instead, the content is arranged in a didactic manner similar to a textbook, although the individual chapters form a self-contained unit that can be read independently of other chapters. The second edition starts with general sections on energy, environment, and resources; electrochemical fundamentals; electrochemical devices; history of electrochemical energy sources; electrochemical terminology; chemistry and electrochemistry of elements; methods, instruments, and techniques; cell components; design of cells and batteries; and is followed by battery sections with a general description of batteries and the different battery types. This is then followed by sections on fuel cells, electrolyzers, and fuels, as well as supercapacitors and photoelectrochemical cells. Finally, there are sections on portable, stationary, and transportation applications. Whereas the first edition of the Encyclopedia contained 351 chapters in five volumes, the second edition contains 366 chapters in seven volumes.

Preface

xiii

Of these 366 chapters, 138 chapters are new and 200 chapters have been revised partly by the old authors and partly also by new authors where the original authors were no longer available for various reasons. 28 chapters have been taken over from the first edition without any changes, as there were practically no changes, or in few cases no authors were available. A few chapters have been omitted because some of them are no longer up to date or because their content has been incorporated into other chapters. The 366 chapters were contributed independently by the authors. The Editorial Board has endeavored to preserve the message given by the author(s), while making adjustments to assist the general reader and to ensure good style and accuracy. The use of symbols and abbreviations has not been standardized throughout the work; they are consistent within a given article and are summarized in an appended list. Given the large number of contributions and the somewhat different treatment of the topics, some overlap between certain articles has proved unavoidable. However, such overlap should help the reader to gain a deeper understanding of the more complex issues and also allows him to study individual topics without referring to other articles. It was sometimes difficult to find authors for the chapters, as many authors are more interested to publish original R&D works than in summarizing overviews on specific topics. Even more reason for the editors to thank the authors of the chapters in the second edition who undertook this work to contribute to the Encyclopedia as a standard reference on electrochemical power sources for a wide readership - or as the Nobel prize winners M.S. Whittingham and A. Yoshino wrote in the foreword as the to-have bible in the electrochemical storage field. The editors would also like to thank Elsevier Science for their great help with the administrative contributions of Blerina Osmanaj, Snehil Sharma, and Kiruthigadevi N. Editor-in-Chief: Jürgen Garche Ulm, Germany Editorial Board: Yuichi Aihara Yokohama, Japan Klaus Brandt Wiesbaden, Germany Sylvain Brimaud Ulm, Germany Angelika Heinzel Freiburg, Germany Ludwig Jörissen Ulm, Germany Peter Kurzweil Amberg, Germany Joachim Scholta Ulm, Germany

This page intentionally left blank

CONTENTS Volume 1 Section Editors - Volume 1

xxxiii

Contributors to Volume 1

xxxv

Energy, Environment, and Resources ENERGY, ENVIRONMENT, AND RESOURCES: Energy Transition and Sector Coupling Martin Wietschel

1

ENERGY, ENVIRONMENT, AND RESOURCES: Renewable Energies Martin Kaltschmitt

8

ENERGY, ENVIRONMENT, AND RESOURCES: Energy Storage ENERGY, ENVIRONMENT, AND RESOURCES: H2-Economy Ilaria Tutore

Jelto Lange and

S Koohi-Fayegh and MA Rosen

19

Pasquale Marcello Falcone and 50

ENERGY, ENVIRONMENT, AND RESOURCES: Circular Economy Bettina Rutrecht, Aleksander Jandric, Thomas Nigl, and Florian Part

Eva Gerold, Stefanie Prenner,

ENERGY, ENVIRONMENT, AND RESOURCES: EU - Battery Regulation

57 Casas Ocampo Andrea

68

Electrochemical Fundamentals ELECTROCHEMICAL FUNDAMENTALS: Thermodynamics of Electrode Reactions Thomas J Schmidt

76

ELECTROCHEMICAL FUNDAMENTALS: Electrochemical Double Layer - Electrified Interface W Schmickler

85

ELECTROCHEMICAL FUNDAMENTALS: Kinetics of Electrode Reactions

97

ELECTROCHEMICAL FUNDAMENTALS: Mixed Potential

Thomas J Schmidt

Angel Cuesta and Alan J Gibson

ELECTROCHEMICAL FUNDAMENTALS: Electrochemical Charge Storage J Messinger, and Alberto Varzi ELECTROCHEMICAL FUNDAMENTALS: Electrocatalysis

Theresa Schoetz, Robert 121

Enrique Herrero and Rosa M Arán-Ais

ELECTROCHEMICAL FUNDAMENTALS: Hydrogen Evolution and Oxidation S Trasatti ELECTROCHEMICAL FUNDAMENTALS: Oxygen Evolution and Reduction S Trasatti ELECTROCHEMICAL FUNDAMENTALS: Corrosion

137

Peter Kurzweil and 149 Peter Kurzweil and 158

Roger C Newman and Desmond Williams

ELECTROCHEMICAL FUNDAMENTALS: Ionic Diffusion in Electrolytes Allan M Torres, and Yuichi Aihara

110

167

William S Price, 172

xv

xvi

Contents

ELECTROCHEMICAL FUNDAMENTALS: Computational Modeling of Electrode Reactions and Cells Birger Horstmann, Arnulf Latz, and Felix K Schwab

186

History of Electrochemistry HISTORY OF ELECTROCHEMISTRY: Electrochemistry

Peter Kurzweil

HISTORY OF ELECTROCHEMISTRY: Primary and Secondary Batteries HISTORY OF ELECTROCHEMISTRY: Fuel Cells

200 Peter Kurzweil

Peter Kurzweil

HISTORY OF ELECTROCHEMISTRY: Supercapacitors

223 249

Peter Kurzweil

267

Electrochemical Devices ELECTROCHEMICAL DEVICES: Primary and Secondary Batteries ELECTROCHEMICAL DEVICES: Metal-Air Batteries ELECTROCHEMICAL DEVICES: Fuel Cells

Peter Kurzweil

278

Hajime Arai and Atsunori Ikezawa

Peter Kurzweil

ELECTROCHEMICAL DEVICES: Redox Flow Batteries ELECTROCHEMICAL DEVICES: Concentration Cells

301

Peter Kurzweil

322

Peter Kurzweil

ELECTROCHEMICAL DEVICES: Temperature Cells and Reserve Systems

339 Peter Kurzweil

ELECTROCHEMICAL DEVICES: Bioelectrochemical Cells and Microbial Fuel Cells ELECTROCHEMICAL DEVICES: Supercapacitors ELECTROCHEMICAL DEVICES: Electrolyzers

292

U Schröder

Peter Kurzweil

349 359 369

Peter Kurzweil

383

ELECTROCHEMICAL DEVICES: Electrochemical Sensors and Actuators Amit Kumar, Arnas Majumder, Santhosh Paramasivam, Giancarlo Cappellini, and Gianluca Gatto

395

ELECTROCHEMICAL DEVICES: Photoelectrochemical Cells

409

ELECTROCHEMICAL DEVICES: Electrochromic Windows

Peter Kurzweil L Niklaus, M Schott, and U Posset

ELECTROCHEMICAL DEVICES: Electrochemical Low-Energy Nuclear Systems

Peter Kurzweil

423 439

Electrochemical Terminology ELECTROCHEMICAL TERMINOLOGY: Electrochemical Cell Nomenclature ELECTROCHEMICAL TERMINOLOGY: Capacity

Peter Kurzweil

H Wenzl, R Benger, and I Hauer

ELECTROCHEMICAL TERMINOLOGY: Capacitance

H Wenzl, R Benger, and I Hauer

449 465 473

ELECTROCHEMICAL TERMINOLOGY: Energy

H Wenzl, R Benger, and I Hauer

479

ELECTROCHEMICAL TERMINOLOGY: Power

H Wenzl, R Benger, and I Hauer

487

ELECTROCHEMICAL TERMINOLOGY: Efficiency

H Wenzl, R Benger, and I Hauer

ELECTROCHEMICAL TERMINOLOGY: Self-Discharge ELECTROCHEMICAL TERMINOLOGY: Lifetime

H Wenzl, R Benger, and I Hauer

H Wenzl, R Benger, and I Hauer

ELECTROCHEMICAL TERMINOLOGY: State-of-Charge (SoC)

H Wenzl, R Benger, and I Hauer

498 507 517 526

ELECTROCHEMICAL TERMINOLOGY: Adaptive State-of-Charge (SoC) Determination HJ Bergveld, D Danilov, PHL Notten, V Pop, and PPL Regtien

534

ELECTROCHEMICAL TERMINOLOGY: State-of-Health (SoH)

553

H Wenzl, R Benger, and I Hauer

ELECTROCHEMICAL TERMINOLOGY: Safety and State-of-Safety (SoS) I Hauer

H Wenzl, R Benger, and 559

Contents

ELECTROCHEMICAL TERMINOLOGY: Techno-Economic Assessment of Hydrogen Derivatives P Mock and SA Schmid

xvii

565

Chemistry and Electrochemistry CHEMISTRY AND ELECTROCHEMISTRY: Aluminum

Qingfeng Li, Bingliang Gao, and David Aili

CHEMISTRY AND ELECTROCHEMISTRY: Carbon

T Takamura

CHEMISTRY AND ELECTROCHEMISTRY: Hydrogen

585 598

Tarun Parangi

639

CHEMISTRY AND ELECTROCHEMISTRY: Iron

James A Behan and Frédéric Barrière

655

CHEMISTRY AND ELECTROCHEMISTRY: Lead

Krzysztof Maksymiuk and Jadwiga Stroka

667

CHEMISTRY AND ELECTROCHEMISTRY: Lithium Zawar Alam Qureshi, Tasneem Elmakki, Jeffin James Abraham, Hanan Abdurehman Tariq, Buzaina Moossa, Leena Al-Sulaiti, Dong Suk Han, and Rana Abdul Shakoor

680

CHEMISTRY AND ELECTROCHEMISTRY: Magnesium Pinto Bautista, and Marcel Weil

Zhenyou Li, Liping Wang, Sebastián 700

CHEMISTRY AND ELECTROCHEMISTRY: Manganese Jiban K Das, and Bankim Ch Tripathy

K Kordesch, W Taucher-Mautner,

CHEMISTRY AND ELECTROCHEMISTRY: Nickel

716 E Cattaneo and B Riegel

CHEMISTRY AND ELECTROCHEMISTRY: Oxygen

AJ Appleby

762

CHEMISTRY AND ELECTROCHEMISTRY: Platinum Group Elements CHEMISTRY AND ELECTROCHEMISTRY: Silver

AJ Appleby

804

Marek Orlik

CHEMISTRY AND ELECTROCHEMISTRY: Sodium CHEMISTRY AND ELECTROCHEMISTRY: Zinc

740

826

Sonia Dsoke and Noha Sabi

838

J-Y Huot, Atsunori Ikezawa, and Hajime Arai

852

Volume 2 Section Editors - Volume 2

xxiii

Contributors to Volume 2

xxv

Methods and Instruments METHODS AND INSTRUMENTS: Reference Electrodes

Gareth Hinds

METHODS AND INSTRUMENTS: Potential and Current Steps

1

Rudolf Holze

METHODS AND INSTRUMENTS: Linear Sweep and Cyclic Voltammetry G Bontempelli, and R Toniolo METHODS AND INSTRUMENTS: Electrochemical Impedance Spectroscopy W Strunz, and C-A Schiller METHODS AND INSTRUMENTS: Electrochemical Quartz Microbalance HJ Gores

19

N Elgrishi, 27 J Odrobina, 42 F Wudy, C Stock, and 60

METHODS AND INSTRUMENTS: Differential Electrochemical Mass Spectrometry and R Jürgen Behm METHODS AND INSTRUMENTS: Scanning Electrochemical Microscopy Marius Muhle, and Monika Wilamowska-Zawłocka METHODS AND INSTRUMENTS: Karl-Fischer-Titration

Zenonas Jusys 77

Gunther Wittstock,

Balwant S Chohan and Dan G Sykes

105 119

xviii

Contents

METHODS AND INSTRUMENTS: IR and Raman Spectroscopy

Peter Kurzweil

METHODS AND INSTRUMENTS: Atomic Force Microscopy

Sergey Yu Luchkin

METHODS AND INSTRUMENTS: X-Ray and Neutron Diffraction

H Dittrich and A Bieniok

METHODS AND INSTRUMENTS: Scanning Electronic Microscopy

F Nobili and A Staffolani

METHODS AND INSTRUMENTS: Transmission Electron Microscopy METHODS AND INSTRUMENTS: X-Ray Computed Tomography

F Nobili and A Staffolani

135 148 164 189 206

Roland Brunner

230

Mika Lindén

243

METHODS AND INSTRUMENTS: Porosity and Pore Size Evaluation METHODS AND INSTRUMENTS: X-Ray Absorption Spectroscopy

AV Chadwick and SLP Savin

METHODS AND INSTRUMENTS: Thermal Analysis

Pei Lay Yap and Dusan Losic

METHODS AND INSTRUMENTS: Machine Learning and Weihan Li

Satish Rapol, Runyang Lian, Dirk Uwe Sauer,

249 263 281

Cell Components CELL COMPONENTS – ELECTRODES: Overview

Huaihu Sun, Shuhui Sun, and Gaixia Zhang

295

CELL COMPONENTS – ELECTRODES: Active Materials - Microstructures and Interphases Xuewei Fu and Yu Wang

327

CELL COMPONENTS – ELECTRODES: Electronic Additives and Binders Mariam Odetallah, and Christian Kuss

339

Anh Ngoc Tram Mai,

CELL COMPONENTS – ELECTRODES: Current Collectors

Futoshi Matsumoto and Mika Fukunishi

351

CELL COMPONENTS – ELECTRODES: Porous Electrodes

S Santhanagopalan and RE White

360

CELL COMPONENTS – ELECTRODES: 3-D Electrodes

H-S Min and B Dunn

CELL COMPONENTS – ELECTRODES: Nanoelectrodes

Najmeh Karimian and Paolo Ugo

CELL COMPONENTS – ELECTRODES: Ion-Selective Electrodes CELL COMPONENTS – ELECTRODES: Semiconductor Electrodes

R De Marco and G Clarke Franco Decker and Danilo Dini

CELL COMPONENTS – ELECTROLYTES: Overview - Liquid Electrolytes

M Ue

371 379 397 406 436

CELL COMPONENTS – ELECTROLYTES: Aqueous Liquid Electrolyte Ahmad Azmin Mohamad, Nor Azmira Salleh, Zulfirdaus Zakaria, Siti Salwa Alias, and Soorathep Kheawhom

443

CELL COMPONENTS – ELECTROLYTES: Non-Aqueous Liquid Electrolyte and George Z Chen

467

CELL COMPONENTS – ELECTROLYTES: Ionic Liquid Electrolytes

Hiroyuki Ohno

CELL COMPONENTS – ELECTROLYTES: Overview - Solid Electrolytes Jaimini Parmar, and Indrajit Mukhopadhyay CELL COMPONENTS – SEPARATORS: Overview

Lan Xia, Jialing Zhu, 480

Atul Kumar Mishra,

Dan Yang

489 512

Cell and Battery Design CELL AND BATTERY DESIGN – CELLS: Overview CELL AND BATTERY DESIGN – CELLS: Form Factor Domenic Klohs, Natalia Soldan, and Timon Elliger

Steffen Link and Christoph Neef

522

Heiner Hans Heimes, Moritz Frieges, 532

CELL AND BATTERY DESIGN – CELLS: Flexible Jia-Qi Huang, and Qiang Zhang

Long Kong, Cheng Tang, Hong-Jie Peng,

CELL AND BATTERY DESIGN – CELLS: Printed

Martin Krebs and Gunter Hübner

540

CELL AND BATTERY DESIGN – BATTERIES: Cell Connections

Haifeng Dai and Bo Jiang

555 564

Contents

CELL AND BATTERY DESIGN – BATTERIES: Bipolar Plates and Batteries CELL AND BATTERY DESIGN – BATTERIES: Cell Balancing

Mareike Partsch

Yevgen Barsukov

xix

585 594

CELL AND BATTERY DESIGN – BATTERIES: Thermal Management Kai Shen, Shiding Zhou, Xinyuan Wang, Yong Wang, Xiaodan Liu, Zhiyong Liu, Wenhe Liu, Jianhua Li, Xuning Feng, and Yuejiu Zheng

619

CELL AND BATTERY DESIGN – BATTERIES: Electrical Management Xuebing Han, and Minggao Ouyang

656

CELL AND BATTERY DESIGN – BATTERIES: Hardware

Yanan Wang, Xuning Feng,

Waleri Milde and Stephan Lux

667

Batteries General BATTERIES – BATTERIES GENERAL – INTRODUCTION: Overview Juergen Garche, and Bernhard Riegel

Eduardo Cattaneo, 685

BATTERIES – BATTERIES GENERAL – CHARGE-DISCHARGE: Charging methods Carlos Fernandez and Shunli Wang

702

BATTERIES – BATTERIES GENERAL – CHARGE-DISCHARGE: Capacity and Power Determination Olaf Böse

717

BATTERIES – BATTERIES GENERAL – CHARGE-DISCHARGE: Charge-Discharge Curves Rüdiger-A Eichel and Dirk Uwe Sauer

723

BATTERIES – BATTERIES GENERAL – CHARGE-DISCHARGE: Cell Dynamics BATTERIES – BATTERIES GENERAL – CHARGE-DISCHARGE: Cell Reversal

A Jossen L Binder

734 748

BATTERIES – BATTERIES GENERAL – CHARGE-DISCHARGE: Alternating Currents and Ripples Erik Goldammer, Dominik Droese, Marius Gentejohann, Michael Schlüter, Daniel Weber, Clemens Gühmann, Sibylle Dieckerhoff, and Julia Kowal

758

BATTERIES – BATTERIES GENERAL – SAFETY: General Aspects Knut Bjarne Gandrud, and Linda Berg Aas

769

Martin Gilljam, Helge Weydahl,

Volume 3 Section Editors - Volume 3

xxiii

Contributors to Volume 3

xxv

Aluminium Systems BATTERIES – BATTERY TYPES – ALUMINIUM BATTERIES: Aqueous Manuel Baumann, and Stefano Passerini

Xu Liu, Hüseyin Ersoy, 1

BATTERIES – BATTERY TYPES – ALUMINIUM BATTERIES: Non Aqueous Hamideh Darjazi, Matteo Gastaldi, and Alessandro Piovano

Giuseppe Antonio Elia, 17

Iron Systems BATTERIES – BATTERY TYPES – IRON BATTERIES: Iron-Air Cinthia Alegre

Rachel McKerracher and 30

Lead-Acid Systems BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Overview DAJ Rand

PT Moseley, J Garche, and

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Negative Electrode

35 Geno Papazov

66

xx

Contents

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Positive Electrode Albena Aleksandrova, Maria Matrakova, and Kathryn R Bullock

Plamen Nikolov, 84

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Expander Additives Albena Aleksandrova, and Maria Matrakova BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Carbon Jakub Lach, Kamil Wróbel, and Andrzej Czerwi nski

Plamen Nikolov, 106

Anthony Frank Hollenkamp, 133

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Electrode Design BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Electrolyte Detchko Pavlov

R Wagner

Albena Aleksandrova and 161

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Phosphorus Additives Albena Aleksandrova, Plamen Nikolov, and Kathryn R Bullock BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Separators Werner Boehnstedt

Maria Matrakova, 176

J Kevin Whear and 187

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Lead Alloys BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Grids

147

E Cattaneo

202

E Cattaneo

223

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Curing and Formation BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Flooded Batteries

R Wagner

R Wagner

243 260

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Valve-Regulated Batteries: Oxygen Cycle Eberhard Meissner

280

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Valve-Regulated Batteries: AGM MJ Weighall and A Kirchev

298

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Valve-Regulated Batteries: Gel and H Niepraschk BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Catalytic Valves E Cattaneo

315

Bernhard Riegel and 324

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Design for Performance T Hildebrandt BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Bipolar

F Kramm

GJ May and 338

Edward O Shaffer II

354

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Supercap Hybrid (Ultrabattery) J Furukawa and LT Lam

364

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Charging Methods

388

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Coup de Fouet

SS Misra and J Garche M Perrin and A Delaille

405

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: State-of-Charge (SoC), State-of-Health (SoH) W Waag and DU Sauer

420

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Partial State-of-Charge Operation (PSoC) E Dickinson and P Kramer

433

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Lifetime Determining Processes DU Sauer

442

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Modeling

454

M Cugnet and BY Liaw

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Automotive Batteries: Products T Hildebrandt and Eberhard Meissner

471

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Automotive Batteries: Applications and Market Trends Eckhard Karden

498

Contents

xxi

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Industrial Batteries: Stationary and Transportation Types GJ May, Bernhard Riegel, and E Cattaneo

510

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Regulation, Codes, Standards, and Testing Olaf Sielemann and Wilhelm Giller

522

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Production Aaron Bollinger, Aaron Kirschman, and Luke Hoffman

535

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Market

Norbert Maleschitz, Bernhard Riegel

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Safety Aaron Kirschman

545

Norbert Maleschitz and 558

BATTERIES – BATTERY TYPES – LEAD-ACID BATTERY: Recycling

Mark W Stevenson

563

Magnesium Systems BATTERIES – BATTERY TYPES – MAGNESIUM BATTERIES: Primary Systems Daniel Höche

Min Deng and 585

BATTERIES – BATTERY TYPES – MAGNESIUM BATTERIES: Secondary Systems Vito Di Noto

Gioele Pagot and 594

Nickel Systems BATTERIES – BATTERY TYPES – NICKEL BATTERIES: Overview

E Cattaneo and Bernhard Riegel

BATTERIES – BATTERY TYPES – NICKEL BATTERIES: Nickel Electrode B Hariprakash

AK Shukla and 625

BATTERIES – BATTERY TYPES – NICKEL BATTERIES: Nickel–Cadmium Kurzweil BATTERIES – BATTERY TYPES – NICKEL BATTERIES: Nickel–hydrogen

608

P Bernard and Peter 634 Taoli Jiang and Wei Chen

657

BATTERIES – BATTERY TYPES – NICKEL BATTERIES: Nickel-Metal Hydride: Overview B Hariprakash, AK Shukla, and S Venugoplan

674

BATTERIES – BATTERY TYPES – NICKEL BATTERIES: Nickel-Metal Hydride: Metal Hydrides PHL Notten and M Latroche

682

BATTERIES – BATTERY TYPES – NICKEL BATTERIES: Nickel-Metal Hydride: Cutting Edge Technologies Jun Ishida, Masaru Kihara, Dag Noréus, and Yang Shen

701

BATTERIES – BATTERY TYPES – NICKEL BATTERIES: Nickel–Iron

AK Shukla and B Hariprakash

711

BATTERIES – BATTERY TYPES – NICKEL BATTERIES: Nickel–Zinc Shadi Mirhashemihaghighi

Fabrice Fourgeot and 719

BATTERIES – BATTERY TYPES – NICKEL BATTERIES: Memory Effect

Y Sato

729

Zinc Systems BATTERIES – BATTERY TYPES – ZINC BATTERIES: Overview

Simon Clark

BATTERIES – BATTERY TYPES – ZINC BATTERIES: Zinc Electrode

LS Cao and D Li

BATTERIES – BATTERY TYPES – ZINC BATTERIES: Zinc–Carbon Mautner

760

K Kordesch and W Taucher778

BATTERIES – BATTERY TYPES – ZINC BATTERIES: Zinc–Manganese Daniel Ivad BATTERIES – BATTERY TYPES – ZINC BATTERIES: Zinc–Silver

747

Gautam G Yadav and Josef

Alexander P Karpinski

791 825

xxii

Contents

BATTERIES – BATTERY TYPES – ZINC BATTERIES: Zinc–Bromine Flow Batteries Frank C Walsh, and Carlos Ponce de León BATTERIES – BATTERY TYPES – ZINC BATTERIES: Zinc–Air Jean-Francois Drillet

Sabrina Berling, 840

Hajime Arai, Atsunori Ikezawa, and 851

Volume 4 Section Editors - Volume 4

xxiii

Contributors to Volume 4

xxv

Li Batteries – Introduction BATTERIES – BATTERY TYPES – LITHIUM BATTERIES: Overview BATTERIES – BATTERY TYPES – LITHIUM BATTERIES: Electrolytes

Klaus Brandt and Yuichi Aihara J Ho and K Xu

1 14

Li-Primary Systems BATTERY TYPES – LITHIUM BATTERIES – LITHIUM PRIMARY BATTERIES: Overview

K Nishio

61

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM PRIMARY BATTERIES: Lithium–IodinePolyvinylpyridine CF Holmes

69

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM PRIMARY BATTERIES: Lithium–Manganese Dioxide K Nishio

77

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM PRIMARY BATTERIES: Lithium-Carbon Fluoride Battery Guiming Zhong and Weimin Zhao

89

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM PRIMARY BATTERIES: Lithium–Vanadium/ Silver Oxides ES Takeuchi, KJ Takeuchi, and AC Marschilok

99

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM PRIMARY BATTERIES: Lithium–Sulfur/ Chlorine Arden P Johnson

111

Li-Secondary Systems – Li-Metal Systems LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-METAL BATTERY: Overview Klaus Brandt and Peter Kurzweil

122

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-METAL BATTERY: Lithium–Sulfur Natarajan Angulakshmi, Kwon-Koo Cho, Jou-Hyeon Ahn, and Hyo-Jun Ahn

141

Li-Secondary Systems – Li-Ion Systems LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Overview Hikari Sakaebe and Jun-ichi Yamaki

149

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Negative Electrodes: Graphite and Carbon Minoru Inaba

159

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Negative Electrode: Silicon, Silicon-Carbon, and Silicon Oxides Yuichi Aihara, Klaus Brandt, Peter Kurzweil, and Jürgen Garche

171

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Negative Electrode: Pre-Lithination Laurin Profanter, Lukas Stolz, Martin Winter, and Johannes Kasnatscheew

187

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Solid Electrolyte Interphase (SEI) Ujwal Shreenag Meda, Charanya Adaguru Rudregowda, and Harika Rajashekaraiah

194

Contents

xxiii

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Negative Electrode: Lithium Titanium Oxides Kingo Ariyoshi

207

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Positive Electrode: Lithium Cobalt Oxide Hajime Arai and Atsunori Ikezawa

218

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Positive Electrode: Lithium Nickel Oxide Ryoji Kanno

226

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Positive Electrode: Lithium Manganese Oxides Masaki Okada

239

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Positive Electrode: Layered Metal Oxides Yoshinari Makimura

255

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Positive Electrode: Lithium Iron Phosphates Karim Zaghib, MV Reddy, A Mauger, F Gendron, CM Julien, and John B Goodenough

267

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Positive Electrode: High-Voltage Materials Margret Wohlfahrt-Mehrens and Peter Axmann

305

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Positive Electrode: Conversion Materials Albert W Xiao

314

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Current Collector Futoshi Matsumoto and Mika Fukunishi

329

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Organic and Inorganic Electrolytes Karim Zaghib, MR Anil Kumar, and MV Reddy

348

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Ionic Liquid Electrolytes Henry Adenusi and Stefano Passerini

370

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Electrolyte Additives Ujwal Shreenag Meda, Nidhi Bhat, Om Madan Raikar, Tribikram Gupta, and Kalpana Sharma

383

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Separators Takashi Ikemoto

397

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Aging Mechanisms and Lifetime Predictions Margret Wohlfahrt-Mehrens and Thomas Waldmann

412

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Material Informatic and Atomistic Simulations Taku Watanabe

426

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Organic Electrolyte Cells John B Goodenough, Yuichi Aihara, Jürgen Garche, and Klaus Brandt

437

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Inorganic Electrolyte Cells Laurent Zinck, Christiane Ripp, Christian Pszolla, Markus Borck, and Manuel Weinberger

447

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Production Paul R Shearing, Denis Cumming, and Emma Kendrick

461

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Recycling Takehiko Okui

472

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LI-ION BATTERY: Market Tsuyoshi Nishimura, Masamichi Yamaguchi, and Kai Shimozono

484

Li-Secondary Systems – Lithium All-Solid State Systems LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Overview Till Fuchs, Burak Aktekin, Felix Hartmann, Felix H Richter, and Jürgen Janek

503

xxiv

Contents

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Solid Polymer Electrolytes Peng Zhang, Zhen Liu, Kang Xia, and He Jia

513

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Solid Oxide Electroltyes Kazunori Takada

526

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Solid Sulfide Electroltyes Ryoji Kanno

538

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Halides and Oxy-Halide Electrolytes Artur Tron, Palanivel Molaiyan, Marcus Jahn, and Andrea Paolella

568

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Negative Electrodes Akiko Tsurumaki, Graziano Di Donato, and Maria Assunta Navarra

578

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Negative Electrodes - “Anode Free” Till Fuchs, Burak Aktekin, Felix Hartmann, Simon Burkhardt, and Jürgen Janek

588

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Thin-Film Type Cells Chuangjie Guo and Yaoyu Ren

600

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Solid Polymer Electrolyte Cells Yo Kobayashi and Kumi Shono

614

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Inorganic Solid-Electrolyte Cells Felix Hippauf and Sahin Cangaz

624

LITHIUM BATTERIES – LITHIUM SECONDARY BATTERIES – LITHIUM ALL-SOLID STATE BATTERY: Production Mareike Partsch, Benedikt Stumper, Jonas Dhom, and Julian Schwenzel

643

Lithium Battery Safety BATTERY TYPES – LITHIUM BATTERIES – LITHIUM BATTERY SAFETY: Overview

Junxian Hou

656

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM BATTERY SAFETY: Chemical Hazards Changyong Jin

671

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM BATTERY SAFETY: Thermal Hazards Yu Wang

677

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM BATTERY SAFETY: Cell Level - Safety Related Material and Design Engineering BY Liaw and G Zhang

688

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM BATTERY SAFETY: Cell and Battery Safety Devices Daniel Wesolowski, David Enos, Noah B Schorr, Josefine McBrayer, and Brian Perdue

701

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM BATTERY SAFETY: Regulations, Codes, and Standards: Safety Testing Joris Jaguemont

717

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM BATTERY SAFETY: Advanced Safety Testing Mark Buckwell, Julia S Weaving, Matilda Fransson, and Paul R Shearing

737

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM BATTERY SAFETY: Hazards During Transport and Storage Chengshan Xu and Enhong Liu

753

BATTERY TYPES – LITHIUM BATTERIES – LITHIUM BATTERY SAFETY: Fire Risks and Fire Extinguishing Huaibin Wang

760

Sodium Systems BATTERY TYPES – SODIUM BATTERIES – LOW-TEMPERATURE SODIUM BATTERIES: Overview Ivana Hasa, Jerry Barker, Giuseppe Elia, and Stefano Passerini

767

Contents

xxv

BATTERY TYPES – SODIUM BATTERIES – LOW-TEMPERATURE SODIUM BATTERIES: Anode Active Materials Xing-Long Wu, Kai-Yang Zhang, Han-Hao Liu, Meng-Yuan Su, and Hao-Jie Liang

785

BATTERY TYPES – SODIUM BATTERIES – LOW-TEMPERATURE SODIUM BATTERIES: Cathode Active Materials Maider Zarrabeitia, Wenhua Zuo, and Stefano Passerini

797

BATTERY TYPES – SODIUM BATTERIES – LOW-TEMPERATURE SODIUM BATTERIES: Electrolytes Juan Forero-Saboya and Sathiya Mariyappan

817

BATTERY TYPES – SODIUM BATTERIES – LOW-TEMPERATURE SODIUM BATTERIES: Sodium-Ion Cells Yan Yu, Xianhong Rui, and Meng Wu

829

BATTERY TYPES – SODIUM BATTERIES – LOW-TEMPERATURE SODIUM BATTERIES: Solid Sulfide Electrolytes Kota Motohashi, Atsushi Sakuda, and Akitoshi Hayashi

842

BATTERY TYPES – SODIUM BATTERIES – HIGH-TEMPERATURE SODIUM BATTERIES: Sodium–Sulfur Tomio Tamakoshi

849

BATTERY TYPES – SODIUM BATTERIES – HIGH-TEMPERATURE SODIUM BATTERIES: Sodium–Nickel Chloride Matthias Schulz

858

BATTERY TYPES – SODIUM BATTERIES – HIGH-TEMPERATURE SODIUM BATTERIES: Safety C-H Dustmann and A Bito

872

BATTERY TYPES – SODIUM BATTERIES – HIGH-TEMPERATURE SODIUM BATTERIES: Alkali Metal Thermal To Electric Energy Converter MAK Lodhi

885

Exploratory Battery Systems BATTERIES – BATTERY TYPES – EXPLORATORY BATTERIES: Potassium-Ion Batteries Xing-Long Wu, Yong-Li Heng, and Yan Liu

898

BATTERIES – BATTERY TYPES – EXPLORATORY BATTERIES: Anion Shuttle Systems Ran Han, and Dong Zhou

Zuxin Wen, 912

BATTERIES – BATTERY TYPES – EXPLORATORY BATTERIES: Lithium–Air Systems Steven ViscoJoseph, Eugene Nimon, and Lutgard De Jonghe (C.)

930

BATTERIES – BATTERY TYPES – EXPLORATORY BATTERIES: Organic Radical Systems Kenichi Oyaizu

941

Volume 5 Section Editors - Volume 5

xxiii

Contributors to Volume 5

xxv

Redox-Flow Systems BATTERIES – BATTERY TYPES – REDOX-FLOW BATTERIES: Overview Eduardo Sánchez-Dí ez

Edgar Ventosa and 1

BATTERIES – BATTERY TYPES – REDOX-FLOW BATTERIES: Vanadium Flow Battery Systems Maria Skyllas-Kazacos

12

BATTERIES – BATTERY TYPES – REDOX-FLOW BATTERIES: Iron Systems Huan Zhang, and Kai Song

24

Chuanyu Sun,

BATTERIES – BATTERY TYPES – REDOX-FLOW BATTERIES: Organic Reactant Systems Ruiyong Chen, Muhammad Mara Ikhsan, Dirk Henkensmeier, Peng Zhang, Zhifeng Huang, Sangwon Kim, and Rolf Hempelmann

37

xxvi

Contents

Fuel Cells - Introduction FUEL CELLS: Overview

Werner Tillmetz

50

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Overview: PEMFC Graham Smith, Billie Sherin, Jonathan Goh, and Dipak V Shinde

57

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Anodes Zheng Li

Lin Zeng and 71

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Cathodes Elisabeth Hornberger, Sebastian Ott, and Peter Strasser

Thomas Merzdorf, 88

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Catalysts: Life Limitations Rebecca Pittkowski, Alessandro Zana, Masanori Inaba, and Matthias Arenz

96

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Catalysts: NonPlatinum Konrad Eiler, Jordi Sort, and Eva Pellicer

111

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Membranes: Fluorinated L Merlo, A Ghielmi, and V Arcella

121

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Membranes: NonFluorinated Maximilian Wagner and Jochen Kerres

143

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Membranes: Design and Characterization Hien Nguyen, Carolin Klose, and Nada Zamel

157

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Membrane: Life Limitations Aryaman Shah and Jay Pandey

169

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Membranes: Modeling B D’Aguanno, DWM Hofmann, LN Kuleshova, and L Pisani

188

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Membrane–Electrode Assemblies Ramaraja P Ramasamy

200

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Gas Diffusion Layers Dominic F Gervasio

221

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Flow Fields Irfan Karagoz

Erman Celik and 228

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Bipolar Plates M Grundler, J Karstedt, A Kayser, and L Kühnemann FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Cells

M Gretzki, 245

Vijay Sethuraman

254

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Water Management EC Kumbur, MM Mench, and Joachim Scholta

266

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Life Limitations Vijay Sethuraman

289

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Modeling Pang-Chieh Sui, Wen-Quan Tao, and Ned Djilali FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Stacks FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Systems

Li Chen, 309 Joachim Scholta Z Qi

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Performance and Operational Conditions Joachim Scholta

324 339 352

Contents

xxvii

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Dynamic Operational Conditions J Mitzel, N Wagner, J Sanchez-Monreal, and KA Friedrich

365

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Freeze Operational Conditions Joachim Scholta

386

FUEL CELLS – POLYMER-ELECTROLYTE MEMBRANE FUEL CELL: Impurities in Fuels and Air KA Friedrich, J Mitzel, S Prass, and Joachim Scholta

398

Fuel Cells – Alkaline Fuel Cells FUEL CELLS – ALKALINE FUEL CELL: Overview

Grietus Mulder

FUEL CELLS – ALKALINE FUEL CELL: Anion-Exchange Membranes Assunta Navarra FUEL CELLS – ALKALINE FUEL CELL: Cells and Stacks Karl Kordesch

411 Nicholas Carboni and Maria 421

Viktor Hacker, Simon D Fraser, and 437

FUEL CELLS – ALKALINE FUEL CELL: Performance and Operational Conditions and Qingfeng Li

San Ping Jiang 448

Fuel Cells – Phosphoric-Acid Fuel Cells FUEL CELLS – PHOSPHORIC ACID FUEL CELL: Overview and Amit C Bhosale

Mohammad Saquib, Akshay Sharma, 460

FUEL CELLS – PHOSPHORIC ACID FUEL CELL: Anodes and High-Temperature Membranes Peter Kurzweil

483

FUEL CELLS – PHOSPHORIC ACID FUEL CELL: PAFC Cathodes

496

T Kuwabara and Peter Kurzweil

FUEL CELLS – PHOSPHORIC ACID FUEL CELL: PAFC and High Temperature PEMFC Cathodes Julia Müller-Hülstede, Henrike Schmies, Dana Schonvogel, and Peter Wagner

504

FUEL CELLS – PHOSPHORIC ACID FUEL CELL: Electrolyte and Separator Peter Kurzweil

516

FUEL CELLS – PHOSPHORIC ACID FUEL CELL: Cells and Stacks FUEL CELLS – PHOSPHORIC ACID FUEL CELL: Systems

T Murahashi and

Sridhar Kanuri

Sridhar Kanuri

FUEL CELLS – PHOSPHORIC ACID FUEL CELL: Life-Limiting Considerations and Sridhar Kanuri

521 535

Timothy Patterson 545

Fuel Cells – Molten Carbonate Fuel Cells FUEL CELLS – MOLTEN CARBONATE FUEL CELL: Overview R Venkataraman FUEL CELLS – MOLTEN CARBONATE FUEL CELL: Anodes R Venkataraman FUEL CELLS – MOLTEN CARBONATE FUEL CELL: Cathodes

C Yuh, A Hilmi, and 557 A Hilmi, C Yuh, M Farooque, and 571 C Yuh and A Hilmi

FUEL CELLS – MOLTEN CARBONATE FUEL CELL: Cells and Stacks R Venkataraman

581

C Yuh, A Hilmi, and

FUEL CELLS – MOLTEN CARBONATE FUEL CELL: Systems - Performance and Life-Limiting Considerations CY Yuh

589 600

Fuel Cells – Solid Oxide Fuel Cells FUEL CELLS – SOLID OXIDE FUEL CELL: Overview FUEL CELLS – SOLID OXIDE FUEL CELL: Anodes

X-D Zhou and SC Singhal

622

KV Galloway and NM Sammes

641

xxviii

Contents

FUEL CELLS – SOLID OXIDE FUEL CELL: Cathodes

NM Sammes and BR Roy

FUEL CELLS – SOLID OXIDE FUEL CELL: Membranes

650

Mihails Kusnezoff

661

FUEL CELLS – SOLID OXIDE FUEL CELL: Gas Distribution

LGJ de Haart and M Spiller

FUEL CELLS – SOLID OXIDE FUEL CELL: Cells and Stacks Mihails Kusnezoff

RJ Kee, AM Colclasure, H Zhu, and 690

FUEL CELLS – SOLID OXIDE FUEL CELL: Cell Interconnection

Zhenguo Yang and SC Singhal

FUEL CELLS – SOLID OXIDE FUEL CELL: Internal and External Steam Reforming and Roland Peters FUEL CELLS – SOLID OXIDE FUEL CELL: System Concepts

705

LGJ de Haart 720

L Blum and E Riensche

FUEL CELLS – SOLID OXIDE FUEL CELL: Life-Limiting Considerations FUEL CELLS – SOLID OXIDE FUEL CELL: Modeling Steven B Beale, Felix Kunz, and Roland Peters

679

A Weber

731 754

Shidong Zhang, Shangzhe Yu, Kai Wang, 769

Fuel Cells – Fuel Cells with Liquid Fuels FUEL CELLS – FUEL CELLS WITH LIQUID FUELS: Direct Alcohol Fuel Cells: Overview Claude Lamy and Ludwig Jörissen FUEL CELLS – FUEL CELLS WITH LIQUID FUELS: Direct Methanol Fuel Cell Viktor Gogel, and Claude Lamy

785 Ludwig Jörissen,

FUEL CELLS – FUEL CELLS WITH LIQUID FUELS: Direct Formic Acid and Formaldehyde Fuel Cells Ludwig Jörissen

806 825

Fuel Cells – Exploratory Fuel Cells FUEL CELLS – EXPLORATORY FUEL CELLS: Direct Carbon Fuel Cells GG Botte FUEL CELLS – EXPLORATORY FUEL CELLS: Ammonia Fuel Cells Shanwen Tao

M Muthuvel, X Jin, and 834 Georgina Jeerh and 849

FUEL CELLS – EXPLORATORY FUEL CELLS: Micro Fuel Cells A Kundu, K Karan, BA Peppley, Y Sahai, YD Premchand, A Bieberle-Hutter, LJ Gauckler, and U Schröder

859

FUEL CELLS – EXPLORATORY FUEL CELLS: Regenerative Fuel Cells

882

F Barbir

Volume 6 Section Editors - Volume 6

xxiii

Contributors to Volume 6

xxv

Electrolyzer – Introduction ELECTROLYZER – INTRODUCTION: Overview

Ashkan Makhsoos and Bruno G Pollet

1

Electrolyzer – Alkaline Electroyzer ELECTROLYZER – ALKALINE ELECTROLYZER: Overview and RB Ferreira

DS Falcão, DFM Santos, AMFR Pinto, 15

ELECTROLYZER – ALKALINE ELECTROLYZER: Performance, Efficiency, and Lifetime Martin Müller, Irina Galkina, Fabian Scheepers, and Felix Lohmann-Richters

33

ELECTROLYZER – ALKALINE ELECTROLYZER: Alkaline Anion Exchange Membrane Electrolysis Shanmugam Ramakrishnan and Mohamed Mamlouk

42

Contents

xxix

Electrolyzer – Polymer-Electrolyte Membrane Electrolyzer ELECTROLYZER – POLYMER-ELECTROLYTE MEMBRANE ELECTROLYZER: Overview A Grigoriev

Sergey 65

ELECTROLYZER – POLYMER-ELECTROLYTE MEMBRANE ELECTROLYZER: State of the Art Technique and Systems Shiva Kumar Sampangi and Lars Röntzsch

79

ELECTROLYZER – POLYMER-ELECTROLYTE MEMBRANE ELECTROLYZER: Small and Large Scale Units Eugen Hoppe

95

Electrolyzer – Solid Oxide Electrolyzer ELECTROLYZER – SOLID OXIDE ELECTROLYZER: Overview K Andreas Friedrich

Matthias Riegraf, Rémi Costa, and 109

Fuels – Introduction FUELS – INTRODUCTION: Overview

M Conte

FUELS – INTRODUCTION: Hydrogen as a Fuel E Boettcher, A Winkler, and S Spitzer

123 J Töpler, T Bregulla, A Stephan, H Fricke,

FUELS – INTRODUCTION: Hydrogen Industrial Use

140 Ludwig Jörissen

FUELS – INTRODUCTION: Hydrogen Global Transport

161

Ludwig Jörissen

171

FUELS – INTRODUCTION: Hydrogen Safety Thomas Jordan, Enis Askar, Kai Holtappels, Manuela Jopen, Uwe Stoll, Ernst-Arndt Reinecke, Ulrich Krause, Michael Beyer, and Detlev Markus

184

FUELS – INTRODUCTION: Hydrogen Non-Conventional Storage Options Ralph-Uwe Dietrich, Sandra Adelung, Felix Habermeyer, Nathanael Heimann, Simon Maier, Moritz Raab, and Yoga Rahmat

199

Fuels – Hydrogen FUELS – HYDROGEN – HYDROGEN PRODUCTION: Fossil Fuels Based Suwimol Wongsakulphasatch, Sakhon Ratchahat, Pattaraporn Kim-Lohsoontorn, Worapon Kiatkittipong, Nopphon Weeranoppanant, Merika Chanthanumataporn, Sumittra Charojrochkul, Navadol Laosiripojana, and Suttichai Assabumrungrat

232

FUELS – HYDROGEN – HYDROGEN PRODUCTION: Biomass Based A Kruse

246

FUELS – HYDROGEN – HYDROGEN PRODUCTION: Electrolysis

N Dahmen, E Dinjus, and T Smolinka

FUELS – HYDROGEN – HYDROGEN PRODUCTION: Thermochemical Vishnu Budama, Martin Roeb, and Christian Sattler FUELS – HYDROGEN – HYDROGEN STORAGE: Compressed

Dorottya Kriechbaumer, 278

Klaas Kunze

FUELS – HYDROGEN – HYDROGEN STORAGE: Cryo- and Cryo-Compressed FUELS – HYDROGEN – HYDROGEN STORAGE: Physical Adsorption L Jiménez-López, A Celzard, and V Fierro FUELS – HYDROGEN – HYDROGEN STORAGE: Chemical

256

288 Klaas Kunze

R Morales-Ospino, 319

F Cuevas, R Moury, and T Belmonte

FUELS – HYDROGEN – HYDROGEN STORAGE: Regulations, Codes, and Standards Reinhold Wurster FUELS – HYDROGEN – HYDROGEN STORAGE: Hydrogen Alternatives

296

330 346

Ludwig Jörissen

360

Supercapacitors SUPERCAPACITORS: Electrochemical Double-Layer Capacitor

Peter Kurzweil

368

xxx

Contents

SUPERCAPACITORS: Carbon Technologies

Peter Kurzweil

SUPERCAPACITORS: Metal Oxide Technologies SUPERCAPACITORS: Ionic Liquids Electrolytes SUPERCAPACITORS: Hybrid Technologies SUPERCAPACITORS: Solid-State Technologies SUPERCAPACITORS: Aging and Service Life SUPERCAPACITORS: Production

402

Peter Kurzweil

426

M Mastragostino and F Soavi

445

DP Chatterjee, U Basak, and AK Nandi

456

Peter Kurzweil

468

Peter Kurzweil

476

Peter Kurzweil and Ch Schell

SUPERCAPACITORS: Performance, Cost, and Application

495

Andrew F Burke

506

Photoelectrochemical Cells PHOTOELECTROCHEMICAL CELLS: Integration in Renewable Energy Systems PHOTOELECTROCHEMICAL CELLS: Water Splitting Liuyang Zhang

S Calnan

525

Jiaguo Yu, Chuanbiao Bie, and 532

PHOTOELECTROCHEMICAL CELLS: Photoelectrocatalysis Weilai Yu, Zhenhua Pan, Pakpoom Buabthong, Paul A Kempler, Julie Tournet, and Siva Karuturi

540

PHOTOELECTROCHEMICAL CELLS: Biological Redox Systems

550

L Kurzweil

Volume 7 Section Editors - Volume 7

xxiii

Contributors to Volume 7

xxv

Application – Portable Applications APPLICATIONS – PORTABLE APPLICATIONS: Battery Based

Martin Krebs and Yuichi Aihara

APPLICATIONS – PORTABLE APPLICATIONS: Fuel-Cell Based

Christoph Hartnig

APPLICATIONS – PORTABLE APPLICATIONS: High Power Use A Gonser

1 12

WJ Weydanz, P Roeder, and 25

APPLICATIONS – PORTABLE APPLICATIONS: Military: Batteries and Fuel Cells J Tübke, and M Krausa

C Cremers, 37

Application – Stationary Applications APPLICATIONS: Overview of Energy Storage Systems Francesco Piraino, and Petronilla Fragiacomo

Viviana Cigolotti, Matteo Genovese,

APPLICATIONS – STATIONARY APPLICATIONS: Batteries

47 Christian Bussar

APPLICATIONS – STATIONARY APPLICATIONS: Fuel Cells - Residential

60

Ludwig Jörissen

APPLICATIONS – STATIONARY APPLICATIONS: Fuel Cells - Larger Size and Hybrids Francesco Calise, Massimo Dentice d'Accadia, Francesco Liberato Cappiello, Luca Cimmino, and Maria Vicidomini APPLICATIONS – STATIONARY APPLICATIONS: Central Storage - Hydrogen MSA Perera

73

90

T Amirthan and 107

APPLICATIONS – STATIONARY APPLICATIONS: Uninterruptible and Backup Power Supply Batteries and Fuel Cells Ulrike Trachte and Benjamin Fumey

116

APPLICATIONS – STATIONARY APPLICATIONS: Remote Area Power Supply - Batteries and Fuel Cells Marion Perrin, Elisabeth Lemaire-Potteau, and Thierry Cortassa

131

Contents

xxxi

Application – Transportation Applications APPLICATIONS – TRANSPORTATION APPLICATIONS: Electric Vehicles - Batteries Yingchen Xie, and Yifan Wang

Xuning Feng, 146

APPLICATIONS – TRANSPORTATION APPLICATIONS: Electric Vehicles - Fuel Cells Clark Hochgraf

170

APPLICATIONS – TRANSPORTATION APPLICATIONS: Hybrid Electric Cars - Batteries and Fuel Cells Parag Jose Chacko, S Akshaya, and R Imran Jafri

184

APPLICATIONS – TRANSPORTATION APPLICATIONS: Hybrid Electric Buses and Trucks Batteries Sebastian Wolff, Jakob Schneider, Georg Balke, Maximilian Zähringer, Steffen Büttner, Maximilian Schuckert, and Malte Jaensch

202

APPLICATIONS – TRANSPORTATION APPLICATIONS: Electric buses – Fuel cells Oliver Hoch, Steffen Schulze, Janosch Rauer, Christopher Borger, Maximilian Lohrer, Maria Renkert, Malte Siemen, and Eugene Müller

215

APPLICATIONS – TRANSPORTATION APPLICATIONS: Light Traction - Batteries and Fuel Cells M Schier, M Brost, and I Dasgupta

227

APPLICATIONS – TRANSPORTATION APPLICATIONS: Vehicle Batteries - End of Life and H-G Schweiger

Y Kotak 244

APPLICATIONS – TRANSPORTATION APPLICATIONS: Material Handling and Forklift Trucks Batteries Riegel Bernhard, Eduardo Cattaneo, and Johannes Büngeler

254

APPLICATIONS – TRANSPORTATION APPLICATIONS: Rail Vehicles - Fuel Cells and Batteries Sebastian Herwartz-Polster, Mathias Böhm, Sebastian Stickel, Christoph Streuling, Benedikt Hertel, and Johannes Pagenkopf

268

APPLICATIONS – TRANSPORTATION APPLICATIONS: Ships - Fuel Cells and Batteries Ligang Wang, Hangyu Yu, Xinyi Wei, and Hua Liu

280

APPLICATIONS – TRANSPORTATION APPLICATIONS: Submersibles - Fuel Cell and Batteries Stefan Krummrich, Marc Pein, Jessica Lück, and Knud Lämmle

301

APPLICATIONS – TRANSPORTATION APPLICATIONS: Aviation Batteries Venkatasubramanian Viswanathan and Shashank Sripad

316

APPLICATIONS – TRANSPORTATION APPLICATIONS: Aviation - Fuel Cells V Ahilan, and W Resende

P Nehter, H Geisler, 324

APPLICATIONS – TRANSPORTATION APPLICATIONS: Space - Batteries and Fuel Cells Ankit Kumar Chourasia, Keerti M Naik, Naga Keerthana Apparla, Surya Babu S., Amit Mall, Srijina M., and Chandra S Sharma

347

APPLICATIONS – TRANSPORTATION APPLICATIONS: Auxiliary Power Units - Fuel Cells Matteo Genovese, Viviana Cigolotti, G Monteleone, Francesco Piraino, O Corigliano, and Petronilla Fragiacomo

358

APPLICATIONS – TRANSPORTATION APPLICATIONS: Battery Charging Technologies Cuili Chen, Xueyuan Wang, Jiangong Zhu, Kailong Liu, and Alois Christian Knoll

373

APPLICATIONS – TRANSPORTATION APPLICATIONS: Hydrogen Refueling Stations Andrii Klymovskyi, Alexander Kabza, Günther Schlumberger, Ludwig Jörissen, and Markus Jenne

391

Conversion Tables

401

Glossary Index

Peter Kurzweil

Jürgen Garche

407 447

This page intentionally left blank

SECTION EDITORS - VOLUME 1 Section Title: Energy, Environment, and Resources Section Editors: Ludwig Jörissen, Center for Solar Energy and Hydrogen Research (ZSW), Ulm, Germany Jürgen Garche, Ulm University, Ulm, Germany Section Title: Electrochemical Fundamentals Section Editors: Sylvain Brimaud, Center for Solar Energy and Hydrogen Research (ZSW), Ulm, Germany Peter Kurzweil, Technical University of Applied Sciences (OTH), Amberg-Weiden, Germany Section Title: History of Electrochemistry Section Editor: Peter Kurzweil, Technical University of Applied Sciences (OTH), Amberg-Weiden, Germany Section Title: Electrochemical Devices Section Editor: Peter Kurzweil, Technical University of Applied Sciences (OTH), Amberg-Weiden, Germany Section Title: Electrochemical Terminology Section Editor: Peter Kurzweil, Technical University of Applied Sciences (OTH), Amberg-Weiden, Germany Section Title: Chemistry and Electrochemistry of Elements Section Editor: Jürgen Garche, Ulm University, Ulm, Germany

xxxiii

This page intentionally left blank

CONTRIBUTORS TO VOLUME 1 Jeffin James Abraham Center for Advanced Materials, Qatar University, Doha, Qatar Yuichi Aihara Nanoscale Organisation and Dynamics Group, School of Science, Western Sydney University, Penrith, NSW, Australia; Nissan Motor Corporation, Yokohama, Japan David Aili Department of Energy Conversion and Storage, Technical University of Denmark, Lyngby, Denmark Leena Al-Sulaiti Department of Mathematics, Statistics and Physics, College of Arts and Sciences, Qatar University, Doha, Qatar Casas Ocampo Andrea CIC energiGUNE, Vitoria, Spain AJ Appleby Texas A&M University, College Station, TX, USA Hajime Arai Tokyo Institute of Technology, Tokyo, Japan Rosa M Arán-Ais Instituto de Electroquí mica, Universidad de Alicante, Alicante, Spain Frédéric Barrière Université de Rennes, CNRS, Institut des Sciences Chimiques de Rennes, Rennes, France Sebastián Pinto Bautista Helmholtz Institute Ulm for Electrochemical Energy Storage (HIU), Ulm, Germany; Institute for Technology Assessment and Systems Analysis (ITAS), Karlsruhe Institute of Technology, Karlsruhe, Germany James A Behan Université de Rennes, CNRS, Institut des Sciences Chimiques de Rennes, Rennes, France R Benger Research Center Energy Storage Technologies, Clausthal University of Technology, Clausthal-Zellerfeld, Lower Saxony, Germany

HJ Bergveld NXP Semiconductors, Eindhoven, The Netherlands Giancarlo Cappellini Department of Physics, University of Cagliari, Cagliari, Italy E Cattaneo Hoppecke Batterien GmbH & Co. KG, Brilon, Germany Angel Cuesta Advanced Centre for Energy and Sustainability (ACES), School of Natural and Computing Sciences, University of Aberdeen, Aberdeen, United Kingdom; Centre for Energy Transition, University of Aberdeen, Aberdeen, United Kingdom D Danilov Eindhoven University of Technology, Eindhoven, The Netherlands Jiban K Das Hydro & Electrometallurgy Department, CSIR-Institute of Minerals and Materials Technology, Bhubaneswar, India Sonia Dsoke Institute for Applied Materials (IAM), Karlsruhe Institute of Technology (KIT), Eggenstein-Leopoldshafen, Germany Tasneem Elmakki Center for Advanced Materials, Qatar University, Doha, Qatar; Materials Science & Technology Program, College of Arts & Sciences, Qatar University, Doha, Qatar Pasquale Marcello Falcone Department of Business and Economics, University of Naples Parthenope, Napoli, Italy Bingliang Gao School of Metallurgy, Northeast University, Shengyang, China Gianluca Gatto Department of Electrical and Electronic Engineering, University of Cagliari, Cagliari, Italy xxxv

xxxvi

Contributors to Volume 1

Eva Gerold Department Metallurgy, Montanuniversitaet Leoben, Leoben, Austria Alan J Gibson Advanced Centre for Energy and Sustainability (ACES), School of Natural and Computing Sciences, University of Aberdeen, Aberdeen, United Kingdom; Centre for Energy Transition, University of Aberdeen, Aberdeen, United Kingdom Dong Suk Han Center for Advanced Materials, Qatar University, Doha, Qatar; Department of Chemical Engineering, College of Engineering, Qatar University, Doha, Qatar; Materials Science & Technology Program, College of Arts & Sciences, Qatar University, Doha, Qatar I Hauer Chair of Electrical Energy Storage Technology, Institute of Electrical Power Engineering, Clausthal University of Technology, Clausthal-Zellerfeld, Germany Enrique Herrero Instituto de Electroquí mica, Universidad de Alicante, Alicante, Spain Birger Horstmann German Aerospace Center (DLR), Ulm, Germany; Helmholtz Institute Ulm (HIU), Ulm, Germany; Ulm University, Ulm, Germany J-Y Huot Technical Consultant, Austin, Canada Atsunori Ikezawa Tokyo Institute of Technology, Tokyo, Japan Aleksander Jandric Department of Water-Atmosphere-Environment, Institute of Waste Management and Circularity, University of Natural Resources and Life Sciences, Vienna, Austria Martin Kaltschmitt Hamburg University of Technology, Institute of Environmental Technology and Energy Economics, Hamburg, Germany S Koohi-Fayegh Faculty of Engineering, University of Alberta, Edmonton, AB, Canada K Kordesch Graz University of Technology, Graz, Austria Amit Kumar Department of Electrical and Electronic Engineering, University of Cagliari, Cagliari, Italy P Kurzweil Technical University of Applied Sciences (OTH), Amberg-Weiden, Germany

Jelto Lange Hamburg University of Technology, Institute of Environmental Technology and Energy Economics, Hamburg, Germany Arnulf Latz German Aerospace Center (DLR), Ulm, Germany; Helmholtz Institute Ulm (HIU), Ulm, Germany; Ulm University, Ulm, Germany Qingfeng Li Department of Energy Conversion and Storage, Technical University of Denmark, Lyngby, Denmark Zhenyou Li Helmholtz Institute Ulm for Electrochemical Energy Storage (HIU), Ulm, Germany; Qingdao Institute of Bioenergy and Bioprocess Technology, Chinese Academy of Sciences, Qingdao, Shandong, China Arnas Majumder Department of Electrical and Electronic Engineering, University of Cagliari, Cagliari, Italy Krzysztof Maksymiuk University of Warsaw, Warsaw, Poland Robert J Messinger Department of Chemical Engineering, The City College of New York, CUNY, New York, NY, United States P Mock German Aerospace Center (DLR), Institute of Vehicle Concepts, Stuttgart, Germany Buzaina Moossa Center for Advanced Materials, Qatar University, Doha, Qatar; Department of Chemical Engineering, College of Engineering, Qatar University, Doha, Qatar Roger C Newman Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, ON, Canada Thomas Nigl Department of Environmental and Energy Process Engineering, Montanuniversitaet Leoben, Leoben, Austria L Niklaus Fraunhofer Institute for Silicate Research ISC, Würzburg, Germany PHL Notten Eindhoven University of Technology, Eindhoven, The Netherlands; Philips Research Laboratories, Eindhoven, The Netherlands Marek Orlik Laboratory of Electroanalytical Chemistry and Electrocatalysis, Faculty of Chemistry, The University of Warsaw, Warsaw, Poland

Contributors to Volume 1

Santhosh Paramasivam Department of Electrical and Electronic Engineering, University of Cagliari, Cagliari, Italy Tarun Parangi Shri N. A. Patel Science College, Kakanpur, Shri Govind Guru University, Godhra, Gujarat, India Florian Part Department of Water-Atmosphere-Environment, Institute of Waste Management and Circularity, University of Natural Resources and Life Sciences, Vienna, Austria V Pop Holst Centre/IMEC-NL, Eindhoven, The Netherlands U Posset Fraunhofer Institute for Silicate Research ISC, Würzburg, Germany Stefanie Prenner Brimatech Services GmbH, Vienna, Austria William S Price Nanoscale Organisation and Dynamics Group, School of Science, Western Sydney University, Penrith, NSW, Australia Zawar Alam Qureshi Center for Advanced Materials, Qatar University, Doha, Qatar; Department of Chemical Engineering, College of Engineering, Qatar University, Doha, Qatar PPL Regtien University of Twente, Enschede, The Netherlands B Riegel Hoppecke Batterien GmbH & Co. KG, Brilon, Germany

xxxvii

Institute for Molecular Physical Science, ETH Zurich, Zurich, Switzerland Theresa Schoetz Department of Chemical and Biomolecular Engineering, University of Illinois Urbana-Champaign, Urbana, IL, United States M Schott Fraunhofer Institute for Silicate Research ISC, Würzburg, Germany U Schröder Institute of Biochemistry, University of Greifswald, Greifswald, Germany Felix K Schwab German Aerospace Center (DLR), Ulm, Germany; Helmholtz Institute Ulm (HIU), Ulm, Germany Rana Abdul Shakoor Center for Advanced Materials, Qatar University, Doha, Qatar; Department of Mechanical and Industrial Engineering, College of Engineering, Qatar University, Doha, Qatar Jadwiga Stroka University of Warsaw, Warsaw, Poland T Takamura Harbin Institute of Technology, Harbin, China Hanan Abdurehman Tariq Center for Advanced Materials, Qatar University, Doha, Qatar W Taucher-Mautner Graz University of Technology, Graz, Austria

MA Rosen Faculty of Engineering and Applied Science, Ontario Tech University, Oshawa, ON, Canada

Allan M Torres Nanoscale Organisation and Dynamics Group, School of Science, Western Sydney University, Penrith, NSW, Australia

Bettina Rutrecht K1-MET GmbH, Linz, Austria

S Trasatti University of Milan, Milan, Italy

Noha Sabi Institute for Applied Materials (IAM), Karlsruhe Institute of Technology (KIT), Eggenstein-Leopoldshafen, Germany; High Throughput Multidisciplinary Research (HTMR), Mohammed VI Polytechnic University (UM6P), Ben Guerir, Morocco

Bankim Ch Tripathy Hydro & Electrometallurgy Department, CSIR-Institute of Minerals and Materials Technology, Bhubaneswar, India

W Schmickler Institute of Theoretical Chemistry, Ulm University, Ulm, Germany

Ilaria Tutore Department of Management and Quantitative Studies, University of Naples Parthenope, Napoli, Italy

SA Schmid German Aerospace Center (DLR), Institute of Vehicle Concepts, Stuttgart, Germany

Alberto Varzi Helmholtz Institute Ulm (HIU), Ulm, Germany; Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany

Thomas J Schmidt Electrochemistry Laboratory, Energy and Environment Division, Paul Scherrer Institute, Villigen, Switzerland;

Liping Wang Helmholtz Institute Ulm for Electrochemical Energy Storage (HIU), Ulm, Germany

xxxviii

Contributors to Volume 1

Marcel Weil Helmholtz Institute Ulm for Electrochemical Energy Storage (HIU), Ulm, Germany; Institute for Technology Assessment and Systems Analysis (ITAS), Karlsruhe Institute of Technology, Karlsruhe, Germany H Wenzl Consulting for Batteries and Power Engineering, Osterode, Germany

Martin Wietschel Competence Center Energy Technology and Energy Systems, Fraunhofer Institute for Systems and Innovation Research ISI, Karlsruhe, Germany Desmond Williams Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, ON, Canada

Energy, Environment, and Resources | Energy Transition and Sector Coupling Martin Wietschel, Competence Center Energy Technology and Energy Systems, Fraunhofer Institute for Systems and Innovation Research ISI, Karlsruhe, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 3 3.1 3.2 3.3 3.4 3.5 3.6 4 5 5.1 5.2 6 References

Introduction The challenges of the today’s energy system Transition to achieve the necessary greenhouse gas reduction Energy efficiency improvements Decarbonization of the electricity sector mainly driven by renewables Increasing direct electrical utilization and modern bioenergy Utilization of green hydrogen and synthesis products Negative emissions by direct air carbon capture and storage Other measures Transition to integrate the renewables into the energy system Transition to avoid energy and raw material supply risks Energy supply risks Critical raw material supply risks Summary

1 1 3 3 3 3 5 5 5 5 6 6 6 7 7

Abstract Today’s energy sector is facing a range of challenges such as the climate crisis, the energy crisis with high, volatile energy prices and a critical supply of raw materials. This requires a comprehensive transformation of the energy sector. Among other things, energy efficiency must be increased, renewable energies must be rapidly expanded, electrification must be driven forward and a hydrogen economy must be established. The ongoing expansion of fluctuating renewable electricity generation requires increased flexibility measures in the energy system of the future. As many of the new technologies require critical raw materials, strategies for dealing with these must be developed and implemented. The challenges and transformation paths are discussed in the chapter.

Key points

• • • • •

1

Challenges of today’s energy system. Transition to achieve the necessary greenhouse gas reduction Energy transition. Transition to integrate the renewables in the energy system. Transition to avoid energy and raw material supply risks Sector coupling and flexibility for integration of renewable energy.

Introduction

The various crises such as climate change, the energy crisis with high energy prices or the limited availability of critical raw materials for clean energy technologies, make it necessary to restructure the energy system. This is described in this book chapter. First, the various challenges facing the energy sector today are discussed. The next sub chapter shows which developments are necessary to drastically reduce greenhouse gas emissions. Then the changes to the system resulting from the imminent increase in renewable energies for power generation will be subject of discussion. The final sub chapter describes the changes to the energy system from the perspective of securing the energy supply and the supply of critical raw materials.

2

The challenges of the today’s energy system

There are currently a number of major challenges that are closely linked to the energy issue. Climate change is one of the greatest challenges facing humankind. According to the new IPCC report,1 the alarm bells are ringing:

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00312-1

1

2

•

•

Energy, Environment, and Resources | Energy Transition and Sector Coupling “Human activities, principally through emissions of greenhouse gases, have unequivocally caused global warming, with global surface temperature reaching 1.1 C above 1850–1900 in 2011–20. Global greenhouse gas emissions have continued to increase, with unequal historical and ongoing contributions arising from unsustainable energy use, land use and land-use change, lifestyles and patterns of consumption and production across regions, between and within countries, and among individuals.”.1 “Widespread and rapid changes in the atmosphere, ocean, cryosphere and biosphere have occurred. Human-caused climate change is already affecting many weather and climate extremes in every region across the globe. This has led to widespread adverse impacts and related losses and damages to nature and people (high confidence). Vulnerable communities who have historically contributed the least to current climate change are disproportionately affected.”1

The Paris Agreement on climate targets calls for complete greenhouse gas neutrality between 2040 and 2060.2 This means that 85% of the remaining fossil energy resources must remain in the ground. The current energy system is a major driver of global warming, accounting for about 75% of total greenhouse gas emissions.3 CO2 emissions are the main driving force behind climate change, but methane emissions also play a significant role and it is here that the energy sector is the largest user of methane. Despite the rapid recent growth in clean energy technologies in the last years, the world is still predominantly dependent on fossil fuels for its energy supply. Oil remains the single source of global primary energy demand, accounting for 29% in 2021, followed by coal at 26% and natural gas at 23%.4 This means that a transformation of energy production and consumption towards clean technologies is essential to achieve net zero greenhouse gas emissions by mid-century. However, the current regulations are not sufficient and the appropriate policy framework needs to be put in place.. According to the EIA4 in most cases, energy-related CO2 emissions continue to rise until 2050 under current laws and regulations. Population and income growth offset the effects of declining energy and carbon intensity on emissions. The longer the delay in taking decisive action to reduce greenhouse gas emissions on a sustainable basis, the greater the economic and social costs are likely to be. Government support for the energy transition has increased in a number of countries, including the passing of the Inflation Reduction Act in the US and the Green Deal in the EU. However, the scale of the decarbonization challenge suggests that greater support is required globally, including policies to facilitate faster permitting and approval of low-carbon energy and infrastructure (see BP 2023). There is also a need of innovation. Many of the clean energy technologies required to get to net zero by mid-century are not available at scale today. This includes key technologies, such as hydrogen based steel production or direct air carbon capture and storage. Bringing new technologies to market takes time. Experience shows that in the past the innovation process typically has taken between 20 and 70 years from prototype to commercialization.4 Russia’s invasion of Ukraine in 2022 and the Covid-19 pandemic have had far-reaching impacts on the global energy system, disrupting supply and demand patterns and fracturing long-standing trade relationships. Prices for spot purchases of natural gas and coal have reached in 2022 and 2023 levels never seen before. High gas and coal prices account for 90% of the upward pressure on electricity costs around the world. Further on, the crisis has stoked inflationary pressures and created a looming risk of recession. With energy markets remaining extremely vulnerable, today’s energy shock is a reminder of the fragility and unsustainability of our current energy system. See.4 The unprecedentedly high energy prices in the last few years require measures to protect consumers and businesses, including vulnerable households and the clean energy technology industry. According to the IEA’s analyses5 the crisis is affecting all countries, but the combination of the Covid pandemic and the current energy crisis means that 70 million people who recently gained access to electricity are likely to lose that access—and 100 million people may no longer be able to cook with clean fuels, returning to unhealthy and unsafe cooking methods. The high energy prices have also led to unprecedented profits for fossil fuel companies such as refineries. A strong focus on oil security will be critical through the clean energy transition. This has made fossil fuel business models more attractive again, preventing the necessary restructuring of the energy sector. This goes hand in hand with energy security. For example, a strong focus on oil security will be critical through the clean energy transition. Also soaring material prices and shortages of critical minerals, semiconductors and other components in the last years are posing potential obstacles to the energy transition. Increasing prices for cobalt, lithium and nickel led to the first ever rise in battery prices, which have risen by nearly 10% globally in 2022.4 The geographical distribution of critical mineral extraction is closely linked to resource endowments. Much of it is very concentrated. As an example, the Democratic Republic of Congo alone produces 70% of the world’s cobalt, and just three countries account for more than 90% of global lithium production.4 Concentration at any point along a supply chain makes the entire supply chain vulnerable to incidents, to an individual country’s political decisions, natural disasters, technical failures or business decisions.4 The long lead times for new mines, which can be well over 10 years from the start of project development to first production, increase the risk that critical minerals supply becomes a major bottleneck in clean technology manufacturing.4 Other challenges are also on the horizon. In general, the concentration of key technologies production for the energy transition is an important issue. For mass-produced technologies such as wind power, batteries, electrolyzers, solar cells and heat pumps, the three largest manufacturing countries account for at least 70% of the production capacity, with China dominating in all of these areas.4 On one side China’s investment in clean energy supply chains has been instrumental in bringing down costs worldwide for key technologies, with multiple benefits for clean energy transitions. On the other side an escalating China-Taiwan conflict could be the next shock for the energy supply after the Ukraine war. This is due to China’s dominant market position in key technologies such as batteries and solar cells, and China is trying to expand its position in other key technologies such as electrolyzers and heat pumps. China also plays an important role as a supplier of critical raw materials such as rare earths, which are highly relevant to the production of energy technologies.

Energy, Environment, and Resources | Energy Transition and Sector Coupling

3

The many crises, some of them overlapping, show that we have clung to fossil fuels for too long and that a rapid transition to clean technologies is now required. The cost advantages of mature clean energy technologies, like wind turbines or photovoltaics, and the prospects for new technologies, such as low-emission hydrogen and hydrogen derivatives, are boosted by new policy measures.5 The result is a boost to the emerging global clean energy economy, with the goals of security, affordability and sustainability. It should be noted that the three goals are partly in conflict with each other and a policy must be well balanced in order to meet the different demands.

3

Transition to achieve the necessary greenhouse gas reduction

In terms of the contribution to GHG reduction and its timing, the following governance structures for the transformation of the energy system can be formulated using a hierarchical principle in five stages (but of course with economic optimization): 1) Top priority of the “energy efficiency first” principle to minimize demand, supplemented by circular economy and material efficiency. 2) Decarbonization of the electricity sector, essentially driven by renewable energies. 3) Direct electrical utilization and use of biomass, taking into account its limited availability and sustainability criteria. 4) Utilization of hydrogen and synthesis products, particularly in those areas where the other three reduction strategies are not possible, economically viable or acceptable. 5) Negative emissions from natural sinks or through capture and utilization or storage of CO2. The individual topics are discussed below. It is clear that there are many measures to reduce greenhouse gases, and only the most important ones can be outlined.

3.1

Energy efficiency improvements

Improving energy efficiency is the single most important measure for reducing greenhouse gases and addressing today’s energy crisis. In 2023, global momentum to target a doubling in the rate of efficiency progress to 4% gathered pace, which could cut today’s energy bills in advanced countries by a third and make up 50% of CO2 reductions by 2030.4 Efficiency measures can have a dramatic impact. For example today’s light bulbs are at least four times more efficient than those on sale two decades ago. An electric vehicle or heat pump uses much less final energy than a conventional car or gas boiler to do the same job. Cooling demand is the second largest contributor to the overall increase in global electricity demand over the coming decades (after electric vehicles), and new and ambitious efficiency standards could lead to significant energy savings here.5 Although the IEA assumes a growing world population and rising economic output in the net emission scenario, final energy demand will decrease slightly worldwide by 2050 due to improved energy efficiency. Other studies even show a significant reduction in global final energy demand in ambitious greenhouse gas scenarios.6 Next to the reduction of greenhouse gases energy efficiency has multiple benefits such as job creation, energy poverty alleviation, decreasing energy supply risks, protection of critical materials, public health and environmental sustainability. Closed-loop material management through recycling is becoming increasingly important, not only to save energy and emissions but also to conserve resources of critical materials. Economic development has historically coincided with increasing demand for materials, resulting in growing energy consumption and carbon dioxide emissions from materials production.7 Clean energy transitions could help to decouple these trends. Material efficiency strategies can contribute to CO2 emissions reduction throughout value chains. Opportunities for material efficiency exist at every stage of the life cycle, from design and fabrication, to use and endof-life.7

3.2

Decarbonization of the electricity sector mainly driven by renewables

The substitution of fossil-based power plants with renewable energy sources is the most relevant measure to reduce greenhouse gas emission from the electricity sector. In the net zero emission scenario of the IEA5 solar PV and wind are the leading means of cutting electricity sector emissions. Their global share of electricity generation increases from 10% in 2021 to 40% by 2030, and 70% by 2050 (see Fig. 1). In the same IEA scenario nuclear power generation more than doubles by 2050, although its share falls from 10% in 2021 to 8% in 2050, as total generation expands rapidly. The use of low-emission hydrogen, and perhaps ammonia, in natural gas and coal-fired power plants and the addition of carbon capture technologies both provide important means of reducing emissions from existing power plants while supporting electricity security.5 This goes hand in hand with the phasing out unabated use of coal to generate electricity and a reduced role of natural gas in the future.

3.3

Increasing direct electrical utilization and modern bioenergy

Today, electricity accounts for about 20% of the world’s total final energy consumption.5 Electricity demand is expected to increase dramatically (see Fig. 2) as a result of a shift from fossil fuels to clean electricity, driven by the reduction of greenhouse gas emissions. In the bp net zero scenario electricity reaches 50% of total final consumption in 2050.6

4

Energy, Environment, and Resources | Energy Transition and Sector Coupling

80000

60000

40000

Hydrogen and ammonia Other renewables Solar PV Wind Hydropower Nuclear Fossil fuels with Carbon capture and use/storage (CCUS) Unabated oil Unabated natural gas Unabated coal

20000

0 2010

2021

2030

2050

Fig. 1 Global electricity supply development in a net zero emission scenario. Own illustration with data form IEA. World Energy Outlook. International Energy Agency (IEA): Paris (2022).

Gloabl electtrictiy demand (TWh)

70000 60000 50000 40000 30000 20000 10000 0 2010 Buildings

2021 Industry

Transport

2030

2050

Hydrogen production

Fig. 2 Global electricity demand development in a net zero emission scenario. Own illustration with data form IEA. World Energy Outlook. International Energy Agency (IEA): Paris (2022).

Furthermore, recent energy shortages and price spikes highlight the importance of an orderly transition away from hydrocarbons, so that the demand for hydrocarbons falls in line with available supply. In the building sector the shift away from oil and gas heating to heat pumps and the increasing use of heat pumps in district heating networks instead of fossil power plants are the major driver here. Also increasing electricity demand due to space cooling will have a significant impact. In developing countries, particularly in Africa, cooking on open fires or basic stoves can be replaced by electric or even gas cooking. In the industry sector electricity is mainly used for electric motors, aluminum smelting and electric arc furnaces. Future potential is offered by new technologies, for example electric steam cracker in the chemical sector, or electric steel production. Electric vehicles are becoming increasingly important in road traffic, both for cars and trucks. Also fuel cell heavy trucks could gain relevant market share by 2050. The increasing electrification of railroad lines also plays a role here. Low-emission hydrogen based on water electrolysis and hydrogen derivatives such as eKerosene or feedstocks such as ammonia or methanol for the chemical sector are additional drivers of future electricity demand. In addition, the increased use of sustainable biomass such as modern solid biomass, biofuels and biomass in the demand sectors can contribute to the reduction of greenhouse gas emissions. However, the potential of sustainable biomass is limited by, among other things, competition with food production. Therefor the use of sustainable biomass should focus on hard-to-abate sectors, especially in cement production, aviation and shipping and power generation.

Energy, Environment, and Resources | Energy Transition and Sector Coupling 3.4

5

Utilization of green hydrogen and synthesis products

The production and use of green hydrogen via electrolysis using renewable electricity in demand sectors is another major source for the increasing electricity demand (see Fig. 2). Green hydrogen and its derivatives will be used in the future particularly in applications where direct electrification is difficult or impossible, particularly in.

• • • • •

the steel sector with direct reduced iron, the production of basic chemicals via Fischer-Tropsch-synthesis and methanol or ammonia production, international aviation (renewable kerosene) and shipping (renewable ammonia or methanol), refineries due to the substitution of hydrogen production by electrolysis instead of gas steam reforming, and in. residual power generation with hydrogen storage and electrification as a flexibility option.

The restriction to certain applications is justified by the fact that the production of hydrogen and its derivatives is associated with considerable conversion losses and will continue to be comparatively expensive. Green hydrogen and its derivatives will only play a significant role in the future energy system if greenhouse gas reduction targets of more than 80% are achieved.8 But for them to play a role, the technologies need to be developed and the infrastructure put in place early. There is currently a great deal of uncertainty about the development of hydrogen prices. Due to the market situation—currently only a limited supply and the expectation of rising demand even if this is only just becoming apparent—green hydrogen and derivatives are likely to be in short supply and expensive in the short to medium term. In addition, it takes time to the buildup production capacities and transport infrastructure. The production of hydrogen and hydrogen derivatives calls for large investments, and comparatively high energy losses occur across the entire production, transportation and use chain. As a result, hydrogen and its derivatives will tend to be more expensive than other energy sources. A strategy should therefore set clear priorities on areas in which hydrogen and its derivatives are essential to achieve climate targets. If other sectors such as building heating or road transport are included on a much larger scale, this could lead to misallocations and unnecessarily drive up the prices in other sectors.

3.5

Negative emissions by direct air carbon capture and storage

In spite of increasingly ambitious reduction targets for greenhouse gases, existing strategies to mitigate climate change are likely to be insufficient and the window of opportunity to achieve greenhouse gas neutrality is shrinking. Therefore, discussions are increasingly focusing on negative emissions—the removal of CO2 from the air and its storage—which would reduce the amount of CO2 accumulated in the atmosphere. Various natural and technical processes are available for this purpose. These processes are seen as a possible option for offsetting CO2 emissions in sectors that are difficult to decarbonize, such as the cement sector. But there are still some challenges, including technological developments, costs and acceptance. According to current knowledge, it is a rather expensive reduction measure compared to other greenhouse gas measures.

3.6

Other measures

It is clear that there are also other important measures to reduce greenhouse gases. In particular, the avoidance of methane emissions has been coming more and more into focus in recent times and is considered important and feasible in the short term at comparatively low cost. Methane emissions from fossil fuels, responsible for almost a third of the rise in global temperatures since the Industrial Revolution, remain far too high to meet international climate targets.9 To limit global warming to 1.5  C, a key goal of the Paris Agreement, methane emissions from fossil fuels need to decline by 75% in this decade, according to IEA. Not directly linked to the energy sector but of high relevance is the issue of land. Land is both a source and a sink of GHGs. Land use plays a key role in the exchange of energy, water and aerosols between the land surface and atmosphere, and human use directly affects more than 70% of the world’s ice-free land surface.10 Many land-related responses that contribute to climate change adaptation and mitigation can also combat desertification and land degradation and enhance food security. The protection of existing ecosystems, for example by avoiding slash-and-burn and overexploitation, and the creation of new ecosystems, e.g. through reforestation, are important climate protection measures.

4

Transition to integrate the renewables into the energy system

In the traditional electricity system, production followed the pattern of demand. Most of the flexibility required to maintain power system reliability today is provided by dispatchable thermal power plants and hydropower. However, the today’s power system has to be adopted to multiple changing boundary conditions. Power system flexibility needs are driven primarily by the rising share of variable wind and solar PV in electricity generation and reduction of thermal power plants as shown in chapter 3.2. The integrating of millions of very small photovoltaic and wind generators operated by private households, small companies, and communes is a challenging task. As shown in chapter 3.3 the electrification of additional end uses, e.g. electric heating and loading electric vehicles, raises peaks and increases the hourly, daily and seasonal variability of electricity demand. Short term electricity storage and longer term storage by thermal heat (including the short term function of buildings) or hydrogen, ammonia or methanol is required.

6

Energy, Environment, and Resources | Energy Transition and Sector Coupling

Questions arise as to how to manage these non-controllable energy sources, how to deal with excess electricity generation, and how to use it efficiently in terms of economic, ecologic and social welfare aspects. To efficiently integrate these variable primary energies, the traditional coupling of the energy sector to the residential, transport, industry, and commercial sector has to be abandoned. And a policy needs to be well balanced in order to meet the different demands. Although the terms “sector coupling” (SC) and “integrated energy” are frequently used in the current energy policy debate, they are often not used unequivocally. Following one of the first definitions by,11 SC is seen as the substitution of fossil fuels in conventional technologies with alternative primary energies (e.g. renewables including wind, solar, hydro, biomass, geothermal) in new applications or technologies. This can be done either by directly using electricity

• • • •

in Power-to-Heat PtH. e.g. heat pumps, electro-thermal industrial processes, in Power-to-Move PtM, e.g. vehicles driven by electrical motors, or by converting electricity into synthetic fuels. Power-to-Gas PtG (e.g. hydrogen) as substitution of conventional fossil gases, or. Power-to-Liquid PtL (e.g. green ammonia, methanol or e-fuels) as substitution of fossil fuels.

These electricity-based final energies are subsumed as Power-to-X (PtX) energies. In addition, the focus here is on the use of new or alternative technologies and less on classical power applications such as electrical motors, night storage heating, or electric trains and trams. This view of SC concentrates on techno-economic issues. The broadly defined aspects of SC also include new standards, new business segments, IT issues (including cyber security), as well as legal and regulatory aspects. A major advantage of some of the relevant new applications for electricity is that they are easy to control, such as heat pumps and electric vehicles. Load shifting in very electricity-intensive production processes such as manufacturing electrical steel, aluminum and chlorine has been around for a long time, but increasing digitization makes it economically feasible to include smaller industrial and commercial electricity consumers in load management. Additionally, new storage technologies are still needed. Stationary batteries, but also the vehicle batteries, can be used to store electricity for short periods. New thermal stores may also become increasingly important in the future. For long-term storage, hydrogen and its derivatives are particularly suitable. Another important measure is the expansion of electricity networks, especially across national borders.

5 5.1

Transition to avoid energy and raw material supply risks Energy supply risks

Russia’s war in Ukraine has led many to realize that the formerly held certainty that close trade relations result in a stable energy supply should be questioned. According to a recent analysis of the U.S. Energy Information Administration3 energy security concerns are accelerating the transition from fossil fuels in some countries, as they focus on energy security through domestically produced renewables and reduce the role of oil and natural gas imports. These goes hand in hand with reaching climate targets. In other countries, however, the opposite reaction may occur. Major crude oil and natural gas producers will continue producing to keep up with growing demand from consumers such as China, India, Southeast Asia, and Africa under prevailing policy, which hinders the achievement of ambitious climate targets. Energy transitions offer the chance to build a safer and more sustainable energy system that reduces exposure to fuel price volatility. Political decision makers should not automatically favor the most cost-effective import pathway as has often been the case in the past, but should also consider the following aspects:

• • • • • • 5.2

addressing the demand side and focusing on efficiency-related demand reduction (the recent gas crisis in Germany and the EU demonstrated the large potential here). synchronizing the expansion of clean energy technologies with the phasing out of fossil fuels (IEA 2022). investments in flexibility like battery-storage and demand side response due to increasing variability in electricity supply and demand.5 diversification via different suppliers, routes and transport options. resilience to shocks by storing energy (e.g. gas or hydrogen storage in salt caverns). long-term supply contracts to safeguard supply, possibly featuring flexibility with regard to purchase quantities.

Critical raw material supply risks

The increasing demand for minerals and materials associated with the energy transition comes from across the low-carbon energy system, including the construction of wind and solar facilities, batteries, fuel cells, hydrogen and heat pumps. Critical materials for the energy transition include copper, lithium, cobalt and nickel. For example, copper is needed in wind offshore and onshore as well as for solar PV in much higher content compared to the conventional technologies such as gas power plants. The same applies to heat pumps compared to gas boilers. Larger quantities of lithium are required in conventional batteries for electric vehicles. The Polymer electrolyte membrane electrolysis (PEM) for the production of hydrogen needs iridium. Here Iridium is used as an example to illustrate the challenges. The dependence on iridium leads to high supply risk; there is only a very low level of production worldwide, and the global market is highly concentrated. Iridium is one of the so-called platinum group metals (PGMs). South

Energy, Environment, and Resources | Energy Transition and Sector Coupling

7

Africa is the largest producer, followed by Russia. The market for this technology, and therefore the demand for iridium, is expected to grow strongly and, against this background, together with the supply situation as a by-product, it is subject to significantly increased price and supply risks. A significant increase in iridium production does not seem possible at present. Therefore, reducing the need for critical materials will be important for supply chain sustainability, resilience and security. The following policy recommendations can be made:

• • • • •

energy efficiency: slowing growth in electricity demand and reducing need for raw material; circular Economy: Closing material cycles (Increase material reuse, recyclability and recycling rates); substitution of critical materials; adopting standards for clean technologies and traceability. International collaboration and strategic partnerships to address security of supply risks geographical concentration.

An example of a successful critical material substitution strategy is the change in battery technology. Here new alternatives to conventional lithium-ion batteries (Nickel-Mangan-Cobalt (NMC) are on the rise. The share of lithium-iron-phosphate (LFP) chemistries which don’t need the critical mineral cobalt and nickel reached an all-time high. The supply chains for (lithium-free, nickel-free, graphite-free) sodium-ion batteries are also being established.

6

Summary

The energy sector is facing a number of challenges today. The climate crisis requires a drastic reduction in greenhouse gases and a shift away from fossil resources. The energy supply crisis caused by Russia’s war of aggression against Ukraine leads to high and volatile energy prices and disruptions in supply relationships. Furthermore, critical materials that are limited in supply and whose production is sometimes concentrated in a few countries, are a challenge, because new clean energy technologies usually need more of these materials than conventional technologies. The relationship between climate targets, affordable energy and resilient energy as well as the supply of raw materials needs to be rebalanced. This requires a comprehensive transformation of the energy sector. The future of global energy will be shaped by the following trends: energy efficiency improvements and circular economy, decarbonisation of the electricity sector mainly driven by renewables, increasing direct electrical utilization and use of modern bioenergy, utilization of green hydrogen and hydrogen derivatives, and negative emissions by direct air carbon capture and storage. The expected significant increase in intermittent renewables in electricity generation worldwide requires the installation of new flexibilities. The development of load management, new storage systems such as batteries, and the expansion of electricity grids need be driven forward. With a focus on energy security of supply, additional diversification through different suppliers, routes and transport options, energy storing strategies, and long-term supply contracts are necessary. Raw material risk can be managed among others by closed-loop material managements, substitution of critical materials, adaptation of clean technology and traceability standards, and international collaboration and strategic partnerships.

References 1. IPCC. Summary for Policymakers. In Climate Change 2023: Synthesis Report, Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; Core Writing Team, Lee, H., Romero, J., Eds.; IPCC: Geneva, 2023; pp. 1–34. https://doi.org/10.59327/IPCC/AR6-9789291691647.001. 2. UN: The Paris Agreement (2024) Online available at: https://unfccc.int/process-and-meetings/the-paris-agreement (Last checked on April 03, 2024). 3. EIA: International Energy Outlook (2023) U.S: Energy Information Administration (EIA). Online available at https://www.eia.gov/outlooks/ieo/ (Last checked on April 03, 2024). 4. IEA. Energy Technology Perspectives; International Energy Agency (IEA): Paris, 2023. 5. IEA. World Energy Outlook; International Energy Agency (IEA): Paris, 2022. 6. British Petrol (bp): Energy Outlook 2023 Edition. 2023, Online available at https://origin.iea.org/reports/world-energy-outlook-2023 (Last checked on April 03, 2024). 7. IEA. Material Efficiency in Clean Energy Transitions; International Energy Agency (IEA): Paris, 2019. 8. Riemer, M.; Zheng, L.; Eckstein, J.; Wietschel, M.; Pieton, N.; Kunze, R.: Future Hydrogen Demand: A Cross-Sectoral, Global Meta-Analysis. HyPat Working Paper 04/2022. Karlsruhe: Fraunhofer ISI. 9. IEA. Global Methane Tracker 2024; International Energy Agency (IEA): Paris, 2024. 10. IPCC. Summary for Policymakers. In An IPCC Special Report on Climate Change, Desertification, Land Degradation, Sustainable Land Management, Food Security, and Greenhouse Gas Fluxes in Terrestrial Ecosystems; IPCC, 2019. 11. Wietschel M, Plötz P, Pfluger B, Klobasa M, Eßer A, Haendel M, Müller-Kirchenbauer J, Kochems J, Hermann L, Grosse B, Nacken L, Küster, M, Pacem J, Naumann D, Kost C, Kohrs R, Fahl U, Schäfer-Stradowsky S, Timmermann D, Albert D: Sektorkopplung – Definition, Chancen und Herausforderungen. ISI Working Paper Sustainability and Innovation. Band: S 01/2018. Fraunhofer-Institut für System- und Innovationsforschung ISI, Karlsruhe.

Energy, Environment, and Resources | Renewable Energies Jelto Lange and Martin Kaltschmitt, Hamburg University of Technology, Institute of Environmental Technology and Energy Economics, Hamburg, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 3 4 References

Introduction Global energy demand Technical potential of renewable energies Summary and conclusion

8 9 13 16 17

Abstract The steady increase in global primary energy demand in the past has led to a substantial rise in greenhouse gas emissions, which contribute significantly to climate change. To mitigate the latter, the energy demand must be met more sustainably, which requires an unprecedented expansion of renewable energies. In this context, the question arises whether renewable sources of energy have sufficient global potential to meet the steadily increasing demand in the future. Against this background, this chapter contrasts the necessary expansion of renewable energies with the technical potentials available worldwide and attempts to answer the question of whether a mostly renewable energy supply would be possible on the basis of technical potentials.

Glossary Carbon budget The carbon budget refers to the total amount of carbon dioxide (CO2) emissions from anthropogenic sources that may be emitted as a maximum, starting from a reference point in time, if global warming beyond a defined limit is to be avoided with a certain probability. Carbon intensity Carbon intensity describes the ratio of carbon dioxide (CO2) emissions to production (e.g., energy supply) Input equivalent The input equivalent describes the amount of primary energy that a conventional power plant would need to supply the same amount of electricity as a system with direct electrical energy output (such as hydroelectricity, wind energy). Primary energy Primary energy describes the energy derived from natural energy sources that have not yet been further processed (e.g., coal, oil, and natural gas). Resource potential The resource or theoretical potential of a (renewable) source of energy describes the energy that can theoretically be physically utilized within a defined region and period of time. It describes the theoretical maximum of available energy of the particular source. Technical potential The technical potential is the share of the resource or theoretical potential that can be used, taking into account the given technical restrictions (e.g., system performance, topographic constraints, land-use limitations).

Key points

• • • • •

1

Provide an overview of global primary energy demand and its (possible) development Outline the need for the expansion of renewable energies to achieve climate protection goals Quantitative assessment of the scale of global energy demand and required renewable energy expansion Inform about the quantitative scale of technical potentials of renewable energies including their uncertainties Comparison of global primary energy demand and expansion requirements as well as technical potentials of renewable energies

Introduction

Impacts of global climate change are becoming increasingly apparent. The number of severe droughts, floods, heat waves, storms and other extreme weather events, at least the severity of which can be attributed to climate change, has increased significantly in recent years.1 Anthropogenic emissions of greenhouse gases (GHG) – with CO2 accounting for the largest share – are the main reason for this rapid and alarming development. In order to prevent the occurrence and continuously accelerating uncontrollable

8

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00137-7

Energy, Environment, and Resources | Renewable Energies

9

intensification of such weather extremes and their impact on human civilizations, their cause has to be mitigated. Therefore, to limit global climate change, suitable and effective actions are needed urgently to reduce anthropogenic greenhouse gas emissions (GHG) on a global scale. The greenhouse gas (GHG) emissions related to global energy consumption – especially the CO2 emissions from the combustion of fossil fuels like oil, natural gas and coal – are of major importance in this regard. They are the cause of roughly three quarters of global anthropogenic greenhouse gas (GHG) emissions.2 Reducing carbon intensity, and in particular absolute CO2 emissions from energy consumption, is therefore of great importance to limit global warming and to achieve (legally binding) climate protection goals. However, the overall energy demand as well as the energy-related greenhouse gas (GHG) and especially the CO2 emissions steadily increased over the last decades globally.2,3 Both, the increase in final energy consumption per capita and the growth of global population – most recently exceeding 8 billion people on earth4 – contributed strongly to this development. As global population is expected to grow beyond 10 billion by 20505 and different countries still lack sufficient energy availability for all of their citizens, these trends are expected to continue – at least as long as cheap fossil fuels are sufficiently available. Thus, a substantial net reduction of global energy consumption and of the corresponding greenhouse gas (GHG) emissions seems rather unlikely for the near future. Therefore, supplying energy more sustainably without causing greenhouse gas (GHG) emissions seems to be the only way to reduce – and eventually eliminate – climate change impacts of energy consumption. Renewable energies like solar energy, wind power, hydro energy as well as geothermal energy offer a possible solution to this challenge. When using energy from the sun directly (i.e., solar energy) or indirectly (i.e., wind and hydro), a growing demand of energy can be supplied without increasing the emissions of greenhouse gases (GHG). In addition, other locally active airborne emissions that stem from the combustion of fossil fuels are avoided. In this context, however, the important question arises as to how much energy could be provided from renewable sources alone. Here, it is of primary importance whether the renewable energies possess sufficient technical potential to enable a comprehensive sustainable supply of the growing global energy demand. Based on this insight, it could be assessed whether the switch to renewable energies could generally nullify the energy-related climate impacts, or whether other additional measures are required and would, therefore, need to be developed immediately. Against this background, this chapter aims to provide an overview of the potentials of renewable energies for supplying the growing global primary energy demand. For this purpose, the discussed issues of a sustainable energy supply are approached from two directions. First, the current and (potential) future development of global primary energy consumption will be analyzed. This will give a more detailed definition of the overall requirements for global energy supply. Subsequently, by assessing the technical potential of various renewable energy sources, the perspectives for a sustainable energy supply are shown and contrasted with the demand side. While there are various – in some cases widely diverging – methods and results for quantifying the technical renewable energy potentials, this comparison nevertheless enables a basic assessment of the technical feasibility of a sustainable global energy supply (at least from the perspective of available technical potentials). This chapter is structured as follows. Section 2 gives an overview over the current and (potential) future global energy demand, discussing both the main drivers that influence energy consumption levels and the emerging development of global primary energy demand. This elaboration already emphasizes the need for reducing global greenhouse gas (GHG) emissions and consequentially the need for increasing the energy supply from renewable sources. Based on this, Section 3 gives a quantitative overview over the values for the technical potentials of different renewable energy sources in comparison to consumption levels. Finally, Section 4 closes with a conclusion.

2

Global energy demand

At the beginning of the third decade of the 21st century, the annual global energy consumption amounts to roughly 600 EJ a−1, while at the beginning of the 2000s 400 EJ a−1 of energies were used globally.6 Therefore, the total global energy consumption increased by about 50% in just 20 years respectively two decades.3,7 This rise in energy consumption is both driven by an increase in global population, which recently surpassed 8 billion people living on planet earth,4 as well as an increase in an energy consumption per capita. The latter is generally correlated with an increase in economic growth (of gross domestic product, GDP).8,9 Even though some more wealthy economies have partially decoupled GDP development respectively growth from energy consumption, especially for lower income countries economic growth is generally directly coupled to a corresponding rise in primary energy demand.10 Thus, it is unlikely, that reducing energy availability and consumption will become a major societal goal, especially for poorer regions that strive toward an increase in wealth, prosperity and opulence. Therefore, it has to be assumed that global energy consumption will increase further significantly and substantially in the coming years and decades. However, primary energy consumption contributes strongly to emissions of greenhouse gases (GHG). As large shares of the energy supply come from fossil fuels, which release carbon dioxide (CO2) during combustion, an increase in energy consumption has generally been related to a corresponding increase in CO2 emissions in the past.11,12 The emitted fossil fuel based CO2 – being the most common greenhouse gas (GHG) – contributes strongly to global warming and climate change. So, in order to limit global warming, emissions of CO2 and other greenhouse gases (GHG) need to be reduced consistently on a global level. However, this is in contrast to the increasing energy demand, which has generally led to a distinct increase in CO2 emissions in the past. This results in the overarching challenge to reduce the climate impact of a steadily growing energy consumption.

10

Energy, Environment, and Resources | Renewable Energies

Fig. 1 summarizes the discussed information, showing per capita emissions of CO2 over per capita primary energy consumption of 82 selected countries including their average income levels (i.e., an information on average wealth). Judging by this diagram, it is clearly visible, that countries with stronger economies (i.e., higher income countries) generally consume more energy and consequentially show larger annual CO2 emissions per capita compared to countries with less strong economies (i.e., lower income countries). Since the poorer countries also want to reach the standard of living of the wealthier regions, they will probably strive for a similar development as the wealthy countries – and thus a significant increase in energy consumption over time. If no additional measures are taken, this development consequentially leads to an additional increase in CO2 emissions on a global level. In order to stop countries from moving to the right and upwards in Fig. 1, the overall carbon intensity of energy consumption has to be reduced. This could then enable higher consumption levels while reducing total global emissions of CO2. Despite increasing awareness for the necessity to reduce greenhouse gas (GHG) respectively CO2 emissions, the global energy demand – and particularly the consumption of fossil fuels like oil, natural gas as well as hard coal and lignite – increased steadily over recent history (Fig. 2).7 Even though the growth of fossil fuel consumption partially slowed down and halted temporarily (e.g., coal since the beginning of the 2010s), overall the consumption of carbon intensive fossil fuels increased. Only nuclear energy, which could at least supply energy with low CO2 emissions (despite the potentially negative implications of nuclear energy like the production of dangerous nuclear waste), seems to show a consistent trend of slow and steady reduction since the beginning of the 21st century.7 In parallel, a strong increase of the energy supply from renewable sources of energy is visible in the most recent development. However, despite this clearly visible growth, renewable energies were not yet able to substitute existing fossil fuel consumption but rather reduced their further expansion.7 Thus, renewable energies have not yet been able to exert a substantial impact on CO2 emission reduction; they just slowed down the ongoing increase of greenhouse gas (GHG) emissions. This is also visible from energy respectively fossil fuel related CO2 emissions, that continuously increased with only temporal interruptions during the world economic crisis and the first year of the COVID-19 pandemic.2,7,13 100

Low Income

Per capita CO2-emissions in t a-1

Quatar

Russia

Lower Middle Income

USA

China

Upper Middle Income

10

High Income

India

Germany Iceland

Pakistan

Sweden

Brazil

1 Bangladesh

Indonesia

South Sudan

0.1

Ethiopia

Niger

0.01 0.1

1 10 Per capita primary energy consumpƟon in MWh a-1

100

Fig. 1 Per capita CO2 emissions over per capita primary energy consumption of 82 selected countries (Low Income: average income below 2 $ d−1 per capita, Lower Middle Income: average income between 2 and 8 $ d−1 per capita, Upper Middle Income: average income between 8 and 32 $ d−1 per capita, High Income: average income above 32 $ d−1).11,12

Natural Gas Nuclear Energy Hydroelectricity

200

80

150

60

100

40

50

20

0 1990

1995

2000

2005 Year

2010

2015

2020

0

Carbon dioxide emissions in Gt/a

Primary energy consumpƟon in EJ

Oil Coal Renewables Carbon dioxide (fossil fuels)

Fig. 2 Development of the consumption of various energy carriers and sources as well as the CO2 emissions from fossil fuels from 1990 to 2021.2,3,7,13

Energy, Environment, and Resources | Renewable Energies

11

While other anthropogenic greenhouse gas emissions also impact climate change (e.g., methane (CH4), nitrous oxide (NOx)), CO2 is by far the most relevant greenhouse gas – and the energy sector is the largest emitter of CO2, causing more than 85% of global anthropogenic CO2 emissions.2,13 Thus, a major task of the energy industry is to reduce its carbon intensity and consequentially its emissions of CO2. Currently, the latter amount to more than 35 Gt annually.2,13 Despite the fact that a high energy consumption typically results in large amounts of CO2 emissions, it seems rather unlikely, that global energy consumption will stop growing in the near future. As eliminating poverty is an important goal with high priority, energy consumption and consequentially also the CO2 emissions will continue to increase in the future, due to the abovementioned contexts. This is also reflected by different scenarios of the development of global energy consumption until 2050. These scenarios describe different possible pathways for the evolution of global economies and how this might impact the global supply and consumption of energy.14–16 While not claiming to be a prediction of future developments, these scenarios can help to assess the implications of different possible pathways of economic development on energy and consequentially environmental aspects like CO2 emissions. The consensus between these scenarios is, that overall (final) energy consumption will increase in one way or another (Fig. 3). While primary energy consumption is often accounted for differently (e.g., accounting for renewable power production calculating an input equivalent an average conventional power plant would need for supplying the same amount of electricity3,19 or directly treating renewable electricity as primary energy15), generally an intensification of energy end use is expected consistently by recently published scenario analyses. When applying the same accounting methodology for the different energy sources to the scenarios (with an input equivalent consideration19) by 2050 the total global primary energy consumption is expected to rise to an average of roughly 780 EJ a−1 with a minimum value3 of 653 EJ a−1 and a maximum value16 of 900 EJ a−1. Compared to today’s consumption6 of about 600 EJ a−1 this would correspond to an increase of at least 50 and on average 180 EJ a−1, which means that energy demand would increase by another third (between 10% and 50%) within only 30 years.

Global primary energy consumption in EJ

History bp: Net Zero IEA: Net Zero Shell: Sky 1.5

bp: Accelerated IEA: Announced Pledges Shell: Islands

bp: New Momentum IEA: Stated Policies Shell: Waves

1000 900 800 700 600 500 400 300 200

Future projections

Recent development

100 0 1990

2000

2010

2020 Year

2030

2040

2050

Fig. 3 Development of global primary energy consumption until 2020 and projections of the possible future development until 20503,7,14–18 (and own calculations, harmonizing accounting methods) (IEA: International Energy Agency).

Under the simplifying assumption of a continuation of the current composition of the global primary energy supply this increase in energy consumption will lead to a corresponding growth in global greenhouse gas (GHG) emissions. With today’s carbon intensity of primary energy, annual CO2 emissions from fossil fuels would rise by roughly 3 to 17.5 Gt a−1 until 2050.2,7,13 However, if the rise in global temperatures due to climate change is to be contained, only a limited amount of greenhouse gases (GHG) can still be emitted into the atmosphere. The quantity of permissible emissions – the carbon budget – amounts to roughly between 1150 and 1700 GtCO2e for halting global warming at 2  C with a probability of 66% and 33%, respectively. If the temperature increase should be limited to 1.5  C with a probability of 66% and 33%, only around 400 to 650 GtCO2e could still be emitted (all carbon budgets are given in CO2-equivalents for the year 2020).20 Thus, the rise in global energy consumption must not be accompanied by an increase of CO2 emissions but rather has to go hand in hand with a distinct and continuing reduction of greenhouse gas (GHG) emissions. Therefore, the carbon intensity of a growing energy demand has to be reduced substantially and over proportionately in a timely manner. For this reason, CO2 emissions are gaining more and more attention when discussing the development of global energy consumption. Therefore, in addition to the expected development of the energy demand, the scenarios also include the

12

Energy, Environment, and Resources | Renewable Energies

quantification of annual CO2 emissions that would result from the changed energy supply.14–16 However, it seems obvious that global energy planning does not yet broadly incorporate the importance of reducing CO2 emissions as various development pathways for the global primary energy consumption would not be in line with targets to limit global warming to 1.5 or 2.0  C. Only four of the nine paths depicted in Fig. 3 would not exceed the permissible annual emissions to stay on track of a 1.5  C- or 2.0  C-target. Allowable annual emissions are quantified be the International Panel on Climate Change (IPCC).21 Using climate models, the IPCC calculates various outcomes for the temperature development in the 21st century, considering different amounts of annual CO2 respectively greenhouse gas (GHG) emissions. Based on this, the IPCC defines different development pathways with varying outcomes for global temperature (e.g., staying below 1.5  C with a likelihood of 66%, not exceeding 2.  C with a likelihood of 50 to 66 %) and publishes the 25th percentile, median and 75th percentile for total greenhouse gas (GHG) and particularly CO2 emissions for the development paths.21 Fig. 4 shows these ranges of permissible annual CO2 emissions to stay in line with 1.5 and 2  C warming limits. Furthermore, it shows the recent development of greenhouse gas (GHG) emissions and the development of CO2 emissions for the scenarios that would be in line with the available carbon budgets and show possible pathways toward a sustainable energy future14,15 (the scenarios consider CO2 emissions from energy alone, while there are further sources of CO2 that would also have to be cut down, for the scenarios to stay in the range defined by IPCC scenarios). It can be seen that merely two of the nine overall scenarios depicted in Fig. 3 would be (roughly) in line with the 1.5  C-range. Comparing the four scenarios that would be in line with IPCC emission ranges to the rest of the scenarios depicted in Fig. 3 it becomes apparent, that while renewable energies generally grow strongly in every development scenario (reflecting the current development), they are much more dominant in the ones achieving overall low energy-related CO2 emissions. Thus, a major driver for reducing CO2 emissions in the scenarios with limited global warming is the broad implementation of renewable energies. Especially the scenarios in line with the 1.5  C-target reach shares of at least 75% of renewable energies on total energy supply (further emission reduction is achieved by slightly increasing nuclear energy as well as fossil energy with carbon capture and utilization as well as carbon capture and storage). Fig. 5 shows the current values and expected future average, minimum and maximum values of the total energy consumption and different renewable supply options for the scenarios that would likely limit global warming to 1.5 and 2.0  C, respectively. Most noticeably, total renewable energy grows drastically, increasing by roughly 600% until 2050. This growth results in an even more drastic development of modern renewable energies (e.g., solar and wind), since current values are strongly influenced by traditional

Fig. 4 Development of global annual CO2 emissions until 2020 and IPCC emission ranges for limiting global warming to 1.5  C and 2.0  C as well as the potential further development of energy-related CO2 emissions in line with IPCC ranges until 20502,13–18,21 (lower end of the 1,5  C-range: 25th percentile of annual CO2 emissions for paths keeping temperatures strictly below 1.5  C in the 21st century; upper end of the 1.5  C-range: 75th percentile of annual CO2 emissions for paths limiting the temperature increase in the 21st century to 1.5  C but with a high overshoot; lower end of 2.0  C-range: 25th percentile of annual CO2 emissions of paths limiting global warming to 2  C throughout the 21st century with a likelihood of more than 66%; upper end of 2.0  C-range: 75th percentile of annual CO2 emissions of paths limiting global warming to 2  C throughout 21st century with a likelihood of between 50 and 60%) (IPCC: International Panel on Climate Change, IEA: International Energy Agency).

Energy, Environment, and Resources | Renewable Energies

13

Fig. 5 Current values for annual total energy consumption, renewable energy supply and supply from different renewable energy sources as well as average, minimum and maximum values for their development in the development scenarios in line with 1.5  C- and 2.0  C-global warming limits14–18 ( including traditional use of biomass and biofuels).

biomass, which is expected to decline instead of increase. Therefore, during the coming three decades solar and wind energy would grow roughly 30- and 14-fold, respectively. Hydroelectricity and bioenergy would each grow by about 65% in the same period. Since traditional biomass is expected to (strongly) decline in the depicted scenarios, modern biomass would need to grow much stronger than the stated 65%, as it has to overcompensate the reduction of traditional biomass.15 However, despite being important contributors to a sustainable global energy supply, the development of hydroelectricity and bioenergy will be dwarfed by wind and especially solar energy expansion, if the real development comes even close to the depicted one. Therefore, the transition to large shares of renewables in the scenarios in line with a 1.5 and 2.0  C limit on global warming will be strongly based on the development of solar and wind energy. In this context, the important question arises as to how much potential these most important renewable sources of energy have on a global scale and whether this potential is large enough to enable the outlined growth necessary to limit significant negative impacts of global climate change.

3

Technical potential of renewable energies

Against this background, this chapter focusses on the available potentials of renewable energies. Thus, the following discussion outline the technical potential of the most important technologies for using renewable sources of energy on a global scale – namely solar energy, wind energy, hydroelectricity and bioenergy. The technical potential extends the quantification of resource potentials – that primarily focusses on physical constraints and the theoretical physical potential – by including system and topographic constraints as well as land-use constraints and system performances.22 Therefore, the technical potential defines an upper boundary for the technically realizable development of an energy resource.22 Thereby, it can help to estimate how restricted the development of renewable energies might be through technical infeasibilities. However, economics and market aspects – and thus costs – are not considered in these definitions and quantifications of the potential. While offering a way to assess the possible development of an energy resource, the quantification of global technical potentials for renewable energies is subject to large variations. Aspects of the technical constraints (e.g., land availability) are interpreted differently (e.g., conservative or progressive estimation of the share of land available for energy supply), which often leads to results that differ by orders of magnitude. Therefore, the following elaborations incorporate the calculated values and ranges of various sources to represent a broad view on technical potential for renewable energies. Fig. 6 shows the technical potentials of solar energy (upper left), wind energy (upper right), hydroelectricity (lower left) and bioenergy (lower right). Furthermore, it presents the expected average total consumption and average, minimum and maximum demand for energy from the different renewable sources in 2050 for the development scenarios in line with 1.5  C- and 2.0  C-global warming limits (all values that are defined as electric potential are transformed to input equivalent energy for 2050 according to 19). Additionally, Table 1 lists the values for the technical potential. It is clearly visible that solar energy (upper left) shows by far the largest potential. Since the earth receives roughly 5.5 million EJ of energy from the sun annually (i.e., 173,000 TW of continuous power31), solar energy can access a vast energy resource resulting in high technical potentials. However, the quantified values differ strongly. The mean value of the technical potential over all

14

Energy, Environment, and Resources | Renewable Energies

Fig. 6 Technical potentials of solar energy (upper left),23–27 wind energy (upper right),23,24,26–28 hydroelectricity (lower left)23,24,26,28,29 and bioenergy (lower right)23,27,30 as well as the average total consumption and average, minimum and maximum demand for energy from the different renewable sources in 2050 for the development scenarios in line with 1.5  C- and 2.0  C-global warming limits14–18 (all values that are defined as electric potential are transformed to input equivalent energy for 2050 according to 19).

Table 1

Technical potentials for solar energy, wind energy, hydroelectricity and bioenergy.

Technology

Energy

Stated value in EJ a−1

Input equivalenta in EJ a−1

Source

Solar energyb

Electricity

Wind energyc

Electricity

Hydroelectricity

Electricity

Bioenergyd

Electricity Heat

1125.9–1689.0 2206.8 14,778.0 10,722.2 1693.0 1500.0 394.2–506.6 219.6 401.0 1261.4–2680.6 165.0–3000.0 20.4–109.7 47.7–50.0 50.5 50.0 52.0 212.4 102.0–184.1 194.0–567.0

2502.0–3753.3 4904.0 32,840.0 23,827.2 3762.2 3333.3 876.0–969.8 488.0 891.1 2803.2–5956.8 366.7–6666.7 45.4–243.8 105.6–111.1 112.1 111.1 115.6 472.0 102.0–184.1 194.0–567.0

26 25 27 24 23 28 26 27 23 24 28 29 26 24 23 27 27 30 23

a Electricity is transferred to an input energy equivalent that an average power plant would need in 2050 for supplying the same amount of electric energy.19 b Potentials mainly include solar photovoltaics. c Potentials express the sum of on- and offshore wind if stated separately. d Potentials express the sum of residue and energy crop biomass if separately stated.

evaluated sources amounts to roughly 10,700 EJ a−1, while the median lies at 3760 EJ a−1. This difference largely stems from a single rather high estimation of the technical potential of solar energy27 that exceeds 30,000 EJ a−1. But also the range between the 25th and 75th percentile of values for the technical potential spans from roughly 3000 to about 15,000 EJ a−1 (i.e., roughly a factor of five), indicating a broad range of conclusions regarding technical availability of solar energy on a global scale. However, in comparison to the expected expansion needs for solar energy in the scenarios in line with 1.5 and 2.0  C warming limits and even the average expected total energy consumption in 2050 (i.e., roughly 750 EJ a−1), solar energy potentials clearly dwarf the

Energy, Environment, and Resources | Renewable Energies

15

demand for solar energy expansion even for the most conservative estimations. Theoretically, it would even be possible to supply all energy demand in 2050 solely from solar energy.23–27 As wind is caused by temperature differences throughout the atmosphere due to solar irradiation and is, thus, a consequence of it, wind energy utilization can generally access a smaller energy resource than solar energy, which directly uses solar irradiation. This translates to wind energy potentials being lower than solar energy potentials. The mean and median of the assessed values for the technical potential of wind energy amount to roughly 2350 and 930 EJ a−1, respectively. Therefore, wind energy potentials are roughly 75% lower than those of solar energy. Again, single sources assign rather high values to wind energy potential resulting in a range between the 25th and 75th percentile of roughly 780 to 3590 EJ a−1. Thus, quantifications of the technical wind energy potential also come to strongly differing conclusions, depending on assumptions on technical performance and land availability. However, as it is the case for solar energy, technical wind energy potentials clearly dwarf the expansion needs for wind power until 2050. Nevertheless, in comparison to the average expected total energy consumption in 2050 wind energy potentials seem more limited than solar energy. Some sources even state values (lower estimates) that would fall short of expected consumption levels.27,28 Thus, supplying all of the global energy demand solely from wind energy could be highly challenging even in theory. However, even the most pessimistic estimates exceed the needed wind energy expansion to realize an energy system development that would be in line with 1.5  C- and 2.0  C-global warming limits by more than 100%. Therefore, the stated technical potentials seem to be sufficient also if economic and market aspects are factored in. Hydroelectricity (lower left) shows the lowest total potentials. With a mean technical potential of 120 EJ a−1 for all evaluated sources and a median of 110 EJ a−1, the potentials are much less scattered. This is also visible from the box plot, which shows a small range between the 25th and 75th percentile from roughly 108.3 to about 113.8 EJ a−1. The quantified potentials commonly range from 105 to 115 EJ a−1. The outliers come from one estimation giving lower and upper boundaries of 45.4 and 243.8 EJ a−1, respectively (recalculated as input equivalent).29 Overall, the technical potential for hydroelectricity is far lower than the expected total consumption level in 2050. Thus, a global energy supply largely based on hydroelectricity is technically unfeasible. However, in comparison to the expected expansion of hydroelectricity in the development scenarios that would be in line with 1.5  C- and 2.0  C-global warming limits (i.e., an average of roughly 63 EJ a−1), the technical potentials will likely not be a limiting factor. However, if economic and market aspects are factored in a sufficient realization of new hydroelectricity generation capacities might become more challenging. For example, when only considering hydroelectricity that would be available up to a certain maximum cost threshold, this economic potential generally ends up being much smaller than the technical potential (e.g., economic potentials being just a little higher than 50% of technical potentials29). If also environmental and other sustainability criterion are included, potentials decline further.29 Therefore, it is unclear, whether hydroelectricity development can lead to a sufficient expansion to meet the necessary capacities. However, the globally available technical potentials are not a limiting factor for achieving the needed expansion of hydroelectricity. Estimations of technical bioenergy potentials (lower right) show a mean value of roughly 300 and a median of about 190 EJ a−1. Overall, the potentials exhibit a much stronger variation in comparison to hydroelectricity technical potentials with the 25th and 75th percentile ranging from 184 to 472 EJ a−1. Therefore, the expansion of bioenergy according to the development scenarios that would be in line with 1.5  C- and 2.0  C-global warming limits (i.e., an average of roughly 96 EJ a−1) likely is technically feasible. However, the technical potentials are not drastically higher than the needed expansion. Therefore, an according development of bioenergy can be challenging – especially if economic and market constraints are taken into account. Furthermore, the quantified potentials are generally stated with higher uncertainty. It is therefore unclear, whether bioenergy potentials are sufficient to enable an increase to roughly 100 EJ a−1. If sustainability criteria and possible competition for the utilization of biomass (e.g., use of biomass as a feedstock for industry) are considered, this becomes even more uncertain. Nevertheless, based on the stated technical potentials, bioenergy would be able to contribute the necessary amount to the overall energy supply. Judging from the technical potentials of solar energy, wind energy, hydroelectricity, and bioenergy it seems that renewable energy potentials are unlikely to become a limiting factor for their further development in the foreseeable future. Especially, the technical potentials of solar and wind energy are high enough to be considered as rather unrestricted in view of global energy needs. For hydroelectricity and bioenergy, the technical limits are much more in reach. Especially when considering cost for the exploitation of these resources (as well as sustainability criteria), it is not unlikely that available economic potentials might restrict their development on a global scale in the coming decades. However, besides the discussed technologies for energy supply based on renewable energies, there are additional technologies that could contribute to supplying global energy demands sustainably and in line with 1.5  C- and 2.0  C-global warming limits in the future. One example is geothermal energy, which is already intensively used in some regions (e.g., Iceland). Estimations of the global technical potential of geothermal energy spread over large ranges from roughly 14 to 624 EJ a−1 depending on the exploitability of known resources and the availability of additional ones.32 If the true technical potentials lie somewhere in the middle of this range, geothermal energy could contribute substantially to a renewable energy supply globally. Harvesting marine or ocean energy (i.e., energy from waves, tides, and currents) can be another option for supplying energy based on renewable resources. Estimates of the available potential are subject to uncertainties but are expected26 to be well over 100 EJ a−1 (input equivalent). Therefore, marine energy also shows a significant potential to support a sustainable energy supply system. Overall, the discussed technical potentials appear to be more than sufficient for a comprehensive expansion of renewable energies not to be considered impossible. Adding up the mean and median values of the technical potential of only solar energy, wind energy, hydroelectricity, and bioenergy – thus disregarding all potential further renewable resources and technologies – the

16

Energy, Environment, and Resources | Renewable Energies

Fig. 7 Recent development of total energy consumption as well as consumption from renewable energy sources, average, minimum and maximum values for their development in the development scenarios in line with 1.5  C- and 2.0  C-global warming limits14–18 (left axis) as well as minimum, average and maximum cumulated potential of solar energy, wind energy, hydroelectricity and bioenergy in 2050 (right axis).23–30

cumulated technical potential amounts to 13,470 (mean value) and 4990 EJ a−1 (median value), respectively. This is at least 6.6 times the expected total global primary energy demand in 2050. Although the estimates of the potentials are subject to considerable uncertainties, even for the technologies with the lowest remaining potential (i.e., hydroelectricity), a doubling of the currently provided amounts of energy seems feasible from a technical perspective. For solar and wind energy in particular, there are no discernible limits to a comprehensive expansion from today’s perspective – at least not due to limitations of the technical potential. This is finally underlined by Fig. 7, which shows the recent development of total energy consumption and consumption from renewable sources of energie as well as average, minimum and maximum values for their possible expansion in the development scenarios in line with 1.5  C- and 2.0  C-global warming limits. Furthermore, the cumulated minimum, mean, and maximum estimated technical potential for solar energy, wind energy, hydroelectricity, and bioenergy in 2050 are shown. From this depiction it can be concluded, that the cumulated maximum estimated technical potential roughly amounts to 50 times the total global primary energy consumption in 2050. The cumulated mean of the estimated global technical potential of the discussed renewable sources of energy is still almost 20 times as large as the global energy demand in 2050. Finally, even the cumulated minimum of expected technical potentials amounts to more than four times the expected average consumption level in 2050. Looking at the dimensions to which the energy supply on the basis of renewable energies would need to be expanded in order to not exceed a 1.5  C- and 2.0  C limit of global warming, these comparisons would result in even larger ratios between available technical potentials and the necessary development. So even in the most conservative respectively pessimistic estimations, the technical potentials are presumably by far sufficient to not hinder the further expansion of the utilization of renewable sources of energie. Even though it remains unclear, whether the true development of global energy consumption and especially the development of renewable energies will follow a path comparable to the ones depicted in Fig. 7, it can be concluded, that limited potentials are not an issue for this development from today’s perspective. Thus, technical potentials do not stand in the way of a mostly renewable energy supply by the mid of the century. Therefore, the primary focus should now lie on the installation and development of technical systems that utilize the existing vast potential and, thus, turn potentials to a practical reality.

4

Summary and conclusion

Ongoing global warming is leading to noticeable changes in the earth’s climate. To mitigate the effects of this climate change on global civilizations, it is imperative that action is taken to mitigate and halt global warming. In this context, the energy sector plays a vital role. Energy consumption is responsible for roughly three quarters of total greenhouse gas (GHG) emission globally. Furthermore, the energy sector is by far the largest emitter of carbon dioxide (CO2) from fossil sources, causing more than 85% of anthropogenic emissions of CO2 – the most important greenhouse gas. Therefore, in order to mitigate climate change, the total emissions of the energy sector have to be reduced drastically and urgently. However, total energy consumption levels will most likely increase further substantially in relative and absolute terms, as the global population as well as the per capita energy consumption will likely keep on growing significantly. Until the mid of the 21st century, total energy consumption might well rise beyond 750 EJ a−1 – this would equate to an increase of 25% compared to today’s global consumption levels. To ensure that this

Energy, Environment, and Resources | Renewable Energies

17

increase in demand is not accompanied by a corresponding increase in greenhouse gas (GHG) and CO2 emissions, respectively, energy has to be supplied more sustainably on a global scale. The utilization of renewable energies – like solar energy, wind energy, hydroelectricity and bioenergy – is an option to enable such a sustainable energy system. However, to make a large and substantial contribution to meeting total global primary energy demand, renewable energies would have to be able to supply hundreds of EJ a−1 of energy. Therefore, the technical potential of renewable energies to contribute substantially to the global energy supply is of great interest. Technical potentials of renewable energies have been quantified many times, often producing widely varying outcomes. This results in a wide range of values for the technical potentials of the particular renewable source of energy as well as the respective energy technology. However, despite the fact that the quantified values for their potentials vary strongly, solar and wind energy seem rather unrestricted from a technical perspective. Only judging from the technical potential, solar energy could be expanded to supply several thousands of EJ a−1 (i.e., roughly 3000 to about 15,000 EJ a−1; 25th and 75th percentile of the assessed sources). Wind energy could supply energy in the lower four digit range (i.e., roughly 780 to about 3590 EJ a−1; 25th and 75th percentile of the assessed sources). Therefore, compared to total consumption levels in 2050 and the necessary expansion for a sustainable energy supply, solar and wind energy will likely not face any restrictions due to insufficient technical potentials. For hydroelectricity and bioenergy, the picture looks slightly different. Technical potentials for bioenergy (i.e., roughly 184 to about 472 EJ a−1; 25th and 75th percentile of the assessed sources) and especially hydroelectricity (roughly 108 to about 114 EJ a−1; 25th and 75th percentile of the assessed sources) are far more limited than it is the case for solar and wind energy. Thus, solely supplying global energy demands based on bioenergy and hydroelectricity would be infeasible. However, they still show significant remaining potentials, allowing for broad further expansion. Therefore, hydroelectricity and bioenergy can also contribute to a sustainable energy future. If additional technologies for exploiting renewable energy sources are considered (e.g., geothermal energy, ocean respectively marine energy), the cumulated technical potential for renewable energies increases even further. However, to reach a conclusive statement on the feasibility of a global energy supply from renewables, additional aspects (e.g., economics, market aspects, environmental constraints, regulatory constraints) would have to be taken into account, which would likely severely reduce the available potentials. However, since technical potentials by far exceed the necessary expansion of renewable energies, they still appear to be highly sufficient for generally enabling a sustainable energy supply. Therefore, in terms of technical potential alone, it appears feasible to cover large parts of the world’s energy demand solely from renewable energies. This makes a sustainable energy supply generally possible, which would allow to mitigate the climate impact of the energy sector in the future.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

World Meteorological Organization. WMO Provisional State of the Global Climate. n.d. Ritchie, H.; Roser, M.; Rosado, P. CO₂ and Greenhouse Gas Emissions; 2020. https://ourworldindata.org (Accessed 11 November 2022). bp. Statistical Review of World Energy 2022. https://www.bp.com/ n.d. (accessed 2022-11-11). United Nations. Day of Eight Billion. https://www.un.org/ n.d. (accessed 2022-11-17). United Nations Department of Economic and Social Affairs, Population Division. World Population Prospect 2022: Summary of Results. n.d. UN DESA/POP/2022/TR/NO. 3. Enerdata World Energy & Climate Statistics - Yearbook 2022: Total energy consumption. https://yearbook.enerdata.net/ n.d.(accessed 2022-11-17). bp Statistical Review of World Energy 2022: Database. http://www.bp.com/ n.d. (accessed 2022-11-11). Faisal, F.; Tursoy, T.; Ercantan, O. The Relationship between Energy Consumption and Economic Growth: Evidence from Non-Granger Causality Test. Procedia Computer Science 2017, 120, 671–675. https://doi.org/10.1016/j.procs.2017.11.294. Our World in Data. GDP Per Capita Vs. Energy Use; 2015. https://ourworldindata.org/. Ritchie, H. A Number of Countries Have Decoupled Economic Growth From Energy Use, Even If We Take Offshored Production into Account; 2021. https://ourworldindata.org/ (Accessed 17 November 2022). Carbon Dioxide Information Analysis Center Fossil-Fuel CO2 Emissions by Nation n.d. Gapminder Foundation. Based on Free Material from gapminder.org, CC-BY LICENSE. www.gapminder.org, n.d. (accessed 2022-11-10). Global Carbon Atlas. Global Carbon Atlas: A Platform to Explore and Visualize the most Up-To-Date Data on Carbon Fluxes Resulting from Human Activities and Natural Processes. www.globalcarbonatlas.org n.d. (accessed 2022-11-11). bp Energy Outlook: 2022 Edition. https://www.bp.com/ n.d. (accessed 2022-11-11). IEA. World Energy Outlook 2022; 2022. https://www.iea.org/ (Accessed 11 November 2022). Shell International B.V. The Energy Transformation Scenarios. https://www.shell.com/ n.d. (accessed 2022-11-11). bp. bp Energy Outlook: Database. https://www.bp.com/ n.d. (accessed 2022-11-11). IEA. World Energy Outlook 2022: Database. https://www.iea.org/ n.d. (accessed 2022-11-11). bp. Updated Methodology for Converting Non-Fossil Electricity Generation to Primary Energy. https://www.bp.com/ n.d. (accessed 2022-11-29). IPCC Summary for Policymakers: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change n.d. IPCC. Global Warming of 1.5 C: An IPCC Special Report on the Impacts of Global Warming of 1.5 C above Pre-Industrial Levels and Related Global Greenhouse Gas Emission Pathways, in the Context of Strengthening the Global Response to the Threat of Climate Change, Sustainable Development, and Efforts to Eradicate Poverty; 2018. Lopez, A.; Roberts, B.; Heimiller, D.; Blair, N.; Porro, G. U.S. Renewable Energy Technical Potentials: A GIS-Based Analysis: Technical Report; 2012. Hoogwijk, M.; Graus, W. Global Potential of Renewable Energy Sources: A Literature assessment: Background Report; 2008. Jacobson, M. Z.; Delucchi, M. A. Providing all Global Energy with Wind, Water, and Solar Power, Part I: Technologies, Energy Resources, Quantities and Areas of Infrastructure, and Materials. Energy Policy 2011, 39 (3), 1154–1169. https://doi.org/10.1016/j.enpol.2010.11.040. Korfiati, A.; Gkonos, C.; Veronesi, F.; Gaki, A.; Grassi, S.; Schenkel, R.; Volkwein, S.; Raubal, M.; Hurni, L. Estimation of the Global Solar Energy Potential and Photovoltaic Cost with the Use of Open Data. International Journal of Sustainable Energy Planning and Management 2016, 9, 17–30. https://doi.org/10.5278/ijsepm.2016.9.3. Krewitt, W.; Nienhaus, K.; Kleßmann, C.; Capone, C.; Stricker, E.; Grauss, W.; Hoogwijk, M.; Supersberger, N.; Winterfeld, U. V.; Samadi, S. Role and Potential of Renewable Energy and Energy Efficiency for Global Energy Supply. Climate Change 2009, 18/2009, ISSN 1862-4359.

18

Energy, Environment, and Resources | Renewable Energies

27. de Vries, B. J.; van Vuuren, D. P.; Hoogwijk, M. M. Renewable Energy Sources: Their Global Potential for the First-Half of the 21st Century at a Global Level: An Integrated Approach. Energy Policy 2007, 35 (4), 2590–2610. https://doi.org/10.1016/j.enpol.2006.09.002. 28. IPCC, Ed. Renewable Energy Sources and Climate Change Mitigation: Special Report of the Intergovernmental Panel on Climate Change; Cambridge University Press, 2011. 29. Gernaat, D. E. H. J.; Bogaart, P. W.; van Vuuren, D. P.; Biemans, H.; Niessink, R. High-Resolution Assessment of Global Technical and Economic Hydropower Potential. Nat Energy 2017, 2 (10), 821–828. https://doi.org/10.1038/s41560-017-0006-y. 30. Seidenberger, T.; Thrän, D.; Offermann, R.; Seyfert, U.; Buchhorn, M.; Zeddies, J. Global Biomass Potentials. Investigation and assessment of data, remote sensing in biomass potential research, and country-specific energy crop potentials. In Energy [R]evolution - A Sustainable World Energy Outlook; 3; 2010; pp. 166–168. 31. Department of Energy. Top 6 Things You Didn’t Know About Solar Energy. https://www.energy.gov/ n.d. (accessed 2022-11-30). 32. Stefansson, V. World geothermal assessment. In Geothermal Energy: The Domestic, Renewable, Green Option: Proceedings of the World Geothermal Congress 2005, Antalya, Turkey, 24–29 April 2005; WGC 2005; International Geothermal Association, 2005.

Energy, Environment, and Resources | Energy Storage S Koohi-Fayegha and MA Rosenb, aFaculty of Engineering, University of Alberta, Edmonton, AB, Canada; bFaculty of Engineering and Applied Science, Ontario Tech University, Oshawa, ON, Canada © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of L. Jörissen, H. Frey, ENERGY | Energy Storage, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 215–231, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00090-3.

1 2 2.1 2.1.1 2.1.2 2.2 2.3 2.4 2.5 2.5.1 2.5.2 2.5.3 2.6 2.7 3 3.1 3.2 3.3 3.4 3.5 4 5 5.1 5.2 5.3 5.4 6 7 References

Introduction Energy storage technologies Electrochemical energy storage Batteries and supercapacitors Fuel cells Thermal energy storage Thermochemical energy storage Mechanical energy storage—Dynamic Mechanical energy storage—static Compressed air energy storage Liquified air Pumped hydro Magnetic energy storage Chemical energy storage Energy storage applications General applications Energy utilities Renewable energy utilization Buildings and communities Transportation Categorizations of energy storage technologies Comparisons of energy storage technologies Technical performance Storage duration: From short-term to seasonal Economics Advantages and disadvantages Future perspectives and needs Closing remarks

20 21 21 21 25 26 27 28 29 29 29 30 31 31 32 32 34 34 35 37 38 40 40 41 41 43 44 45 45

Abstract An overview and critical review is provided of available energy storage technologies, including electrochemical, battery, thermal, thermochemical, flywheel, compressed air, pumped, magnetic, chemical and hydrogen energy storage. Storage categorizations, comparisons, applications, recent developments and research directions are discussed. Significant performance parameters are described, such as energy density, power density, cycle efficiency, cycle life, charge/discharge characteristics and cost, making different storage technologies suitable for particular applications. Recent research on new energy storage types as well as important advances and developments in energy storage, are also included, as are future perspectives and needs.

Glossary Aquifer thermal energy storage (ATES) A type of thermal energy storage that uses underground aquifers as a storage medium. Compressed air energy storage (CAES) A type of energy storage that uses contained air as a storage medium. Charging is accomplished by compressing air, and discharging by expanding it. Electrical energy storage (EES) Energy storages that store electricity or the potential for generating electricity. Hybrid flow battery (HFB) A kind of flow battery in which the electrolyte contains one or more dissolved electroactive elements. Their chemical energy is converted to electricity when these elements flow through an electrochemical cell. Metal hydride (MH) Compounds containing one or more metal cations and one or more hydride anions. Used to store hydrogen by pressurising, as the metals bind with hydrogen.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00228-0

19

20

Energy, Environment, and Resources | Energy Storage

Phase change material (PCM) A material that can be used for latent thermal energy storage by undergoing phase change. Pumped hydro energy storage (PHES) A type of energy storage that uses water elevation differences for energy storage. Charging is accomplished by moving a water against gravity to a higher elevation, while discharging is accomplished as the water descends in a controlled manner. Photovoltaic (PV) The conversion of light to electricity using semiconducting materials that exhibit the photovoltaic effect. PV panels are used for converting solar energy to electricity. Redox flow battery (RFB) A flow battery that uses reduction and oxidation in the form of an electrochemical cell. Superconducting magnetic bearing (SMB) A type of bearing that supports a load using magnetic levitation with very little losses due to its superconducting nature. Superconducting magnetic energy storage (SMES) A type of thermal energy storage that stores energy in a magnetic field created by the flow of direct current in a superconducting coil. Thermal energy storage (TES) A type of energy storage that allows heat to be stored in a storage medium. Ultracapacitor (UC) A device for energy storage that involves positively and negatively charged ions and a liquid electrolyte to permit energy flow. UCs can store and release electricity faster than batteries. Vanadium redox battery (VRB) A type of flow battery that uses vanadium ions as charge carriers. Zero energy building (ZEB) A building that generates as much energy as it utilizes.

Key points

• • • •

A general review of energy storage types is performed to provide insights on their similarities and differences, potential application and integration opportunities, and needed policy development. Innovative energy storage advances, including new types of energy storage systems and recent developments, are covered throughout. To allow a broad review of relevant topics, most work that focuses on a detailed area of energy storage is not considered in detail. Future perspectives and needs are described and examined.

Abbreviations ATES CAES EES HFB MH O&M PCM PHES PSB PV RFB SMB SMES TES UC VRB ZEB

1

Aquifer thermal energy storage Compressed air energy storage Electrical energy storage Hybrid flow battery Metal hydride Operating and maintenance Phase change material Pumped hydro energy storage Polysulphide bromide Photovoltaic Redox flow battery Superconducting magnetic bearing Superconducting magnetic energy storage Thermal energy storage Ultracapacitor Vanadium redox battery Zero energy building

Introduction

Energy systems play a key role in harvesting energy from various sources and converting it to the energy forms required for applications in various sectors, e.g., utility, industry, building and transportation. Energy sources like fossil fuels can be used to provide energy according to customer demand, i.e., they are readily storable when not required. But other sources such as solar and wind energy need to be harvested when available and stored until needed. Applying energy storage can provide several advantages

Energy, Environment, and Resources | Energy Storage

21

for energy systems, such as permitting increased penetration of renewable energy and better economic performance. Also, energy storage is important to electrical systems, allowing for load leveling and peak shaving, frequency regulation, damping energy oscillations, and improving power quality and reliability. Energy storage systems come in various types and have many applications, and several reviews and overviews of these have been published.1–3 Guney and Tepe4 present a description of energy storage systems with detailed classifications, features, advantages, environmental impacts, and implementation/application possibilities. Aneke and Wang5 provide a detailed analysis of applications and performances of various energy storage technologies. Luo et al.6 provide an overview of various types of electrical energy storage technologies and provide a detailed comparison based on technical and economic data. Scientific and engineering requirements of some storage technologies are reviewed by Hall and Bain,7 who describe the state of technologies in 2008 and anticipated developments for superconducting magnetic energy storage (SMES), flywheel energy storage and electrochemical energy storage. Olabi et al.8 provide a review of energy storage systems focusing on several technical aspects of energy storage systems. The previous reviews are often limited in terms of the types of energy storage covered. For example, some reviews focus only on energy storage types for a given application such as those for utility applications. Other reviews focus only on electrical energy storage systems without reporting thermal energy storage types or hydrogen energy systems and vice versa. It is important that more general reviews covering all energy storage types are performed to provide better insights on their differences, potential integration opportunities, and needed policy development. Furthermore, with the area of energy storage being very broad and numerous articles being published on them every year from technical and economical perspectives, the currency of reviews is particularly important for articles aiming to provide a review on a broad range of topics. In the current chapter, a broader and more recent review of each storage classification type is provided. More than 500 articles on various aspects of energy storage were considered and the most informative ones in terms of novelty of work or extent of scope have been selected and briefly reviewed. Several review articles in the literature provide a more detailed review of a single energy storage topic, such as reviews on thermal energy storage, whereas the current chapter aims to provide a more general review of various energy storage types to compare their characteristics. As a result, several noteworthy articles may not be included due to their high level of detail that does not serve the purpose of the current chapter. This chapter reviews energy storage types, focusing on operating principles and technological factors (Section 2). In addition, a critical analysis of the various energy storage types is provided by reviewing and comparing the applications (Section 3), providing their categorization basis (Section 4), and comparing technical and economic specifications of energy storage technologies (Section 5). Innovative energy storage advances, including new types of energy storage systems and recent developments, are covered throughout Future perspectives and needs are also examined (Section 6) before closing remarks are provided (Section 7). This chapter cites many articles on energy storage, selected based on factors such as level of currency, relevance and importance (as reflected by number of citations and other considerations). The manner in which the various energy storage topics are categorized in this chapter is summarized in Fig. 1.

2

Energy storage technologies

The various types of energy storage can be divided into many categories, and here most energy storage types are categorized as electrochemical and battery energy storage, thermal energy storage, thermochemical energy storage, flywheel energy storage, compressed air energy storage, pumped energy storage, magnetic energy storage, chemical and hydrogen energy storage. Other types of energy storage such as biological energy storage are not focused on in this chapter since they have not been the object of extensive research from a storage point of view. Note that the focus in the following sections is on the various energy storage types; details on applications as well as technical and economical specifications are provided in Sections 3 and 5, respectively.

2.1

Electrochemical energy storage

Electrical energy can be stored electrochemically in batteries and capacitors. Batteries are mature energy storage devices with high energy densities and high voltages. Various types exist including lithium-ion (Li-ion), nickel-cadmium (NiCd), lead–acid (Pb-acid), lead-carbon batteries, as well as sodium-sulfur (NaS), zebra batteries (Na-NiCl2) and flow batteries. Capacitors store and deliver electrical energy, and can be classified as electrostatic capacitors, electrolytic capacitors, and electrochemical capacitors Among these three types, electrochemical capacitors, also called supercapacitors or ultracapacitors (UCs), have the greatest capacitance per unit volume due to having a porous electrode structure and, therefore, a relatively high surface area.

2.1.1 Batteries and supercapacitors Several new electrode materials and electrolytes have been reviewed and suggested to improve the cost, energy density, power density, cycle life, and safety of batteries. Hall and Bain7 provide a review of electrochemical energy storage technologies including flow batteries, Li-ion batteries, sodium-sulfur and the related zebra batteries, nickel-cadmium and the related nickel-metal hydride batteries, lead–acid batteries, and supercapacitors. Some of these electrochemical energy storage technologies are also reviewed by Baker,9 while performance information for supercapacitors and Li-ion batteries are provided by Hou et al..10 Nitta et al.11 review fundamental properties, opportunities, challenges, and recent progress of anode and cathode material research for lithium batteries. As strategies to improve the performance of Li-ion batteries, Nitta et al. suggest (a) reducing dimensions of active materials,

22

Energy, Environment, and Resources | Energy Storage

Fig. 1 Categorization of energy storage topics in the current chapter.

(b) formation of composites, (c) doping and functionalization, (d) tuning particle morphology, (e) formation of coatings or shells around active materials, and (f ) modification of the electrolyte. Among the various battery types, lithium batteries are playing an increasingly important role in electrical energy storage because of their high specific energy (energy per unit weight) and energy density (energy per unit volume). A charged Li-air battery provides an energy source for electric vehicles rivalling that of gasoline in terms of usable energy density (Fig. 2). The fundamental battery chemistry during discharge is the electrochemical oxidation of lithium metal at the anode and the reduction of oxygen from air at the cathode. Before Li–air batteries can achieve high performance and become commercially viable, numerous technical challenges need to be addressed: designing cathode structures, optimizing electrolyte compositions and elucidating the complex chemical reactions during charge and discharge.12–14 The most significant developments and the main limiting factors for Li–air batteries, as well as the current understanding of their chemistry, have been summarized in the literature.12,13 The Li-ion battery is a type of lithium battery that uses an intercalated lithium compound as an electrode material. Bruce et al.15 examine the energy that can be stored in Li–air (based on aqueous or non-aqueous electrolytes) and lithium–sulfur (Li-S) batteries and compare it with that for

Energy, Environment, and Resources | Energy Storage

23

Fig. 2 Energy densities for various types of rechargeable batteries compared to gasoline. Note: NiCd: Nickel-cadmium; Ni-MH: Nickel-metal hydride; Li-ion: Lithium-ion; Zn-Air: Zinc-air; LiS: Lithium-Sulphur; Li-Air: Lithium-air. Adapted from Christen, T.; Carlen, M. W.; Theory of Ragone Plots. J. Power Sources 2000; 91(2), 210–6.

Li-ion batteries, and discuss cell operation and development challenges. They suggest that both batteries offer improved specific energy compared to Li-ion batteries and could also be more cost-competitive than Li-ion batteries. They suggest that more research on the fundamental chemistry involved in the Li–air and Li-S cells is needed before they can reach markets. It should be noted that the energy efficiency of Li-air batteries is lower than 70%. Thackeray et al.16 provide a historical overview of Li-ion batteries, the status of current ones, and a description of advances in lithium-air batteries. The performance of Li-ion batteries is affected by the solid electrolyte interphase, a protecting layer formed on the negative electrode of the battery due to the reaction of the electrolyte with the anode material (carbon) during the first charge-discharge cycle. Factors that affect the solid electrolyte interphase and how they impact battery performance are discussed by Verma et al..17 Li et al.18 review use of silicon active materials incorporated with graphite frameworks for use as anodes in Lithium batteries. They note improved storage capacity and prolonged cycling stability of this integration. Janek and Zeier19 suggest that the energy density of conventional Li-ion batteries will soon reach a physicochemical limit and solid-state batteries that use Li metal anodes and solid electrolytes instead of liquid ones could meet the need for higher energy and power densities, although technical issues such as slow kinetics limit commercialization of solid-state systems. Due to the widespread availability and low price of sodium, and the similarity of Li and Na insertion chemistries, Na-ion batteries could become the future low-cost batteries for smart electric grids that integrate renewable energy sources but also for low-cost electric vehicles with limited range. Much work has to be done in the Na-ion field to catch up with Li-ion technology. Cathodic and anodic materials must be optimized, and new electrolytes will likely be the key for Na-ion success. Palomares et al.20 and Zhao et al.21 describe Na-ion battery materials, to provide a broad view of already explored systems and a platform for future research. Among the Na insertion cathodic materials, the authors suggest phosphates and fluorophosphates as promising options, but only after structural characteristics and Na insertion-extraction mechanisms are further studied and well understood. They also suggest a number of electrolytes as promising for Na-ion batteries, including sodium b”-alumina solid electrolyte and gel polymer electrolytes. Zhao et al.22 focus on recycling aspects of sodium-ion batteries and suggest that design for recyclability rather than only performance is a necessary step to promote wide adoption of sodium-ion batteries. However, they note challenges in developing recycling methodologies that are related to the uncertainty of materials used in Sodium-Ion batteries at their current stage of development. Watanabe et al.23 review various application of ionic liquid electrolytes, i.e., liquids consisting entirely of ions, and focus on their use as electrolyte materials for Li/Na ion batteries, Li-S batteries, and Li-air batteries. They focus on the unique properties of ionic liquids such as non-volatility, high thermal stability, and high ionic conductivity and suggest that they could provide solutions to some of the current barriers to further development of batteries, although ionic liquid electrolytes are currently still quite expensive. Ru et al.24 suggest aluminum-ion batteries as the most suitable candidate to replace Li-ion batteries due to their abundant resources, cost-effectiveness and eco-friendliness as well as their potential for fast charging speed and long life. Such advantages could make them suitable to support power generation from renewable energy sources. However, their energy density, cell capacity and cycle stability may still need to be improved before commercialization. Ru et al. review development challenges for such batteries, such as selection of the most suitable electrolyte and positive electrode materials; these challenges result in the batteries remaining in the conceptual stage. The authors suggest acidic AlCl3-based electrolytes and transition metal oxides, metal sulfides, and carbonaceous materials for positive electrodes. Details on battery technologies and their thermal management is provided by Rosen and Farsi.25 Zinc-based rechargeable batteries are promising battery options due to their low cost, environmental friendliness, safety, and good stability in aqueous electrolytes (alkaline, neutral, and weakly acidic).Wan et al.26 review the fundamentals of zinc energy storage including aqueous electrolytes (alkaline, neutral, and weakly acidic), various electrolytes for cathodes and anodes, and

24

Energy, Environment, and Resources | Energy Storage

Fig. 3 Efficiency/lifetime properties of some energy storage technologies. Note: SMES: superconducting magnetic energy storage; Li-ion: Lithium-ion battery; NaS: Sodium-Sulphur battery; Batt.: Flow battery; NiCd: Nickel-Cadmium battery. Reproduced from Hall, P. J.; Bain, E. J. Energy-Storage Technologies and Electricity Generation. Energy Policy 2008, 36, 4352–5.

electrolyte selection and optimization, partly to solve the cycle stability of zinc. Their energy density, however, is well below that of Li-ion batteries. Exploring materials with higher specific capacities and electrolytes with wider voltage ranges (water-in-salt electrolyte and the hybrid electrolyte battery using ion exchange membranes) seem to be effective approaches. Electrochemical capacitors have high storage efficiencies (>95%) and can be cycled hundreds of thousands of times without loss of energy storage capacity (Fig. 3) as there are no chemical changes in the active masses (no Faraday current flow). Energy efficiency for energy storage systems is defined as the ratio between energy delivery and input. The long life cycle of electrochemical capacitors is difficult to measure directly. Therefore, capacitance retention rate is used to estimate indirectly the cycle life by measuring and comparing the capacitance after a given number of cycles with that of the first cycle.27 Although their efficiency and life cycle are very high, electrochemical capacitors are susceptible to self-discharge (discussed in detail by Shang et al.27,28), and their operating voltages cannot exceed the potential at which the electrolyte undergoes chemical reactions. For high-voltage applications, they can be used in combination with batteries. Much research and development is focused on these energy storage options and their commercialization. Enhancing the kinetics of ion and electron transport within the electrochemical capacitor electrodes and increasing the rate of charge transfer at the interface of the electrode and the electrolyte help increase the storage capacity of electrochemical capacitors. They currently store 1–2 orders of magnitude less energy compared with batteries.23 A recent development in electrochemical capacitor energy storage systems is the use of nanoscale research for improving energy and power densities. Kötz and Carlen29 review fundamental principles, performance measures, characteristics, and present and future applications of electrochemical capacitors. Also, Lu et al.30 examine recent progress in energy storage mechanisms and supercapacitor prototypes, the impacts of nanoscale research on the development of electrochemical capacitors in terms of improved capacitive performance for electrode materials, and significant advances in electrode and device configurations. Electrochemical capacitors are classified according to the charge storage mechanism and the electrode materials used: electrochemical double-layer capacitors, pseudocapacitors and a combination of the two types. In electrochemical double-layer capacitors, the electrode material rapidly attracts solvated ions in the electrolyte which creates a double-layer acting as two capacitors, connected in series by the electrolyte, that remain charged after the circuit is opened. Since double-layer charge storage is a surface process, the electrochemically active surface area of the electrode greatly influences cell capacitance. Materials such as carbon, metal oxides, conducting polymers, hybrid and conducting polymers are used for the electrode. Various aspects of electrochemical double-layer capacitor technology including their historical background, classification, construction, modelling, testing, and voltage balancing are discussed by Sharma and Bhatti.31 They suggest that manufacturing tolerances, the temperature gradient in the system, and cell aging are affected by unequal capacitance that is often observed within the cell series in double-layer capacitors. Voltage equalization circuits have to be employed to balance the voltage among cells. Several strategies to design the architecture of micro-supercapacitors are reviewed by Qi et al..32 Pseudocapacitors operate based on a Faradic charge transfer process on or near the electrode surface in which metal oxides transition. Electrically conducting polymers are often used as electrochemically active materials. Batteries and supercapacitors are often compared for various storage applications. Batteries can store up to 30 times more charge per unit mass than supercapacitors. This high energy density is achieved by storing charge in the bulk of a material. However, supercapacitors can deliver up to thousands of times the power of a battery of the same mass as they only store energy by surface

Energy, Environment, and Resources | Energy Storage

25

adsorption reactions of charged species on an electrode material. Electrochemical capacitors can be cycled more than batteries. The redox reactions in batteries usually produce volume changes that limit energy storage cycles in batteries. Batteries and supercapacitors are further compared by Miller and Simon.33 Lukatskaya et al.34 suggest that the utilization of multi-electron chemistries of both electrolyte and electrode materials can increase the amount of energy stored. They expect hybrid devices that combine the useful features of metal-ion batteries and electrochemical capacitors to provide the improved performance that is needed to meet future demands for electrical energy storage. The energy of batteries with electrodes with solid active masses is limited. If the active masses are, however, liquid, the energy can be significantly increased because the liquid active mass can be permanently fed into the cell.. The flow battery, another type of electrochemical energy storage, uses liquid active masses (mostly salts in an aqueous solution). Flow batteries consist of two electrolyte reservoirs (that determine primarily the energy) from which the electrolytes are circulated through an electrochemical cell comprising a cathode, an anode and a membrane separator (that determines the power). The energy density of such systems is mainly dependent on the stored electrolyte volume and is independent of the size and design of the electrochemical cell, which defines power density, i.e., energy and power can be selected independently of each other by variation of the electrolyte reservoir and the electrochemical cell size which is particularly useful for long-term storage, where high energy is required at a comparatively low power. The redox flow battery is suitable for utility-scale renewable energy storage applications. The main flow battery designs are polysulfide bromide (PSB), vanadium redox (VRB) and zinc bromide (ZnBr). Since flow battery operation involves pump systems and flow control with external storage, its operation has increased capital and operating costs in comparison to batteries. Materials issues are a significant cause of the high costs of flow batteries, particularly those using redox-active metals (especially vanadium) and precious metal electrocatalysts. A class of energy storage materials that exploits the favorable chemical and electrochemical properties of a family of molecules known as quinones are described by Huskinson et al..35 This is a metal-free flow battery based on the redox chemistry that undergoes extremely rapid and reversible two-electron two-proton reduction on a glassy carbon electrode in sulfuric acid. Cycling of this quinone–bromide flow battery demonstrates a greater than 99% storage capacity retention per cycle. This flow battery may be able to provide large electrical energy storage at a greatly reduced cost. Increasing the energy and power density of flow batteries is another challenge associated with the development of flow batteries. Developing monolithic electrodes with increased specific and volumetric surface areas, increasing electrode wetting by electrolytes and creating an open pore structure to allow increased mass transfer are some research topics for flow battery design aimed at overcoming some of these challenges. Studies on various redox flow battery (RFB) technologies focus on addressing issues regarding cell design, including cell-level components of electrolytes, electrodes, and membranes, and chemistry for both aqueous and non-aqueous systems.36–38 Wang et al.39 highlight the importance of advancing the understanding of the complex charge transfer and redox reaction kinetics on the electrode surface, transport in membranes, and fluid mechanics through the electrode. In another type of battery, the hybrid flow battery (HFB), features of conventional batteries and redox flow batteries are combined. One of the electrochemically active elements is stored within the electrochemical cell while the other is dissolved in the liquid electrolytes held in a tank. Novel redox flow battery concepts have been introduced including a solid oxide electrochemical cell integrated with a redox-cycle unit,40 a zinc hybrid-flow battery with a stable potential window of up to 2 V,41 hybrid membranes for VRB42 and a trapezoid-shaped flow battery.43 Such concepts introduce improvements in various aspects such as energy capacity, power density, cost, efficiency, self discharge time, electrolyte utilization, membrane structure stability against strong acidic and oxidizing conditions, utilization of non-toxic material, and utilization of less expensive heavy metals. To increase the energy capacity, not only can liquid active masses be used, but also gaseous active masses (e.g., hydrogen at the negative electrode and oxygen at positive electrode of a fuel cell). Opposite to the redox-flow cell, the fuel cell is practically nonrechargeable. There exists also the so-called regenerative fuel cell which allows discharging (water generation in the fuel cell mode) and charging (water splitting in the electrolyzer mode) in one cell.44 However, these cells have a relatively low performance because they use a “compromise” catalyzer (e.g., one that enables oxygen reduction and oxygen generation). Higher performance is achieved when catalysts are used that have been specially optimized for reduction or oxidation. Oxygen reduction and oxygen generation then take place in two different cells, i.e., in the fuel cell and the electrolysis cell. More about hydrogen technologies and fuel cells is provided in Section 2.7.

2.1.2 Fuel cells Fuel cells are low power density devices like batteries that convert chemical energy to electricity. They exhibit energy efficiencies of approximately 70–80%, while some power plants (e.g., combined cycle units) can achieve efficiencies as high as 60%. Fuel cells use oxygen (mostly oxygen from air) and a fuel such as hydrogen. They can be combined with supercapacitors and batteries to improve their power densities. Research to exploit unique features of graphene to produce supported catalysts with enhanced electrocatalytic activity, increased durability, and high performance electrode architectures in fuel cells is discussed by Hou et al..10 Panahi et al.12 analyze technical and economical aspects of using metal hydride storage coupled with fuel cells as a suitable alternative for batteries for local or small-scale applications. Califano et al.45 study reversible solid oxide cells (providing electric power in fuel cell mode or hydrogen gas in electrolyser mode) for use in a renewable energy system based microgrid that is equipped with hydrogen tanks and thermal energy storage systems. They show that the reversible solid oxide cells offset the high costs of energy storage technologies by making the most of the energy provided by the hydrogen energy system, thermal energy storage and renewable energy systems.

26 2.2

Energy, Environment, and Resources | Energy Storage Thermal energy storage

Thermal energy storage refers to storage of heat or “cold” in a storage medium. Thermal storage systems typically consist of a storage medium and equipment for heat injection and extraction to/from the medium. The storage medium can be a naturally occurring structure or region (e.g., ground) or it can be artificially made using a container that prevents heat loss or gain from the surroundings (water tanks). There are three main thermal energy storage (TES) modes: sensible, latent and thermochemical. Traditionally, heat storage has been in the form of sensible heat, raising the temperature of a medium. Examples of such energy storage include hot water storage (hydro-accumulation), underground thermal energy storage (aquifer, borehole, cavern, ducts in soil, pit),46 and rock filled storage (rock, pebble, gravel). Latent heat storage is a developing technology that involves changing the phase of a storage material, often between solid and liquid phases although solid-gas, liquid-gas and solid-solid phase changes are also available. Latent heat storage has attracted considerable attention recently, primarily due to the isothermal nature of the phase-change process, and its lower weight per unit of storage capacity and compactness. Its improved thermal properties compared to sensible heat storage materials, such as stable phase-change temperature and a high latent heat, are also factors that contribute to its emergence. Typical phase change materials (PCMs) used as the storage media include paraffin waxes, esters, fatty acids and salt hydrates, eutectic salts, and water.9 PCMs are classified in Table 1. A detailed analysis and comparison of thermal energy storage systems is provided by Dincer and Rosen.45,47 Similar to other energy storage types, thermal energy is stored when the source of thermal energy does not provide energy at a continuous rate and/or a fixed cost. The fluctuations in thermal energy supply can occur seasonally or in shorter time periods. In seasonal energy storage, a larger energy storage system is required that is able to retain heat for its use after several months. An example is a ground heat storage system coupled to a building to store the heat that is removed from the building in the summer in the ground and use it in cooler seasons when heating is needed in the building. A similar concept can be applied by storing solar thermal energy over the summer for use in the winter. Short-term energy storage systems often have smaller capacities and retain heat for a period of a few hours to a few days. Such systems can also be used to store solar thermal energy during the day for use during cooler hours when heating is needed. In buildings where electrical heating and/cooling is used during the day, thermal energy storage systems can be used to reduce cost of electricity by storing thermal energy, produced using electricity during low-rate periods, and using it at peak times. Research on thermal energy storage technologies is very extensive and several detailed reviews are available of various aspects of these technologies.48–52 These include TES modes, material thermal properties, formulation and modeling approaches, thermal enhancement techniques for sensible and latent thermal storage systems and design configurations of heat storage facilities. Research on latent heat storage is mostly focused on the development and introduction of new storage media and enhancing thermodynamic properties of the existing ones.53 A recently investigated PCM is fatty acids derived from vegetable and animal oils.54 Nazir et al.55 focus on the application of various phase change materials based on their thermophysical properties such as the melting point, thermal energy storage density and thermal conductivity. They suggest that the application of PCMs in smart thermal grid systems along with intermittent renewable energy sources is promising. Alhuyi Nazari et al.56 review applications of nanotechnology in PCMs and suggest that it can reduce charge and discharge times of the storage units equipped with the PCMs as a result of enhancing heat transfer rates. Ground thermal storage is increasingly common method of sensible thermal energy storage. It often involves using a circulating medium (usually water or air) to extract heat from a building in summer and store it in the ground for winter use. Ground heat exchangers convey the circulating medium to the deeper ground. Models of ground heat exchangers and their applications are reviewed by Florides and Kalogirou.57 Developments in using underground spaces for sensible heat storage include aquifer, borehole, cavern, pit and water tank thermal energy storages. Water tanks are suggested as the most favorable option from the thermodynamic Table 1

Classifications of solid-liquid phase change materials.

Type of phase change material

Operating temperatures ( C)

Compound groups

Examples

Organic

4–150

Paraffin compounds Non-paraffin compounds

Inorganic

8–900

Eutectic

12–600

Salts Salt hydrates Metals Organic-organic Inorganic-inorganic Inorganic-organic

Paraffin waxes Fatty acids Esters Alcohols Glycols

Adapted from Nazir, H.; Batool, M.; Bolivar, Osorio, F. J.; Isaza-Ruiz, M.; Xu, Z.; Vignarooban, K.; Phelan, P.; Inamuddin, Kannan AM. Recent Developments In Phase Change Materials for Energy Storage Applications: A Review. Int. J. Heat and Mass Transfer 2019, 129, 491-523; Sharma, A.; Tyagi, V. V.; Chen, C. R.; Buddhi, D. Review on Thermal Energy Storage With Phase Change Materials and Applications. Renew. Sust. Energy Rev. 2009, 13, 318–45.

Energy, Environment, and Resources | Energy Storage

27

point of view due to the high specific heat of water and their high capacity rates for energy charge and discharge.51,58 Aquifer thermal energy storage (ATES) systems (Fig. 4) use natural water in a saturated and permeable underground layer as the storage medium.46,58 Based on a country-by-country statistical analysis, Feluchaus et al.46 identify market barriers for entering a commercialization level from the perspective of an emerging market phase, a growth phase, and a maturity phase. They suggest technical feasibility, lack of awareness and mistrust in technology some as of the barriers in the emerging market phase. As the technology moves towards the growth phase, high investment costs, policy and legislation, and lack of knowledge among national and local consultants become important barriers. In established energy markets, lower financial savings in smaller applications and a scarcity of subsurface space with an increasing number of implemented systems can be limiting factors. Mahon et al.59 suggest unfamiliarity with the subsurface, presumed limited compatibility with existing energy systems, energy imbalances, and groundwater contamination as some of the limiting factors to wider adoption of aquifer thermal energy storage. Cheng et al.60 propose combining both shallow and deep borehole heat exchangers as a method to provide stability of system operation when using borehole heat exchangers to store heat in the ground. Sliwa et al.61 determine the best borehole heat exchanger design from five basic possibilities studied. Details in borehole heat exchanger modelling is provided by Rosen and Koohi-Fayegh.62

2.3

Thermochemical energy storage

Thermochemical energy storage systems utilize chemical reactions that require or release thermal energy. They have three operating stages: endothermic dissociation, storage of reaction products, and exothermic reaction of the dissociated products (Fig. 5). The final step recreates the initial materials, allowing the process to be repeated. Thermochemical energy storage systems can be classified in various ways, one of which is illustrated in Fig. 6. Thermochemical energy storage systems exhibit higher storage densities than sensible and latent TES systems, making them more compact. This is a beneficial characteristic in applications where storage space is limited or expensive. Since energy losses during storage are smaller for thermochemical energy storage than for sensible or latent

Fig. 4 Aquifer heat storage.

Fig. 5 Processes involved in a thermochemical energy storage cycle.

28

Energy, Environment, and Resources | Energy Storage

Fig. 6 Chemical storage and sorption storage classification (reproduced from66).

TES, thermochemical energy storage has good potential for long-term storage applications.63,64 Thermochemical energy storage systems nonetheless face various challenges before they can achieve efficient operation. Suitable materials or combinations of materials are needed that store energy with low heat loss and release it readily when it is needed. Potential thermochemical storage materials include MgSO4
7H2O (for which the solid reactant is MgSO4 and the working fluid is H2O), Ca(OH)2 (for which the solid reactant is CaO and the working fluid is H2O), CaSO4
2H2O (for which the solid reactant is CaSO4 and the working fluid is H2O), and FeCO3 (for which the solid reactant is FeO and the working fluid is CO2). Lefevbre and Tezel65 suggest composite materials and materials with salt impregnations as suitable for use in thermochemical storage systems. They suggest that ensuring the stability of the salt addition to the adsorbent material for repeated consistent long-term applications is one of the areas in which further research is needed. Haji Abedin and Rosen67 review principles of thermochemical energy storage and recent developments, and compare thermochemical storage systems with other TES systems. Due to the high cost of materials and operating problems, few long-term sorption or thermochemical energy storages are in operation. Several studies describe the physicochemical and thermodynamic properties of materials that are suitable for long-term storage of thermal energy.48,66 The feasibility of a solar-driven thermochemical cycle for dissociating H2O and CO2 using nonstoichiometric ceria (CeO2), yielding CO and H2, respectively, is demonstrated by Chueh et al.68 in terms of materials, reaction rates, cyclability, reactor technology, and energy conversion efficiency. A new technology for energy storage, based on microwave-induced CO2 gasification of carbon materials, is proposed by Bermúdez et al..69 Various carbon materials are tested to examine the amount of energy consumed. Two microwave heating mechanisms, a single-mode oven and a multimode device, are evaluated to test their efficiencies in terms of energy consumption and recovery. The technology has achieved energy efficiencies of 45% at the laboratory scale, and seems improvable so that it becomes competitive with other energy storage technologies.

2.4

Mechanical energy storage—Dynamic

Flywheel energy storage, also known as kinetic energy storage, is a form of mechanical energy storage that is a suitable to achieve the smooth operation of machines and to provide high power and energy density at relatively low specific cost ($/kWhstored). In flywheels, kinetic energy is transferred in and out of the flywheel with an electric machine acting as a motor or generator depending on the charge/discharge mode. Permanent magnet machines are commonly used for flywheels due to their high efficiencies, high power densities, and low rotor losses.70 Other electrical machines such as induction, bearing-less and variable-reluctance machines vary in terms of limitations in application speed, idling losses, vibration, noise and cost. Charging energy is input to the rotating mass of a flywheel and stored as kinetic energy. This stored energy can be released as electric energy on demand. The rotating mass is supported by magnetic bearings which operate in a vacuum to eliminate frictional losses during long-term storage and safety issues.71 The rotor bearing system can be mechanical or magnetic or a hybrid system of both to take advantage of the strengths of each type. The magnetic bearing has no lubrication requirements as it has no frictional loss, but it has complicated control systems and some types require energy to operate. Superconducting magnetic bearings (SMBs) are suitable for high-speed applications, but require energy to operate a cryogenic cooling system. Achieving high rotational velocity, with high power density, in flywheels is desirable since the energy stored is proportional to the square of the velocity but only linearly proportional to the mass. The key enabling technologies are in systems engineering and material science.9 Steel, alloys (e.g., titanium or aluminum alloys) and more recently strong materials such as composites are used for the flywheel rotor and the housing that contains it. Much research is focused on rotor materials and design and speeds of up to 10,000 rpm can now be

Energy, Environment, and Resources | Energy Storage

29

achieved.70 The use of composite materials enables high rotational speeds with greater power densities than chemical batteries. High power density is desirable in vehicles where a large peak power is needed when accelerating and a large power becomes available for storage in a short time when braking. In addition to high energy and power density, high cycle life (many tens of thousands), long operational life, high round-trip efficiency, and low environmental impacts are also attributed to flywheel energy storage systems.72 Compared to batteries and supercapacitors, lower power density, cost, noise, maintenance effort and safety concerns are some of the disadvantages of flywheel energy storage systems.73,74 To improve their power density, Toodeji74 proposes a novel design for a combined system in which supercapacitors are located inside the flywheel rotating disk. This allows exchanging pulsed power as well as storing large amounts of energy.

2.5

Mechanical energy storage—static

In this section, storage devices following a static mechanical energy storage principle, i.e., applying force to compress or displace a medium, are reviewed.

2.5.1 Compressed air energy storage In compressed air energy storage (CAES) systems, air is compressed and stored in an underground cavern or an abandoned mine when excess energy is available. Upon energy demand, this pressurized air can be released to a turbine to generate electricity. Caverns can either be drilled in salt or rock formations, or existing cavities such as aquifer strata can be utilized. Such geological formations do not exist everywhere and large steel tanks that can maintain high pressures are sometimes installed under the ground at a higher system cost. Compressed air energy storage systems can be economically attractive due to their capacity to shift time of energy use, and more recently due to the need for balancing effects of intermittent renewable energy penetration in the grid.75 Another option is to use available energy to store liquefied air at cryogenic temperatures in low-pressure insulated reservoirs. Compared to compressed air, liquid air has lower losses since it can be maintained at moderate pressures. Therefore, it may be a better option than compressed air for long-term storage. Liquid air also is denser and can be stored in smaller reservoirs. For a given amount of liquid air in a tank of 5000 m3, it is shown in a case study that the CAES volume would be approximately 310,000 m3.76 A comparison between compressed air and liquefied air energy storage systems indicates a higher efficiency for the latter.76 Use of liquified air energy storage integrated with multigeneration systems has also been investigated.77 Depending on how heat is handled during compression (i.e., heat discharge) and prior to the expansion stage (i.e., heat intake), there are three types of CAES: isothermal, diabatic and adiabatic. The thermal energy resulting from the charging compression process is dissipated in the diabatic type and needs to be provided during discharging, but is retained in a thermal storage in the adiabatic type for use during discharging. This results in lower efficiencies of diabatic CAES systems as they require a source of heat, often natural gas, to heat the compressed air before it is sent to the turbine for energy discharge. This also makes the economics of using diabatic CAES dependant on fossil fuel prices.78 The Huntorf gas turbine plant in Germany was the first utility-scale CAES plant (1978) and is of the diabatic type, using 1.6 kWh in terms of heating value of natural gas for every 1 kWh of electricity generation. In isothermal CAES systems, the temperature during compression and expansion is maintained close to ambient temperature, making the required power for compression the lowest that is thermodynamically possible and the generation power during expansion the highest. This makes the compression and expansion processes slow which can best be handled using piston machines. Further details on the three CAES types are described by Budt et al..75 Although CAES systems are mature technologies, they are still subject of studies that aim to identify how their current efficiencies (42–55%) can be improved.5,75 The efficiencies of a charging and discharging cycle for several adiabatic CAES configurations are analyzed using energy balances by Hartmann,79 who also examines the main factors affecting CAES efficiency. An accurate dynamic simulation model for diabatic CAES inside caverns, which involves formulating the mass and energy balances inside the storage, is developed by Raju and Khaitan.80 A typical daily operation schedule of the Huntorf gas turbine plant and its CAES is used to validate the model. Further insights are provided by comparing the results obtained using adiabatic and isothermal assumptions inside the cavern.

2.5.2 Liquified air To produce liquid air when additional energy is available, the simplest approach is based on the Linde-Hampson cycle in which a Joule-Thompson effect valve is used for expansion. Other variations of this cycle may also include cryogenic turbines for expansion (e.g., Claude and Collins cycles) which result in lower operating pressures, and higher liquid air production rates and efficiencies.81 At discharge, liquid air is pumped to high pressure, evaporated and heated to provide high pressure air. Heat can be provided from any ambient-temperature medium such as air, but can additionally be provided from a higher-temperature medium such as gases from combustion of natural gas. Increasing the temperature of the air improves the specific work output and efficiency of the system, making it comparable to other energy storage technologies. Another option to increase the temperature is to use air directly for combustion. The air, or gas, from a liquefied container can be expanded in turbines to generate electricity. Methods to reduce wastes of liquefaction and external energy requirements of regasification of liquefied air to improve the system efficiency have been proposed.82–84 For example, the use of the waste cooling power from the liquid air evaporation stage in other cycles (e.g., Rankine) can generate additional work and improve the system efficiency to higher than 80%.82 The cooled air can be used in a Brayton cycle or in a cryogenic organic Rankine cycle, as a heat sink. The waste heat that is generated when compressing the air before it is liquefied can be stored and used to reheat the air as it passes through turbines as well as to act as a heat source in a

30

Energy, Environment, and Resources | Energy Storage

Brayton cycle. She et al.83 propose a Brayton cycle that uses the heat from air liquefaction and releases heat to the evaporator of a liquefied natural gas storage system, thus coupling the two systems for improved efficiency. The authors show that system round-trip efficiency is approximately 70%. Peng et al.85 suggest packed beds as direct contact heat exchangers to collect the excess heat in the compression stage of liquefaction and release it to the air in the expansion stage while discharging. Xie et al.86 suggest that economical feasibility is unlikely for liquefied air energy storage systems without using waste heat, and that the feasibility is improved with larger plant installations.

2.5.3 Pumped hydro Pumped hydro energy storage (PHES) is a resource-driven facility that stores electric energy in the form of hydraulic potential energy by using an electric pump to move water from a water body at a low elevation through a pipe to a higher water reservoir (Fig. 7). The energy can be discharged by allowing the water to run through a hydro turbine from a high elevation to a lower elevation. The turbine is connected to a generator that can produce electricity as energy is discharged from the turbine. The inlet flow of water to the turbine can be controlled using gates to allow a variable power output. Variable-speed drives can also be used to provide regulation during charging. Pumped hydro energy storage systems require specific conditions such as availability of locations with a difference in elevation and access to water. If conditions are met, it is a suitable option for renewable energy storage as well as the grid. The energy efficiency of PHES systems varies between 70% and 80% and they are commonly sized at 1000–1500 MW.87 Other characteristics of PHES systems are long asset life, i.e., 50–100 years, and low operation and maintenance costs. Some of the disadvantages of pumped hydro electricity are large unit sizes, high capital costs and topographic limitations, i.e., available elevation difference between both reservoirs, and environmental ones. Underground PHES systems are considered a technically feasible option to avoid some of these challenges by using underground reservoirs, e.g., abandoned mines.88 Research is needed regarding methods and tools for identification and selection of feasible sites for PHES that are technically feasible, and commercially and socially acceptable.87 Deane et al.89 review locations and proposed timelines for new PHES development, and comprehensively review development trends. They suggest that the exploitable resources available for economically viable PHES are decreasing. Deane et al.89 review existing and proposed PHES plants and discuss their technical and economic drivers. To achieve greater operational flexibility and efficiencies than conventional PHES, variable speed PHES technologies and/or by-pass pump/turbine arrangements are being developed to increase the number of operation hours. Variable speed PHES technologies, while incurring slightly higher capital costs, offer a greater range of operation and efficiency than conventional PHES. At small capacities, PHES systems can vary design pumping capacity from 60% to full capacity and generation capacity from 20% to full capacity.90 While single machines may be limited in efficiency when capacity is varied, options to use multiple machines in various configurations have also been explored. For example, various dynamic-response by-pass arrangements are analyzed by Beevers et al.90 and their capacity flexibility and efficiency are compared. Various control strategies corresponding to different levels for variable speed operation of PHESs have also been developed.91 The fast power response of variable-speed PHES systems was proved in a comparison that is made between the variable- and constant-speed PHES systems for wind power regulation.92 The results are compared based on average and standard deviation of power difference between the two cases, penalty energy and power delay, and show improvements up to one order of magnitude in the variable-speed PHES case compared to the constant-speed case. The use of power converters also provides a quick response (i.e., within 2 s) in both pumping and generation modes of PHES systems, compared to the mechanically controlled, fixed speed system, allowing their application in micro-grids. To investigate the ability of power converters and controllers for the provision of stability within a PHES system and grid in several modes of operation as observed in practice, a prototyping environment is developed.93

Fig. 7 Illustration of pumped hydro storage with the pumping energy supplied by wind turbines: (a) charging at off-peak hours, (b) discharging at peak hours.

Energy, Environment, and Resources | Energy Storage

31

Yang and Jackson94 review the historical development of pumped-hydro energy storage facilities in the United States, including new development activities and approaches in PHES technologies. To mitigate environmental issues of PHES systems, developers are proposing innovative ways of addressing the environmental impacts, including the potential use of waste water in PHES applications. With the increasing need for energy storage, these new methods can lead to increased use of PHES in coupling intermittent renewable energy sources such as wind and solar power. New PHES designs are addressing the major challenges associated with conventional PHES. Vasel-Be-Hagh et al.95 introduce a new design, which does not require tall water tank towers or long piping and has scalable operation over a wide range of capacities depending on the electrical surpluses. The design provides constant-pressure and faster discharge, permitting quick response to instantaneous demand fluctuations.

2.6

Magnetic energy storage

Superconducting magnetic energy storage (SMES) can be accomplished using a large superconducting coil which has almost no electrical resistance near absolute zero temperature and is capable of storing electric energy in the magnetic field generated by dc current flowing through it. The superconducting coil is kept at a cryogenic temperature by using liquid helium or nitrogen vessels. Some energy losses are associated with the cooling system that maintains the cryogenic temperature, but energy losses in the coil are almost zero because superconductors offer no resistance to electron flow. SMES coils can discharge large amounts of power almost instantaneously, and can undergo an unlimited number of charging and discharging cycles at high efficiency. Coil configuration, energy capability, structure and operating temperature are some of the main parameters in SMES design that affect storage performance. Low temperature superconductor devices are currently available while high temperature ones are still in development due to their high costs. SMES applications include load leveling, system stability, voltage stability, frequency regulation, transmission capability enhancement, power quality improvement, automatic generation control, and uninterruptible power supplies. Configurations of thyristor-based, voltage-source-converter-based, and current-source-converter-based SMES are reviewed by Hassan Ali et al..96 They suggest categorizing the cost of SMES technologies based on the cost of the energy storage capacity (i.e., costs of conductor, coil structure components, cryogenic vessel, refrigeration, protection, and control equipment) and the cost of power handling capability. They suggest a wide cost variation exists in the latter, and that focusing research on it could considerably reduce the overall SMES cost. Sutanto and Cheng97 review SMES systems for power systems. They emphasize the importance of the development of practical applications of SMES for power systems as opposed to several studies performed through computer simulations or in laboratories. They also suggest the development of efficient control strategies is needed to integrate small ratings of SMES systems at various locations to improve their power capacities. Khaleel et al.98 suggest that future research should focus on integration of SMES systems with power systems including reliability and safety of SMES, economic feasibility, market integration and field testing of SMES.

2.7

Chemical energy storage

A reversible chemical reaction that consumes a large amount of energy may be considered for storing energy. Chemical energy storage systems are sometimes classified according to the energy they consume, e.g., as electrochemical energy storage when they consume electrical energy, and as thermochemical energy storage when they consume thermal energy. Many chemical energy storage systems are based on hydrogen energy storage. In hydrogen energy storage, hydrogen is produced via direct (e.g., photoconversion) or electrolytic methods, stored for a period of time, and then oxidized or otherwise chemically reacted to recover the input energy (Fig. 8). The hydrogen results from a chemical reaction, but is not the source of energy. For many decades, electricity has been a primary energy carrier for many of society’s energy technologies. Hydrogen energy exhibits characteristics complementary to those of electricity. Some have proposed a “hydrogen economy” involving all aspects of hydrogen energy systems, including production, storage, distribution and utilization.99 Winter100 describes the hydrogen economy, its environmental and climatic relevance, its positive influence on the energy quality of the system, its effect on decarbonizing fossil fueled power plants, and the novel non-heat-engine-related electrochemical energy

Fig. 8 Production of hydrogen using renewable energy sources. Adapted from Abbasi, T.; Abbasi, S.A. ‘Renewable’ Hydrogen: Prospects and Challenges. Renew. Sustain. Energy Rev. 2011, 15(6), 3034–40.

32

Energy, Environment, and Resources | Energy Storage

converter fuel cell in portable electronics, in stationary and mobile applications. In this section, processes in which energy is stored by producing hydrogen and hydrogen storage techniques are both described. One common method of hydrogen production is by splitting water. The energy required for this process can be provided from fossil fuels and renewable or other energy sources. Energy from renewable sources is often intermittent and needs to be stored before it is needed. Abbasi and Abbasi101 discuss the production of hydrogen from solar energy with the following processes: (i) a combination of a solar cell with an electrolyser, (ii) a combination of a concentrated solar thermal system with a turbine and an electrolyser, (iii) a combination of a solar concentrating system with a thermochemical water-splitting cycle, and (iv) direct photoconversion (photocatalytic, photoelectrochemical, photobiological water splitting). Production of hydrogen from other renewable sources such as wind, hydroelectric, geothermal, ocean thermal energy conversion, anaerobic digestion of biomass and biowastes are also discussed in this work.101 Since photosynthetic and photovoltaic processes harvest the energy in sunlight, they are sometimes compared. But the two processes operate differently and produce different products: biomass or chemical fuels in the case of natural photosynthesis and non-stored electrical current in the case of photovoltaics. Blankenship et al.102 compare natural photosynthesis to present technologies for photovoltaic-driven electrolysis of water to produce hydrogen, and show photovoltaic-driven electrolysis is more efficient on an annual basis. Ways in which new developments in synthetic biology may be used to improve solar energy conversion efficiency of natural photosynthesis are discussed. The storage of hydrogen is a substantial challenge, especially for automotive applications. Hydrogen has a low energy density on a volume basis compared to the other fuels, requiring a much larger fuel tank for a vehicle operating on hydrogen rather than petrol/diesel. Furthermore, hydrogen is the lightest of all elements and harder to liquefy than methane and propane. Due to its low density and also its small molecular size, it can leak from containment vessels. Hydrogen can be stored in its pure form as a compressed gas or as a cryogenic liquid or in a mixed-phase (hydrogen slush). Liquefaction of hydrogen and pressurization of a cryogenic liquid in order to achieve higher density in a liquid or mixed-phase state both require a large amount of energy and specialized infrastructure. In adsorptive storage, hydrogen is absorbed to and released from porous networks such as zeolites, metal −organic frameworks, clathrate hydrates, various carbon materials (e.g., nanotubes, fullerenes and graphene), and conventional organic polymers. This process can occur for several cycles without decomposition of the solid or loss of gas. In chemical storage, hydrogen is stored in chemical bonds with other elements in a hydrogen-rich material, in solid or liquid phases. Solid-phase systems include metal and non-metal hydrides, amines, amides, and ammonia-like complexes. Liquid carriers include N-ethylperhydrocarbazole, alcohols and formic acid. Details of specific storage types are compared by Pruestre et al..103 Lamb et al.104 explore materials and technologies for hydrogen purification from decomposed ammonia gas streams, and suggest that energy-efficient decomposition of ammonia and subsequent separation and purification of the hydrogen product are two key challenges in using ammonia as a hydrogen storage intermediate. They show that defect-free dense-metal membranes can achieve satisfactory product purity. Various aspects of hydrogen storage methods have been reviewed.105–107 These include various hydrogen storage methods, including high-pressure107 and cryogenic-liquid storage, adsorptive storage on high-surface-area adsorbents, chemical storage in metal hydrides and complex hydrides and intermetallic compounds,108 and storage in boranes. Hosseini et al.107 thermodynamically model the filling phase of compressed hydrogen storage and analyze it based on the second law of thermodynamics.

3

Energy storage applications

Energy storage is an enabling technology for various applications such as power peak shaving, renewable energy utilization, enhanced building energy systems, and advanced transportation. Energy storage systems can be categorized according to application. Hybrid energy storage (combining two or more energy storage types) is sometimes used, usually when no single energy storage technology can satisfy all application requirements effectively. Storage mass is often an important parameter in applications due to weight and cost limitations, while storage volume is important when the system is in a space-restricted or costly area such as an urban core. Energy storage applications are continuously expanding, often necessitating the design of versatile energy storage and energy source systems with a wide range of energy and power densities. In this section, we focus on various applications of energy storage such as utilities, renewable energy utilization, buildings and communities and transportation. Table 2 provides examples of energy storage systems currently in operation or under construction and includes some of the features of such storage systems.

3.1

General applications

Some energy storage systems find broad and general applications. For instance, the fact that PCMs melt and solidify at a wide range of temperatures makes them attractive in numerous applications, including solar water heating, solar air heating, solar cooking, solar greenhouses, space heating and cooling in buildings, off-peak electricity storage, and waste heat recovery.55,121 Also, applications of flywheels, as discussed by Liu and Jiang,122 include uses in the International Space Station, Low Earth Orbits in earth observation missions, overall efficiency improvement and pulse power transfer for hybrid electric vehicles, and power quality assurance. Finally, asphalt concrete pavements have been considered for use as solar heat collectors and storage systems by Hall

Energy, Environment, and Resources | Energy Storage Table 2

33

Examples of current energy storage systems in operation or under development.

Storage type

Example

Power capacity/ duration

Application

System specifications

Pumped hydro

Bath County Pumped Storage Station, US

3003 MW/10 h 18 min

Electric energy time shift

La Muela Pumped-Storage Plant, Spain Huntorf, Germany

2000 MW

Renewable energy capacity firming

Consists of two large reservoirs with 385 m difference in height, a power house and the tunnels that connect them. At high demand, water is passed through the tunnel at a rate of up to 852 m3/s to drive six generators.109 Provides 5000 GWh of energy storage.

290 MW/2 h

Arbitrage Spinning reserve Black start applications

McIntosh, US

110 MW/26 h

Refine base-load electricity from a nuclear power plant, producing peak load electricity

Thermal

Planta Solar 20, Spain

20 MW/1 h

Thermal, molten salt

Solana Solar Generating Plant, US

280 MW/6 h

Thermal, ice

University of Arizona, US

3 MW/6 h

Renewable energy time shift Renewable energy capacity firming Renewable energy time shift Renewable energy capacity firming Electricity use time shift

Thermal, chilled water

University of Nebraska-Lincoln, US

100 MW/6.24 h

Electricity use time shift

Flywheel

Hazle, Pennsylvania, US

20 MW/15 min

Frequency regulation Increase renewable energy use (wind and solar)

Railway Technical Research Institute (RTRI), Japan Endesa STORE, Spain Endesa STORE, Spain Kaheawa Wind Power Project II, US

0.3 MW/20 min

Increase renewable energy use (solar)

4 MW/5 s 1 MW/3 h 10 MW/45 min

Frequency regulation114 Frequency regulation114 Frequency regulation114 Ramping Renewable energy time shift Renewable energy capacity firming Renewable energy capacity firming118 Renewable energy time shift

Compressed air

Supercapacitor Battery, Li-ion Battery, lead acid

Battery, Vanadium Redox flow Chemical, hydrogen Gravitationala

Hokkaido Electric Power, Japan 140-MW wind Park, Germany Advanced Rail Energy Storage, US

15 MW/4 h 1 MW/27 h 50 MW/15 min

Frequency regulation Electric supply reserve capacity Voltage support

Uses two cylindrical 150,000-m3 salt caverns at a depth of 600–800 m. Pressure tolerance is 50–70 bar. With 42% efficiency,110 it uses 0.8 kWh electricity and 1.6 kWh in natural gas heating value for every 1 kWh of electricity generation. Uses two cylindrical 538,000-m3 salt caverns at depth of 450–750 m. Pressure tolerance is 45–76 bar.111 Uses heat from turbine exhaust gases to preheat compressed air. With 54% efficiency,110 it uses 0.69 kWh electricity and 1.17 kWh in natural gas heating value for every 1 kWh of electricity generation. Integrated with solar field of 150,000-m2 containing 1255 heliostats. Provides 48 GWh of stored energy per year.112 Integrated with a parabolic-trough solar plant. Provides 944 GWh of stored energy per year.112 Utilizes nighttime-produced electricity to generate and store daytime cooling. Comprised of a total of 205 ice storage tanks.113 A chilled water storage tank of 2.8 MG provides 314 MWh of cooling capacity at a maximum chilled water flow rate of 0.5 m3/s.114,115 Plant comprises 200 flywheels rated at 0.1 MW and 25 kWh.116 Flywheel spins at a rate of up to 15,500 rpm. Flywheels are able to operate at more than 100,000 full charge/discharge cycles. Uses superconducting magnetic rotor and bearing, the rotor being 2 m in diameter and weighing 4 tons.117

Integrated with a 21 MW wind power plant.

Can produce 210 Nm3/h of hydrogen. It is connected to a 140 MW wind farm.119 Has an efficiency of 86% and ramps to full power in seconds.120

a Utilizes a single uphill track with a central queue of loaded shuttle-trains that travel up and down grade in response to an independent system operator command to provide frequency adjustment.

34

Energy, Environment, and Resources | Energy Storage

et al..123 Asphalt concrete pavements that incorporate aggregates and additives (e.g., limestone, quartzite, lightweight aggregate, copper slag, and copper fiber) are designed to become more conductive, or more insulating, or to store more thermal energy.

3.2

Energy utilities

The use of energy storage systems in utility networks has become increasingly important and focused on as more storage options become available. Energy storage deployed at any of the five major subsystems in the electric power systems, i.e., generation, transmission, substations, distribution, and final consumers, can help balance customer demand and generation. Intermittent power generation, such as that provided by many renewable energy sources, results in power instability which can damage grid equipment such as generators and motors. By combining renewable energy systems with energy storage technology, renewable energy penetration is increased and overall system performance improves, while flexibility is provided for grid control and maintenance. Some of the applications of energy storage systems include124:

• • • • •

reduction of congestion in the transmission system, storing energy during periods of low demand for use during periods of high demand, maintaining voltage and frequency within normal operating ranges, compensating for unexpected contingencies such as failure of a generating unit, and maintaining a real-time balance between generation and load.

Electricity can be stored in electric fields (capacitors) and magnetic fields (SMES), and via chemical reactions (batteries) and electric energy transfer to mechanical (flywheel) or potential (pumped energy storage) energy or pressure (compressed air energy storage) energy forms. Pumped energy storage has been the main storage technique for large-scale electrical energy storage (EES). Battery and electrochemical energy storage types are the more recently developed methods of storing electricity at times of low demand. Battery energy storage developments have mostly focused on transportation systems and smaller systems for portable power or intermittent backup power, although system size and volume are less critical for grid storage than portable or transportation applications. Future utility applications of batteries could be focused on providing peak distribution capacity deferral and peak shaving at the substation as well as reliability enhancement.125 Research into reliable battery storage at the grid scale is focused on durability for large numbers of charge/discharge cycles and lifetime, high round-trip efficiency, ability to respond rapidly to changes in load or input, and reasonable capital costs. Moseley and Garche126 provide an overview of electrochemical energy storage applications in grid balancing and renewable energy storage. Koohi-Kamali et al.127 review various applications of electrical energy storage technologies in power systems that incorporate renewable energy, and discuss the roles of energy storage in power systems, which include increasing renewable energy penetration, load leveling, frequency regulation, providing operating reserve, and improving micro-intelligent power grids. Flywheel storage, electrochemical storage, pumped hydroelectric storage, and compressed air storage, as well as their operating principles and applications, are described. Vazquez et al.128 review the main applications and the power converters used to operate some energy storage technologies, and describe various storage technologies, including batteries, electrochemical double-layer capacitors, regenerative fuels cells, CAES, flywheel, SMES, and thermoelectric energy storage, and their applications. Roberts and Sandberg129 review new types of storage being utilized for grid support, and emphasize the growing importance of energy storage systems in smart grids with more dynamic loads and sources. Yang et al.130 examine electrochemical storage technologies used in grids, such as redox flow batteries, Na-beta alumina membrane batteries, unique Li-ion chemistries, and lead-carbon technologies, and the needs to reduce costs and improve performance for these technologies to increase their market penetration. Dunn et al.131 review sodium-sulfur batteries, redox-flow batteries and Li-ion batteries for use in the grid and their potentials. Xue et al.97 describe applications of SMES in improving power quality and system stability. Mousavi et al.70 and Li and Palazzolo132 review applications of flywheel energy storage systems. Mousavi et al.70 highlight the potential of flywheel energy storage systems compared to other energy storage technologies for power leveling, grid frequency support/control, and voltage sag mitigation based on their fast recharge time and high power density. Khodadoost et al.133 suggest that future developments in increasing flywheel energy density and investigating the feasibility of its modular application, will improve the power system stability.

3.3

Renewable energy utilization

Renewable energy use is growing rapidly, helping provide electricity to satisfy the world’s demand and mitigate environmental impacts, especially related to the electricity sector. However, the variability of these resources creates technical and economic challenges for their operation and use when integrated on a large scale. An important means of addressing the intermittency of renewable energy sources is energy storage. A higher penetration of renewable energy generation is typically achieved with storage, as it permits excess energy produced from renewable energy sources to be stored and dispatched later when needed. In addition to intermittency of renewable energy sources, storages are also distributed. Beaudin et al.134 suggest that energy storage technologies that are scalable, modular, durable, and low maintenance may be suitable options for distributed renewable energy harvesting. Such technologies include flywheels, capacitors, SMES and most batteries (excluding lead-acid batteries). Beaudin et al.134 review the technology status and installations for a broad range of EES, focusing on advantages and disadvantages for integrating large-scale, variable renewable electricity sources, and discusses external factors affecting numerous

Energy, Environment, and Resources | Energy Storage

35

EES applications, such as mineral availability and geographic limitations. The article indicates that addressing each challenge imposed by variable renewable electricity sources requires a different set of EES characteristics, and that no single EES technology consistently outperforms all others in all applications. Technologies that couple a solar energy source with energy storage are discussed and/or reviewed by many researchers.16,19,135 Examples include solar stills and solar dryers, which are used in drying agricultural food products.54,55 Arjunan et al.136 experimentally investigate various storage materials, such as black granite gravels, pebbles, blue metal stone, and paraffin wax, to examine the productivity of storing excess heat in solar stills during the day and releasing the stored heat to the basin at night. Agrawal and Sarviya137 also review the use of various thermal storage materials in solar air heaters and dryers (e.g., rock, water, sand and granite, metal scrap, pure paraffin wax, and a mixture of aluminum power and paraffin wax). Due to their high heat storage capacity per unit volume, paraffin and salt hydrates have received considerable attention as storage materials for solar air dryers and heaters. Another way to store solar energy is to convert it directly into chemical fuels by methods such as photon-driven electrolysis of water to produce hydrogen and oxygen. Research on new metal oxide visible light-absorbing semiconductors could help improve this technology. Osterloh and Parkinson138 review developments of semiconductor light absorbers and co-catalysts. They emphasize the importance of low cost and material stability of photoactive materials against photocorrosion to allow system technical and financial feasibility. Díaz-González et al.139 review several energy storage technologies for wind power applications, including gravitational potential energy with water reservoirs, compressed air, electrochemical energy in batteries and flow batteries, chemical energy in fuel cells, kinetic energy in flywheels, magnetic fields in inductors, and electric fields in capacitors. Mousavi et al.70 suggest flywheel energy storage systems as the best systems for wind energy storage due to their quick response times and favorable dynamics. They provide several examples of wind-flywheel pairing studies and their control strategies to achieve smooth power control. Khodadoost et al.133 suggest that flywheels are favorable options for integration with wind and PV systems compared to battery energy storage systems since variations in their output power occur in a short period of time. Although the use of compressed air energy storage (CAES) has for some time been for grid management applications such as load shifting and regulation, CAES is expected to increase flexibility when integrating renewable energy sources such as wind, solar and tidal with the power grid. Succar and Williams140 review CAES systems that are combined with wind turbines to produce electricity, including technical and geologic requirements for widespread CAES deployment and paying attention to relevant geologies in wind-rich regions of North America. Konrad et al.141 investigate factors that affect site selection and planning of CAES facilities to assist in renewable energy harvesting in Ontario, Canada. The authors provide details of the groundwork needed for the feasibility study and identify and characterize influences such as mechanics of the ground rock and locations of renewable energy resources. Esmaeilion et al.142 perform an exergoeconomic assessment of a CAES system within a multigeneration system of combined cooling, heat and power, and a desalination unit. They show that the proposed system offers great potential for reliable operation during peak demand hours. Rehman et al.87 review several PHES hybrid systems such as wind-hydro, solar PV-hydro, and wind-PV-hydro. The importance, necessity and contribution of wind-hydro pumped storage systems in meeting Turkey’s electric energy demand as well as the current status and potential of using pumped hydro in wind energy applications in Turkey are investigated by Dursun and Alboyaci.143 They found that, in generating systems with limited flexibility, the application of PHES for providing standing reserve could significantly reduce the amount of wind curtailed and reduce the amount of energy produced by a conventional plant. A question that arises when integrating energy storage with renewable energy systems is what configuration provides the most technically and economically viable method to supply electricity in stand-alone systems. To address this, Askari and Ameri144 perform a feasibility analysis of renewable energy systems for supplying the electrical load requirements of a typical community in a remote location in Kerman, Iran, considering various combinations of PV modules and wind energy conversion systems supplemented with battery storage (e.g., photovoltaic/battery, wind/battery and hybrid photovoltaic/wind/battery). The assessment criterion is taken to be the total net present cost of each system configuration, and the results show that, because of the sudden decreases in wind speeds in Kerman, the total net present cost of the wind/battery and hybrid renewable energy systems is increased, making the PV/battery system the most advantageous for supplying the electrical load requirements. Energy and exergy analyses are used to assess a hybrid solar hydrogen system with activated carbon storage for residential power generation in a novel study by Hacatoglu et al..145 Exergy flows and efficiencies are calculated for individual devices and the overall system, and show that solar photovoltaic-based sub-systems have the lowest exergy efficiencies and significant potential for improvement.

3.4

Buildings and communities

As research aimed at nearing or achieving net-zero energy buildings and communities intensifies, governments are promoting the adoption of renewable energy sources in buildings in the commercial, institutional, industrial and residential sectors. Energy storage is recognized as an important way to facilitate the integration of renewable energy into buildings (on the generation side), and as a buffer that permits the user-demand variability in buildings to be satisfied (on the demand side). Pero et al.146 suggest a number of key performance indicators to facilitate the comparison of various storage technologies in the decision-making/design phase and the assessment of technical solutions. The indicators include storage capacity, maximum charge and discharge power, depth of charge, durability, specific cost of storage, maximum self discharge rate, storage weight, and generated energy/cost savings.

36

Energy, Environment, and Resources | Energy Storage

Thermal energy storage is a relatively common storage technology for buildings and communities and extensive research is available on storage materials and their classifications, recent developments, thermal storage usage conditions and performance, limitations and possible improvements for buildings uses.48,121,147–154 Buildings and communities can benefit from short-term (up to a few days) and long-term (up to a few months) storage. For example, thermal energy storage is capable of shifting electrical loads from peak to off-peak hours, providing a powerful tool in demand-side management programs.147 Although this technology is a relatively mature type of energy storage, research and development is ongoing to overcome technical issues such as subcooling, segregation and materials compatibility,148 and to develop more efficient and economic TES systems in buildings, e.g., building thermal mass utilization, PCMs used to increase the thermal capacity of storage while operating at a fixed temperature (Fig. 9), underground thermal energy storage and storage tanks. Alavy et al.155 develop a phase change material-based thermal caisson system to assess the long-term energy performance of ground-source heat pump systems. Pavlov et al.149 review developments during the last four decades on seasonal TES in the ground, including aquifer, borehole, water tank and water gravel-pit thermal energy storage systems. They consider various storage concepts coupled with natural and renewable energy sources such as solar and waste thermal energy. They suggest that various parameters such as building peak thermal loads, thermal load profiles, availability of waste or excess thermal energy, availability of natural and renewable energy sources, type of thermal generating equipment, and building type and occupancy impact the feasibility of use of TES in buildings. Feluchaus et al.46 suggest small system size as one of the barriers to market growth of ATES systems. They suggest a combination of district heating or cooling with ATES as a promising option for integration of ATES and building heating and cooling systems. Studies on the dynamic performance and control strategies of energy storage systems for various building types, weather conditions, and user behavior are needed to understand how TES systems can best support the development of low-energy and zero-emission buildings. Among renewable energy sources, storage of solar thermal energy in building heating and cooling supply have been extensively reviewed.17,32,63 A good example of systems utilizing thermal energy storage in solar buildings is the Drake Landing Solar Community in Okotoks, Alberta, Canada, which incorporates a borehole seasonal storage to supply space heating to 52 detached energy-efficient homes through a district heating network. Sibbitt et al.150 describe the system and its operation and presents 5 years of performance data. Thermal analyses of PCM uses in building envelope (e.g., walls, floors, ceilings and windows) demonstrate that they can be effective in shifting heating and cooling loads from peak electrical demand periods to off-peak periods or in storing solar energy for use in hours when solar radiation is not available.152 However, due to various PCM thermophysical properties and incorporation methods, investigations are needed to evaluate and compare cost, efficiency, environmental impact, life cycle, and practicability of various options under various weather and experimental conditions.151 To store electricity in buildings, batteries are most commonly used. Examples include lead-acid, molten salt (sodium-sulfur, sodium-metal chloride), Li- ion and flow batteries. Smolinski et al.156 suggest that the use of flywheels in buildings that have solar

Active Heating

Passive Solar Heating

Walls

Roof

PCM PCM

PCM With daytime solar radiation

With daytime solar radiation

PCM Heat pump

PCM With solar collector system

Night Cooling

Floor

air

PCM With nighttime ventilation

With nighttime cheap electricity

PCM

With daytime solar radiation

PCM

Electricity

With nighttime cheap electricity

air air

With nighttime ventilation

PCM With nighttime ventilation

Fig. 9 Forms and effect of PCM applications in building. Reproduced from Zhang, Y.; Zhou, G.; Lin, K.; Zhang, Q.; Di, H. Application of Latent Heat Thermal Energy Storage in Buildings: State-of-the-Art and Outlook. Build. Environ. 2007, 42(6), 2197–209.

Energy, Environment, and Resources | Energy Storage

37

PV panels installed significantly reduces the costs of the system, and they design and operate a prototype system. Other promising electrical energy storage technologies such as CAES and hydrogen storage technologies still face issues such as low efficiency, safety and cost for use in building-scale applications. Zero Energy Buildings (ZEBs) are viewed by many as the future target for the design of buildings and have attracted considerable attention during the past decade. Thermal energy storage is a particularly attractive option for the development of zero energy buildings by reducing the energy consumption of the buildings, improving system efficiency, and reducing the peak load.157 Reduction of building energy consumption can be achieved by increased renewable energy use which can be achieved by overcoming the time mismatch between demand and availability of solar and aero-thermal energy. The efficiencies of heating and cooling systems can be improved by avoiding partial-load operation. Lastly, the reduction of peak loads by the use of stored energy at peak times could result in smaller power capacity requirements for heating and cooling. Applications in the built environment of fuel cells that utilize hydrogen from renewable sources are reviewed by Abdel-Wahab and Ali,158 based on information from the literature and industry experts. While they provide a structured approach for evaluation of such systems, Singh et al.159 focus on the modelling and simulation of a hydrogen system for performance and cost feasibility estimation. Such analyses can be considered as preliminary steps towards more detailed analyses of the use of hydrogen fuel cells in buildings.

3.5

Transportation

The evolution of ground, water and air transportation technologies has resulted in the need for advanced energy storage systems. Compared to conventional transportation technologies that are driven by internal combustion engines and utilize gasoline tanks for energy storage, hybrid electric vehicles use onboard energy-storage systems such as flywheels, ultra-capacitors, batteries and hydrogen storage tanks for fuel cells. The requirements for the energy storage devices used in vehicles are high power density for fast discharge of power, especially when accelerating, large cycling capability, high efficiency, easy control and regenerative braking capacity. The primary energy-storage devices used in electric ground vehicles are batteries. Electrochemical capacitors, which have higher power densities than batteries, are options for use in electric and fuel cell vehicles. In these applications, the electrochemical capacitor serves as a short-term energy storage with high power capability and can store energy from regenerative braking. A combination of a battery and an electrochemical capacitor can enhance the characteristics desired in land-based vehicles, aircraft and ships, including engine starting, high current for fast preheating of catalysts, electric power steering, and local power for actuators and distributed power systems. However, a system consisting of a battery of reduced size and an electrochemical capacitor has to be commercially competitive with battery-only systems to penetrate the market. Farsi and Rosen160,161 propose integration of a solid-oxide fuel cell and Li-ion batteries for a hybrid electric aircraft and electric vehicles. Flywheels have also been used for a long time in transportation systems. They can be used for regenerative braking and load averaging or as the prime energy source for propulsion; the latter option is only a theoretical probability due to flywheel low energy density.70 To improve energy storage energy density, hybrid systems using flywheels and batteries can also be attractive options in which flywheels, with their high power densities, can cope well with the fluctuating power consumption and the batteries, with their high energy densities, serve as the main source of energy for propulsion.133 However, for large vehicles such as trains, a larger flywheel needs to be used to serve such a purpose and its weight becomes a disadvantage.70 The need for a storage unit to recapture vehicular braking energy can be achieved in railway systems by installing an energy storage device at the supply substations, along the railway track or on board the train. Designing and optimizing a train timetable to allow the interchange of energy among accelerating and decelerating trains with energy storage options is another approach that is proving to be effective for high traffic conditions.162 Wayside energy recovery systems store energy along the railway tracks from decelerating vehicles and discharge it to accelerating ones. This increases overall system efficiency and voltage stability within the grid, and lowers peak power demands, costs and potentially CO2 emissions depending on the energy mix. Flywheels, batteries and supercapacitors are suitable options for wayside energy storage.73 Pneumatic accumulators are also available options for regenerative braking energy storage, but often not considered due to their low energy density and efficiency.70 Some additional benefits of such installations are load leveling and support of the mains voltage, lower energy costs, reduced investment costs since fewer substations are needed, and emergency supply in case of power failures. Several investigations have been made regarding energy storage applications in transportation.128,163–165 Hannan et al. suggest that, currently, limitations in electric vehicle energy storage and powering lies in raw material support and proper disposal, energy management, power electronics interface, sizing, safety measures. Khaligh and Li163 suggest that hybrid energy storage systems with large capacity, fast charging/discharging, long lifetime, and low cost could be more feasible and increase competitiveness with conventional vehicles in the near future. Several challenges and limitations exist in using lithium batteries in transportation. Methods for improving Li-ion batteries to meet demands for powering electric vehicles and storing renewable energy, including new ways to prepare electrode materials via eco-efficient processes and the use of organic rather than inorganic materials and new chemistries for Li-ion batteries, are suggested.166 Thackeray et al.16 suggest that while Lithium-based batteries have considerable potential for improved energy densities (e.g., factors of five or more may be possible for Li–oxygen systems), major breakthroughs, and not incremental advances, in materials and chemistries are required for their adoption in transportation systems to become widespread. Currently, most commercial electric and hybrid vehicles do not have hybrid energy storage systems on board. Since one type of energy storage systems cannot meet all electric vehicle requirements, a hybrid energy storage system composed of batteries, electrochemical capacitors, and/or fuel cells could

38

Energy, Environment, and Resources | Energy Storage

be more advantageous for advanced vehicular energy storage systems. Such hybrid energy storage systems, with large capacity, fast charging/discharging, long lifetime, and low cost are currently being investigated for electric vehicles.163,167 Also, Yang et al.165 describe the application of other energy storage candidates such as flywheels in automotive applications. Cao et al.168 propose a new battery/ultracapacitor hybrid energy storage system for electric drive vehicles including electric, hybrid electric, and plug-in hybrid electric vehicles. This design can fully utilize the power capability of the UCs without requiring a matching power dc/dc converter to satisfy the real-time peak power demands. It uses a smaller dc/dc converter working as a controlled energy pump to keep the ultracapacitor voltage higher than the battery voltage for most city driving conditions, improving the battery load profile and vehicle drivability. A global research effort focusing on the development of physical and chemical methods for storing hydrogen in condensed phases has recently emerged due to the need to store hydrogen onboard at high volumetric and gravimetric densities when using hydrogen as a vehicular fuel. Thus, new materials with improved performance, or new approaches to the synthesis and/or processing of existing materials, are highly desirable. Desirable characteristics for hydrogen storage materials are investigated by Yang et al.165 and Winter,100 accounting for fuel cell vehicle requirements. Yang et al.165 also introduce candidate storage materials, such as conventional metal hydrides, chemical hydrides, complex hydrides and sorbent systems, and describe their performances and improvement prospects.

4

Categorizations of energy storage technologies

Energy storage systems have been used for centuries and undergone continual improvements to reach their present levels of development, which for many storage types is mature. Many types of energy storage systems exist, and they can be categorized in various ways. For example, storage characteristics of electrochemical energy storage types, in terms of specific energy and specific power, are often presented in a “Ragone plot”,169 which helps identify the potentials of each storage type and contrast them for applications requiring varying energy storage capacities and on-demand energy extraction rates. The plot also aids in selecting the most appropriate energy storage for specific applications or needs (Fig. 10). Storage energy density is the energy accumulated per unit volume or mass, and power density is the energy transfer rate per unit volume or mass. When generated energy is not available for a long duration, a high energy density device that can store large amounts of energy is required. When the discharge period is short, as for devices with charge/discharge fluctuations over short periods, a high-power density device is needed. Note that only a few energy storage types are shown in Fig. 10 as the Ragone plot is traditionally used only for batteries, capacitors and fuel cells. However, others have presented this chart for/including other storage types such as thermal energy storage170 and flywheels171,172 as well as combustion engines171 for comparison purposes. In the current chapter, a more comprehensive comparison of specific energy and power as well as other technical details of several energy storage types are provided in Table 3 for better comparison. Energy storage systems also can be classified based on storage period. This is discussed in more detail in Section 5.2. In summary, the energy storage types covered in this section are presented in Fig. 11. Note that other categorizations of energy storage types have also been used such as electrical energy storage vs thermal energy storage, and chemical vs. mechanical energy storage types, including pumped hydro, flywheel and compressed air energy storage.

Fig. 10 Energy storage Ragone plot Reproduced from Hall, P. J.; Bain, E. J. Energy-Storage Technologies and Electricity Generation. Energy Policy 2008, 36, 4352–5.

Energy, Environment, and Resources | Energy Storage Table 3

39

Technical characteristics of energy storage technologies.

Storage type

Power density (volumetric) (kW/m3)

Energy density (volumetric) (kWh/m3)

Energy density (mass) (Wh/kg)

Cycle efficiency (%)

Lifetime (cycles)

Capacitor Supercapacitor

>100,000173 40,000–120,000174,a >100,000173 15–4500175

2–10173 1074 10–20174,a 10–3073,173 1–35175

0.05–5173 1–573 2–574 1–15174,a 2.5–15173

60–70173 90–10073 85–98174,a 90–97173 90–95169 65–99175

>5  104173 105173 104–106175

1300–10,000174 1500–10,000173 150–36073 60–800175

300–75073 200–400174 200–500173 90–500175 80–200174 40–300175 50–80173,174 25–90175

NiCdb NiCd vented NiCd sealed

40–140175 75–700174 80–600173

15–150175 15–80174 80–110174 60–150173

85–98174 90–97173 954,134 85–95124 >9573 70–100175 65–75174 50–80175 65–80134 75–90124,174 70–80173 804 60–90175 60–90175 60–80174 60–70173,174 804

500–104174 1000–1044,173 3000134 250–104175

500–3000174 8–600175 90–700174 10–400173,175

100–30073 60–200174 75–200173 200134 150–2004 30–300175 40–80174 30–90175 30–45174 30–50173 35–504 10–50175

NaNiCl

250–270174 15–260175

150–200174 100–200175

NaS

120–160174 140–180173 1–50175 50–100174 10–200175 1–25174 0.5–2174 1000174 >500175 1000–3650109 >104174 >1.2  104173 800–1.6  104175 2  104–107174 >2  104173 104–105134,175

75–80174 95–97173 90–95169 80–99175 41–75174 60–90175

>105173 >2  104134 104–105175

65–85134 70–80174 70–85173 50–85169 75–78124 65–90175 20–50169,173 30–60173 75–90175

>0.5  104174 104–3  104134 104–6  104175

Battery Li-ion

NiMH sealed Lead-acid

Zinc- air Hybrid flow-battery Vanadium redox-flow battery

10–80175 15–40174 30–45174 50–75173 754 100–200174 1254 85–140175 100–250174 150–2404,173 100–240175 130–200174 10–500175 75–85174 15–50174 10–30173 10–50175 5–10073 5–30174 10–30173 5–200175 0.5–5173 0.3–75175

Flywheel

5000174 1000–2000173 40–2000175

Magnetic

2600174 300–4000175

6174 0.2–2.5173 0.2–14175

Compressed air

0.2–0.6174 0.5–2173 0.04–10175

30–60173 3–60175

Pumped hydro

0.1–0.2174 0.5–1.5173 0.01–0.10175

2–6174 3–6173 0.4–20175 0.5–0.8176,c 0.2–2174 0.5–1.5173 0.5–1.3175 0.4–1.1176

Hydrogen fuel cell Thermal Latent Sensible

>500173

500–3000173 80–500173 100–370175 25–120175

800–10,000173 80–250173 150–250175 10–120175

a

Double-layer capacitor. Vented versus sealed is not specified in the reference. c Energy density evaluated at 60 bars. b

0.2–2174 0.5–1.5173 0.3–1.3175

600–1200174 300–3000175 250–1500174 500–1000173 500–15004 100–2000175 300–104175 1500–3000174 500–800174 25004

>104174 104–3  104134,175

>1000173

40

Energy, Environment, and Resources | Energy Storage

Fig. 11 Classification of energy storage types.

5

Comparisons of energy storage technologies

In this section several energy storage types are described and/or compared from technical and economic perspectives, rather than their classifications and principles. Similar analyses and comparisons have been reported in the past and shown to be of great interest.177–179 The analysis in this section aims to provide an updated comparison.

5.1

Technical performance

Energy storage technologies are reviewed and compared in this section from a technical viewpoint, focusing on parameters that can improve the design and performance of energy storage systems, rather than their classifications and principles.166,173,180–184 Some comparisons are also made in previous sections of various energy storage technologies, for example, the advantages of electrochemical double-layer capacitors over other storage technologies as discussed by Sharma and Bhatti.31 To assess the technical performance of various energy storage types, design parameters such as efficiency, energy capacity, energy density, run time, capital investment costs, response time, lifetime in years and cycles, self-discharge and maturity are often considered.173,180,181 Here, technical characteristics of energy storage technologies are summarized in Table 3. Note that the values in this table are collected from references that are published over various years, since the literature on energy storage technologies lacks data for recent energy storage technologies in some cases. Differences that are noticed in technical information regarding a given energy storage technology

Energy, Environment, and Resources | Energy Storage

41

may be due to various factors such as different applications or technical developments in a technology that have caused improvements to its technical characteristics. It is observed that energy storage systems with higher power density are often used for short-duration applications requiring fast response such as grid voltage maintenance. Storage systems with higher energy density are often used for long-duration applications such as renewable energy load shifting.185 Raising power and energy densities of energy storage units significantly depends on advances in storage materials and the development of new materials for various energy storage types, including thermal, mechanical, electromagnetic, hydrogen and electrochemical.166,182–184 Strategies for developing advanced energy storage materials in electrochemical energy storage systems include nano-structuring, pore-structure control, configuration design, surface modification and composition optimization.182 An example of surface modification to enhance storage performance in supercapacitors is the use of graphene as graphene anodes, graphene-based hybrid anodes and electrode additives. A single layer of graphene with little agglomeration is expected to exhibit high surface area and thus yield higher specific capacitance in a supercapacitor application. Graphene is also applied in other energy conversion and storage devices such as fuel cells and Li-ion batteries.10 Flexible electrodes based on carbonaceous nanomaterials can also improve such technologies as supercapacitors and Li-ion batteries.183 Gogotsi and Simon184 suggest that the most viable materials for electrochemical capacitors are biomass-derived and polymer-derived activated carbons. Hall et al.186 discuss how the use of metal oxides can improve electrochemical capacitor performance. The use of a combination of pseudo-capacitive nanomaterials, including oxides, nitrides and polymers, with the latest generation of nanostructured lithium electrodes for enhancing the energy density of electrochemical capacitors, allows them to perform more like batteries.187 The utilization of carbon nanotubes has further advanced micro-electrochemical capacitors, enabling flexible and adaptable devices.187 Ways to enhance the power rate in batteries so they are more comparable to those of supercapacitors are also proposed using a material with high lithium bulk mobility (LiFePO4) that can create a fast ion-conducting surface phase through controlled off-stoichiometry.188 Various design aspects of latent thermal energy storage technologies such as material, encapsulation, heat transfer, applications and new PCM technology innovation have been extensively reviewed.53,55,189 Experimental/computational efforts are made to enhance the thermal conductivity of PCMs.190 These include the placement of fixed, stationary high conductivity inserts made from copper, aluminum, nickel, stainless steel and carbon fiber in various forms (fins, honeycomb, wool, brush, etc.). Various development possibilities also exist for high-temperature thermal storage including embedding metal foams within PCMs. Mathematical modelling need to be used to predict how such options affect the stability and energy flux through the composite at high temperatures.191 Other TES techniques and their applications are discussed in Sections 2 and 3, respectively.

5.2

Storage duration: From short-term to seasonal

Energy storage systems can be classified based on storage period, as mentioned in Section 4. Short-term energy storage typically involves the storage of energy for hours to days, while long-term storage (also known as seasonal energy storage) refers to storage of energy from a few months to a season (3–6 months). For instance, a long-term thermal energy storage retains thermal energy in the ground over the summer for use in winter. Mahon et al.59 review four methods of seasonal energy storage: tank, pit, borehole, and aquifer. They suggest that these systems might become more poplar as district energy systems move towards lower distribution temperatures and more waste heat options become viable. Guerra et al.192 investigated the operational performance and cost benefit comparison of three seasonal energy storage types: pumped hydro, compressed air, and hydrogen storage, in the context of high shares of wind and solar photovoltaic power sources. While they do suggest that potential grid benefits must be considered in addition to the technology cost assessment when assessing cost effectiveness of energy storage technologies, from a technology cost perspective, they suggest that pumped hydro energy storage has the lowest cost for low discharge durations and hydrogen is the lowest cost technology for high discharge durations, due to its significantly lower energy storage capital costs compared to the other two technologies considered.

5.3

Economics

Various economic advantages and challenges exist regarding the use of energy storage technologies for the various applications included in Section 3. The cost of an energy storage system is often application-dependent. Carnegie et al.124 identify applications that energy storage devices serve and compare costs of storage devices for the applications. In addition, costs of an energy storage system for a given application vary notably based on location, construction method and size, and the cost effectiveness depends on the price of the source of energy such as natural gas. For example, Marean193 report capital costs of CAES systems for bulk energy storage applications based on various geologic formations: from $1/kWh for salt cavern (solution mined) to $30/kWh for hard rock (excavated and existing mines). For this reason, economic analyses comparing a wide range of energy technologies often have a degree of uncertainty, which needs to be taken into account. Nonetheless, estimated capital costs for various energy storage systems are listed in Table 4. Note that the costs listed are obtained from the literature that are published in different years. The costs of a number of energy storage technologies, that have not yet reached a mature development stage at the time of publication, are expected to be currently lower due to technology development and economies of scale. Examples of such technologies include hydrogen fuel cells, Li-ion batteries for utility and residential applications, and vanadium-redox flow batteries. On the other hand, pumped hydro and lead-acid batteries have reached a mature status and have exhibited smaller variations in cost over the past two decades.194 As a result, their current costs are not expected to vary much from the values listed in Table 4. The capital costs are reported in terms of cost per unit power output ($/kW) and cost per unit energy stored ($/kWh). In applications where energy is to

42

Energy, Environment, and Resources | Energy Storage

Table 4

Estimated capital cost ranges for various energy storage systems.

Storage type

Capital cost (power based) ($/kW) 173

Battery (various types)

600–2000 (2009) 500–4600195 2000–4000196 400–800134,173 500–1500195 400–1550196 300–3500195

Li-ion

1200–4000173 (2009)

Lead-acid

463–966196 300–600134,173

NiCd

615–750196 5000–1500173

NaS

1000–3000173

VRB Capacitor Supercapacitor

600–1500173 200–400173 100–300173 130–515195

Pumped hydro Compressed air

196

Magnetic

Flywheel

Hydrogen fuel cell Thermal Solid media (demand) Underground Molten salts (supply) Pit (supply) Ice (demand) Cold water (demand) Thermochemical (supply and demand)

a

286–331 200–300173 130–515195 1000–10,000134 303–761196 300–100073 250–350173 (2009) 130–500195 1950–2200124 (2013) 590–1446196 500–10,000134 2400–4700196 200–300173 500–3000195 3400–4500195 400–700195 100–300195 6000–15,000195 300–600195 1000–3000195

Capital cost (energy based) ($/kWh) 173

5–100 (2009) K+ > Na+ > Li+.69 Also, for the CO2RR, the yield of the different products is affected by the cation in the solution, which points out certain mechanistic roles of cations at the rds. The activity trend is the same as that found for the ORR.70,71 Assuming that a proton-coupled electron transfer (PCET) is the rds, the cation effect can be explained as a result of variations in the interfacial local pH. The pKa for cation hydrolysis decreases with increasing cation size. This means that larger M+ serves as buffering agents and lowers the pH at the cathode surface, thus increasing the concentration of dissolved CO2 at the interface and favoring the CO2RR over the HER. Another explanation for this activity tendency relies on the so-called field effect.72 The solvated M+ present at the outer Helmholtz plane (OHP) stabilizes the adsorbed intermediates via electrostatic interactions. A higher concentration of M+ is achieved with increasing cation size, which displays a way to modulate the stability of the reaction intermediates. Finally, water activity also modifies the reactivity. In solution, the change in the water activity requires the addition of a second solvent, which also interacts with the electrode, so it is difficult to elucidate which effect is determining the change in reactivity. In solid polymer electrolytes, the change in the water activity can be easily modified by the change in humidity in the thin layer of the polymer.73 Studies on model Pt(111) surfaces with different polymers and different humidity degrees indicate that low water activity increases oxygen reduction rates.

4

Conclusions

The electrocatalysis requires the specific interaction between some of the species participating in the reaction and the electrode surface, also known as electrocatalysts. As has been shown, for simple reactions, the rates for the electrocatalytic reactions depend on the adsorption energy of the relevant species with the surface, giving rise to the appearance of volcano curves. Although this model provides fruitful insights into the electrocatalysis of the HER/HOR and ORR/OER, its applicability to more complex reactions should be taken with care. Finally, it has been shown that the interfacial structure and properties fine-tune the reaction rates, providing ways to improve the electrocatalysis of the desired reactions.

References 1. Mc Naught, A. D.; Wilkinson, A. Compendium of Chemical Terminology-Gold Book; IUPAC, 2012; pp. 1–1670. https://doi.org/10.1351/goldbook. 2. Noftle, R. E.; Pletcher, D. An Interpretation of the Formal Potential for the Ferricyanide/Ferrocyanide Couple as a Function of Solvent Composition. J. Electroanal. Chem. Interfacial Electrochem. 1990, 293 (1–2), 273–277. https://doi.org/10.1016/0022-0728(90)80070-M. 3. Marcus, R. A. On the Theory of Oxidation-Reduction Reactions Involving Electron Transfer. I. J. Chem. Phys. 1956, 24 (5), 966–978. https://doi.org/10.1063/1.1742723. 4. Frumkin, A. Wasserstoffüberspannung und Struktur der Doppelschicht. Z. Phys. Chem. 1933, 164A (1), 121–133. https://doi.org/10.1515/zpch-1933-16411. 5. Capon, A.; Parson, R. The Oxidation of Formic Acid at Noble Metal Electrodes: I. Review of Previous Work. J. Electroanal. Chem. Interfacial Electrochem. 1973, 44 (1), 1–7. https://doi.org/10.1016/S0022-0728(73)80508-X. 6. Ferre-Vilaplana, A.; Perales-Rondón, J. V.; Buso-Rogero, C.; Feliu, J. M.; Herrero, E. Formic Acid Oxidation on Platinum Electrodes: A Detailed Mechanism Supported by Experiments and Calculations on Well-Defined Surfaces. J. Mater. Chem. A 2017, 5 (41), 21773–21784. https://doi.org/10.1039/C7TA07116G. 7. Brönsted, J. N.; Pedersen, K. Die Katalytische Zersetzung des Nitramids und Ihre Physikalisch-Chemische Bedeutung. Z. Phys. Chem. 1924, 108U (1), 185–235. https://doi.org/ 10.1515/ZPCH-1924-10814. 8. Exner, K. S. Is Thermodynamics a Good Descriptor for the Activity? Re-Investigation of Sabatier’s Principle by the Free Energy Diagram in Electrocatalysis. ACS Catal. 2019, 9 (6), 5320–5329. https://doi.org/10.1021/acscatal.9b00732. 9. Trasatti, S. Work Function, Electronegativity, and Electrochemical Behaviour of Metals: III. Electrolytic Hydrogen Evolution in Acid Solutions. J. Electroanal. Chem. Interfacial Electrochem. 1972, 39 (1), 163–184. https://doi.org/10.1016/S0022-0728(72)80485-6. 10. Ferre-Vilaplana, A.; Herrero, E. Why Nitrogen Favors Oxygen Reduction on Graphitic Materials. Sustain. Energy Fuel 2019, 3 (9). https://doi.org/10.1039/c9se00262f. 11. Nørskov, J. K.; Rossmeisl, J.; Logadottir, A.; Lindqvist, L.; Kitchin, J. R.; Bligaard, T.; Jónsson, H. Origin of the Overpotential for Oxygen Reduction at a Fuel-Cell Cathode. J. Phys. Chem. B 2004, 108 (46), 17886–17892. https://doi.org/10.1021/jp047349j. 12. Stephens, I. E. L.; Bondarenko, A. S.; Gronbjerg, U.; Rossmeisl, J.; Chorkendorff, I. Understanding the Electrocatalysis of Oxygen Reduction on Platinum and Its Alloys. Energy Environ. Sci. 2012, 5 (5), 6744–6762. https://doi.org/10.1039/c2ee03590a. 13. Man, I. C.; Su, H. Y.; Calle-Vallejo, F.; Hansen, H. A.; Martínez, J. I.; Inoglu, N. G.; Kitchin, J.; Jaramillo, T. F.; Nørskov, J. K.; Rossmeisl, J. Universality in Oxygen Evolution Electrocatalysis on Oxide Surfaces. ChemCatChem 2011, 3 (7), 1159–1165. https://doi.org/10.1002/CCTC.201000397. 14. Greeley, J.; Jaramillo, T. F.; Bonde, J.; Chorkendorff, I.; Nørskov, J. K. Computational High-Throughput Screening of Electrocatalytic Materials for Hydrogen Evolution. Nat. Mater. 2006, 5 (11), 909–913. https://doi.org/10.1038/nmat1752.

Electrochemical Fundamentals | Electrocatalysis

147

15. Wang, H. F.; Liu, Z. P. Formic Acid Oxidation at Pt/H2O Interface from Periodic DFT Calculations Integrated with a Continuum Solvation Model. J. Phys. Chem. C 2009, 113 (40), 17502–17508. https://doi.org/10.1021/JP9059888. 16. Li, Y.; Chan, S. H.; Sun, Q. Heterogeneous Catalytic Conversion of CO2: A Comprehensive Theoretical Review. Nanoscale 2015, 7 (19), 8663–8683. https://doi.org/10.1039/ C5NR00092K. 17. Shi, C.; Hansen, H. A.; Lausche, A. C.; Nørskov, J. K. Trends in Electrochemical CO2 Reduction Activity for Open and Close-Packed Metal Surfaces. Phys. Chem. Chem. Phys. 2014, 16 (10), 4720–4727. https://doi.org/10.1039/C3CP54822H. 18. Peterson, A. A.; Nørskov, J. K. Activity Descriptors for CO2 Electroreduction to Methane on Transition-Metal Catalysts. J. Phys. Chem. Lett. 2012, 3 (2), 251–258. https://doi.org/ 10.1021/jz201461p. 19. Bagger, A.; Ju, W.; Varela, A. S.; Strasser, P.; Rossmeisl, J. Electrochemical CO2 Reduction: A Classification Problem. ChemPhysChem 2017, 18 (22), 3266–3273. https://doi. org/10.1002/CPHC.201700736. 20. Ringe, S. The Importance of a Charge Transfer Descriptor for Screening Potential CO2 Reduction Electrocatalysts. Nat. Commun. 2023, 14 (1), 1–14. https://doi.org/10.1038/ s41467-023-37929-4. 21. Stratton, S. M.; Zhang, S.; Montemore, M. M. Addressing Complexity in Catalyst Design: From Volcanos and Scaling to More Sophisticated Design Strategies. Surf. Sci. Rep. 2023, 78 (3), 100597. https://doi.org/10.1016/J.SURFREP.2023.100597. 22. Nørskov, J. K.; Abild-Pedersen, F.; Studt, F.; Bligaard, T. Density Functional Theory in Surface Chemistry and Catalysis. Proc. Natl. Acad. Sci. 2011, 108 (3), 937–943. https:// doi.org/10.1073/pnas.1006652108. 23. Hammer, B.; Norskov, J. K. Why Gold Is the Noblest of All the Metals. Nature 1995, 376 (6537), 238–240. https://doi.org/10.1038/376238a0. 24. Gorzkowski, M. T.; Lewera, A. Probing the Limits of D-Band Center Theory: Electronic and Electrocatalytic Properties of Pd-Shell-Pt-Core Nanoparticles. J. Phys. Chem. C 2015, 119 (32), 18389–18395. https://doi.org/10.1021/acs.jpcc.5b05302. 25. Batchelor, T. A. A.; Pedersen, J. K.; Winther, S. H.; Castelli, I. E.; Jacobsen, K. W.; Rossmeisl, J. High-Entropy Alloys as a Discovery Platform for Electrocatalysis. Joule 2019, 3 (3), 834–845. https://doi.org/10.1016/j.joule.2018.12.015. 26. Xin, Y.; Li, S.; Qian, Y.; Zhu, W.; Yuan, H.; Jiang, P.; Guo, R.; Wang, L. High-Entropy Alloys as a Platform for Catalysis: Progress, Challenges, and Opportunities. ACS Catal. 2020, 11280–11306. https://doi.org/10.1021/acscatal.0c03617. 27. Stamenkovic, V. R.; Mun, B. S.; Mayrhofer, K. J. J.; Ross, P. N.; Markovic, N. M. Effect of Surface Composition on Electronic Structure, Stability, and Electrocatalytic Properties of Pt-Transition Metal Alloys: Pt-Skin versus Pt-Skeleton Surfaces. J. Am. Chem. Soc. 2006, 128 (27), 8813–8819. https://doi.org/10.1021/ja0600476. 28. Vidal-Iglesias, F. J.; Solla-Gullón, J.; Herrero, E.; Aldaz, A.; Feliu, J. M. Formic Acid Oxidation on Pd-Modified Pt(100) and Pt(111) Electrodes: A DEMS Study. J. Appl. Electrochem. 2006, 36 (11), 1207–1214. https://doi.org/10.1007/s10800-006-9179-y. 29. Vidal-Iglesias, F. J.; Solla-Gullón, J.; Herrero, E.; Aldaz, A.; Feliu, J. M. Pd Adatom Decorated (100) Preferentially Oriented Pt Nanoparticles for Formic Acid Electrooxidation. Angew. Chem. Int. Ed. 2010, 49 (39), 6998–7001. https://doi.org/10.1002/anie.201002501. 30. Lu, G. Q.; Waszczuk, P.; Wieckowski, A. Oxidation of CO Adsorbed from CO Saturated Solutions on the Pt(111)/Ru Electrode. J. Electroanal. Chem. 2002, 532 (1–2), 49–55. https://doi.org/10.1016/S0022-0728(02)00952-X. 31. Maillard, F.; Lu, G. Q.; Wieckowski, A.; Stimming, U. Ru-Decorated Pt Surfaces as Model Fuel Cell Electrocatalysts for CO Electrooxidation. J. Phys. Chem. B 2005, 109 (34), 16230–16243. https://doi.org/10.1021/jp052277x. 32. Perales-Rondón, J. V.; Ferre-Vilaplana, A.; Feliu, J. M.; Herrero, E. Oxidation Mechanism of Formic Acid on the Bismuth Adatom-Modified Pt (111) Surface. J. Am. Chem. Soc. 2014, 136 (38), 13110–13113. https://doi.org/10.1021/ja505943h. 33. Ferre-Vilaplana, A.; Perales-Rondón, J. V.; Feliu, J. M.; Herrero, E. Understanding the Effect of the Adatoms in the Formic Acid Oxidation Mechanism on Pt(111) Electrodes. ACS Catal. 2015, 5 (2), 645–654. https://doi.org/10.1021/cs501729j. 34. Leiva, E.; Iwasita, T.; Herrero, E.; Feliu, J. M. Effect of Adatoms in the Electrocatalysis of HCOOH Oxidation. A Theoretical Model. Langmuir 1997, 13 (23), 6287–6293. https:// doi.org/10.1021/la970535e. 35. Cuesta, A.; Escudero, M.; Lanova, B.; Baltruschat, H. Cyclic Voltammetry, FTIRS, and DEMS Study of the Electrooxidation of Carbon Monoxide, Formic Acid, and Methanol on Cyanide-Modified Pt(111) Electrodes. Langmuir 2009, 25 (11), 6500–6507. https://doi.org/10.1021/la8041154. 36. Briega-Martos, V.; Ferre-Vilaplana, A.; Herrero, E.; Feliu, J. M. Why the Activity of the Hydrogen Oxidation Reaction on Platinum Decreases as pH Increases. Electrochim. Acta 2020, 354, 136620. https://doi.org/10.1016/J.ELECTACTA.2020.136620. 37. Schouten, K. J. P.; van der Niet, M.; Koper, M. T. M. Impedance Spectroscopy of H and OH Adsorption on Stepped Single-Crystal Platinum Electrodes in Alkaline and Acidic Media. Phys. Chem. Chem. Phys. 2010, 12 (46), 15217–15224. https://doi.org/10.1039/c0cp00104j. 38. Rizo, R.; Sitta, E.; Herrero, E.; Climent, V.; Feliu, J. M. Towards the Understanding of the Interfacial pH Scale at Pt(111) Electrodes. Electrochim. Acta 2015, 162, 138–145. https://doi.org/10.1016/j.electacta.2015.01.069. 39. Garcia-Araez, N.; Climent, V.; Feliu, J. M. Temperature Effects on Platinum Single Crystal/Aqueous Solution Interphases. Combining Gibbs Thermodynamics with Laser-Pulsed Experiments. In Interfacial Phenomena in Electrocatalysis; Vayenas, C. G., Ed.; Modern Aspects of Electrochemistry, vol. 51, White, R. E., Vayenas, C. G., Eds.; Springer: New York, 2011; pp. 1–105. 40. Sarabia, F. J.; Sebastián, P.; Climent, V.; Feliu, J. M. New Insights into the Pt (hkl)-Alkaline Solution Interphases from the Laser Induced Temperature Jump Method. J. Electroanal. Chem. 2020, 872, 114068. https://doi.org/10.1016/j.jelechem.2020.114068. 41. Ledezma-Yanez, I.; Wallace, W. D. Z.; Sebastian-Pascual, P.; Climent, V.; Feliu, J. M.; Koper, M. T. M. Interfacial Water Reorganization as a pH-Dependent Descriptor of the Hydrogen Evolution Rate on Platinum Electrodes. Nat. Energy 2017, 2 (4). https://doi.org/10.1038/nenergy.2017.31. 42. Subbaraman, R.; Tripkovic, D.; Strmcnik, D.; Chang, K. C.; Uchimura, M.; Paulikas, A. P.; Stamenkovic, V.; Markovic, N. M. Enhancing Hydrogen Evolution Activity in Water Splitting by Tailoring Li+-Ni(OH)2-Pt Interfaces. Science (1979) 2011, 334 (6060), 1256–1260. https://doi.org/10.1126/SCIENCE.1211934. 43. Prieto, A.; Hernández, J.; Herrero, E.; Feliu, J. M. The Role of Anions in Oxygen Reduction in Neutral and Basic Media on Gold Single-Crystal Electrodes. J. Solid State Electrochem. 2003, 7 (9), 599–606. https://doi.org/10.1007/s10008-003-0362-3. 44. Briega-Martos, V.; Herrero, E.; Feliu, J. M. Recent Progress on Oxygen and Hydrogen Peroxide Reduction Reactions on Pt Single Crystal Electrodes. Chin. J. Catal. 2020, 41 (5), 732–738. https://doi.org/10.1016/S1872-2067(19)63325-5. 45. Briega-Martos, V.; Herrero, E.; Feliu, J. M. The Inhibition of Hydrogen Peroxide Reduction at Low Potentials on Pt(111): Hydrogen Adsorption or Interfacial Charge? Electrochem. Commun. 2017, 85, 32–35. https://doi.org/10.1016/j.elecom.2017.10.016. 46. Rizo, R.; Herrero, E.; Feliu, J. M. Oxygen Reduction Reaction on Stepped Platinum Surfaces in Alkaline Media. Phys. Chem. Chem. Phys. 2013, 15 (37), 15416–15425. https:// doi.org/10.1039/c3cp51642c. 47. Sebastián-Pascual, P.; Petersen, A. S.; Bagger, A.; Rossmeisl, J.; Escudero-Escribano, M. pH and Anion Effects on Cu–Phosphate Interfaces for CO Electroreduction. ACS Catal. 2021, 11, 1128–1135. https://doi.org/10.1021/acscatal.0c03998. 48. Sebastián-Pascual, P.; Sarabia, F. J.; Climent, V.; Feliu, J. M.; Escudero-Escribano, M. Elucidating the Structure of the Cu-Alkaline Electrochemical Interface with the Laser-Induced Temperature Jump Method. J. Phys. Chem. C 2020, 124 (42), 23253–23259. https://doi.org/10.1021/acs.jpcc.0c07821. 49. Trasatti, S. Work Function, Electronegativity, and Electrochemical Behaviour of Metals. II. Potentials of Zero Charge and “Electrochemical” Work Functions. J. Electroanal. Chem. 1971, 33 (2), 351–378. https://doi.org/10.1016/S0022-0728(71)80123-7. 50. Orts, J. M.; Gómez, R.; Feliu, J. M.; Aldaz, A.; Clavilier, J. Potentiostatic Charge Displacement by Exchanging Adsorbed Species on Pt(111) Electrodes—Acidic Electrolytes with Specific Anion Adsorption. Electrochim. Acta 1994, 39 (11–12), 1519–1524. https://doi.org/10.1016/0013-4686(94)85129-8.

148

Electrochemical Fundamentals | Electrocatalysis

51. Herrero, E.; Franaszczuk, K.; Wieckowski, A. Electrochemistry of Methanol at Low Index Crystal Planes of Platinum. An Integrated Voltammetric and Chronoamperometric Study. J. Phys. Chem. 1994, 98 (19), 5074–5083. https://doi.org/10.1021/j100070a022. 52. Maciá, M. D.; Campiña, J. M.; Herrero, E.; Feliu, J. M. On the Kinetics of Oxygen Reduction on Platinum Stepped Surfaces in Acidic Media. J. Electroanal. Chem. 2004, 564 (1–2), 141–150. https://doi.org/10.1016/j.jelechem.2003.09.035. 53. Colmati, F.; Tremiliosi-Filho, G.; Gonzalez, E. R.; Berná, A.; Herrero, E.; Feliu, J. M. Surface Structure Effects on the Electrochemical Oxidation of Ethanol on Platinum Single Crystal Electrodes. Faraday Discuss. 2008, 140, 379–397. https://doi.org/10.1039/b802160k. discussion 417–37. 54. Perales-Rondón, J. V.; Herrero, E.; Feliu, J. M. Effects of the Anion Adsorption and pH on the Formic Acid Oxidation Reaction on Pt(111) Electrodes. Electrochim. Acta 2014, 140, 511–517. https://doi.org/10.1016/j.electacta.2014.06.057. 55. Hara, M.; Linke, U.; Wandlowski, T. Preparation and Electrochemical Characterization of Palladium Single Crystal Electrodes in 0.1 M H2SO4 and HClO4 Part I. Low-Index Phases. Electrochim. Acta 2007, 52 (18), 5733–5748. https://doi.org/10.1016/j.electacta.2006.11.048. 56. Arán-Ais, R. M.; Scholten, F.; Kunze, S.; Rizo, R.; Roldan Cuenya, B. The Role of in Situ Generated Morphological Motifs and Cu(i) Species in C2+ Product Selectivity During CO2 Pulsed Electroreduction. Nat. Energy 2020, 5 (4), 317–325. https://doi.org/10.1038/s41560-020-0594-9. 57. Climent, V.; Gómez, R.; Orts, J. M.; Feliu, J. M. Thermodynamic Analysis of the Temperature Dependence of OH Adsorption on Pt(111) and Pt(100) Electrodes in Acidic Media in the Absence of Specific Anion Adsorption. J. Phys. Chem. B 2006, 110 (23), 11344–11351. https://doi.org/10.1021/jp054948x. 58. Rizo, R.; Fernández-Vidal, J.; Hardwick, L. J.; Attard, G. A.; Vidal-Iglesias, F. J.; Climent, V.; Herrero, E.; Feliu, J. M. Investigating the Presence of Adsorbed Species on Pt Steps at Low Potentials. Nat. Commun. 2022, 13 (1), 2550. https://doi.org/10.1038/s41467-022-30241-7. 59. Mekazni, D. S.; Arán-Ais, R. M.; Ferre-Vilaplana, A.; Herrero, E. Why Methanol Electro-Oxidation on Platinum in Water Takes Place Only in the Presence of Adsorbed OH. ACS Catal. 2022, 12 (3), 1965–1970. https://doi.org/10.1021/acscatal.1c05122. 60. Mekazni, D. S.; Arán-Ais, R. M.; Herrero, E.; Feliu, J. M. On the Oxidation of Isopropanol on Platinum Single Crystal Electrodes. A Detailed Voltammetric and FTIR Study. J. Power Sources 2023, 556 (5), 232396. https://doi.org/10.1016/j.jpowsour.2022.232396. 61. Rizo, R.; Ferre-Vilaplana, A.; Herrero, E.; Feliu, J. M. Ethanol Electro-Oxidation Reaction Selectivity on Platinum in Aqueous Media. ACS Sustain. Chem. Eng. 2022, 11 (13), 4960–4968. https://doi.org/10.1021/acssuschemeng.2c02663. 62. Grozovski, V.; Climent, V.; Herrero, E.; Feliu, J. M. Intrinsic Activity and Poisoning Rate for HCOOH Oxidation at Pt(100) and Vicinal Surfaces Containing Monoatomic (111) Steps. ChemPhysChem 2009, 10 (11), 1922–1926. https://doi.org/10.1002/cphc.200900261. 63. Grozovski, V.; Climent, V.; Herrero, E.; Feliu, J. M. Intrinsic Activity and Poisoning Rate for HCOOH Oxidation on Platinum Stepped Surfaces. Phys. Chem. Chem. Phys. 2010, 12 (31), 8822. https://doi.org/10.1039/b925472b. 64. Herrero, E.; Feliu, J. M. Understanding Formic Acid Oxidation Mechanism on Platinum Single Crystal Electrodes. Curr. Opin. Electrochem. 2018, 9, 145–150. https://doi.org/ 10.1016/j.coelec.2018.03.010. 65. Yang, Y.; Agarwal, R. G.; Hutchison, P.; Rizo, R.; Soudackov, A. V.; Lu, X.; Herrero, E.; Feliu, J. M.; Hammes-Schiffer, S.; Mayer, J. M.; Abruña, H. D. Inverse Kinetic Isotope Effects in the Oxygen Reduction Reaction at Platinum Single Crystals. Nat. Chem. 2023, 15, 271–277. https://doi.org/10.1038/s41557-022-01084-y. 66. Climent, V.; Garcia-Araez, N.; Feliu, J. M. Influence of Alkali Cations on the Infrared Spectra of Adsorbed (Bi)Sulphate on Pt(111) Electrodes. Electrochem. Commun. 2006, 8 (10), 1577–1582. https://doi.org/10.1016/j.elecom.2006.07.027. 67. Garcia-Araez, N.; Climent, V.; Feliu, J. M. Evidence of Water Reorientation on Model Electrocatalytic Surfaces from Nanosecond-Laser-Pulsed Experiments. J. Am. Chem. Soc. 2008, 130 (12), 3824–3833. https://doi.org/10.1021/ja0761481. 68. Chen, X.; McCrum, I. T.; Schwarz, K. A.; Janik, M. J.; Koper, M. T. M. Co-Adsorption of Cations as the Cause of the Apparent pH Dependence of Hydrogen Adsorption on a Stepped Platinum Single-Crystal Electrode. Angew. Chem. Int. Ed. 2017, 56 (47), 15025–15029. https://doi.org/10.1002/anie.201709455. 69. Strmcnik, D.; Kodama, K.; van der Vliet, D.; Greeley, J.; Stamenkovic, V. R.; Markovic, N. M. The Role of Non-Covalent Interactions in Electrocatalytic Fuel-Cell Reactions on Platinum. Nat. Chem. 2009, 1 (6), 466–472. https://doi.org/10.1038/nchem.330. 70. Resasco, J.; Chen, L. D.; Clark, E.; Tsai, C.; Hahn, C.; Jaramillo, T. F.; Chan, K.; Bell, A. T. Promoter Effects of Alkali Metal Cations on the Electrochemical Reduction of Carbon Dioxide. J. Am. Chem. Soc. 2017, 139 (32), 11277–11287. https://doi.org/10.1021/jacs.7b06765. 71. Singh, M. R.; Kwon, Y.; Lum, Y.; Ager, J. W.; Bell, A. T. Hydrolysis of Electrolyte Cations Enhances the Electrochemical Reduction of CO2 over Ag and Cu. J. Am. Chem. Soc. 2016, 138 (39), 13006–13012. https://doi.org/10.1021/jacs.6b07612. 72. Ringe, S.; Clark, E. L.; Resasco, J.; Walton, A.; Seger, B.; Bell, A. T.; Chan, K. Understanding Cation Effects in Electrochemical CO2 Reduction. Energy Environ. Sci. 2019, 12 (10), 3001–3014. https://doi.org/10.1039/C9EE01341E. 73. Kunimatsu, K.; Bae, B.; Miyatake, K.; Uchida, H.; Watanabe, M. ATR-FTIR Study of Water in Nafion Membrane Combined with Proton Conductivity Measurements during Hydration/Dehydration Cycle. J. Phys. Chem. B 2011, 115 (15), 4315–4321. https://doi.org/10.1021/jp112300c.

Electrochemical Fundamentals | Hydrogen Evolution and Oxidation Peter Kurzweila and S Trasattib,⁎, aTechnical University of Applied Sciences (OTH), Amberg-Weiden, Germany; bUniversity of Milan, Milan, Italy © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of S. Trasatti, ELECTROCHEMICAL THEORY | Hydrogen Evolution, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 41–48, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00022-8.

1 2 3 4 5 6 6.1 6.2 7 7.1 7.2 8 9 10 References

Introduction Thermodynamic aspects Hydrogen adsorption Hydrogen absorption Electrode poisoning Reaction mechanisms Rate-determining steps Tafel slopes Electrocatalysis Activation energy and bond strength Volcano curve Materials for cathodes Factors of electrocatalysis Conclusion

150 150 151 152 152 152 152 153 154 154 155 156 156 157 157

Abstract The hydrogen evolution reaction is of strategic importance in various fields of electrochemistry. Adsorption of hydrogen atoms is a necessary step that depends on the interaction of metal surfaces with solvents. Thus, a correlation is found between M–H bond strength and hydrophilicity of metal surfaces. This point is discussed with the help of voltammetric curves for platinum-group metals. Besides being adsorbed, hydrogen atoms can also penetrate beneath the surface, thus affecting the adsorption energy on the surface. Hydrogen evolution is often accompanied by cathodic poisoning owing to the presence of metallic impurities. Ways to minimize this are outlined. The three most popular mechanisms of the electrode reaction for hydrogen liberation are discussed and kinetic parameters reported. The effect of the coverage of the electrode with the intermediate (Had) is illustrated in detail. The effect of the nature of the electrode material (in electrocatalysis) is discussed using the so-called volcano curves depicting the dependence of the activity on M–H bond strength. A detailed analysis of electronic and geometric factors in electrocatalysis is also carried out. Finally, the findings of a survey of the most active materials in acidic and alkaline solutions are presented.

Key points

• • •

Hydrogen evolution reaction: importance, reversibility, mechanisms, poisoning Voltammetric study of the adsorption of hydrogen atoms on platinum-group metals Volcano curves in electrocatalysis

Nomenclature

Symbols and Units b E E6¼ E0

Tafel slope (mV) Electrode potential (V) Activation energy (J mol−1) Standard electrode potential (V)

⁎

Posthumous.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00145-6

149

150

Electrochemical Fundamentals | Hydrogen Evolution and Oxidation

EH EO Had j R T z a d DG u

Potential of incipient adsorption of hydrogen (J mol−1) Potential of incipient adsorption of oxygen (J mol−1) Adsorbed hydrogen Current density (A m−2) Ohmic resistance (O) Temperature (K) Stoichiometric number Transfer coefficient Thickness of the diffusion layer (m) Gibbs free energy change (J mol−1) Surface occupancy (0. . .1)

Abbreviations rds RHE SHE UPD

1

Rate-determining step Reversible hydrogen electrode Standard hydrogen electrode Underpotential deposition

Introduction

The evolution of hydrogen gas at the cathode is one of the reactions most frequently occurring in industrial cells.1–9 Besides the obvious case of water electrolysis, hydrogen evolution occurs at the cathode of chlor-alkali and chlorate cells, which are among the most intensive electrochemical processes. On the other hand, hydrogen liberation is the cathodic process in cells containing aqueous solutions or protonated solvents where the wanted reaction takes place at the anode. Also, formation of the gaseous hydrogen accompanies free-metal corrosion in an acidic aqueous environment (e.g., zinc dissolution in acids), as well as the application of anodic protection technologies. Hydrogen has acquired strategic importance as an energy vector in fuel cells at a time when pollution from the combustion of fossil fuels has become quite crucial. The combination of electrolytic cells and renewable energy sources would constitute the cleanest technology to produce hydrogen of the highest purity, which is stringently required in H2–O2 fuel cells, for avoiding poisoning of electrodes.

2

Thermodynamic aspects

The electrode reaction of hydrogen formation involves two electrons per mole of the product. In acid solutions, H+, or still better H3O+, is the reacting particle: 2 H+ + 2 e− ! 2 H2 with E0 ¼ 0 V (SHE, standard hydrogen electrode), while in alkaline solution the reacting particle is H2O: 2 H2 O + 2 e− ! H2 + 2 OH− with E0 ¼ −0.828 V (SHE). The above reactions show that the liberation of hydrogen is accompanied by alkalinization of the solution near the cathode. This is precisely the process that produces concentrated sodium hydroxide solutions in the cathodic compartment of chlor-alkali membrane cells. With older mercury cells, hydrogen is not a primary product of the process as H+ discharge is so much retarded on a mercury surface that Na+ is reduced first, thus forming sodium amalgam. The decomposition of Hg(Na) in water then produces sodium hydroxide with liberation of hydrogen. Hydrogen formation is catalytically a facile reaction. This implies that the above reactions exhibit a high degree of reversibility, so much that a reversible hydrogen electrode can easily be realized. This does not mean that every material is suitable for hydrogen evolution or ionization. However, with some specific materials, notably metals of the platinum group, reversibility is easily observed experimentally. But it can be fully realized only with platinum. A necessary condition is that hydrogen atoms interact directly with the bare metal surface on which a sizable covering of the intermediate must exist at the equilibrium. Hydrogen liberation as a rule does not show diffusion limitations. While this seems obvious in alkaline solutions, thanks to the high population of reacting particles (water) it can be explained in acids by the specific mechanism (Grotthuss) of H+ migration in

Electrochemical Fundamentals | Hydrogen Evolution and Oxidation

151

aqueous solution that exalts the diffusion rate of H+ by at least one order of magnitude and allows water molecules to act as bridges between the H+ coming from the bulk of the solution and the H+ discharged on the electrode surface.

3

Hydrogen adsorption

Some of the metals of the platinum group can dissociate hydrogen in solution, thus taking the reversible potential of the hydrogen electrode. This is particularly the case for platinum. Dissociating hydrogen molecules implies that the PtdH bond strength is higher than the HdH bond strength. On the other hand, it also implies that PtdOH2 interactions are relatively minor in the potential range of hydrogen adsorption. Actually, adsorption from solution on a solid surface is to be regarded as a solvent replacement reaction: SH2 O + B ! SB + H2 O where S is a surface site and B an adsorbed species. This implies that, as the S–H2O interaction is stronger than the S–B interaction, B is unable to displace water and hence not adsorbed. The above rule holds in particular as ionic or nonionic molecules (e.g., organic molecules) occupy the surface site. Hydrogen/water competition is intrinsic to aqueous solutions. Metal surfaces as a rule interact with water molecules through the oxygen atoms. One would expect that the stronger the metal–water interaction, the weaker the M–H bond strength. This is actually what is observed experimentally in a voltammetric curve. Fig. 1 shows the typical cyclic voltammetric curve of a platinum electrode in an acidic solution, and EH and EO are the potentials of incipient adsorption of hydrogen and oxygen, respectively. The fact that EH is more positive than E0(H+/H2) indicates that the PtdH bond is stronger than the HdH bond. On the other hand, the position of EO can be associated with the magnitude of the affinity of the metal surface for oxygen and therefore also for water (via the oxygen atom). If voltammetric curves for different metals of the platinum group are collated, EO, for instance, of ruthenium is observed to be less positive than that of platinum. This implies that the RudO bond is stronger than the PtdO bond. At the same time, EH of ruthenium is also observed to be less positive than that of platinum, that is, the RudH bond is weaker than the PtdH bond. Since experiments in the gas phase unambiguously show that the hydrogen adsorption heat on ruthenium is higher than that on platinum, it is to be concluded that the M–H bond in solution is weakened by the process of displacement of water molecules from the electrode surface. The amount of atomic hydrogen covering the electrode surface at the equilibrium is also determined by the competition described above. Only for platinum is the electrode surface practically saturated with the intermediate. Adsorption of hydrogen in a potential range more positive than the thermodynamic potential is called underpotential deposition (UPD). The M–H bond strength however is a function of hydrogen coverage (yH). More specifically, it decreases with increasing yH due to lateral interaction between adsorbed species. Thus, on the surface of a metal electrode, there exists strongly adsorbed as well as weakly adsorbed hydrogen. The situation with other metals is more complex. Gold does not show any voltammetric peak of hydrogen adsorption. This is in line with the poor ability to dissociate hydrogen in the gas phase. Other transition metals do not show adsorption peaks simply because their surface is covered by an oxide film inhibiting H+ discharge (e.g., titanium, tantalum, and molybdenum). Nickel, cobalt, and iron do show some hydrogen UPD that however depends on the experimental conditions since hydrogen and oxygen

Fig. 1 Typical current–potential (voltammetric) curve for a polycrystalline Pt electrode in acidic aqueous solution. EH and EO are the potentials of incipient adsorption of hydrogen and oxygen, respectively. E0 is the thermodynamic potential for the indicated electrode reaction. RHE, reversible hydrogen electrode.

152

Electrochemical Fundamentals | Hydrogen Evolution and Oxidation

adsorptions overlap. Finally, sp metals (tin, lead, cadmium, etc.) do not show any hydrogen UPD because of competition with water, but rather because their M–H bond is weaker than the HdH bond.

4

Hydrogen absorption

As atomic hydrogen is adsorbed on a metal surface, it can penetrate into the lattice (absorption) depending on the particular conditions. Hydrogen penetration is tantamount to the formation of a metal hydride. Owing to their small size, hydrogen atoms can diffuse interstitially and do not require a place exchange mechanism as in the case of oxygen penetration. As a consequence, hydrogen atoms are relatively free to move around in the lattice and can recombine, forming hydrogen molecules in the mass of the solid. The onset of an internal pressure can undermine the stability of the metal lattice. The outcome is the phenomenon of hydrogen embrittlement, well known in corrosion, that leads to serious degradation and even to the collapse of the metal structure. Materials prone to embrittlement are, for instance, iron, steel, and titanium. Hydrogen absorption, even if it does not produce embrittlement, affects the adsorption properties of the metal surface. In particular, subsurface hydrogen decreases the M–H bond strength of the metal surface. This is, for instance, the case for nickel whose adsorption capacity decreases after it has absorbed a certain amount of hydrogen. It is presumed that interstitial hydrogen saturates part of the bonding capacity of nickel atoms, which thus show a lower adsorption affinity toward hydrogen on the external surface. This phenomenon of hydrogen absorption is presumably responsible for the time-dependent properties of some electrodes under hydrogen evolution conditions.

5

Electrode poisoning

While poisoning of anodes is rare on account of the strongly oxidizing conditions existing on their surfaces, poisoning of cathodes is routine in technology.

• •

Organic impurities in solutions become adsorbed on metal surfaces in the potential range of hydrogen evolution, thus blocking part of the surface sites where H+ ions discharge. This can decrease the activity of the electrode and can also change the reaction mechanism. Metallic impurities, always present in technological cells, are even more dangerous as they can be reduced to metals on the electrode surface resulting in impurity accumulation with time. The final effect in this case depends on the relative properties of the electrode and the metallic impurity. As a rule, iron ions are present in solutions due to corrosion of steel cases and tubes. The deposition of iron on the cathode will be highly deleterious if the material of the cathode is very active (e.g., platinum). If the material of the cathode is a poorly active metal (e.g., titanium), the deposition of iron can even result in electrode activation in the context of hydrogen evolution.

The most effective way to protect cathodes against poisoning is to work with highly porous and/or rough electrodes. The effectiveness of poisoning depends on the extent of surface coverage with impurities. If the ratio of real surface area to apparent surface area is increased, the same amount of impurities will be diluted on a wider surface, thus becoming less effective. The case of platinized platinum (platinum black) is typical. The real surface area of such an electrode can be up to three orders of magnitude higher than that of smooth platinum. Thus, in the same solution with impurities, the time of deactivation of platinized platinum is up to 1000 times longer. Bubbles of hydrogen gas forming on an electrode surface can be the origin of deactivation. If bubbles stick onto the electrode surface, part of the surface will become inaccessible to reactants. At the same time, a curtain of gas bubbles at the electrode surface can give rise to ohmic drop effects, resulting in the distortion of measured kinetic parameters. On the other hand, the movement of gas bubbles in the laminar layer of solution at the electrode surface may enhance mass transfer in situ, resulting in an increase of current. The accumulation of hydrogen gas in solution at high current densities can even result in supersaturation phenomenon, which is tantamount to locally shifting the overpotential of the reaction with important modifications of the Tafel line in the given current density range.

6

Reaction mechanisms

Hydrogen evolution is a nondemanding (facile) reaction. This is because the intermediates are monoatomic species, and the bonds formed or broken during the reaction are of moderate strength. This implies that moderate activation energies are involved.

6.1

Rate-determining steps

While tens of mechanisms have been proposed for oxygen evolution, only two two-step mechanisms are proposed for hydrogen evolution, with the exception of the introduction of some variants in very few specific cases.

Electrochemical Fundamentals | Hydrogen Evolution and Oxidation

153

The first step is always and necessarily the discharge of H+ (H3O+) or of H2O molecules: H+ + e− ! Had H2 O + e− ! Had + OH− Both reactions show a decrease in acidity and hence an increase in alkalinity of the solution near the electrode surface. The intermediate is a hydrogen atom adsorbed on the electrode surface. This step has not only a phenomenological name (primary discharge), but also a historical name (Volmer reaction). The latter nomenclature is becoming obsolete. (a) Adsorbed hydrogen atoms can follow two alternative pathways depending on the nature of the electrode. These hydrogen atoms can react with further discharging H+ ions producing hydrogen molecules Had + H+ + e− ! H2 Had + H2 O + e− ! H2 + OH− for acid and alkaline solutions, respectively. The above step is known as the ‘ion + atom’ reaction or the Heyrovsky reaction. (b) Alternatively, hydrogen atoms can recombine to give hydrogen molecules, which is an entirely chemical step. This reaction is known as recombination reaction, or Tafel reaction. Had + Had ! H2 : From all the above, it is evident that combination of the various steps can result in only two independent mechanisms, where z is the stoichiometric number:

A. Electrochemical desorption (EE) B. Chemical desorption C. Spillover (ECE)

H+ + e− ! Had Had + H+ + e− ! H2 H+ + e− ! Had Had + Had + e− ! H2 H+ + e− ! H∗ad H∗ad ! Had Had + H+ + e− ! H2

z

b (mV)

1 1 2 1 1 1 1

120 40 ! 120 120 30 ! 1 120 60 40 ! 120

E E E C E C E

Although the first step is the same in both mechanisms, the number of exchanged electrons z can distinguish between the two. Since the hydrogen reaction is highly reversible, determination of z is experimentally possible; b is the Tafel slope, assuming a free surface (yH  0), that is, the intermediate is adsorbed under Langmuir conditions. E and C indicate the nature of the step (electrochemical and chemical, respectively). If the second step (Had removal) is rate determining, the intermediate accumulates on the electrode surface and can eventually saturate it at high current densities. With a fully covered surface (yH ! 1), Tafel slope evolves to the values indicated by the arrow. Thus, in mechanism A, in the second step, H+ is discharged on a surface covered by a carpet of hydrogen atoms. Kinetically, this is exactly equivalent to H+ discharging on a bare metal surface. As a matter of fact, electrochemistry is kinetically insensitive to the nature of the electrode surface, but sensitive only to the sequence of reaction steps. Therefore, discharging H+ on a bare surface or on a Had-covered surface is kinetically equivalent, and the Tafel slope turns out to be the same. Thus, to distinguish which step is governing the mechanism, it would be necessary to probe the electrode surface to determine yH. In mechanism B, if the second step is rate determining and the surface is saturated with yH, Had can be removed only by chemical recombination, whose rate depends only on the surface concentration of Had and not on the electrode potential. Thus, a limiting reaction rate is attained and the current does not increase with overpotential any more (b ! 1).

6.2

Tafel slopes

Tafel slopes of 120 mV are as a rule observed with poorly active electrodes, for instance, lead and mercury, while 30 mV is an indication of very active materials, for instance, platinum. On most active metals, mechanism A is preferred. Sometimes, Tafel lines with two slopes are observed, especially 40 ! 120 mV. As discussed above, this may indicate a change in yH in mechanism A, but it may also indicate change in rate-determining step (rds) of the same mechanism. In other words, at low current densities, electrochemical desorption is rate determining, while at high current densities, the primary discharge becomes rate determining. Although the way the two Tafel lines are joined can give a hint as to which case is the actual one, the most direct way to solve the problem is to probe the surface coverage for yH. In very few cases, a Tafel slope of 60 mV has been claimed. Kinetically, two variants can be proposed. First, a chemical step after the primary discharge determines the reaction rate. A possible mechanism C is, where Had is an unstable adsorbed intermediate

154

Electrochemical Fundamentals | Hydrogen Evolution and Oxidation

that migrates to a more stable position on the electrode surface (spillover by surface diffusion). Further reduction of H+ occurs only with Had and not with Had . Therefore, the second step becomes rate determining. If the rate of spillover is not affected by the electrode potential, the current can reach a limiting value with Had ! 1. Alternatively, a Tafel slope of 60 mV is obtained if, in any of the above mechanisms, the primary discharge proceeds barrierless (transfer coefficient, a ¼ 1) and is rate determining. This implies that the electrode surface adsorbs hydrogen very weakly so that the energy curve of the intermediates lies in a very high position with respect to reactants. This may lead to a barrier-free situation. It is interesting that such a situation has been claimed for hydrogen evolution on silver electrodes at very small current densities. Adsorption of hydrogen on the surface of silver is in fact very weak.

7

Electrocatalysis

It is an experimental fact that the rate of hydrogen evolution depends on the nature of the electrode material even after allowance for surface area effects. Here we enter the realm of electrocatalysis. Since electrochemistry possesses an additional degree of freedom (the electrical state of the interphase), electrocatalytic activity is assessed by comparing the reaction rates at constant electrode potential.

7.1

Activation energy and bond strength

Hydrogen evolution is one of the few electrode processes for which a theory of electrocatalysis has been fully developed. This could be because of the simplicity of the reaction steps and of the reaction intermediates. A seminal paper by Roger Parsons on this topic dates back to 1958,7 although Horiuti and Polanyi pointed out a first idea in that direction in 1932. Fig. 2 illustrates how the activation energies (E6¼) of the steps of the hydrogen evolution reaction depend on the M–H bond strength. In particular, for the general sequence, 2 H+ + 2 e− ! 2 Had ! H2 Increase in hydrogen adsorption decreases the activation energy of the primary discharge while increasing that of the desorption step. The M–H bond strength will bear by a fraction a on E6¼1 and by (1 − a) on E6¼2. If we confine our analysis to the primary discharge, it ensues that current density (ln j) at constant electrode potential E will increase in a way proportional to [a DG(M–H)/RT]. Thus, this argument would predict a linear increase of ln j with increasing change of free enthalpy DG(M–H). It is however necessary to consider another parameter: the coverage with the intermediate (atomic hydrogen). According to concepts of adsorption isotherms, the coverage with the intermediate will depend on DG(M–H), increasing as DG(M–H) increases. In turn, yH is related to DG(M–H) by an adsorption isotherm, increasing exponentially with DG(M–H). The primary discharge of H+ takes place on the sites of the electrode surface free from adsorbed intermediates. Thus, j  (1 − yH). Since yH  exp [DG(M–H)/RT], it follows that ln j  ln(1 − yH) ∝ − [DG(M–H)/RT]. Globally, the reaction rate depends on two

Fig. 2 (A) Energy curves for the indicated reaction path. The curve A is the energy curve of H not stabilized by adsorption on a metal surface. The curve B, DGH, is the free energy gain of H as it is stabilized by adsorption. (B) Graphical presentation of the relationship between electrocatalytic activity (ln j) and free energy of adsorption of the intermediate resulting from the analysis in (A).

Electrochemical Fundamentals | Hydrogen Evolution and Oxidation

155

competing factors: the activation energy decreasing with increasing adsorption energy, and the coverage with the reaction intermediate increasing with increasing adsorption energy. The situation is illustrated in Fig. 2B. Since the two factors operate simultaneously, the reaction rate will be governed by the one that determines the slower reaction rate. The outcome is that ln j depends on the M–H bond strength according to a so-called volcano-shaped curve. On the left-hand side, ln j is low because a weak M–H bond does not depress the activation energy for H+ discharge sizably. On the right-hand side, ln j is again low because the reaction is inhibited by the self-poisoning of the electrode surface with adsorbed intermediates (hydrogen atoms).

7.2

Volcano curve

Intuitively, one may be tempted to understand the existence of a volcano curve in terms of a change in the rate-determining step. In other words, the ascending branch is governed by hydrogen adsorption (that increases as DG(M–H) increases), while the descending branch is governed by hydrogen desorption (that becomes more difficult as DG(M–H) increases). This is however an inadequate picture of the actual situation. Even if the rds always remains the same, a volcano curve would turn up in the end. This is the case for the primary discharge being rds, but it is also the case for, for example, electrochemical desorption being rate determining. In the latter case, as DG(M–H) is low, yH is small and the reaction rate, proportional to yH, will be small. On the other hand, as DG(M–H) is high, yH is high, but the reaction rate will again be small because the activation energy for desorption will be high. The predictions of the theory have been confirmed experimentally, although collecting experimental data under perfectly comparable conditions is difficult. Nevertheless, as Fig. 2B shows, a plot of ln j (activity) at constant potential (or overpotential) against the M–H bond strength gives a volcano-shaped curve as expected. The difficulty is in the assessment of the M–H bond strength. It is in fact difficult to determine DG(M–H) under operating conditions, especially because DG(M–H) depends in turn on yH (non-Langmuirian conditions). It has however been reported that DG(M–H) can be approached by the quantity determined experimentally in gas-phase adsorption. This is valid only to a first approximation owing to solvent displacement effects in solution, which are however not such as to overturn the outcome. In Fig. 3, sp metals are located on the left-hand branch of the volcano. They adsorb hydrogen rather weakly and hydrogen is evolved via a rate-determining primary discharge. This branch includes also the metals of the IB group (copper, silver, and gold). All transition metals are located on the right-hand branch of the volcano. For these metals, the adsorption of hydrogen is strong and yH close to saturation, thus making hydrogen evolution difficult. Precious metals are found in the intermediate range of M–H bond strength. For these, the M–H bond strength is close to the HdH bond strength (neither too weak nor too strong, according to the qualitative principle of Sabatier in catalysis), and the reaction may proceed at high rate with a mechanism producing a low Tafel slope. Platinum is in fact among the best catalysts for electrolytic hydrogen evolution. The position of nickel, iron, and cobalt is a debated issue. If DG(M–H) taken from gas-phase experiments are used, these metals are located on the descending branch. If DG(M–H) is estimated from in situ electrochemical experiments, they appear on the ascending branch. This apparent contradiction can be reconciled if it is considered that DG(M–H) becomes weaker if hydrogen is absorbed, as occurs during electrolytic hydrogen evolution.

Fig. 3 Experimental verification of the predicted volcano-shaped curve (cf. Fig. 2B) for the reaction of H2 evolution on polycrystalline metal surfaces. Adapted from Trasatti, S. Work Function, Electronegativity, and Electrochemical Behaviour of Metals: III. Electrolytic Hydrogen Evolution in Acid Solutions. J. Electroanal. Chem. 1972, 39, 163–184.

156

8

Electrochemical Fundamentals | Hydrogen Evolution and Oxidation

Materials for cathodes

Platinum is the reference electrode material for hydrogen evolution since it is the most active elemental cathode. The high cost of platinum, however, makes its use impractical in routine applications. Actually, traditional cathode materials in technology have for long been iron or steel in acidic solutions, and nickel in (strongly) alkaline solutions. Steel can also be used in moderately basic solutions. Problems with the use of nickel are that its electrocatalytic activity is not good enough; in addition, nickel absorbs atomic hydrogen that depresses the surface M–H bond strength, thus giving rise to time-dependent performance. A corollary is that if nickel is used in industrial cells subject to shutdown (e.g., for maintenance), the cathode potential is driven to more positive values with sizable dissolution of the metal. The above-mentioned shortcomings can be alleviated if nickel cathodes are activated, that is, if they are coated with a thin layer of more active and more stable materials. Activation of cathodes has been attempted with a variety of materials from sulfides to oxides, from alloys to intermetallic compounds. This process is very briefly discussed below. Looking at the volcano curve for elemental metals (Fig. 3), it is intriguing to figure out that a combination of a metal on the ascending branch and one on the descending branch should result in a material with enhanced activity. This is indeed the case and it has been theoretized as the combination of hypo-d-electron metals (on the left-hand side of the volcano) and hyper-d-electron metals (on the right-hand side). Examples are NidMo, ModCo, and NidW alloys. Another combination is constituted by intermetallic compounds generally containing nickel whose electronic structure is sizably modified by a second component. Examples are LaNi5, CeNi3, and Ni3Ti. If the activity of compounds of this type, for instance, nickel with tantalum, titanium, niobium, zirconium, and hafnium, is plotted against the calculated enthalpy for the formation of the related hydride, a sort of volcano-shaped curve is again obtained. Carbides (tungsten carbide, titanium carbide) also belong to this category of cathode materials. Oxides are unique cathode materials. The reason is that they are predicted to be thermodynamically unstable in the conditions of hydrogen evolution. In fact, electronically conducting oxides, such as ruthenium(IV) oxide and iridium(IV) oxide, are very active and satisfactorily stable even in strongly reducing conditions. This is thought to be a consequence of the fact that no penetration of H+ into the lattice is possible, since no electric field gradient is formed in the solid phase owing to the very low electric resistivity.

9

Factors of electrocatalysis

There are two factors governing electrocatalysis: 1. Electronic factors, related to chemical composition and structure of materials influencing primarily the M–H bond strength and the reaction mechanism. 2. Geometric factors, related to the extension of the real surface area influencing primarily the reaction rate at constant electronic factors. Only the former result in true electrocatalytic effects, while the latter give rise to apparent electrocatalysis. The two factors are seldom completely independent; mostly they are interdependent. A typical example is the effect of particle size. Decrease in particle size produces an increase in surface area for constant amount of material. At the same time, as the particle size decreases, the surface areato-volume ratio increases, which may lead to modifications in the electronic properties of surface atoms. However, there are cases where clear electronic factors turn out, in the end, to be in fact geometric effects. Two examples can best illustrate the issue. A coating of nickel sulfide appears to activate smooth nickel considerably. However, a scrutiny of the electrode surface reveals that nickel sulfide is in fact hydrogenated: NiS + H2 ! Ni + H2 S: In the end, the surface becomes covered by a layer of very fine powder of nickel that is responsible for the apparent activation of the underlying metal. The case of Raney nickel is even more intriguing and typical. Raney nickel is obtained by alloying nickel with zinc or aluminium, then leaching out the latter metal in caustic solution. The dissolution of zinc or aluminium results in a very porous layer of nickel exhibiting a high roughness factor. Nevertheless, geometric effect are not the only visible effects in the cathodic performance of Raney–nickel. While the activity is many orders of magnitude higher, the Tafel slope, 120 mV for pure nickel, comes to 40 mV for Raney nickel. A change in Tafel slope, an intensive quantity, is synonymous with a change in the reaction mechanism, that is, these data point to true electrocatalytic effects (electronic factors). It is inferred that the very high porosity of the nickel surface can modify the nature of the active sites (edge effects, among others), thus affecting the M–H bond strength. Other aspects related to electronic factors are, for instance, the effect of the topography of electrode surfaces (e.g., single crystal faces), and the degree of crystallinity of solid surfaces (e.g., amorphous materials). In both cases, the effect is minor but finite. It is however of secondary importance in practical terms.

Electrochemical Fundamentals | Hydrogen Evolution and Oxidation

10

157

Conclusion

The two-step reaction mechanisms of hydrogen evolution requires the adsorption of hydrogen atoms. The best electrode material (platinum) allows both good hydrogen adsorption and desorption due to its medium metal-hydrogen bond strength. Moreover, important cathode materials are nickel, and platinum group metal oxides.

References 1. Breiter, M. W. Reaction Mechanisms of the H2 Oxidation/Evolution Reaction. In Handbook of Fuel Cells: Fundamentals, Technology, and Applications; Vielstich, W., Lamm, A., Gasteiger, H. A., Eds.; Electrocatalysis, Vol. 2; Wiley: Chichester, 2003; pp. 361–367. 2. Enyo, M. Hydrogen Electrode Reaction on Electrocatalytically Active Metals. In Comprehensive Treatise of Electrochemistry; Conway, B. E., Bockris, J. O.’. M., Yeager, E., Kahn, S. U. M., White, R. E., Eds.; Vol. 7; Plenum: New York, 1983; pp. 241–300. 3. Frumkin, A. N. Hydrogen Overvoltage and Adsorption Phenomena. Part I. In Advances in Electrochemistry and Electrochemical Engineering; Delahay, P., Tobias, C. W., Eds.; Vol. 1; Wiley-Interscience: New York, 1961; pp. 65–121. 4. Frumkin, A. N. Hydrogen Overvoltage and Adsorption Phenomena. Part II. In Advances in Electrochemistry and Electrochemical Engineering; Delahay, P., Tobias, C. W., Eds.; Vol. 3; Wiley-Interscience: New York, 1963; pp. 287–391. 5. Krishtalik, L. I. Hydrogen Overvoltage and Adsorption Phenomena. Part III. In Advances in Electrochemistry and Electrochemical Engineering; Delahay, P., Tobias, C. W., Eds.; Vol. 7; Wiley-Interscience: New York, 1970; pp. 283–339. 6. Lasia, A. Hydrogen Evolution Reaction. In Handbook of Fuel Cells: Fundamentals, Technology, and Applications; Vielstich, W., Lamm, A., Gasteiger, H. A., Eds.; Electrocatalysis, Vol. 2; Wiley: Chichester, 2003; pp. 416–440. 7. Parsons, R. The Rate of Electrolytic Hydrogen Evolution and the Heat of Adsorption of Hydrogen. Trans. Faraday Soc. 1958, 54, 1053–1063. 8. Trasatti, S. Electrocatalysis of Hydrogen Evolution: Progress in Cathode Activation. In Advances in Electrochemical Science and Engineering; Gerischer, H., Tobias, C. W., Eds.; Vol. 2; Weinheim: VCH, 1992; pp. 1–85. 9. Wendt, H.; Plzak, V. Electrocatalysis and Electrocatalysts for Cathodic Evolution and Anodic Oxidation of Hydrogen. In Electrochemical Hydrogen Technologies. Electrochemical Production and Combustion of Hydrogen; Wendt, H., Ed.; Elsevier: Amsterdam, 1990; pp. 15–62.

Electrochemical Fundamentals | Oxygen Evolution and Reduction Peter Kurzweila and S Trasattib,⁎, aTechnical University of Applied Sciences (OTH), Amberg-Weiden, Germany; bUniversity of Milan, Milan, Italy © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of S. Trasatti, ELECTROCHEMICAL THEORY | Oxygen Evolution, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 49–55, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00023-X.

1 2 2.1 2.2 2.3 2.4 2.5 2.6 3 3.1 3.2 3.3 3.4 3.5 4 4.1 4.2 5 6 References

Introduction Factors influencing oxygen evolution Effect of solution pH Surface oxidation Hydrous versus dry oxides Anodic dissolution Problems of reversibility Tafel lines and Tafel slopes Reaction mechanisms Possible pathways Multiple Tafel lines Reaction orders Surface acid-base properties Point of zero charge Electrode materials Metal oxides Composite materials Electrocatalysis Conclusion

159 159 159 160 160 161 161 161 161 161 162 163 163 163 164 164 164 164 165 165

Abstract The basic aspects of oxygen evolution at anodes of electrochemical cells containing aqueous solutions are presented and discussed. The effect of solution pH on the features of current-potential curves is illustrated and associated with the nature of the reacting particles. It is explained why O2 evolution always takes place on surfaces significantly modified by the anodic process. In particular, an oxide film before O2 evolution invariably covers metal surfaces. The difference between electrolytic and thermal oxides is discussed. Various mechanisms proposed for the O2 evolution reaction are presented and discussed in terms of Tafel slope, stoichiometric number, and reaction order. The specific acid-base properties of oxide surfaces in aqueous environment are introduced and explained; a few concepts about the point of zero charge are also provided. Finally, the best materials for electrodes evolving oxygen in acid as well as alkaline solution and the problem of electrocatalysis at anodes are discussed.

Key points

• • •

Oxygen evolution reaction: importance, reversibility, mechanisms, poisoning Tafel slopes Volcano curves in electrocatalysis

Nomenclature

Symbols and units b E

Tafel slope (mV) Electrode potential (V)

⁎

Posthumous

158

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00144-4

Electrochemical Fundamentals | Oxygen Evolution and Reduction

E0 j0 z a g h u

159

Standard electrode potential (V) Exchange current density (mA cm−2) Stoichiometric number Transfer coefficient (symmetry factor) for single-step reaction Observable transfer coefficient (multistep reactions) Overpotential (V) Surface occupancy (0. . . 1)

Abbreviations and acronyms DSA OPD pzc rds SHE UPD

1

Dimensionally stable anodes Overpotential deposition Point of zero charge (pH) Rate determining step Standard hydrogen electrode Underpotential deposition

Introduction

The formation of gaseous O2 at the anode of an electrolytic cell containing an aqueous electrolyte solution is the most frequent electrode reaction.1–11 It has immense practical implication because it always accompanies the electrolytic production of H2 in water electrolysis, the cathodic deposition of metals in metal electrowinning processes, and the application of a cathodic current in cathodic protection operations, and it takes place at the negative pole of metal-air power sources upon recharging. The electrode reaction of oxygen formation involves four electrons per mole of O2: 2 H2 O ! O2 + 4 H+ + 4 e− in acid solution, with the potential E0 ¼ 1.229 V versus the standard hydrogen electrode (SHE), and: 4 OH− ! O2 + 2 H2 O + 4 e− in alkaline solution, with E0 ¼ 0.401 V (SHE). Thus, the reacting particles from which molecular O2 originates are water molecules in acid solution and hydroxyl ions in alkaline solution. Oxygen formation is a catalytically demanding reaction. This is because of the involvement of high-energy intermediates with complex molecular structure and complicated reaction pathways. As a consequence, the reversible O2 electrode is not realizable in practice, that is, nothing similar to the reversible H2 electrode has ever been realized. Some authors have claimed approach to equilibrium conditions, but only in a very unstable situation easily affected by a minimum amount of oxidizable impurities. As a consequence of its demanding nature, the rate of O2 evolution at a given electrode potential depends critically on the nature of the electrode material. This phenomenon is known as electrocatalysis of O2 evolution. Oxygen evolution hardly commences at potentials lower than 1.4 V (SHE), in other words, at a lower overpotential than about 150 mV. Thus, at a practical rate of 0.1 A cm−2, the minimum overpotential achievable is around 0.3 V. Usually, the ordinary overpotential is >0.4 V. This explains the wide interest in the search for new or improved electrode materials that can help reduce the cost of the electric power in electrolytic processes.

2 2.1

Factors influencing oxygen evolution Effect of solution pH

Although the overpotential does not differ appreciably for a given material in strongly acid or strongly basic solutions, it generally reaches a maximum in the intermediate pH range. This is related to the different population of reacting particles in different pH ranges. At low pH, where H2O molecules are the reacting particles, there are no problems of mass transfer limitation in that the nominal concentration of H2O species is 55.5 mol dm−3. At high pH, problems of mass transfer may arise because of the limited availability of hydroxide species. The problem is not critical for c(OH−) >1 mol dm−3 at current densities E0(O2) is called ‘overpotential deposition’ (OPD) of oxygen. These phenomena are also observed in the case of hydrogen and (foreign) metal deposition.

2.3

Hydrous versus dry oxides

The thickness and the morphology of an oxide layer depend on the procedure of formation. Those described above are layers formed anodically at constant potential (or current). If layers are formed by means of square potential (or current) pulses, or by repetitive linear potential cycling, porous layers of hydrous material are formed whose thickness may reach some micrometers. These hydrous layers (electrolytic oxides) behave more reversibly with respect to formation/reduction than less hydrated films. Hydrous layers are generally also obtained by anodic oxidation of metal ions in solution. Thus, the anodic oxidation of Ru(III) in aqueous solution produces a film of RuO2xH2O on the electrode surface. If hydrous oxides are heated above 100  C, they lose water. Dry oxides (thermal oxides) can be obtained straightforwardly by thermal decomposition of suitable precursors directly onto inert metallic supports. Thus, RuO2 can be obtained by thermal decomposition of RuCl3 or Ru(NO3)3, and Co3O4 can be obtained by thermal decomposition of Co(NO3)2, and so on. Thermal oxides are usually obtained in the temperature range 200–500  C depending on the nature of precursors and of the oxide. The structure of thermal oxides prepared in the low temperature range is intermediate between hydrous and dry. It is thus possible to obtain a wide range of electrode materials with varying degrees of hydration and nonstoichiometry that enable the modulation of the activity of a given nominal oxide for O2 evolution.

Electrochemical Fundamentals | Oxygen Evolution and Reduction 2.4

161

Anodic dissolution

During oxygen evolution, besides forming a surface oxide, electrodes may also undergo dissolution if the formed oxides are partially or totally soluble. Thus, NiO and Co3O4 can be regarded as insoluble in alkaline solution, although they are easily soluble in acidic solution. On the contrary, although RuO2 is insoluble in acids, it can dissolve if oxidized to Ru(VI) or Ru(VII). This is easier in alkaline solution. In acid solution, if oxidized to Ru(VIII), ruthenium oxide can be lost by formation of volatile RuO4. For a given oxide the electrolytic form is as a rule less stable anodically than the corresponding thermal oxide, in that in the former the metal cation is surrounded by an aqueous structure more closely resembling the solution environment. On the contrary, dissolution of a thermal oxide implies proton penetration into the lattice. Although this is not a problem for semiconducting oxides where an electric field assisting proton penetration is formed in the bulk owing to low conductivity, it does not occur with highly conductive and metallic oxides such as RuO2 and IrO2. This is thought to be the reason for their high resistance to reduction (proton penetration) even under H2 evolution. Although metals or metal alloys were used customarily for O2 evolution since the birth of electrochemistry, the discovery of the favorable properties of thermal oxides by H. Beer in the 1950s led to the introduction of activated electrodes, that is, an inert metallic support (Ti or steel in acids and Ni in bases) covered with a thin layer of thermal oxides. Historically, electrodes of this type were termed dimensionally stable anodes (DSA®). They were originally introduced as anodes for chlor-alkali cells, but were successively extended to O2 evolution anodes and other applications. In view of the fact that a solid surface may undergo some essential modifications under O2 evolution, the kinetic parameters may be time dependent. This is probably the main reason why the reproducibility of experimental data in the literature is sometimes poor, especially in papers dealing with bare metals. In practice, a hysteresis is often observed between forward and backward polarization curves. For the reaction to advance at a constant rate, it is necessary that a stable oxide of constant thickness be formed on the surface. Thus, the electrode conditioning before the experiment is of paramount importance for the reliability of the data. This problem is alleviated with preformed (thermal) oxide layers, but it can still show up in a potential region where severe electrochemical modifications may take place.

2.5

Problems of reversibility

There is no point in determining the exchange current density j0 for the O2 electrode reaction. Because the reaction is highly irreversible, j0 can hardly be obtained from the charge transfer resistance, that is, from d/dj. It can only be extrapolated from the Tafel line, with the risk that owing to the high overpotential, extrapolation does not indicate the true equilibrium conditions. In addition, the apparent exchange current density is an extensive quantity, that is, it depends on the actual (real) surface area. At any rate, apparent j0 values are of the order of 10−10–10−11 A cm−2, that is, of the order of nanoamperes (nA), which would make its direct measurability difficult even if an equilibrium could be effectively established. Conversely, Tafel lines usually extend over several tens of orders of current with no problems of mass transfer in acidic solution because water molecules are the reacting particles. Under these circumstances, Tafel line distortion can come from uncompensated ohmic drops, which can rather easily be allowed for. Thus, the analysis of the reaction mechanism is mostly and principally based on the value of the Tafel slope, sometimes supplemented by the determination of reaction orders, and/or by spectroscopic studies of the nature of surface sites and/or reaction intermediates.

2.6

Tafel lines and Tafel slopes

The whole series of classical Tafel slopes, from 30 mV (rare) up to 120 mV, have been observed experimentally, with cases of intermediate values and even b > 120 mV. Intermediate values are often interpreted in terms of mixed mechanisms, or of intermediate coverage with adsorbed species (0.1 < yi < 0.9). Tafel slopes higher than 120 mV are interpreted as due to a double barrier. For instance, O2 evolution on a growing electrolytic oxide layer proceeds at two electrode interfaces: one at the oxide/ solution interface and the other at the metal/oxide interface. Charge transfer at each of these interfaces proceeds with a ¼ 0.5, so that the global transfer coefficient turns out to be 0.25, that is, a Tafel slope close to 240 mV. Such high Tafel slopes have not been reported with conducting thermal oxides because the penetration of the electric potential into the bulk of the oxide can be neglected.

3 3.1

Reaction mechanisms Possible pathways

In view of the complexity of the O2 reaction, several mechanisms have been proposed, differing in molecular details that can hardly be confirmed experimentally. It is therefore sensible to restrict our analysis to the most popular mechanisms in Table 1. These can be associated with the names of those who proposed them first. Mechanisms are written for acid solution only. They can easily be extended to alkaline solution. A. Chemical oxide path, proposed by Bockris in 1956. The first step is the oxidation of a water molecule on a reaction site (S). As a second step, two OH species condense to produce an adsorbed atom; this is the chemical step characterizing the mechanism.

162

Electrochemical Fundamentals | Oxygen Evolution and Reduction Table 1

Reaction mechanisms proposed for the oxygen evolution reaction.

Mechanism

Reaction steps: S ¼ surface site

Stoichiometric factor z

Tafel slope b (mV)

A. Chemical oxide path

S + H2O ! SOH + H+ + e− 2 SOH ! 2 SO + H2O 2 SO ! O2 + 2 S S + H2O ! SOH + H+ + e− SOH ! SO + H+ + e− 2 SO ! O2 + 2 S S + H2O ! SOH + H+ + e− SOH ! SO− + H+ 2 SO− ! SO + e− 2 SO ! O2 + 2 S Sn + OH− ! SnOH + e− SnOH ! Sn+1OH + e− 2Sn+1OH + 2 OH− ! O2 + 2 Sn + H2O S + H2O ! SOH∗ + H+ + e− SOH∗ ! SOH SOH ! SO + H+ + e− 2 SO ! O2 + 2S MOx + H2O ! MOx+1 + 2 H+ + 2 e− MOx+1 ! MOx + 12 O2

4 2 1 2 2 1 2 2 2 1 2 2 1

120 30 15 120 40 15 120 60 40 15 120 40 15

B. Electrochemical oxide path C. Oxide decomposition path

D. Alkaline path E. Generalized mechanism

B.

C.

D.

E.

The predicted Tafel slope for yi  0 is 30 mV, which shows that this would be one of the most efficient mechanisms for O2 evolution because the transfer coefficient (g ¼ 2) is the highest observed experimentally. In fact, b ¼ 15 mV for the last step is only virtual because it has never been reported. An anode with such a low Tafel slope for O2 evolution remains simply a dream. Electrochemical oxide path, proposed by Bockris in 1956. After the primary oxidation of H2O, the adsorbed OH groups are further oxidized electrochemically to SdO; this is the electrochemical step characterizing the mechanism. If the first step is rate determining, this mechanism can be distinguished from (A) only on the basis of the stoichiometric number, which is however difficult to determine for O2 evolution. Oxide decomposition path, proposed by Krasil’shchikov in 1963. The most distinctive feature of this mechanism is the acid-base dissociation of surface OH groups. This mechanism implies the presence of a true oxide layer on the electrode surface and thus it introduces a chemical view of O2 evolution at electrodes. Alkaline path, proposed by Yeager in 1974. This mechanism was specifically proposed for O2 evolution on thermal oxide electrodes in alkaline solution. For these solutions, the other mechanisms predict a reaction order of 2 with respect to OH−, whereas Yeager reported a reaction order of 1 with a Tafel slope of 40 mV. Thus, it has been proposed that reaction sites exchange two electrons with solution species although involving only one OH−. It is postulated that the second step involves the oxidation of the surface site only. Generalized mechanism. The mechanisms A to C can be generalized into a single mechanism encompassing all steps. The first step is necessarily the discharge of water. The intermediate SdOH must be converted into something more stable before further oxidation. Therefore, a surface rearrangement is expected to take place. Such a rearrangement can even include surface migration from the discharge site to a more active site. The motion of adsorbed species between two different surface sites on catalyst surfaces is often called spillover in catalysis. The stable species is then further oxidized electrochemically, and O2 is finally liberated.

Other kinetically indistinguishable reaction schemes reproducing the experimental data can be proposed. However, the basic chemical message of this generalized mechanism is that O2 is evolved through the decomposition of unstable higher oxides.

3.2

Multiple Tafel lines

In a number of cases two different Tafel slopes can be observed as a function of overpotential: usually, a lower b in the low  range and a higher b in the high  range. Generally, Tafel slopes of 40–60 mV are followed by a Tafel slope of 120 mV. These transitions can be due to (i) change in surface coverage yi for the same rate determining step (rds) of the same mechanism, (ii) change in rds of the same mechanism, or (iii) change in mechanism. Distinction among these three possibilities calls for a specific detailed analysis case by case. A typical case (iii) is illustrated in Fig. 2. In the low  range, step A is slower and its Tafel slope is observed experimentally. Beyond the intersection point, step B becomes slower and determines the Tafel slope. Thus, the observed Tafel line is the bold one. This is

Electrochemical Fundamentals | Oxygen Evolution and Reduction

163

Electrode potential

A

B

ln(current)

Fig. 2 Sketch of a potential-current curve with two Tafel slopes determined by the transition of the rate determining step from step A to step B.

not uncommon for bulk thermal oxides. For instance, with IrO2 the Tafel slope goes from 60 to 120 mV, that is, the rds moves from chemical rearrangement to primary water discharge. With thermal NiO, only a Tafel slope of 40 mV is observed over several tens of orders of current. However, if NiO is doped with Fe oxide, a Tafel slope of 30 mV is observed at low  values, becoming 60 mV at higher overpotentials. Evidently, in this case a true change in mechanism occurs.

3.3

Reaction orders

The sole Tafel slope may sometimes be insufficient to identify a reaction mechanism. Because the stoichiometric number cannot be determined for the O2 electrode, the next kinetic parameter that can help in a diagnosis is the reaction order. Although in acid solution the population of H2O species cannot be changed in principle, the other possibility is to determine the reaction order with respect to H+ that is a product of the reaction. In alkaline solution, the relevant reaction order is the one with respect to OH−. According to the mechanisms discussed above, integral reaction orders are predicted, either positive or negative, between 0 and 2. In practice, particularly with thermal oxides, fractional reaction orders are observed, positive or negative, between 0.5 and 1.5, in case the rds is an electrochemical step, although integral values are still observed if the rds is a chemical step.

3.4

Surface acid-base properties

The above experimental observation can be explained in terms of the acid-base properties of oxide surfaces in aqueous solution. Oxide surfaces are high-energy surfaces that interact strongly with water molecules, becoming covered by a carpet of OH groups. In aqueous solution, these groups behave as weak bases or weak acids, thus giving rise to surface charging: S  OH + H+ $ S  OH2 + S  OH + OH− $ S  O− + H2 O Dissociation of OH groups is a function of the pH of the solution. The pH at which an oxide surface is electrically neutral is termed the point of zero charge (pzc). Thus, the pzc of Co3O4 is around pH 7, and that of RuO2 is about pH 5.5, but it depends on the calcination temperature. The determination of reaction orders for O2 evolution entails modifying the concentration of either H+ or OH−. In both instances, the free charge on the oxide surface changes with pH. As a consequence, the electric potential at the reacting site does not remain constant. In principle, this is a double-layer effect and the resulting reaction orders turn out to be fractional.

3.5

Point of zero charge

The point of zero charge (pzc) is the solution pH at which an oxide surface is electrically neutral. At pH 1 O cm2, and a voltage window of 0.6 V. The energy storage in solid-state capacitors may involve electrochemical redox processes as in batteries. Manganese dioxide and 7,7,8,8-tetracyano quinodimethane (TCNQ) complexes are known as solid electrolytes, which additionally enable redox reactions. Metal oxides. A solid-state thin-film supercapacitor is obtained by coating amorphous ruthenium(IV) oxide15 on a platinum current collector by direct current reactive sputtering deposition at 400  C. The solid electrolyte of Lipon (Li2.94PO2.37N0.75) was deposited by radiofrequency reactive sputtering at room temperature. Nanocrystalline nickel oxide (NiO), for example, prepared from nickel hydroxide, was suggested for an all-solid-state supercapacitor. The potassium hydroxide electrolyte was fixed in a polymer matrix of polyvinyl alcohol (PVA). An all-solid-state supercapacitor with phosphotungstic acid as the proton-conducting electrolyte in a composite with Al2(SO4)318H2O was used in a symmetric supercapacitor based on polyaniline. Capacitances as high as 240 F g−1 at 6 mA were achieved. Table 6 1982 1983 1987 1990 1991 2004

Examples of solid-state capacitors. US 4363079, Matsushita Electric Ind. Co. Ltd (Japan): Solid-state electrode (Cu2S/CuS) in a solid electrolyte (CuCl doped with Cu+, Rb+, I−) US 4414607, Matsushita, Japan: Solid electrolyte and carbon-bound electrodes thereof, e.g., KxRb1−xCu4IyCl5−y (0.1 < x < 0.25, 1.25 0), endothermic (DH > 0) and involuntary (DG0 > 0), i.e. the cell cools down. The contribution of reaction entropy, Q ¼ DH – DG ¼ T DS  49 kJ mol−1, cannot be converted into useful work. Only above Uh ¼ 1.48 V does the cell heat up despite the endothermic cell reaction. In practice, however, waste heat is always released by the internal resistance. Steam electrolysis. The electrolysis of water vapor H2O(g) ! H2(g) + ½ O2(g) surprises by the favorable decomposition voltage U0 ¼ 1.17 V (100  C) to 0.91 V (1000  C) and the high thermodynamic efficiency DH/DG > 1, as shown in Fig. 2. The decomposition voltage at high temperature improves because the contribution of the reaction entropy grows noticeably: T DS/(2F) ¼ −0.09 V (at 100  C) to −0.37 V (1000  C). The unfavorable heat of reaction DH ¼ +242.4 kJ mol−1 (at 100  C) to +249.4 kJ mol−1 (1000  C) makes little impression on the heating value voltage: Uh ¼ 1.26 V (100  C) to 1.29 V (1000  C). Hot steam electrolysis is therefore energetically advantageous, but encounters material problems. Electrolysis under pressure. A pressure increase from 1 bar to 30 bar worsens the decomposition voltage by 0.065 V. With gas electrodes, the entropy effect is noticeable on the cell voltage from 10 bar. It therefore makes sense to carry out the electrolysis under atmospheric pressure.

Electrochemical Devices | Electrolyzers

389

Fig. 2 Contribution of reaction entropy and decomposition voltage of steam electrolysis. Modified from Kurzweil, P. Angewandte Elektrochemie (Applied Electrochemistry). Springer Vieweg: Wiesbaden, 2020.

3.2

Kinetics of electrolysis

Chemistry comprises thermodynamics plus kinetics, i.e. the ‘true’ decomposition voltage cannot be derived from thermodynamic table values alone. The practical decomposition voltage to decompose water into oxygen at the anode and hydrogen at the cathode is significantly higher than 1.23 V. The catalytic activity of the electrodes is determined by: Surface properties (active area, adsorbate layers, dissolution, recrystallization), electrolyte state (pH, temperature, concentration, composition) and electrode poisons (As, Pb. Cu, impurities). To overcome the kinetic inhibitions, especially the overvoltage at the oxygen electrode, high temperature and large electrode area are useful. In Eq. (13), the internal resistance Ri and the overvoltage Z ¼ U(I) – U0 show the deviation from the theoretical decomposition voltage U0 ¼ 1.23 V. U ðIÞ ¼ U 0 + IRi ¼ U 0 + I ðRanode + Rcathode + Re Þ

(13)

For the investigation of cathodes and anodes, the measured current-overpotential curve is evaluated in logarithmic representation, the so-called Tafel diagram, lg I against Z. The catalytic activity of an electrode is expressed by the exchange current I0 and the charge-transfer resistance Rct, the diffusion limiting current Ilim and the diffusion resistance Rd as well as the Tafel slope b ¼ dE/d(lg I). The sum of the charge-transfer and diffusion resistance is called concentration polarization, RP ¼ Rct + Rd.
ct RT ¼ I zFI0

(14)


d RT ¼ I zFIlim

(15)

Rct ¼ Rd ¼

3.3

Cell construction

Monopolar cell construction. One hydrogen or oxygen chamber supplies two adjacent anodes or cathodes simultaneously (Fig. 3). Two spatially separated anodes and cathodes are connected to each other via a lateral electrical connection. Advantage: Defective cells can be bridged or a part of the stack can be operated. Disadvantage: The lateral contacting requires electrolytes with good conductivity, and, in the case of carbon electrodes, metallic frames or current collectors. Otherwise, the current density is distributed unevenly over the electrode area. This design is impracticable for electrode cross-sections of 400 cm2 and more.

390

Electrochemical Devices | Electrolyzers

Fig. 3 (A) Monopolar tank cell, (B) bipolar cell stack. Modified from Kurzweil, P. Angewandte Elektrochemie (Applied Electrochemistry). Springer Vieweg: Wiesbaden, 2020.

Bipolar cell design: Hydrogen chamber, anode, electrolyte matrix, cathode, oxygen chamber, and bipolar plate form a repeatable unit. Advantage: Current flows vertically through large cross-sections, even with less conductive carbon electrodes. Disadvantage: The weakest cell determines the total cell voltage, and, in the worst case, causes failure of the cell stack. Electrolyte bridges and leakages between the individual cells interfere and must be removed. The contact resistances force a high compression of the cell stack.

3.4

Alkaline electrolysis

Alkaline water electrolysis9–14 using nickel electrodes at operating temperatures of 50–80  C, pressures of 1–150 bar, current densities of 0.2–0.45 A cm−2 and cell voltages of 1.8–2.4 V dominates in small devices and megawatt plants. Water is injected at the cathode and additionally formed at the anode. Potassium hydroxide solution has its best conductivity at a concentration of 32.6% or 7.61 molar: k(80  C) ¼ 1.36 S/cm. The lower water vapor partial pressure above 30% KOH, p(H2O) ¼ 1950 Pa, deteriorates the cell voltage by 6.2 mV compared to pure water (3167 Pa) at 25  C. Separator. A hydroxide-permeable, alkali and pressure-resistant diaphragm (made of potassium titanate/PTFE, ZrO2/PSU, glass fiber PPS, sulfonated PEEK, nickel oxide) separates the anode and cathode compartments and prevents gas mixing. Zero-gap technology (Lurgi 1975) uses perforated electrodes with 0.1 mm (H2) and 0.7 mm (O2) holes directly on the diaphragm so that gas bubbles escape into the electrolyte on the back side. 10–50 mm thick cells with 4 m2 cross-section are practicable. Gas bubbles create a resistance greater than the electrolyte and the charge-transfer inhibition. Electrodes hydrophilized with surfactants and ultrasonic irradiation increase bubble release. Catalysts. Conventional electrodes are made of nickel, nickel-plated iron, nickel/nickel sulfide, or Raney nickel. Steel is widely used as cathode material. NiOOH formed under load is reduced to corrosive Ni(OH)2 during current interruptions, so that a protective voltage of 1.3 V in standby operation or the addition of molybdenum is useful. In part-load operation, (1) decreasing gas production despite undiminished gas permeation through the separator and (2) dissolved extraneous gases in the anolyte and catholyte are troublesome. System components: The pumped ‘mobile’ electrolyte transports water to and from the electrodes and removes heat, impurities, carbonate and dissolved gases (Fig. 4A). Gas and liquid are segregated by gravity in gas separators. Heat exchangers and baffles cool the electrolyte; coalescing filters retain aerosol droplets. A scrubber in the H2 electrolyte circuit is fed with deionized water, which then dilutes the KOH solution added to the stack. Without further purification, the gas purity is 99.8–99.9% H2 and 99.3–99.8% O2. Hydrogen is catalytically deoxidized below an oxygen concentration of 10−4%, dried by pressure swing absorption, and is stored in tanks with a purity of 99.999%.

3.5

SPE-electrolysis

The solid polymer electrolyte (SPE)15,16 works as an ion conductor, catalyst support and gas separator. General Electric, in 1966, introduced 100–400 mm thin perfluorosulfonic acid (PFSA) films. Proton exchange membranes (PEM) such as Nafion™ are also used for fuel cells. The redox equations in acidic solution apply for SPE electrolysis. Water is injected at the anode and is also used for cooling. By electroosmosis of the hydrated proton [H(H2O)2 to 3]+, water migrates through the membrane to the cathode and enters the

Electrochemical Devices | Electrolyzers

391

Fig. 4 (A) Alkaline electrolyzer, (B) solid-polymer electrolysis (SPE), (C) solid-oxide steam electrolysis (SOE). Modified from Kurzweil, P. Angewandte Elektrochemie (Applied Electrochemistry). Springer Vieweg: Wiesbaden, 2020.

hydrogen compartment. If the O2 stream is to be dry, the cathode is flooded with water (or water vapor), which diffuses through the membrane to the anode and determines the highest possible current density. Bipolar stacks in filter press design consist of 10 to 120 electrode-electrolyte units arranged electrically in series with up to 2500 cm2 cross-sectional area and an electrolyte supply connected hydraulically in parallel. The membrane electrode assembly (MEA) is a supported catalyst layer (H2: platinum, PtRu; O2: IrO2, RuO2) hot pressed with the membrane. Titanium powder sintered discs or niobium (anode) or titanium or carbon fabric cathode) serve as current collectors. Titanium tends to hydrogen embrittlement. Advantages of this zero-gap geometry are the small electrolyte resistance and the fast kinetics on platinum metals. Anodes and cathodes are often mounted in metal or plastic frames for better sealing and pressing of the individual cells. Bipolar plates (titanium, tantalum, zirconium, niobium, coated metals) with milled flow channels separate the gas spaces. System components. Small-scale systems produce 20 m3 of hydrogen per hour with about 86% efficiency, which is determined by slow oxygen kinetics and membrane resistance. Increased operating pressure on the cathode side facilitates storage of the 99.99% pure hydrogen; the oxygen side remains safely at atmospheric pressure. Water from the storage tank (with gas separator) is cleaned from heavy metal ions (iron, chromium, nickel) in the ion exchanger. The energy consumption of 4.1–5.0 kWh/(m3 H2) in the stack depends little on the gas volume flow produced. Gas purification and ancillary units make production volumes economical only above 10 m3/h (10,000

3000 ( 450) @ 0.1 Hz 329,500 ( 20) @ 5 Hz

3

Electrochemical Devices | Electrochemical Sensors and Actuators

405

Fig. 4 Process of fabricating ionogel based actuator.

decomposition because of their unique features like chemical and thermal robustness, negligible volatility and low melting point.84 The process of fabricating ionogel based actuator is shown the Fig. 4. The preferable material for ionic EAP actuators would be photo polymerization ionic gels, which has a simple preparation method, photolithography,85 through which soft actuators can be produced faster and larger in quantity. Moreover, with the aid of 3D printers, sophisticated actuators can be built by extending the fabrication process (2). Apart from actuators, the ionogels finds their applications in catalysis, solar cell, fuel cell capacitors, medicine, etc.86 Due to their small actuation amplitude, long response time, and expensive fabrication, electrochemical actuators are still in the experimental research stage and have yet to be widely produced. In addition, various aspects must be researched and examined to develop a mature system for actual use.

7

Challenges and future perspective

Many unique directions are being explored in electrochemical research. Microfluidic platforms, for instance, could be enhanced with electrochemical technology to facilitate on-site and in-situ monitoring of environmental conditions. A large scale effort to address environmental monitoring needs will also depend on network communication refinement and commercialization improvements.87 The development of electrochemical sensors grows exponentially with various research groups adopting different combinations of nanomaterials to achieve better possibilities due to their surface-to-volume ratio88–90 allowing faster electron transfer kinetics, improved catalytic activity, and biocompatibility.88,90,91 Graphene nanoribbons, graphene flowers, carbon nanotubes, metallic nanoparticles, and quantum dots are a few examples. The constraints attached to the nanomaterial composition for electrochemical sensing and actuation are that it requires sophisticated equipment for each type of composition for visual and chemical characteristics, and more than the equipment, skilled human resources are required to handle the equipment. Another challenge is maintaining stability, as these are sensitive and prone to thermal heating, UV exposure, and other environmental factors.92 In short, the limitations and challenges involved can be categorized into three, on a broader scale: (i) Obtaining a low limit of detection; (ii) Suppressing non-specific absorption of interfering species; and (iii) Maintaining the reproducibility and stability.89 Low detection limit is a critical property when electrochemical sensing is concerned as it corresponds to the lowest concentration of an analyte, with which it can detect a signal within the acceptable signal-to-noise ratio. Most electrochemical sensing is involved in biosensing applications, where suppressing non-specific absorption of interfering species is of significant concern. The presence of interfering analytes affects the theragnostic accuracy of the trace-level concentrations of the targeted biomarker.93 Another challenge is maintaining reproducibility and stability. The sensor-to-sensor reproducibility becomes essential as the modified structure is concerned; it becomes difficult to ensure the same structure is reproduced, adding to it in mass production cases. It is difficult to test the manufactured sensors individually; they are tested only for a batch. Stability is a concern in electrochemical sensors because of the nature of the modified nanomaterials employed in them, i.e., aggregation and flaking. It is essential to build sensors and actuators that last longer in the field, as, in general, they are characterized only by their shelflife.94,95 Still, the recent advancements in combining various materials like sol-gel, ceramics, hydrogels, novel polymers, and organogels have resulted in positive outcomes in increasing the stability of the electrochemical sensors/actuators.96,97

7.1

Emerging trends and technologies in the field

Including materials with sol-gel, ceramics with nanomaterials, and designing stretchable, water-processable, and self-healable miniatures are emerging trends in electrochemical sensing, resulting in appreciable electrical conductivity with superior analytical performance characteristics. This breakthrough achieved in the field through the inclusion of nanomaterials and design

406

Electrochemical Devices | Electrochemical Sensors and Actuators

modifications facilitated the electrochemical sensing and actuation with enormous potential, enabling it to suit many application areas of sensing and actuation across domains. This results in it being used in biosensing, environmental monitoring, forensic analysis, food analysis for quality and safety, agriculture, electronics, avionics, and military applications.98

7.2

Potential future advancements and applications

The progressive research in the field has widened the application of electrochemical sensing and actuation. Here, areas such as bio-sensing/health care, food analysis, and unmanned aerial vehicles have more potential for future advancements than the other application area. Biosensing is a promising field for these sensors and actuators as these sensors can diagnose infectious diseases in a shorter time, and these are employed in the theragnostics of breast cancer. This was possible because of the various sensing methods/sensors like voltammetric sensing, impedance sensing, amperometric sensing, potentiometric sensing, redox Mediators, sensors based on enzymes, and detection of guanine residues.98–100 Food quality and safety is very much essential for the well-being of any living organism. When it comes to human health, the change in the ecosystem, i.e., lifestyle and work environment, there is a decline in the number of people having homemade food, wherein the concern about the quality of food eaten is increasing as well. The evolution in electrochemical sensing has derived ways of analyzing food quality with inexpensive methods with high sensitivity, speed, accuracy, and ease of downsizing capable of capturing targets specifically and effectively.101–103 Unmanned aerial vehicle (UAV) systems are one of the marvels of the 20th century made for the purpose of military applications, through which human interventions in precarious places can be limited/avoided. Later its application spread to other purposes as well. Advancement in electrochemical sensing has ensured its place in UAVs. It has embedded itself in applications like pollution detection and air quality monitoring and could be seeding for creating artificial rain to start with.104,105

8

Conclusion

Significant improvements and advances have been made in electrochemical sensors and actuators. This chapter summarizes an overview of electrochemical sensors and actuator’s capabilities and their broader applicability in several sectors. Electrochemical sensing techniques and sensor types include amperometric, conductometric, potentiometric, impedimetric, and voltammetric sensors with their applications in environment and food safety. Novel electrochemical actuation principles and actuator materials include conducting polymers and ionogels. Finally, there are challenges, future directions and emerging trends that continue to advance the field and drive innovation. As a whole, continued research, invention, and collaboration will lead to even more sophisticated and adaptable electrochemical devices capable of addressing complex challenges and improving the quality of our lives as we move forward.

References 1. Ray, T. R.; Choi, J.; Bandodkar, A. J.; Krishnan, S.; Gutruf, P.; Tian, L.; Ghaffari, R.; Rogers, J. A. Bio-Integrated Wearable Systems: A Comprehensive Review. Chem. Rev. 2019, 119 (8), 5461–5533. 2. Sivaranjanee, R.; Senthil Kumar, P.; Saravanan, R.; Govarthanan, M. Electrochemical Sensing System for the Analysis of Emerging Contaminants in Aquatic Environment: A Review. Chemosphere 2022, 294, 133779. 3. Kaur, H.; Siwal, S. S.; Chauhan, G.; Saini, A. K.; Kumari, A.; Thakur, V. K. Recent Advances in Electrochemical-Based Sensors Amplified with Carbon-Based Nanomaterials (CNMs) for Sensing Pharmaceutical and Food Pollutants. Chemosphere 2022, 304, 135182. 4. Sumitha, M. S.; Xavier, T. S. Recent Advances in Electrochemical Biosensors—A Brief Review. Hybrid Adv. 2023, 2, 100023. 5. Ning, Z. H.; Huang, J. Q.; Guo, S. X.; Wang, L. H. A Portable Potentiostat for Three-Electrode Electrochemical Sensor. J. Phys. Conf. Ser. 2020, 1550 (4), 042049. 6. Ryu, H.; Thompson, D.; Huang, Y.; Li, B.; Lei, Y. Electrochemical Sensors for Nitrogen Species: A Review. Sens. Actuators Rep. 2020, 2 (1), 100022. 7. Crispi, S.; Nocito, G.; Nastasi, F.; Condorelli, G.; Ricciardulli, A. G.; Samorì, P.; Conoci, S.; Neri, G. Development of a Novel C-Dots Conductometric Sensor for NO Sensing. Sens. Actuators B 2023, 390, 133957. 8. Ferlazzo, A.; Espro, C.; Iannazzo, D.; Moulaee, K.; Neri, G. A Novel Yttria-Doped ZrO2 Based Conductometric Sensor for Hydrogen Leak Monitoring. Int. J. Hydrogen Energy 2022, 47 (16), 9819–9828. 9. Ghosh, P.; Biswas, S.; Kushagra, A. Development of Conductometric Glucose Sensor in Nanomolar (nM) Range from Phantom Blood Serum. Mater. Today. Proc. 2023, 80, 1865–1867. 10. Simões, F. R.; Codognoto, L.; Lima, T.d. S.; Fernandes-Filho, C. Pulsed Voltammetry for Monitoring of Emerging Contaminants. In Encyclopedia of Solid-Liquid Interfaces; Wandelt, K., Bussetti, G., Eds.; Elsevier, 2024; pp. 450–460. 11. Zhu, F.; Xue, Y.; Ji, W.; Li, X.; Ma, W.; Yu, P.; Jiang, Y.; Mao, L. Galvanic Redox Potentiometry for Fouling-Free and Stable Serotonin Sensing in a Living Animal Brain. Angew. Chem. Int. Ed. 2023, 62 (11), e202212458. 12. Bahadır, E. B.; Sezgintürk, M. K. A Review on Impedimetric Biosensors. Artif. Cells Nanomed. Biotechnol. 2014, 44 (1), 248–262. 13. Pourali, A.; Rashidi, M. R.; Barar, J.; Pavon-Djavid, G.; Omidi, Y. Voltammetric Biosensors for Analytical Detection of Cardiac Troponin Biomarkers in Acute Myocardial Infarction. TrAC Trends Anal. Chem. 2021, 134, 116123. 14. Sha, R.; Komori, K.; Badhulika, S. Amperometric pH Sensor Based on Graphene–Polyaniline Composite. IEEE Sens. J. 2017, 17 (16), 5038–5043. 15. Manjakkal, L.; Szwagierczak, D.; Dahiya, R. Metal Oxides Based Electrochemical pH Sensors: Current Progress and Future Perspectives. Prog. Mater Sci. 2020, 109, 100635. 16. Awasthi, P.; Mukherjee, R.; O Kare, S. P.; Das, S. Impedimetric Blood pH Sensor Based on MoS2–Nafion Coated Microelectrode. RSC Adv. 2016, 6 (104), 102088–102095.

Electrochemical Devices | Electrochemical Sensors and Actuators

407

17. Vivaldi, F.; Santalucia, D.; Poma, N.; Bonini, A.; Salvo, P.; Del Noce, L.; Melai, B.; Kirchhain, A.; Kolivoška, V.; Sokolová, R.; Hromadová, M.; Di Francesco, F. A Voltammetric pH Sensor for Food and Biological Matrices. Sens. Actuators B 2020, 322, 128650. 18. Zhang, Y.; Li, Z.; Guo, X.; Liu, G.; Zhang, S. Potentiometric Sensor Based on Carbon Paste Electrode for Monitoring Total Residual Chlorine in Electrolytically-Treated Ballast Water. Sensors 2021, 21 (2), 350. 19. Assaifan, A. K.; Hezam, M.; Al-Gawati, M. A.; Alzahrani, K. E.; Alswieleh, A.; Arunachalam, P.; Al-Mayouf, A.; Alodhayb, A.; Albrithen, H. Label-Free and Simple Detection of Trace Pb(II) in Tap Water Using Non-faradaic Impedimetric Sensors. Sens. Actuators A: Chem. 2021, 329, 112833. 20. Saiapina, O. Y.; Dzyadevych, S. V.; Walcarius, A.; Jaffrezic-Renault, N. A Novel Highly Sensitive Zeolite-Based Conductometric Microsensor for Ammonium Determination. Anal. Lett. 2012, 45 (11), 1467–1484. 21. Ilager, D.; Malode, S. J.; Shetti, N. P. Development of 2D Graphene Oxide Sheets-Based Voltammetric Sensor for Electrochemical Sensing of Fungicide, Carbendazim. Chemosphere 2022, 303, 134919. 22. Lee, H.; Lagger, G.; Pereira, C.; Silva, A.; Girault, H. Amperometric Tape Ion Sensors for Cadmium(II) Ion Analysis. Talanta 2009, 78 (1), 66–70. 23. Bobacka, J.; Ivaska, A.; Lewenstam, A. Potentiometric Ion Sensors. Chem. Rev. 2008, 108 (2), 329–351. 24. Alshawi, J. M. S.; Mohammed, M. Q.; Alesary, H. F.; Ismail, H. K.; Barton, S. Voltammetric Determination of Hg2+, Zn2+, and Pb2+ Ions Using a PEDOT/NTA-Modified Electrode. ACS Omega 2022, 7 (23), 20405–20419. 25. Bertocchi, P.; Compagnone, D.; Palleschi, G. Amperometric Ammonium Ion and Urea Determination with Enzyme-Based Probes. Biosens. Bioelectron. 1996, 11 (1–2), 1–10. 26. Imato, T.; Ishibashi, N. Potentiometric Butyrylcholine Sensor for Organophosphate Pesticides. Biosens. Bioelectron. 1995, 10 (5), 435–441. 27. Sandoval Bojórquez, D. I.; Janicijevic, Ž.; Palestina Romero, B.; Oliveros Mata, E. S.; Laube, M.; Feldmann, A.; Kegler, A.; Drewitz, L.; Fowley, C.; Pietzsch, J.; Fassbender, J.; Tonn, T.; Bachmann, M.; Baraban, L. Impedimetric Nanobiosensor for the Detection of SARS-CoV-2 Antigens and Antibodies. ACS Sens. 2023, 8 (2), 576–586. 28. Mikkelsen, S. R.; Rechnitz, G. A. Conductometric Tranducers for Enzyme-Based Biosensors. Anal. Chem. 2002, 61 (15), 1737–1742. 29. Batista, É. A.; Garcia, L. F.; Albuquerque, A. J. C. D.; Ballaminut, N.; Scalize, P. S.; Gil, E. S. Application of a Voltammetric Enzymatic Biosensor Based on Crude Extract of Marasmiellus Colocasiae for the Detection of Phenolic Compounds in Drinking Water. Revista Ambiente e Agua – Interdiscip. J. Appl. Sci. 2020, 15 (6), e2610. 30. Mizutani, F.; Yabuki, S.; Sato, Y. Voltammetric Enzyme Sensor for Urea Using Mercaptohydroquinone-Modified Gold Electrode as the Base Transducer. Biosens. Bioelectron. 1997, 12 (4), 321–328. 31. Liang, R.; Chen, L.; Qin, W. Potentiometric Detection of Chemical Vapors Using Molecularly Imprinted Polymers as Receptors. Sci. Rep. 2015, 5 (1), 12462. 32. Wilson, D.; Gutiérrez, J. M.; Alegret, S.; del Valle, M. Simultaneous Determination of Zn(II), Cu(II), Cd(II) and Pb(II) in Soil Samples Employing an Array of Potentiometric Sensors and an Artificial Neural Network Model. Electroanalysis 2012, 24 (12), 2249–2256. 33. Abdel-Ghany, M. F.; Hussein, L. A.; El Azab, N. F. Novel Potentiometric Sensors for the Determination of the Dinotefuran Insecticide Residue Levels in Cucumber and Soil Samples. Talanta 2017, 164, 518–528. 34. Vyshniakova, K.; Bai, H.; Rocha Malacco, V. M.; Pavlica, E.; Voyles, R. M.; Donkin, S. S.; Gehman, A.; Nawrocki, R. Electrochemical ZnO-Based Impedimetric Sensor for Aqueous Ammonia Detection for Precision Animal Agriculture. ECS Meet. Abstr. 2021, MA2021-01 (57), 1541. 35. Chen, S.; Chen, J.; Qian, M.; Liu, J.; Fang, Y. Low Cost, Portable Voltammetric Sensors for Rapid Detection of Nitrate in Soil. Electrochim. Acta 2023, 446, 142077. 36. Barhoum, A.; Hamimed, S.; Slimi, H.; Othmani, A.; Abdel-Haleem, F. M.; Bechelany, M. Modern Designs of Electrochemical Sensor Platforms for Environmental Analyses: Principles, Nanofabrication Opportunities, and Challenges. Trends Environ. Anal. Chem. 2023, 38, e00199. 37. Mariani, F.; Gualandi, I.; Schuhmann, W.; Scavetta, E. Micro- and Nano-Devices for Electrochemical Sensing. Microchimica Acta 2022, 189 (12), 459. 38. Mohan, J. M.; Amreen, K.; Javed, A.; Dubey, S. K.; Goel, S. Emerging Trends in Miniaturized and Microfluidic Electrochemical Sensing Platforms. Curr. Opin. Electrochem. 2022, 33, 100930. 39. Zhang, W.; Wang, R.; Luo, F.; Wang, P.; Lin, Z. Miniaturized Electrochemical Sensors and their Point-of-Care Applications. Chin. Chem. Lett. 2020, 31 (3), 589–600. 40. Putzbach, W.; Ronkainen, N. Immobilization Techniques in the Fabrication of Nanomaterial-Based Electrochemical Biosensors: A Review. Sensors 2013, 13 (4), 4811–4840. 41. Tan, H.-M.; Cheong, S.-P.; Tan, T.-C. An Amperometric Benzene Sensor Using Whole Cell Pseudomonas Putida ML2. Biosens. Bioelectron. 1994, 9 (1), 1–8. 42. Zhang, L.; Wen, Y.-P.; Yao, Y.-Y.; Wang, Z.-F.; Duan, X.-M.; Xu, J.-K. Electrochemical Sensor Based on f-SWCNT and Carboxylic Group Functionalized PEDOT for the Sensitive Determination of Bisphenol a. Chin. Chem. Lett. 2014, 25 (4), 517–522. 43. Biran, I.; Babai, R.; Levcov, K.; Rishpon, J.; Ron, E. Z. Online and In Situ Monitoring of Environmental Pollutants: Electrochemical Biosensing of Cadmium. Environ. Microbiol. 2000, 2 (3), 285–290. 44. Hu, X.; Leng, Z. Determination of Cyanide Using a Tyrosinase Amperometric Biosensor with Catechol as Substrate. Analyst 1995, 120 (5), 1555–1557. 45. Han, H. S.; You, J.-M.; Seol, H.; Jeong, H.; Jeon, S. Electrochemical Sensor for Hydroquinone and Catechol Based on Electrochemically Reduced GO–Terthiophene–CNT. Sens. Actuators B 2014, 194, 460–469. 46. Wang, J. Preconcentration and Voltammetric Measurement of Mercury with a Crown-Ether Modified Carbon-Paste Electrode. Talanta 1988, 35 (4), 277–280. 47. Baldwin, R. P.; Christensen, J. K.; Kryger, L. Voltammetric Determination of Traces of Nickel(II) at a Chemically Modified Electrode Based on Dimethylglyoxime-Containing Carbon Paste. Anal. Chem. 2002, 58 (8), 1790–1798. 48. Kalcher, K. A New Method for the Voltammetric Determination of Nitrite. Talanta 1986, 33 (6), 489–494. 49. Gorski, W.; Cox, J. A. Amperometric Determination of N-Nitrosamines in Aqueous Solution at an Electrode Coated with a Ruthenium-Based Inorganic Polymer. Anal. Chem. 2002, 66 (17), 2771–2774. 50. Kariuki, V. M.; Fasih-Ahmad, S. A.; Osonga, F. J.; Sadik, O. A. An Electrochemical Sensor for Nitrobenzene Using p-Conjugated Polymer-Embedded Nanosilver. Analyst 2016, 141 (7), 2259–2269. 51. Wang, J.; Chen, Q. Microfabricated Phenol Biosensors Based on Screen Printing of Tyrosinase Containing Carbon Ink. Anal. Lett. 1995, 28 (7), 1131–1142. 52. Rawal, R.; Pundir, C. S. Development of an Amperometric Sulfite Biosensor Based on SOx/PBNPs/PPY Modified ITO Electrode. Int. J. Biol. Macromol. 2012, 51 (4), 449–455. 53. Švitel, J.; Stanislav, M. Development of Tyrosinase-Based Biosensor and its Application for Monitoring of Bioremediation of Phenol and Phenolic Compounds. Environ. Sci. Technol. 1998, 32 (6), 828–832. 54. Chung, S.; Sicklick, J. K.; Ray, P.; Hall, D. A. Development of a Soluble KIT Electrochemical Aptasensor for Cancer Theranostics. ACS Sens. 2021, 6 (5), 1971–1979. 55. Boonkaew, S.; Yakoh, A.; Chuaypen, N.; Tangkijvanich, P.; Rengpipat, S.; Siangproh, W.; Chailapakul, O. An Automated Fast-Flow/Delayed Paper-Based Platform for the Simultaneous Electrochemical Detection of Hepatitis B Virus and Hepatitis C Virus Core Antigen. Biosens. Bioelectron. 2021, 193, 113543. 56. Marie, M.; Mandal, S.; Manasreh, O. An Electrochemical Glucose Sensor Based on Zinc Oxide Nanorods. Sensors 2015, 15 (8), 18714–18723. 57. Ponnaiah, S. K.; Periakaruppan, P.; Vellaichamy, B. New Electrochemical Sensor Based on a Silver-Doped Iron Oxide Nanocomposite Coupled with Polyaniline and its Sensing Application for Picomolar-Level Detection of Uric Acid in Human Blood and Urine Samples. J. Phys. Chem. B. 2018, 122 (12), 3037–3046. 58. Dutta, G. Electrochemical Biosensors for Rapid Detection of Malaria. Mater. Sci. Energy Technol. 2020, 3, 150–158. 59. Arduini, F.; Forchielli, M.; Scognamiglio, V.; Nikolaevna, K.; Moscone, D. Organophosphorous Pesticide Detection in Olive Oil by Using a Miniaturized, Easy-to-Use, and Cost-Effective Biosensor Combined with QuEChERS for Sample Clean-Up. Sensors 2016, 17 (12), 34. 60. Silva, N. F. D.; Almeida, C. M. R.; Magalhães, J. M. C. S.; Gonçalves, M. P.; Freire, C.; Delerue-Matos, C. Development of a Disposable Paper-Based Potentiometric Immunosensor for Real-Time Detection of a Foodborne Pathogen. Biosens. Bioelectron. 2019, 141, 111317. 61. Muniandy, S.; Dinshaw, I. J.; Teh, S. J.; Lai, C. W.; Ibrahim, F.; Thong, K. L.; Leo, B. F. Graphene-Based Label-Free Electrochemical Aptasensor for Rapid and Sensitive Detection of Foodborne Pathogen. Anal. Bioanal. Chem. 2017, 409 (29), 6893–6905. 62. Son, M.; Kim, D.; Kang, J.; Lim, J. H.; Lee, S. H.; Ko, H. J.; Hong, S.; Park, T. H. Bioelectronic Nose Using Odorant Binding Protein-Derived Peptide and Carbon Nanotube Field-Effect Transistor for the Assessment of Salmonella Contamination in Food. Anal. Chem. 2016, 88 (23), 11283–11287.

408

Electrochemical Devices | Electrochemical Sensors and Actuators

63. Malvano, F.; Pilloton, R.; Albanese, D. Sensitive Detection of Escherichia coli O157:H7 in Food Products by Impedimetric Immunosensors. Sensors 2018, 18 (7), 2168. 64. Helali, S.; Sawelem Eid Alatawi, A.; Abdelghani, A. Pathogenic Escherichia Coli Biosensor Detection on Chicken Food Samples. J. Food Saf. 2018, 38 (5), e12510. 65. Guo, Y.; Wang, Y.; Liu, S.; Yu, J.; Wang, H.; Wang, Y.; Huang, J. Label-Free and Highly Sensitive Electrochemical Detection of E. Coli Based on Rolling Circle Amplifications Coupled Peroxidase-Mimicking DNAzyme Amplification. Biosens. Bioelectron. 2016, 75, 315–319. 66. Idil, N.; Hedström, M.; Denizli, A.; Mattiasson, B. Whole Cell Based Microcontact Imprinted Capacitive Biosensor for the Detection of Escherichia coli. Biosens. Bioelectron. 2017, 87, 807–815. 67. Lu, Y.; Liu, Y.; Zhao, Y.; Li, W.; Qiu, L.; Li, L. A Novel and Disposable Enzyme-Labeled Amperometric Immunosensor Based on MWCNT Fibers for Listeria Monocytogenes Detection. J. Nanomater. 2016, 2016, 1–8. 68. Bhardwaj, J.; Devarakonda, S.; Kumar, S.; Jang, J. Development of a Paper-Based Electrochemical Immunosensor Using an Antibody-Single Walled Carbon Nanotubes Bio-Conjugate Modified Electrode for Label-Free Detection of Foodborne Pathogens. Sens. Actuators B 2017, 253, 115–123. 69. Wang, H.; Xiu, Y.; Chen, Y.; Sun, L.; Yang, L.; Chen, H.; Niu, X. Electrochemical Immunosensor Based on an Antibody-Hierarchical Mesoporous SiO2 for the Detection of Staphylococcus aureus. RSC Adv. 2019, 9 (28), 16278–16287. 70. Jasim, I.; Shen, Z.; Mlaji, Z.; Yuksek, N. S.; Abdullah, A.; Liu, J.; Dastider, S. G.; El-Dweik, M.; Zhang, S.; Almasri, M. An Impedance Biosensor for Simultaneous Detection of Low Concentration of Salmonella Serogroups in Poultry and Fresh Produce Samples. Biosens. Bioelectron. 2019, 126, 292–300. 71. Riu, J.; Giussani, B. Electrochemical Biosensors for the Detection of Pathogenic Bacteria in Food. TrAC Trends Anal. Chem. 2020, 126, 115863. 72. Deng, Q.; Jia, H.; An, C.; Wu, S.; Zhao, S.; Hu, N. Progress and Prospective of Electrochemical Actuator Materials. Compos. A: Appl. Sci. Manuf. 2023, 165, 107336. 73. Sansiñena, J. M.; Gao, J.; Wang, H. L. High-Performance, Monolithic Polyaniline Electrochemical Actuators. Adv. Funct. Mater. 2003, 13 (9), 703–709. 74. Smela, E. Conjugated Polymer Actuators for Biomedical Applications. Adv. Mater. 2003, 15 (6), 481–494. 75. Pei, Q.; Inganäs, O. Electrochemical Muscles: Bending Strips Built from Conjugated Polymers. Synth. Met. 1993, 57 (1), 3718–3723. 76. Wataru Takashima, W. T.; Masanori Fukui, M. F.; Masamitsu Kaneko, M. K.; Keiichi Kaneto, K. K. Electrochemomechanical Deformation of Polyaniline Films. Jpn. J. Appl. Phys. 1995, 34, 3786–3789. 77. Das, T. K.; Prusty, S. Review on Conducting Polymers and their Applications. Polym.-Plast. Technol. Eng. 2012, 51 (14), 1487–1500. 78. Wang, Z.; He, B.; Liu, X.; Wang, Q. Development and Modeling of a New Ionogel Based Actuator. J. Intell. Mater. Syst. Struct. 2017, 28 (15), 2036–2050. 79. Yang, Y. W.; Zhang, L.; Soh, C. K. Ionic Polymer-Metal Composite and its Actuation Characteristics. In Smart Materials in Structural Health Monitoring, Control and Biomechanics; 2012; pp. 501–532. 80. Shahinpoor, M.; Bar-Cohen, Y.; Simpson, J. O.; Smith, J. Ionic Polymer-Metal Composites (IPMCs) as Biomimetic Sensors, Actuators and Artificial Muscles—A Review. Smart Mater. Struct. 1998, 7 (6), R15–R30. 81. Chen, Z.; Hedgepeth, D. R.; Tan, X. A Nonlinear, Control-Oriented Model for Ionic Polymer–Metal Composite Actuators. Smart Mater. Struct. 2009, 18 (5), 055008. 82. Bennett, M. D.; Leo, D. J. Ionic Liquids as Stable Solvents for Ionic Polymer Transducers. Sens. Actuators A. Phys. 2004, 115 (1), 79–90. 83. Bonomo, C.; Fortuna, L.; Giannone, P.; Graziani, S.; Strazzeri, S. A Nonlinear Model for Ionic Polymer Metal Composites as Actuators. Smart Mater. Struct. 2007, 16 (1), 1–12. 84. Holmberg, S.; Holmlund, P.; Nicolas, R.; Wilén, C.-E.; Kallio, T.; Sundholm, G.; Sundholm, F. Versatile Synthetic Route to Tailor-Made Proton Exchange Membranes for Fuel Cell Applications by Combination of Radiation Chemistry of Polymers with Nitroxide-Mediated Living Free Radical Graft Polymerization. Macromolecules 2004, 37 (26), 9909–9915. 85. Allen, N. S. Photopolymerisation and Photoimaging Science and Technology, Springer: Dordrecht, 1989. 86. Sahrash, R.; Siddiqa, A.; Razzaq, H.; Iqbal, T.; Qaisar, S. PVDF Based Ionogels: Applications towards Electrochemical Devices and Membrane Separation Processes. Heliyon 2018, 4 (11), e00847. 87. Hanrahan, G.; Patil, D. G.; Wang, J. Electrochemical Sensors for Environmental Monitoring: Design, Development and Applications. J. Environ. Monit. 2004, 6 (8), 657–664. 88. Jiang, Z.; Feng, B.; Xu, J.; Qing, T.; Zhang, P.; Qing, Z. Graphene Biosensors for Bacterial and Viral Pathogens. Biosens. Bioelectron. 2020, 166, 112471. 89. Ferrag, C.; Kerman, K. Grand Challenges in Nanomaterial-Based Electrochemical Sensors. Fron. Sens. 2020, 1, 583822. 90. Muniandy, S.; Teh, S. J.; Thong, K. L.; Thiha, A.; Dinshaw, I. J.; Lai, C. W.; Ibrahim, F.; Leo, B. F. Carbon Nanomaterial-Based Electrochemical Biosensors for Foodborne Bacterial Detection. Crit. Rev. Anal. Chem. 2019, 49 (6), 510–533. 91. Ligler, F. S.; Gooding, J. J. Lighting Up Biosensors: Now and the Decade to Come. Anal. Chem. 2019, 91 (14), 8732–8738. 92. Ismail, N. S.; Husain, U. S.; Selvan, S. I. S.; Mordani, N. A.; Juhari, N.; Halim, N. H. A. Effect of heating power towards synthesis of carbon dots through microwave pyrolysis method for optical-based biosensor. In The 2nd International Conference on Applied Photonics and Electronics 2019 (InCAPE 2019); 2020. 93. Durai, L.; Badhulika, S. Current Challenges and Developments in Perovskite-Based Electrochemical Biosensors for Effective Theragnostics of Neurological Disorders. ACS Omega 2022, 7 (44), 39491–39497. 94. Wu, Y.; Tilley, R. D.; Gooding, J. J. Challenges and Solutions in Developing Ultrasensitive Biosensors. J. Am. Chem. Soc. 2018, 141 (3), 1162–1170. 95. Kim, J.; Jeerapan, I.; Sempionatto, J. R.; Barfidokht, A.; Mishra, R. K.; Campbell, A. S.; Hubble, L. J.; Wang, J. Wearable Bioelectronics: Enzyme-Based Body-Worn Electronic Devices. Acc. Chem. Res. 2018, 51 (11), 2820–2828. 96. Meng, L.; Turner, A. P. F.; Mak, W. C. Soft and Flexible Material-Based Affinity Sensors. Biotechnol. Adv. 2020, 39, 107398. 97. Li, T.; Li, Y.; Zhang, T. Materials, Structures, and Functions for Flexible and Stretchable Biomimetic Sensors. Acc. Chem. Res. 2019, 52 (2), 288–296. 98. Menon, S.; Mathew, M. R.; Sam, S.; Keerthi, K.; Kumar, K. G. Recent Advances and Challenges in Electrochemical Biosensors for Emerging and Re-Emerging Infectious Diseases. J. Electroanal. Chem. 2020, 878. 99. Chiorcea-Paquim, A.-M. Advances in Electrochemical Biosensor Technologies for the Detection of Nucleic Acid Breast Cancer Biomarkers. Sensors 2023, 23 (8), 4128. 100. Singh, A.; Sharma, A.; Ahmed, A.; Sundramoorthy, A. K.; Furukawa, H.; Arya, S.; Khosla, A. Recent Advances in Electrochemical Biosensors: Applications, Challenges, and Future Scope. Biosensors 2021, 11 (9), 336. 101. Marx, Í. M. G. Emerging Trends of Electrochemical Sensors in Food Analysis. Electrochemistry 2023, 4 (1), 42–46. 102. Umapathi, R.; Ghoreishian, S. M.; Rani, G. M.; Cho, Y.; Huh, Y. S. Review—Emerging Trends in the Development of Electrochemical Devices for the On-Site Detection of Food Contaminants. ECS Sens. Plus 2022, 1 (4), 044601. 103. Baranwal, J.; Barse, B.; Gatto, G.; Broncova, G.; Kumar, A. Electrochemical Sensors and their Applications: A Review. Chem 2022, 10 (9), 363. 104. Guan, R.; Yu, J.; Li, M.; Yan, J.; Liu, Z. Preparation of Electrochemical Sensor Assisted Unmanned Aerial Vehicles System for SO2, O3, NO2, CO and PM2.5/PM10 Detection in Air. Int. J. Electrochem. Sci. 2021, 16 (10), 211021. 105. Mawrence, R.; Munniks, S.; Valente, J. Calibration of Electrochemical Sensors for Nitrogen Dioxide Gas Detection Using Unmanned Aerial Vehicles. Sensors 2020, 20 (24), 7332.

Electrochemical Devices | Photoelectrochemical Cells Peter Kurzweil, Technical University of Applied Sciences (OTH), Amberg-Weiden, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of F. Decker, S. Cattarin, PHOTOELECTROCHEMICAL CELLS | Overview, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 1–9, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00035-6.

1 2 2.1 2.2 3 3.1 3.2 3.3 3.4 3.4.1 3.4.2 3.5 3.5.1 4 5 5.1 5.2 6 References

Introduction Photovoltaics Types of solar cells Semiconductor solar cells Electrochemical solar cells Semiconductor electrodes Photogalvanic cell Dye-sensitized solar cell Solid state solar cell Perovskite technology Inorganic layer structures Photocorrosion Electrolytes for the stability challenge Photoelectrolysis Photoelectrocatalysis Titanium dioxide Photocatalysts for visible light Conclusion

410 411 412 412 414 414 415 415 417 417 417 417 418 419 420 420 421 421 421

Abstract Photoelectrochemical cells extract electrical energy from light. This overview chapter outlines the principle of photoelectrochemical solar cells, photoelectrolysis, photocatalysis and similar applications that combine electrochemistry and semiconductors. The properties and differences of photoelectrochemical cells and photovoltaic cells are compared.

Key points

• • •

Fundamentals of active and passive photoelectrochemical cells Design of semiconductor electrodes and solar cells Insight into photocatalytic and corrosion processes

Nomenclature

Symbols and Units A C e EF El F h I k P T t

area (m2) capacitance (F) elementary charge: 1.602
10–19 C Fermi energy (eV) irradiance (W m−2) fill factor for a solar cell Planck’s constant: 6.626
10−34 Js electric current (A) Boltzmann constant: 1.3806
10−23 J K−1 electric power (W) thermodynamic temperature (K) time (s)

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00033-5

409

410

Electrochemical Devices | Photoelectrochemical Cells

U e0 h n

electric voltage (V) electric constant: 8.854
10−12 F m−1 conversion efficiency of a solar cell frequency of light (Hz)

Abbreviations and Acronyms PEC PV SHE UV

1

photo-electrochemical cell photovoltaic standard hydrogen electrode ultraviolet (electromagnetic radiation)

Introduction

Photoelectrochemical cells extract electrical energy from light, including sunlight. Each cell consists of one or two semiconducting photoelectrodes and also auxiliary metal and reference electrodes immersed in an electrolyte. In addition to, or instead of, electrical energy, useful fuels may be produced in a process such as the electrolysis of water to hydrogen and oxygen. An operating photoelectrochemical cell is generally represented by its energy level diagram. Fig. 1 shows such a diagram for the two most common cells, having one photoelectrode (p- or n-type) and one metal electrode.

Fig. 1 (a) Energy level diagram of a n-type semiconductor photoanode and metal cathode regenerative photoelectrochemical cell, at short circuit. (b) Energy level diagram of a p-type semiconductor photocathode and metal anode regenerative cell, at short circuit.

Electrochemical Devices | Photoelectrochemical Cells

411

In the early attempts, from the early 1970s onward, to use such cells for the solar generation of hydrogen, the conversion efficiency was low (about 1%) because of the large bandgap of the used oxide photoelectrodes and the consequent mismatch with the solar spectrum. Solar cell efficiency has been largely increased by using non-oxide semiconductors with a smaller bandgap, such as II–VI and III–IV compounds and, to a lesser extent, silicon: such semiconductors are more prone to corrosion than oxides, and carefully tailored redox electrolytes have to be used to protect the photoelectrodes from degradation. Photosensitizers applied to large-bandgap nanocrystalline oxides, such as textured titanium dioxide (TiO2), in the so-called ‘Grätzel cell’ or dye-sensitized nanostructured solar cells, have efficiencies on the order of 11% and promise to be cheap and suitable for collecting low-intensity sunlight. High efficiencies have been obtained with hybrid arrangements involving the use of multiple bandgap structures: regenerative photoelectrochemical cells sustaining over 19% solar-to-electrical conversion efficiency and electrolytic cells sustaining over 18% solar conversion efficiency to hydrogen. Considerable progress during the last 40 years is a consequence of improvement of semiconductor materials, better understanding of the connection between the efficiency and stability issues, and the ever-increasing ability to synthesize nanostructures of desired properties. Yet for large-scale solar energy conversion, it is hardly possible to make a sound comparison between photoelectrochemical cells and silicon-based photovoltaic cells for which the conversion efficiency has increased from 10% to over 20%. Photovoltaic cell fabrication is a mature industrial technology today: the estimated yearly silicon solar cell production was 1800 MW in 2006, with a cell life expectation of 25 years. Photoelectrochemical cells are still, on the contrary, mainly prototypes with a shorter life expectation, produced on a small scale. However, they offer an intimate contact between semiconductor and electrolyte, and that this junction can be assembled by mere dipping and disassembled for inspection by simple emersion from the solution. The two most important requirements for semiconducting photoelectrodes in an efficient and stable solar cell are (1) a good match of its bandgap (the energy separation between the valence and conduction band edges) with the spectrum of the incident radiation (usually the solar spectrum) and (2) well-tailored redox processes at both electrodes. Photoelectrochemical cells can be divided into groups according to the basic mode of operation:

• •

regenerative cells, in other words wet photovoltaic cells generating external electrical work with no net change in electrolyte composition (no Gibbs function change in the cell, DG ¼ 0); photoelectrolytic cells, in which two different redox reactions are driven at the two cell electrodes, with an overall endergonic process, DG > 0. Spontaneous photoelectrocatalytic processes do not proceed at significant rates in the absence of suitable conditions.

A number of cells belonging to each group are listed in Table 1, in the order of solar conversion efficiency above 10%, but this selection is by no means exhaustive.

2

Photovoltaics

A solar cell (photovoltaic cell)1,2 converts sunlight directly into electricity. A solar cell is not to be confused with a solar-thermal collector that heats a water reservoir using solar irradiation. Conventional photovoltaic cells use semiconductors to generate electricity, while electrochemical solar cells also use redox systems.

Table 1 Selection of photoelectrochemical cells for solar energy conversion: Solar conversion efficiency reported by S. Licht (1998, 2001, 1990), B. J. Tufts (1987), J. W. Gibbons (1984), K. Kawakami (1997), R. Tenne (1985), O. Khaselev (1998), A. Heller (1981), M. K. Nazeeruddin, G. Kline (1980). Electrodes Regenerative AlGaAs/Si multijunction n-CdSe; metal CE n-GaAs (Os3+ modified)/Pt n-Si; ITO–Au grid Porous n-Si/Pt; Pt CE n-WSe2; metal CE p-InP; C n-TiO2/cis-RuL2(SCN)2; ITO + catalyst CE n-WSe2; metal CE Photoelectrolysis AlGaAs/Si multijunction; RuO2-Pt black p-GaInP2; n-GaAs p-InP (Ru); Pt (Rh)

Redox couple/electrolyte

Stability

Conversion efficiency (%)

Aqueous 10 mol/L HI + 0.01 mol/L I2 Aqueous 0.5 mol/L KOH + hexacyanoferrate(II,III) Aqueous 1 mol/L K2Se + 0.01 mol/L K2Se2 Ferrocene, CH3OH, thin-layer cell Aqueous (8 mol/L HBr)/Br2 Aqueous (6 mol/L KI)/I2 Aqueous HC + V3+/V2+ 0.6 mol/L LiI + 0.1 mol/L I2 in CH3CN Aqueous (1 mol/L KI)/I2

1 h, HF added 3 days, KCN added >3000 C cm−2 1 month, 70,000 C cm−2 2h – 2 months, 30,000 C cm−2 1000 h 41 months, 400,000 C cm−2

19.2 16.4 15 14 14 14 11.5 10.4 10.2

1 mol/L HClO4 Aqueous H2SO4 Aqueous HCl + KCl

14 h 20 h 10,000 C cm−2

18.3 12.4 12

412 2.1

Electrochemical Devices | Photoelectrochemical Cells Types of solar cells

Direct conversion of solar energy into electrical energy is achieved with satisfactory efficiency on semiconductors (Si, GaAs, CdTe) and conductive polymers (polyaniline, polythiophene, polyphenylene). Thick film and thin film cells use crystalline or amorphous silicon (Table 2). Semiconductor solar cells are connected in series to the solar module (panel) with conductors on the front and back. Shaded individual cells must be electrically bridged by bypass diodes. Tandem cells (multi-junction solar cell) stack several light-transmitting thin films with different spectral sensitivity (band gap) on top of each other. Concentrator cells save expensive semiconductor space by concentrating sunlight with lenses, mirrors or light guides and tracking the position of the sun; fluorescent cells do not need to be tracked and convert short wavelengths into longer ones for silicon solar cells.

2.2

Semiconductor solar cells

A solar cell is driven by the photovoltaic effect. A solar cell is a semiconductor diode with a wide space charge region: (1) sun-facing and transparent, a strongly n-doped surface layer, and (2) a weakly p-doped support layer on a mirrored contact (copper, silver), as shown Fig. 2. Light quanta with an energy larger than the band gap of the semiconductor (see Table 3) generate electron-hole pairs by the internal photoelectric effect; the electrons migrate to the n-contact, the holes to the p-contact. The free charge carriers are separated by the electric field of the space charge region. Some of the minority charge carriers recombine under heat release. In the boron-doped p-type silicon crystal, excited electrons (minority charge carriers) diffuse long distances to the n-type layer and separate from the holes (majority charge carriers) retained by the opposing field of the pn-junction. Silicon shimmers silver-gray, mechanically roughened silicon is black, an antireflection layer (Si3N4, SiO2, TiO2) appears bluish-black or in an interference color. Silicon absorbs in the red spectral range, so a blue reflection ensures the best efficiency. Silicon oxide and nitride also act as a passivation layer against the rapid recombination of charge carriers on the surface. A silicon cell delivers 0.64 V open-circuit voltage, and about 0.5 V under peak load with power matching. A 60-cell module delivers 38–30 V and 230-260 W. Performance data. The characteristics of a solar cell are given for normalized test conditions. Solar cells produce roughly 160 W/m2 of power on earth, and 220 W/m2 in space, since no atmosphere absorbs radiation. The photocurrent increases proportionally to the irradiance (Fig. 2c). The irradiance 1000 W/m2 at module level corresponds to a mid-summer radiation spectrum in Central Europe at sea level at 48 latitude (AM 1.5 global). In winter, the sun is lower; the sun rays pass through a light path that is several times the atmospheric altitude (AM 4. . .6). Global radiation includes diffuse and direct radiation from the sun.

Table 2

Types of solar cells.

Material

Layer thickness

Inorganic solar cells Silicon thick thin

GaAs CdTe CuInS2 Cu(In,Ga)Se2 Cu2ZnSnS4 Perovskite

wire – thin thin – – –

Semi-conductor

Efficiency (%)

Application, advantages and challenges

mono-crystalline

19. . .26

poly-crystalline amorphous

17. . .22 6. . .14

microcrystalline

15

0.02–0.05 kW/kg, Waver thickness 200 mm, mature technology, expensive, only for red and infrared light widespread, castable, potential Z > 20% < 2 kW/kg, calculators, watches, inexpensive, vapor-depositable on glass or aluminum, long-lasting, suitable for low and scattered light, high temperatures tandem cell, performance loss when heating

III–V II–VI I-III-VI – – AIIBIVO3

30. . .40 16. . .22 15. . .23 – 10 4. . .20

bendable, economical use of materials (Ga,In)P|GaAs|Ge in space industry, heat and UV resistant. Expensive low-cost bath and vapor phase deposition copper(I)-indium(III)-disulfide Indium, gallium, selene: scarce and expensive band gap 1.5 eV (sulfide), 1.0 eV (selenide) ultrathin, for red., green and blue light; short life, sensitive to moisture; experimental status: CdPbI3, CH3NH3SnI3

0) are environmentally benign and store solar energy in a way convenient for subsequent recovery by allowing the reverse, spontaneous reaction. It is well known, however, that in order to be able to decompose water into hydrogen and oxygen, semiconductor photoelectrodes must have a large bandgap, which is not suitable for harvesting most of the solar spectrum. Moreover, for unassisted water splitting, the semiconductor band edges must encompass conveniently oxygen and hydrogen reactions, sustained by the photon flux and occurring without external bias. The energetic position of these band edges is determined by the chemistry of the semiconductor–electrolyte interface, which is controlled by the composition of the semiconductor, the nature of the surface, and the electrolyte composition. Photoelectrolysis cell. Ideally, no external voltage source is required. Charge separation and water splitting occur by the help of the energy of electron-hole pairs on the same catalytic surface. Stable semiconductors with a large band gap between the valence band and the conduction band are required, and only the short-wave part of the sunlight is used. On the hydrogen side, a metal sheet is sufficient on the hydrogen side; a semiconductor electrode with a different band gap is more suitable in order to make better use of the solar spectrum (Fig. 8). Balanced illumination is difficult. With multilayer electrodes, 30% efficiency is theoretically possible. The electrode processes read: (−) Cathode (p-type or metal), 0.41 V at pH 7: 4 H2O + 4 e− ! 2H2 + 4OH− (+) Anode (n-type), +0.80 V at pH 7: 4 OH− ! O2 + 4 e− + 2 H2O GaN-ZnO and ZnGeN2-ZnO provide only 0.1% quantum yield; months of photolysis produce only micromolar amounts of gases. Copper(I) oxide (Cu2O) with copper thiocyanate CuSCN as a transparent, hole-conducting light transport layer (EPFL Lausanne) makes metallic back contacts unnecessary for electron-hole recombination and filtering of sunlight.

(+) n-Anode

p-Cathode (–)

ZnS

V SHE –2 Si

H+

hv

H2 O2 H2O

TiO2 WO3 D-Fe2O3 BiVO4 Ta3N5 CdS –0.9

hv

+2

–0.9

–0.5

–0.5

+0.6

H2 345 nm 595 nm 3.6 eV 2.1 eV 520 nm 2.4 eV 520 nm O2 568 nm 2.4 eV 415 nm 2.2 eV 3.2 eV 446 nm +1.2 2.8 eV +1.5 +1.8 +2.2 +2.1

0.4 0 0.8

–1.8 –1.8

–0.3

+0.1

0.0

+2.7 +2.9

(a)

(b)

Fig. 8 (a) Principle of an np bipolar photoelectrolysis cell: Energy level diagram of photoanode and cathode made of different semiconductors, both under illumination, working without application of external bias. (b) Band edges of semiconductor powders in water. Modified from Kurzweil, P. Angewandte Elektrochemie (Applied Electrochemistry), Springer Vieweg: Wiesbaden, 2020.

420

Electrochemical Devices | Photoelectrochemical Cells

Examples for photoelectrolysis cells:

• • •

p-GaP or p-Si anode, n-metal sulfide cathode p-GaInP2 anode, n-GaAs cathode, in sulfuric acid, 12% efficiency 12% and some hours lifetime. p-InP (Ru) anode, Pt (Rh) cathode, in HCl/KCl solution, 12% efficiency 12%, and some days lifetime.

Development strategies. In the early 1970s, water photoelectrolysis had been demonstrated to take place only under UV illumination. More recently, water splitting under visible light (although with low solar efficiency) was achieved with hydrated layered perovskites modified with nickel. In the early 1980s, Texas Instruments developed a program based on a hybrid PV photoelectrolytic system that used embedded multilayer silicon spheres acting as photoanode and photocathode yielding bromine and hydrogen, respectively. The system was designed to provide aqueous hydrogen bromide splitting in acidic solution, at the maximum power point of the silicon PV component. The program was discontinued in 1983, but had a positive impact in the field of PECs. Looking for higher efficiency of photoconversion, a two-photoelectrode configuration (each with a different bandgap) can better exploit the solar spectrum than that with one bandgap alone, but requires a careful choice of photoelectrode bandgaps and a well-balanced illumination of both. The balance of illumination can be extremely critical if the irradiation intensity and spectrum are varying during the day, as is often the case with sunlight. The energetics of p/n cells, containing simultaneously illuminated p-type photocathodes and n-type photoanodes, has been investigated by using appropriate combinations of n-type disulfides and p-GaP (or p-Si) semiconductor electrodes. Appropriate combinations of p- and n-type III–V semiconductors were later experimentally demonstrated with p-InP and n-GaAs (8.2% efficiency to generate H2 and O2) and with p-GaInP2 and n-GaAs (12.4% efficiency). An alternative strategy to improve efficiency is the use of multijunction electrodes, made of two or more layers of different semiconductors on the same piece of substrate (‘monolithic’). This was achieved with an integrated, hybrid photoelectrochemical–photovoltaic arrangement involving the use of multiple bandgap structure designs such as those of the electrodes in the two most efficient cells. Small bandgap semiconductors were always modified in order to limit photocorrosion reactions occurring in competition with photoelectrolysis. A thin layer of platinum catalyst, electrochemically deposited on the surface of semiconductor photoelectrodes, reduces the overvoltage losses typical of hydrogen and oxygen evolution reactions on a noncatalytic semiconductor surface. A fundamental analysis predicted that solar-to-chemical energy conversion efficiency in hydrogen generation from water could be as high as 30% under the following conditions: (a) use of efficient monolithic multijunction semiconductor photoelectrodes; (b) use of effective catalysts in the electrolytic process; and (c) separation of photoactive and electrolysis surface area. These conditions mean that the electrolysis cell and the solar, solid-state PV cell must be two different parts of the whole energy conversion system, in order to be able to optimize the two parts separately, each with the most updated and effective materials. In fact, the photocurrent per unit surface area in an efficient PV cell is always much larger than the electrode current density effective for water splitting or for other synthetic reactions taking place at low overpotential. Thus, the PV component should ideally be under concentrated insolation (typically from 100 to 1000 suns to improve cell efficiency) in order to photodrive charge into the electrolysis component. The experimental verification of the above ideas has demonstrated that a monolithic multijunction PV cell (AlGaAs cap layers onto p-Si/n-Si/n-Si multijunction substrate, with a GaAs buffer layer), coupled to a modern water electrolyzer (with Pt black and RuO2 electrodes), sustained water splitting under solar light at 18.3% conversion efficiency (see Table 1 above).

5

Photoelectrocatalysis

The photoelectrochemistry has developed since 1975 into a closely related area—that of photoelectrocatalysis20–24 specifically devoted to environmental tasks. In photoelectrocatalysis, the photon absorption promotes an exergonic reaction with DG < 0, so there is no net storage of chemical energy, but the radiant energy speeds up an otherwise slow reaction. Early research on semiconductor powders pointed out that such small-size solid particles could be used for the photocatalytic generation of useful products as well as for the destruction of pollutants in the cell solvent. The first of such reports was in 1977, in which the photodecomposition of cyanide in the presence of aqueous titanium dioxide suspensions was described and the implications for the field of environmental purification highlighted.

5.1

Titanium dioxide

Photoelectrocatalysis accelerates slow reactions. Holes created in suspended TiO2 particles by light have a strong oxidizing effect, generate hydroxyl radicals from water; the photoelectrons reduce oxygen to hyperoxide O−2. Doping with transition metal ions shifts the band gap into the visible range. Example: TiO2 particles catalyze the decomposition of cyanide, support air purification and disinfection, and provide self-cleaning glass surfaces. The large availability of materials such as titanium dioxide, which is relatively inexpensive and highly stable chemically, makes it a good photocatalyst (sometimes an even better one if activated locally by tiny metal islands) because its photogenerated holes are

Electrochemical Devices | Photoelectrochemical Cells

421

highly oxidizing, yielding the reactive hydroxyl radical (●OH) from water, and its photogenerated electrons are reducing enough to produce superoxide O−2 from O2. Although it is very difficult to confirm experimentally that oxidative and reductive photocatalytic reactions take place simultaneously on titanium dioxide particles, several studies have shown that the role played by superoxide is often (although not always) important and that many reduction reactions can take place not only on the metallic regions of the photocatalytic particle but also on the TiO2 itself. Many applications of such environmental photocatalysis are already at or near the stage of commercialization, for example, photocatalytic indoor and automobile air cleaners, self-cleaning ceramic tiles for bathrooms and kitchens, self-cleaning Venetian window blinds, and glass covers for highway tunnel lamps. The tunnel lamp application is a good example of the appropriate use of photocatalytic cleaning technology, because the light source is already built into the system and the flux of organic contaminants to the surface is more or less balanced with the rate at which the photocatalyst can break them down. Because cleaning processes are time consuming, expensive, and sometimes even dangerous, all inventions allowing a reduction in the need for cleaning, by the use of efficient photocatalysts, are clearly welcome and commercially profitable. The ability of titanium dioxide films to kill bacteria and other microorganisms is also a very active area for research and development, such as that of self-sterilizing ceramic tiles for hospitals and operating rooms. Bacteria are organisms made up of organic compounds, and thus are subject to the same types of decomposition reactions discussed above. Of particular interest is very low-intensity UV light (i.e., nW cm−2 to mWcm−2), similar to the intensity existing in ambient indoor lighting. The efficiency for the photocatalytic decomposition of simple organic compounds was found to depend on the intensity of light, in such a way that one should maximize the number of adsorbed molecules in order to optimize light utilization at low intensity. In other cases, when there is not enough UV intensity in ambient light, it would be desirable to increase titanium dioxide optical absorption of visible light in order to be able to increase the rate of photocatalytic reaction.

5.2

Photocatalysts for visible light

High-energy implantation of transition metal ions and other doping techniques have been used to shift the action spectra of titanium dioxide toward longer wavelengths, with an improvement of solar light absorptivity on the order of 20%. As an alternative to this approach, a variety of semiconductors, including ZnS, CdS, WO3, and WO3/WS2, are currently under investigation to replace titanium dioxide in photocatalytic systems where the absorption of a larger part of the visible spectrum is needed. More versatile materials with a greater absorptivity of visible light would greatly expand the possible range of applications. Particularly important in this regard is the constant interplay between the fundamental understanding of physical and chemical processes occurring on metal oxide particles and the continuous search for new materials aimed at targeted applications.

6

Conclusion

Photoelectrochemical cells are not yet competitive with pure silicon-based photovoltaics in terms of lifetime. Nevertheless, the development of dye-sensitized solar cells and thin-film technologies with new materials is progressing promisingly. Photoelectrochemical cells incorporating in situ energy storage elements in the form of intercalation electrodes or with a third, battery electrode have the potential not only to convert, but also to store incident solar energy inside a single device. Such devices mark an additional important difference between a traditional PV solar cell and a PEC. Photoelectrocatalysis has found a variety of niche applications in the field of environmental cleanup and disinfection.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

Kurzweil, P. Angewandte Elektrochemie (Applied Electrochemistry); Springer Vieweg: Wiesbaden, 2020. Bard, A. J.; Memming, R.; Miller, B. Terminology in Semiconductor Electrochemistry and Photoelectrochemical Energy Conversion. Pure Appl. Chem. 1991, 63, 569–596. Grätzel, M. Photoelectrochemical Cells. Nature 2001, 414, 338–344. Hanna, M. C.; Nozik, A. J. Solar conversion efficiency of photovoltaic and photoelectrolysis cells with carrier multiplication absorbers. J. Appl. Phys. 2006, 100. 074510-1–074510-8. Lewis, N. S. Frontiers of Research in Photoelectrochemical Solar Energy Conversion. J. Electroanal. Chem. 2001, 508, 1–10. Licht, S.; Semiconductor Electrodes and Photoelectrochemistry, Bard A.J. and Stratmann N. (Eds.) Encyclopedia of Electrochemistry, Vol. 6, Wiley-VCH: Weinheim, 2002. Licht, S.; Hodes, G.; Tenne, R.; Manassen, J. A Light-Variation Insensitive High Efficiency Solar Cell. Nature 1987, 326, 863–864. Menezes, S.; Lewerenz, H. J.; Bachmann, K. J. Efficient and Stable Solar Cell by Interfacial Film Formation. Nature 1983, 305, 615–616. Nakamura, R.; Tanaka, T.; Nakato, Y. Mechanism for Visible Light Responses in Anodic Photocurrents at N-Doped TiO2 Film Electrodes. J. Phys. Chem. B 2004, 108, 10617–10620. Parkinson, B. A.; Heller, A.; Miller, B. Effects of Cations on the Performance of the Photoanode in the N-GaAs|K2Se-K2Se2-KOH|C Semiconductor Liquid Junction Solar Cell. J. Electrochem. Soc. 1979, 126, 954–960. Tributsch, H. Layer-Type Transition Metal Dichalcogenides—A New Class of Electrodes for Electrochemical Solar Cells. Ber. Bunsen. Phys. Chem 1977, 81, 361–369. Cahen, D.; Hodes, G.; Grätzel, M.; Guillemoles, J. F.; Riess, I. Nature of Photovoltaic Action in Dye-Sensitized Solar Cells. J. Phys. Chem. B. 2000, 104, 2053–2059. Gerischer, H.; Willig, F. Reaction of Excited Dye Molecules at Electrodes. Top. Curr. Chem. 1976, 61, 31–84. Grätzel, M. Conversion of Sunlight to Electric Power by Nanocrystalline Dye-Sensitized Solar Cells. J. Photochem. Photobiol. A Chem. 2004, 164, 3–14. Park, J. H.; Bard, A. J. Photoelectrochemical Tandem Cell with Bipolar Dye-Sensitized Electrodes for Vectorial electron Transfer for Water Splitting. Electrochem. Solid St. 2006, 9, E5–E8.

422 16. 17. 18. 19. 20. 21. 22. 23. 24.

Electrochemical Devices | Photoelectrochemical Cells

Bard, A. J.; Fox, M. A. Artificial Photosynthesis: Solar Splitting of Water to Hydrogen and Oxygen. Acc. Chem. Res. 1995, 28, 141–145. Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37–38. Lewis, N. S. Light Work with Water. Nature 2001, 414, 589–590. Licht, S.; Wang, B.; Mukerji, S.; Soga, T.; Umeno, M.; Tributsch, H. Efficient Solar Water Splitting, Exemplified by RuO2-Catalyzed AlGaAs/Si Photoelectrolysis. J. Phys. Chem. B 2000, 104, 8920–8924. Gerischer, H.; Heller, A. The Role of Oxygen in Photooxidation of Organic Molecules on Semiconductor Particles. J. Phys. Chem. B. 1991, 95, 5261–5267. Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Environmental Applications of Semiconductor Photocatalysis. Chem. Rev. 1995, 95, 69–96. Inoue, T.; Fujishima, A.; Konishi, S.; Honda, K. Photoelectrocatalytic Reduction of Carbon Dioxide in Aqueous Suspensions of Semiconductor Powers. Nature 1977, 277, 637–638. Ollis, D. F., Al-Ekabi, H., Eds.; In Photocatalytic Purification and Treatment of Water and Air; Elsevier: Amsterdam, 1993. Tryk, D. A.; Fujishima, A.; Honda, K. Recent Topics in Photoelectrochemistry: Achievements and Future Prospects. Electrochim. Acta 2000, 45, 2363–2376.

Electrochemical Devices | Electrochromic Windows L Niklaus, M Schott, and U Posset, Fraunhofer Institute for Silicate Research ISC, Würzburg, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 3 4 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 5 5.1 5.2 5.3 5.4 5.5 5.6 5.7 6 7 8 References

Introduction History of electrochromic windows Overview of smart windows and working mechanism of EC devices Figures-of-merit Lambert-Beer law Contrast ratio Coloration efficiency Visible light transmittance Solar heat gain coefficient (SHGC) CIE L
a
b
system Haze Response time Cycling stability Memory Materials and requirements Transparent conductive substrates Electrolytes EC and ion storage materials Metal oxides Metal complexes Organic materials Hybrid (inorganic/organic) materials Industrial applications of EC windows Future trends and multifunctional ECDs Conclusion

425 425 426 427 427 427 427 428 428 428 428 428 429 429 430 430 430 431 431 432 433 434 434 437 438 438

Abstract This chapter provides an overview of electrochromic (EC) glazing technologies available on the market or under development. EC systems for smart windows that are switchable between a colored and a colorless state or between different colored states by applying a low voltage or current can improve energy efficiency and comfort in buildings. Dynamic modulation in the visible (Vis) and near-infrared (NIR) regions of the electromagnetic spectrum enables an adaption of the thermal and optical behavior of a glazing system to ambient conditions. While the energy consumption for air conditioning and artificial lighting is significantly reduced, smart windows can continuously make optimal use of daylight and be tailored to personal needs using appropriate control algorithms.

Key points

• • • •

Overview of the electrochromic history. Working mechanisms of electrochromic materials and devices. Smart glazing applications in buildings. State-of-the-art of scientific and industrial relevant electrochromic technologies.

Nomenclature

Symbols and Units DE
A a


Color distance Absorbance (a.u.) From green (−128) to red (+127)

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00017-7

423

424

b
c CR d D65 Dl I I0 L
qi Sl T Tb Tc ts V(l) e h hv t(l) te tv

Electrochemical Devices | Electrochromic Windows

From blue (−128) to yellow (+127) Concentration (mol L−1) Contrast ratio Path length (cm) CIE standard illuminant Relative spectral distribution of the standard illuminant D65 Transmitted light intensity (W m−2) Incident light (W m−2) Lightness from black (0) to white (100) Secondary heat dissipation coefficient (W m−2 K−1) Relative spectral distribution of the solar radiation (%) Transmittance (1 ¼ 100%) Bright state transmittance (%) Colored (dark) state transmittance (%) Response time (s) Spectral sensitivity of the human eye (%) Molar absorptivity (m2 mol−1) Coloration efficiency (cm2 C−1) Visible coloration efficiency (cm2 C−1) Spectral transmittance (1 ¼ 100%) Direct radiation transmittance (%) Visible light transmittance (%)

Abbreviations and Acronyms AZO CIE CMY CRI CV CVD DGU EC ECD ECP FTO HOMO ITO IVCT LC LCD LiClO4 LiTf LiTFSI low-e LSPR LUMO MCP MLCT NIR OCP PB PC PDLC PEDOT

Aluminum-doped zinc oxide Commission Internationale de l’Éclairage Cyan, magenta, yellow Color rendering index Cyclic voltammetry Chemical vapor deposition Double glazing unit Electrochromic Electrochromic device Electrochromic polymer Fluorine-doped tin oxide Highest occupied molecular orbital Tin-doped indium oxide Intervalence charge transfer Liquid crystal Liquid crystal display Lithium perchlorate Lithium triflate (Lithium trifluoromethanesulfonate) Lithium bis-(trifluoromethanesulfonyl)imide Low emissivity localized surface plasmon resonance Lowest unoccupied molecular orbital Metal coordination polymer Metal-to-ligand charge transfer Near-infrared Open circuit potential Prussian blue Propylene carbonate Polymer-dispersed liquid crystals Poly(3,4-ethylenedioxythiophene)

Electrochemical Devices | Electrochromic Windows

PEO PET PG PMMA PW PY PV SHGC SPD TCO TGU UV Vis

1

425

Polyethylene oxide Polyethylene terephthalate Prussian green Poly(methyl methacrylate) Prussian white Prussian yellow Photovoltaics Solar heat gain coefficient Suspended particle device Transparent conductive oxide Triple glazing unit Ultraviolet Visible

Introduction

Coping with climate change, one of the current global challenges, is at the center of the scientific and public discussion today. In parallel with the development of renewable energy sources, the reduction and optimization of energy consumption is necessary and urgent. The building sector plays a crucial role in this process, as it accounts for 30–40% (about 1.4 TW) of global primary energy consumption and about 38% of US greenhouse gasses are produced in this sector.1 A major trend in modern architecture is the use of large glass façades and roof elements, not only for aesthetic reasons, but also to make these buildings more comfortable. However, large glass façades require energy-intensive cooling or heating to create a pleasant indoor climate. Although commercially available windows are already well insulated (e.g., double and triple glazing units (DGU/ TGU)), the automized and intelligent control of incident solar radiation to optimize heat and light flows is highly demanded. In addition to indoor comfort, privacy, sun and glare protection as well as generally higher energy efficiency in buildings can be enabled by such materials. Among the possibilities offered by smart functional materials, the selective and dynamic modulation of thermal energy and incident light by chromogenic materials can be key for the energy and environmental performance optimization of buildings. Such innovative glazing systems could be an ideal solution for new buildings as well as for the renovation of existing buildings. Dimmable windows based on electrochromic (EC) materials are smarter than passive solar control coatings, e.g., low emissivity (low-e) coatings or heat mirrors, as the light and heat transmittance can be modulated by an external electrical stimulus independent of the ambient conditions. Electrochromic smart windows increase comfort by adjusting the amount of sunlight and lighting conditions to personal needs or preferences and can improve the energy efficiency of buildings by dynamically adapting heat gain from solar radiation to changing climate conditions. According to a theoretical study,2 energy savings of up to 40% can be achieved with modulation in the visible (Vis) and near-infrared (NIR) range conventional windows (DGU, without additional coatings). According to another study,3 EC windows can reduce the annual peak cooling load by 19–26% and lighting energy by 48–67%.

2

History of electrochromic windows

Modern research in the field of electrochromism began after S. K. Deb published his fundamental work on tungsten trioxide (WO3) thin films in the 1960s.4,5 He had observed that the colorless WO3 thin films could be electrochemically reduced and oxidized between a deep blue state and a colorless state. J. R. Platt introduced the term ‘electrochromism’ in 1961 to describe a color produced by a molecular Stark effect, in which orbital energies are shifted by an electric field.6 Electrochromic devices (ECDs) were first considered for use in information displays, and major research efforts were undertaken by several large companies such as IBM, Philips, and Canon in the first half of the 1970s. These efforts became less important in the 1970s when liquid crystal displays (LCDs) began to dominate the display market. In 1984, EC windows were considered by the U.S. Department of Energy for use in energy-efficient buildings. The term ‘smart window’ was first used by C. G. Granqvist in 1985 to describe windows changing their light transmittance via electrochemical processes.7,8 Since then, the term smart windows has been adapted to describe novel fenestration applications. Since the early 1980s, smart glazing has been a rapidly evolving innovative technology with great market potential, designed to help control energy transfer through the building envelope and avoid unnecessary indoor cooling and heating. Smart windows have a major impact on building envelope performance in terms of thermal management, daylight regulation, glare reduction, and view preservation.

426

3

Electrochemical Devices | Electrochromic Windows

Overview of smart windows and working mechanism of EC devices

In addition to passive technologies such as low-e (static), thermochromic (temperature), and photochromic (light), there are three different technologies with distinct property profiles that are known for their use in smart windows and are commercially available to date: electrochromic (EC), suspended particle (SPD) and liquid crystal (LC)-based devices as shown in Fig. 1. For SPD, a thin film of nanoparticles suspended in a liquid or solid layer is sandwiched between two transparent conductive oxide (TCO)-coated polyethylene terephthalate (PET) sheets. This film can subsequently be laminated between two panes of glass. When no voltage is applied, the nanoparticles consisting of organometallic polyhalides are randomly organized, thus absorbing and blocking light. When a voltage is applied, they align themselves, allowing light to pass through. Varying the voltage of the film changes the orientation of the suspended particles, thereby regulating the amount of transmitted light and haze. In addition, they are colored in the voltage-free state. A well-known manufacturer of these devices is Gauzy, which also offers the next technology, namely polymer-dispersed liquid crystal (PDLC) devices. Here, micro- or nanometer-sized LCs are embedded in a cured polymer matrix. It is also coated between two pieces of PET-ITO to create a film, which is either laminated into glass, or applied as a sticker to existing glass or other transparent materials. During curing, the liquid crystals form droplets that are randomly distributed in the solid polymer matrix, increasing the haze of the device. Once a sufficient electrical field is generated by an external voltage, PDLC films switch from opaque to transparent due to the alignment of the LC droplets along the direction of the electrical field. In contrast, LC devices overcome the haze of the typical laminated PDLC product and exhibit a fully transparent and clear state at all times and viewing angles. One example is Merck’s eyrise® systems (i.e., solar: eyrise® s350, privacy: eyrise® i350), in which a wide range of dye-doped LC mixtures is capable of achieving almost any color and saturation level. SPD can control solar transmittance over a range of approximately 50% (LCG® from Gauzy switches between 20 years under changing environmental conditions such as temperatures below −20  C and above +80  C, high solar irradiation levels, rapid temperature changes, and uneven temperature distribution over the area of the window. Accelerated aging tests have been developed during the last years, e.g., ISO 1854314 to assess the durability and reliability of EC windows. However, their applicability to all types of EC glass is discussed controversially. The long-term cycling stability of EC materials and ECDs can be evaluated by observing the loss of optical contrast and charge upon repetitive electrochemical cycling. Materials capable of maintaining over 95% of their maximum optical contrast during the application of boundary cell voltages (two-electrode setup) or potentials (three-electrode setup) are deemed stable. In addition to monitoring only the loss of optical contrast, electrochemical measurements, e.g., cyclic voltammetry (CV), potentiostatic or galvanostatic steps before and after cycling should be provided together, so that the current state of the entire thin film can be evaluated. The coulombic efficiency (CE), defined as the ratio of discharge density to charge density of the EC/ion storage layer, can be calculated to determine the cycling stability. A stable operating voltage or potential window can be defined as the voltage or potential range in which the EC material undergoes reversible electrochemical redox reactions without irreversible side reactions involving the other components of the ECDs. Degradation of the TCO layer and active materials usually results in longer response times, while loss of EC contrast is due to EC layer degradation.

4.10 Memory Optical memory refers to maintaining the transmittance level once the supply voltage or potential has dropped. This open circuit memory has a positive impact on the overall energy consumption of ECDs. Measurement of the optical memory of an ECD consists

430

Electrochemical Devices | Electrochromic Windows

of an open circuit potential (OCP) step after decoloration or coloration while monitoring the transmittance. This memory (also called bi-stability) is highly dependent on the type of ECD, the applied voltage/current, and side reactions occurring at the electrolyte/EC layer interface.

5

Materials and requirements

For the application in smart windows, ECDs are commonly assembled in a two-electrode (sandwich) configuration. They can be described as a transparent rechargeable electrochemical cell or battery, in which the EC electrode is separated from a charge-balancing ion-storage layer (counter electrode) by a liquid, gel or solid electrolyte. An overview of the requirements and examples of each layer is given in the following.15

5.1

Transparent conductive substrates

Transparent conductive substrates are an essential component for device operation for smart window applications. In addition to a high visible light transmittance, the lowest possible sheet resistance is desirable for ECDs in order to avoid ohmic voltage losses. If the sheet resistance is high, the voltage drops over the surface resulting in longer response times and non-uniform decoloration and coloration. Due to the high demands for transparent electrical conductors and other factors, the conductive substrates can be the most expensive part of an ECD. Even if these substrates usually do not constitute rare earth metals, e.g., cerium or neodymium, the processes involved in manufacturing these components can be both complex and energy demanding, in particular if processing from the ore is taken into account. Thin films of heavily doped (several atom percent) semi-conductive oxides are often used for EC applications. These n-type semiconductors, e.g., In2O3: Sn (ITO, tin-doped indium oxide) or SnO2: F (FTO, fluorine-doped tin oxide), exhibit high visible light transmittance (tv > 90%), and sheet resistances in the range of approximately 10 O sq.−1. Moreover, there are other metal oxides such as aluminum-doped zinc oxide (AZO) that are more sensitive to oxygen or other reagents and do hardly meet the optical and resistivity requirements for ECDs.

5.2

Electrolytes

In addition to the two electrodes, an ECD comprises a third key component, the highly transparent (tv >99%) electrolyte. The electrolyte plays an essential role as an ionic conductor placed between the electrodes. The most important properties that define a well-performing electrolyte for ECDs are ionic conductivity, electronic insulation, ion dissociation, the transport rate of the ions through bulk and interfaces as well as thermal and electrochemical stability during operation. An electrolyte must fulfill several requirements for the use in ECDs, i.e., an ionic conductivity in the range between 10−4 S cm−1 and 10−7 S cm−1 and low electronic conductivity in the range of 10−12 S cm−1. Electrolytes can be divided into liquid, gel, and solid electrolytes, the latter consisting of either solid polymer or inorganic or ceramic electrolytes. Liquid electrolytes exhibit the highest ionic conductivity, but suffer from a major disadvantage in that hydrostatic pressure may build up in large windows, which can ultimately lead to electrolyte leakage. Gel polymer electrolytes are usually produced by adding salts, liquid plasticizers, thickeners, and solvents to a polymer matrix such as polyethylene oxide (PEO) or poly(methyl methacrylate) (PMMA). They typically offer ionic conductivities at the high end, i.e., of the order of magnitude of 10−4 S cm−1. As plasticizers, organic carbonates such as ethylene carbonate or propylene carbonate (PC) are commonly first choice when it comes to (‘green’) solvents with a large electrochemical stability window to adjust the ionic conductivity by increasing the amorphous phase concentration, dissociating possible ion aggregates, lowering the glass transition temperature, and increasing the ionic mobility. Safety concerns related to electrolyte leakage, internal short circuits, and material degradation upon extensive heating or ultraviolet (UV) irradiation may remain an issue for gel electrolytes. Some of the drawbacks of gels can be tackled by the addition of cross-linking components to enhance their mechanical properties. Solid polymer electrolytes are solvent-free systems particularly interesting due to their easy processing on large areas, low manufacturing costs and the possibility of designing leak proof and mechanically stable, yet flexible ECDs. Different compositions have been proposed here, whereby most are based on commercially available polymers, such as the examples given above. The conductivity of this class of electrolytes is typically lower than for liquid or gel electrolytes, i.e., 10−5. . .10−6 S cm−1 at room temperature. Mixed forms are also known, for example, soft polymers or gels with solid fillers or so-called “osmolytes,”, i.e., inorganic-organic hybrid polymers produced by hydrolysis and polycondensation (sol-gel-process). Commonly used salts are, e.g., lithium perchlorate (LiClO4), lithium trifluoromethanesulfonate (LiTf ) or lithium bis-(trifluoromethanesulfonyl)imide (LiTFSI). Small ions, or more precisely ions with a small Stokes radius, are particularly well suited for ECD operation. Because of their small Stokes radius, they are more mobile and can migrate into the EC layer (a process called intercalation) more easily than large ions. In addition to the lithium salts mentioned, sodium, potassium, magnesium, aluminum, or zinc salts with larger ion diameters, can also be suitable and have some advantages over lithium-ion technology, particularly with regard to raw material abundance and cost. It should be mentioned at this point that it is not just the properties of the cations that matter. Some EC materials, such as poly(3,4-ethylene-dioxythiophene) (PEDOT)-type conductive polymers, are positively charged in their bright state and neutral in their dark state. Hence, they insert anions from the electrolyte during decoloration.16 Different types of additives can be considered to tailor the properties for the respective application, e.g., inorganic nanoparticles to enhance mechanical robustness and UV shielding or ionic liquids to enhance conductivity and lower flammability. Although the

Electrochemical Devices | Electrochromic Windows

431

work in the field of electrolytes has significantly progressed over the past years, challenges regarding the electrochemical, thermal and UV stability are still being faced. In addition, the electrolyte must in any case be adapted to the chemistry of the active layers used and show excellent processability, flexibility, adhesion and cohesion in order for it to be used in durable flexible devices.

5.3

EC and ion storage materials

Electrochromic materials are typically classified into three groups, namely inorganic, organic, and hybrid (inorganic/organic). EC thin films are usually sub-micron to micron in thickness. An EC color intense enough to be observed in thin films under normal illumination is ideally achieved using materials with high absorption coefficients. The lower the absorption coefficient, the thicker the layer has to be in order to attain appreciable color (change) impressions.

5.4

Metal oxides

One group of inorganic EC materials are oxides of transition metals, which can adopt different valence states of different optical absorption characteristics. The transition metals, whose oxides have EC properties, are highlighted in the periodic table of elements given in Fig. 3 and described in detail in ref. 19. EC metal oxides are typically deposited as thin films by physical vapor deposition (PVD) processes such as sputter deposition. They typically work via electrochemical ion intercalation and redox processes. In addition, wet-chemical approaches have been developed specifically for the deposition of oxidic nanoparticles in recent years, e.g., spin-coating, dip-coating, slot-die coating, and screen printing. Depending on the synthesis and deposition parameters, the EC properties of the materials can be tuned. These oxides commonly consist of highly distorted MO6 octahedral crystalline structures, where M represents the transition metal atom. The layered structure formed by the edge- and corner-sharing MO6 octahedra facilitates ion transport through the conduction paths or chains of interstitial sites. Metal oxides, e.g., WO3, NiO, MoO3, and V2O5 have been used as EC material where small cations (H+, Li+, etc.) intercalate into the crystal lattice, resulting in color changes caused by an intervalence charge transfer (IVCT) transition induced by the applied voltage. Metal oxide-based ECDs operate at low voltages with low power consumption and exhibit high optical contrast between colored and bright states. However, due to the large electronic bandgap in metal oxides, it remains a challenge to achieve multiple colors with closely adjacent wavelengths. Although ECDs based on metal oxides are the current state-of-the-art in smart windows, further optimization in terms of costs, durability, and functionality (i.e., color and switching time) are desired. Thin metal oxide films have been widely studied in the literature, in large part owing to photochemical stability issues (e.g., many transition metal oxides show photocatalytic activity). In addition, like most solid-state crystalline structures, metal oxide thin films are brittle; they may tend to mechanical cracking due to volume swelling and contraction occurring upon ion intercalation and deintercalation. Strain arises from the corresponding changes in the lattice constants. In recent years, several studies have focused on the aging and rejuvenating mechanism of metal oxides, in particular of tungsten (VI) oxide, e.g., restoring the initial charge density and transmittance change after extensive cycling by means of short voltage pulses. Most EC materials modulate the visible radiation ranging from 380 nm to 780 nm (Vis), but are also able to reduce near-infrared (NIR, 780 nm–2500 nm) transmission through the window. Because NIR transmission is one component contributing to the heat transfer in windows, selectively modulating this radiation provides an interesting option with regard to thermal comfort and energy efficiency improvement. Consequently, nanocrystal in-glass composite windows represent a promising EC system alongside stateof-the-art metal-oxide based ECDs (WO3/NiO) to improve the overall performance and maximize the energetic impact of EC glazing.18 Here, the plasmonic reaction of colloidal nano-crystals controls NIR light transmission without affecting the visible light transmittance, as shown in Fig. 4.19

Fig. 3 EC metal oxides showing both cathodic and anodic coloration. Reproduced from Granqvist, C.G. Handbook of Inorganic Electrochromic Materials. Elsevier: Amsterdam, New York, (1995), with permission from Elsevier.

432

Electrochemical Devices | Electrochromic Windows

Fig. 4 Depiction of the microscopic operation of a nanocrystal-based plasmonic EC film incl. The corresponding optical changes. (A) In the OFF state, a positive potential is applied to the nanocrystals, which are depleted of electrons and lithium ions (Li+) are repelled. (B) In the ON state, a negative potential is applied to the nanocrystals, through which electrons are injected. Lithium ions are attracted to the nanocrystal surface to compensate for the injected charge capacitively. (C) Optical density changes result from electron injection. The increase in charge carrier density causes a blue shift in the LSPR and absorption. (D) Corresponding changes in transmission of the film. Reprinted from Ref. Runnerstrom, E.L.; Llordés, A.; Lounis, S.D.; Milliron, D.J. Nanostructured Electrochromic Smart Windows: Traditional Materials and NIR-Selective Plasmonic Nanocrystals. Chem. Commun. (2014), 50, 10555–10572. https://doi.org/10.1039/C4CC03109A, with permission from the Royal society of chemistry. (E and F) Schematic overview of the different states of an ECD based on plasmonic materials. Reprinted from Ref. Kim, J.; Ong, G.; Wang, Y., Leblanc, G., et al. Nanocomposite Architecture for Rapid, Spectrally-Selective Electrochromic Modulation of Solar Transmittance, Nano Lett. (2015), 15(8), 5574–5579; Tandon, B.; Lu, H.C.; Milliron, D.J. Dual-Band Electrochromism: Plasmonic and Polaronic Mechanisms. J. Phys. Chem. C (2022), 126, 9228–9238. https://doi.org/10.1021/acs.jpcc.2c02155, respectively with permission from the American chemical society.

This approach allows the transmittance of the Vis or NIR to be regulated separately and independently, creating a ‘dual-band’ dynamic window. This novel EC system was first developed by researchers at the University of California, Berkeley, using ITO nanocrystals embedded in a glassy matrix of niobium oxide (NbOx).18 It represents a starting point for NIR-switching EC materials that allow control of NIR radiation without blocking visible light transmittance, similar to non-modulating selective glass. The system usually works in three states: bright (Vis and NIR transmitted), cool (only Vis transmitted), and dark (Vis and NIR blocked).20,21 Plasmonic EC materials can be used to achieve multiple colors at low voltage due to the formation of different size-dependent optical states and localized surface plasmon resonance (LSPR). Therefore, color stabilization in ECDs based on metallic plasmonic materials relies on precise control and stability of desired shape/size of the nanoparticles, which is achieved by applying continuous voltage pulses and introducing a suitable capping ligand. However, the long-term stability of plasmonic nanoparticles, in terms of size and shape, within the ECD remains an issue due to their aggregation or agglomeration tendency.

5.5

Metal complexes

Metal complexes are another class of EC materials utilizing metal ions and ligands that attract considerable interest because of their intense coloration and exceptional redox behavior. Their chromophoric properties arise from low-energy metal-to-ligand charge

Electrochemical Devices | Electrochromic Windows

433

transfer (MLCT), IVCT, intra-ligand excitation, and related visible-region electronic transitions, which lead to color changes upon oxidation or reduction of the complex, i.e., redox reactions can occur at the metal-ion and/or the ligand. Iron(II)-hexacyanoferrate (III) (Fe4[Fe(CN)6]3), commonly known as Prussian blue (PB), is a mixed-valence coordination complex forming a blue pigment that was first studied as an EC material by Neff and coworkers in the late 1970s.22 PB can be electrochemically reduced to colorless “Everett’s Salt” or Prussian white (PW), which is iron(II), or it can be completely oxidized to Prussian yellow (PY), i.e., iron(III)-hexacyanoferrate(III). Partial oxidation results in Prussian green (PG) (Fig. 5; Eq. 11).  2 −  − h     i1=3 −   FeII FeII ðCNÞ6 >FeIII FeII ðCNÞ6 > FeIII FeIII ðCNÞ6 2=3 FeII ðCNÞ6 1=3 >FeIII FeIII ðCNÞ6 PW

PB

PG

PY

(11)

Besides, there are PB analogs, in which either iron(II) and/or iron(III) are replaced by other metal ions, e.g., Cu, Co, Ni, and Ru. Since changing the metal ion changes the absorbed energy required for the IVCT, PB analogues switch between completely different tints compared to PB, making ECDs in other colors possible. For instance, cobalt-based analogues switch between a pale green and a reddish-brown state, copper-based ones between a pale yellow and red state, and nickel-based ones between an almost colorless and a yellow state. Among the PB analogues, ruthenium purple promises the most pronounced (and from a window application perspective probably most interesting) color change from transparent to purple. Noteworthy, these metal complexes are deposited by wet-chemical processes, e.g., electrodeposition, hydrothermal, and photochemical deposition.

5.6

Organic materials

Electrochromic characteristics are shown by a wide range of organic materials, including organic molecules, e.g., viologens, carbazoles, triphenylamines, and conjugated polymers. Typically, organic EC materials offer multiple colors under different applied biases, which are desirable for multicolor displays. Viologens, i.e., 1,10 -disubstituted-4,40 -bipyridinium salts, are the most intensively investigated small molecule-based EC compounds. Viologens exhibit three reversible redox states (Fig. 6): a dication (V2+), a radical cation (V+), and a neutral form (V), thus yielding differently colored species.23 The most distinct color is visible upon reduction of the viologen dication resulting in a strongly absorbing radical cation state. The substitution groups on the nitrogen of the bipyridinium salt mainly control the colors of their reduction states. The most important representatives of conjugated electrochromic polymers (ECPs) include polyaniline, polypyrrole, polyfuran, or polythiophene, which are based on electron-rich conjugated five-or six-membered ring polyheterocycles and have a lower oxidation potential and good solubility. Polythiophenes are the most widely investigated conjugated polymers because they differ from other conjugated polymers in their higher environmental stability, ease of synthesis and processability, and enormous structural diversity and electronic properties. Switchable transmittance in the visible range has been the subject of research for over 50 years. In recent years, there has been an increasing trend towards switchable materials that can be fabricated on plastic film using wet chemical roll-to-roll deposition processes and at low temperatures. A wide range of organic materials can be used in wet-chemical processes. The group of J. Reynolds has been particularly active in the field of conjugated conducting polymers.24 Thiophene-based EC polymers that are soluble in common solvents have been synthesized and studied to create a broad color palette of EC materials (Fig. 7).

Fig. 5 Overview of the redox processes of Prussian blue.

Fig. 6 Redox and color states of methyl viologen. Reprinted from Ref. Stolar, M. Organic Electrochromic Molecules: Synthesis, Properties, Applications and Impact. Pure Appl. Chem. (2020), 92, 717–731. https://doi.org/10.1515/pac-2018-1208, with permission from deGruyter.

434

Electrochemical Devices | Electrochromic Windows

Fig. 7 (A) Chemical formula and (B) photographic images of CMY (C ¼ cyan, M ¼ magenta, Y ¼ yellow) representative EC polymers in solution and thin films. (C) Solutions of ECPs-C, -M, and -Y at concentrations of 2 mg mL−1 and 1:1 w/w ratios of these solutions. Reprinted from Bulloch, R.H.; Kerszulis, J.A.; Dyer, A.L.; Reynolds, J.R. An Electrochromic Painter’s Palette: Color Mixing Via Solution Co-Processing. ACS Appl. Mater. Interfaces (2015), 7, 1406–1412. https://doi.org/10.1021/am507514z, with permission of the American Chemical Society.

Conjugated polymers are characterized by their high coloration efficiencies, broad color palette, fast response, and ease of fabrication via solution processing. The band gap between the valence band (HOMO) and conduction band (LUMO) determines the optical properties of the polymer. With suitable functionalization (or derivatization), conjugated polymers can be suitable for large-scale and high-throughput processes, e.g., roll-to-roll deposition. They may then represent a cost-effective solution for ECDs, since the use of wet-chemical coating or printing techniques is also economically feasible for small lot sizes and small quantities as well as for complex device geometries. However, compared to sputter-deposited inorganic EC materials, the organic EC materials can differ in their stability under thermal or UV stress when not properly encapsulated, which may be one reason why their industrial application is yet limited to date.

5.7

Hybrid (inorganic/organic) materials

Metal-organics is another class of EC materials, where organic species act as ligands, hence contribute to a wide color range following oxidation and reduction of the metal centers, due to their MLCT. This class includes metal organic complexes, such as metal phthalocyanines and triphenylamine complexes, as well as metal coordination polymers (MCPs). As the EC mechanism of both subclasses is identical, the following part will focus on MCPs. In recent years, MCPs have attracted special attention due to their outstanding optoelectronic properties, in particular for those of transition metal ions with terpyridine ligands. MCPs differ from conventional organic polymers as the bonds connecting the monomers are not covalent bonds, but coordinated covalent bonds.25 Polypyridyl ligands in combination with different metal ions, e.g., Fe2+, Ru2+, and Os2+, show purple, red, and orange colors, respectively.26,27 Various colors can be realized among MCPs by either changing the metal ion and/or the ligand (Fig. 8) or by mixing different MCPs. Moreover, it is possible to obtain linear, branched and ring-like MCPs depending on the ligands. The ground state of most MCPs is typically colored, with the metal center in the (M2+) reduced state and the color arising from the MLCT transition. The oxidation of M2+ to M3+ prohibits the MLCT causing a decrease in light absorption and the corresponding color intensity in the visible range. This implies that ECDs based on MCPs are colored under open circuit conditions, with the colorless state or different colored states being obtained by applying a potential under which electrochemical oxidation of the metal center occurs.

6

Industrial applications of EC windows

To date, EC technology has been implemented in large-area glazing (windows and façades) for buildings combining energy efficiency with improved indoor comfort. The color of commercially available EC glazing systems (Fig. 9) is usually blue or grey-blue to almost color-neutral in the dark state mainly due to WO3 commonly used as an EC material showing a color modulation from colorless (oxidized/clear state) to blue (reduced/dark state). Smart windows, in which metal oxides are primarily used as EC materials, are known for several decades and manufactured, e.g., by SageGlass (USA), View (USA), ChromoGenics (SE), EControl-Glas (D), Halio (USA), Heliotrope (USA), and Miru (CA). SageGlass started in 1989 in New York and became one of the first companies to develop EC smart windows. By 2012 Saint-Gobain had acquired 100% of SageGlass. View is an American glass company that was founded in 2007 in the Silicon Valley. Another EC window based on sputtered metal oxide EC materials and a polymer electrolyte has been developed by EControl-Glas in

Electrochemical Devices | Electrochromic Windows

435

Fig. 8 (A) Scheme of the synthetic route and possibilities for variation (spacer unit, side chains of the bis-terpyridine ligands from L1 to L5). Adapted from Ref. Han, F.S.; Higuchi, M.; Kurth, D.G. Metallosupramolecular Polyelectrolytes Self-Assembled From Various Pyridine Ring-Substituted Bisterpyridines and Metal Ions: Photophysical, Electrochemical, and Electrochromic Properties. J. Am. Chem. Soc. (2008), 130, 2073–2081. https://doi.org/10.1021/ja710380a with permission of the American Chemical Society. (B) Comparison of the absorbance spectra of Fe-MCP thin films with three different bis-terpyridine ligands on FTO glass. Adapted from Ref. Schott, M.; Niklaus, L.; Clade, J.; Posset, U. Electrochromic Metallo-Supramolecular Polymers Showing Visible and Near-Infrared Light Transmittance Modulation. Solar Energy Mat. Solar Cells (2019) 200, 110001. https://doi.org/10.1016/j.solmat.2019.110001 with permission of Elsevier. (C) In situ UV-Vis spectra of a MCP with Os, Ru, and Fe metal centers at different applied potentials of 0.70, 0.85, and 1.20 V for stepwise oxidation of Os(II), Fe(II), and Ru(II), respectively. Adapted from Ref. Bera, M.K.; Ninomiya, Y.; Higuchi, M. Stepwise Introduction of Three Different Transition Metals in Metallo-Supramolecular Polymer for Quad-Color Electrochromism. Commun.Chem. (2021), 4, 1–12. https://doi.org/10.1038/s42004-021-00495-1 with permission from Springer Nature.

Germany. While SageGlass and View employ all-solid-state technology with sputter-deposited electrolytes, for the EControl-Glas product an electrolyte is injected in fluid form between the glass panes by means of vacuum filling. To produce TCO, EC, and ion-storage layers, various fabrication methods have been studied, such as chemical and physical vapor deposition (CVD and PVD), and sol-gel processes. Because the deposition process often requires high temperatures or vacuum, high process costs can be an issue. Sputtering processes deliver high-quality products even with large surfaces (windows), but are complex in terms of equipment and, therefore, expensive, which is only put into perspective when coating very large surfaces. In addition, the switching times of such windows are slow due to the low diffusion rate of ionic charge carriers in inorganic

436

Electrochemical Devices | Electrochromic Windows

Fig. 9 Electrochromic windows from (A) SageGlass, (B) View, (C) ChromoGenics, and (D) Halio.

Fig. 10 Principle design of a film-based ECD manufactured by ChromoGenics. Arrows indicate ion transport when a voltage is applied between the transparent electrical conductors. The entire foil can be employed to laminate glass panes, as shown in the left-hand part. Reprinted from Ref. Granqvist, C.G.; Arvizu, M.A.; Bayrak Pehlivan, I.; Qu, H.Y.; Wen, R.T.; Niklasson, G.A. Electrochromic Materials and Devices for Energy Efficiency and Human Comfort in Buildings: A Critical Review. Electrochim. Acta (2018), 259, 1170–1182. https://doi.org/10.1016/j.electacta.2017.11.169, with permission from Elsevier.

solids. Because of these limitations, which apply to most inorganic EC layers, there is currently a trend towards EC materials that can be processed by wet chemistry and at low temperatures. The company Halio offers an EC glass product called Halio® (tv changes from 2% to 67%) and Halio® Black (tv from 0.1% to 54%), in which wet-chemically processed WO3 and NiO nanoparticles are used. In contrast, ChromoGenics from Sweden developed an EC film with roll-to-roll sputter-deposited WO3 and NiO as active layers on ITO-coated PET that can be laminated between glass (Fig. 10).28 Here, the visible light transmittance can be changed either from 33% to 61% (ConverLight® Dynamic 75) or from 14% to 56% (ConverLight® Dynamic 65). In addition to the integration of EC windows in new buildings, it is also possible to upgrade existing windows to smart windows by retrofit. At present, the costs for this still exceed the benefits of possible energy and cost savings. For buildings constructed or renovated after 2010, replacing the new windows or glass façade elements is not desirable from an environmental and cost perspective.

437

Electrochemical Devices | Electrochromic Windows Table 1

Overview of state-of-the-art EC windows for building applications.

Manufacturer

Product name

EC materials (deposition process)

Max. size/cm2

tv (clear-dark state)/%

SHGC (clear-dark state) /%

Switching time /min

SageGlass

SageGlassW LightzoneW HarmonyW View Smart Glass

WO3/NiO (sputtered)

152  305 182  305

36–3 40–5 n.a. 41–9

5–15

WO3/NiO (sputtered) WO3/NiO (sputtered) WOx/TiVOx (sputtered) WO3/NiO (wet-chemical) ITO-nanocrystal (wet-chemical) WO3/NiO (wet-chemical)

54–1 60–1 n.a. 58–1 61–33 56–14 56–10

40–14 36–6 42–10

10–20

67–2 54–0.1 63–2

46–8 39–8 36–1

4He (0.08 MeV) + gamma ray (23.77 MeV).

The primary fuel used in ‘cold fusion’ experiments is ‘heavy’ hydrogen, which has a single proton as well as a neutron in its nucleus. Deuterium is found abundantly in, and can be extracted from, ocean water. One in every 6000 hydrogen atoms in water molecules is in the form of deuterium. When water is made from deuterium rather than hydrogen, it is called heavy water (D2O) because the extra neutron makes it 10% heavier than regular water (H2O). One distinct characteristic of low-energy nuclear reactions (LENR) is that they can occur at or near ordinary room temperature in a relatively simple apparatus. This distinguishes the reactions from thermonuclear fusion and other high-energy nuclear reactions, which require very high temperatures and complex containment facilities. Deuterium gas (D2) and normal hydrogen are also used in this research as an alternative to heavy water.

1.1

Thermonuclear fusion

The most significant initial argument against the hypothesis of thermonuclear fusion by electrolysis of heavy water was based largely on theoretical grounds: the expected neutron emissions were not present at the expected rates. That M. Fleischmann and S. Pons3–5 survived was proof of a negligible neutron flux relative to the energy produced in their experiment. In the years that followed, the most significant initial argument for the hypothesis of a new kind of fusion process was based largely on the somewhat quantitative correlation of excess heat and generation of helium-4. Between 1995 and 2002, SRI International, the US Navy China Lake laboratory, and the Italian National Agency for New Technologies, Energy and the Environment (ENEA) laboratory in Italy reported a wide range of energy values for the helium-4 from 22.85 to 103 MeV—that evolved at the same time as the excess heat. Only the 22.85 value from SRI International was close to the value expected from the third branch of deuterium–deuterium thermonuclear fusion, 23.77 MeV. The current rejection of thermonuclear fusion as the mechanism for the observed phenomena is also supported by a variety of other differences between the experimental results and thermonuclear fusion. In addition to the heat-producing reactions discovered by M. Fleischmann and S. Pons, the field encompasses normal hydrogen and transmutation reactions that clearly are not the result of thermonuclear fusion processes—all the more reason to refer to low-energy nuclear reaction (LENR) processes rather than ‘cold fusion.’

1.2

Low-energy nuclear reactions

In 2004, the Department of Energy (DoE) conducted a ‘Review of Low-Energy Nuclear Reactions’ of the subject formerly called cold fusion. The DoE selected 18 scientists as reviewers for 8 papers to read, and 11 of the reviewers attended a daylong presentation and discussion with 6 LENR researchers on 23 August 2004. The panel did not make any laboratory visits. Some of the reviewers recognized the possibility of novel phenomena. However, the consensus was not strong enough to support a funding commitment by the DoE. In the scope of the review, low-energy nuclear transmutation with heavy elements and nickel-hydrogen research were omitted. The use of the term ‘cold fusion’ was applied to this field not by its discoverers but by the media, which confused the Fleischmann–Pons work with that of Brigham Young University physicist S. E. Jones, who was researching a different type of reaction known as muon-catalyzed fusion. The term ‘cold fusion’ was never ideal to describe LENR research, because it implied that the reactions were just a colder form of thermonuclear fusion, which they are not. The term ‘condensed matter nuclear science’ (CMNS) has been adopted by researchers studying LENRs to identify the new field represented by this work. A key requirement is the reliance on condensed matter, such as palladium, as an integral component of these reactions. In some circumstances, liquid media act as the condensed matter and take the place of the host metals.

2 2.1

Excess heat and calorimetry Excess heat

The primary claim of excess heat is perhaps the most significant phenomenon of low-energy nuclear reactions.3–33 For a closed electrolytic cell, calculating the amount of energy coming out of the system is based on Joule heating: Z Q ¼ U I dt ¼ m cp DT M. Fleischmann and S. Pons found that the amount of heat energy coming out of their electrochemical cell was up to 1000 times greater than it should have been, based on their knowledge of possible chemical reactions. A set of experiments performed at SRI International from 2006 to 2007 showed 10% to 300% energy gain and production of a few watts. An example is shown in Fig. 1.

2.2

Calorimetry

Experiments with palladium and deuterium consistently show significant excess heat. Control experiments with palladium and normal hydrogen show almost no excess heat (Fig. 1). Calorimetry has been consistently useful and reliable as a tool to measure the

Electrochemical Devices | Electrochemical Low-Energy Nuclear Systems

441

Fig. 1 (a) Significant excess power (heat) shown with D2O as compared with H2O experiment. Stepped current shows causative effect relative to heat production. The figure depicts simultaneous series operation of D2O test cell, H2O control cell, and current density vs time (SRI experiments P14 and P13). (b) Schematic of SRI International type mass-flow calorimeter. (c) An inexpensive yet effective enclosure-type calorimeter designed and built by E. Storms. Photo: Steven B. Krivit, 2002. Reproduced from 1st edition Krivit, S.B., Cold fusion—Precursor to Low-Energy Nuclear Reactions, Encyclopedia of Electrochemical Power Sources, 1st edn.; Elsevier: Amsterdam, 2009; Vol. 2. pp. 255–270.

power and energy released from these experiments. Over the years, many experimenters have shifted to smaller and shorter experiments, which have generally yielded smaller absolute values of power and energy. The cost of expensive materials such as palladium and heavy water, as well as safety issues, has encouraged researchers to work with smaller palladium samples. Concurrent with the general reduction of more grossly dramatic results, the precision and reliability of all methods—isoperibolic, mass flow, and enclosure—has improved. Isoperibolic calorimetry. In the simplest configuration, the cell is immersed in a constant temperature bath, and two sensors measure the temperature, one inside the cell and another outside the cell in the water bath. The rate of heat transfer dQ/dt can then be measured. Other isoperibolic calorimeters surround the cell with an insulating layer that provides additional thermal isolation of the cell from the water bath. The temperature within this insulating layer, called the thermal barrier, is measured in addition to or in place of the temperature directly within the cell. Isoperibolic calorimetry is not intrinsically complex; however, it becomes so when a mixture of radiative, conductive, and convective heat flows must be accounted for. Many critics distrusted M. Fleischmann and S. Pons’ isoperibolic calorimetry. Mass-flow calorimetry. Mass-flow calorimeters enclose the experiment in a chamber filled with a recirculating fluid or use a closely contacting heat exchanger to extract heat (Fig. 1b). The temperature of the fluid is measured when it enters the chamber and when it exits the chamber. The difference in the temperatures DT along with the flow rate can be used to calculate accurately the heat coming from the reaction. Mass-flow calorimeters have the advantage of being much easier to calibrate, and errors are easier to recognize. However, this method constantly circulates a fluid around the cell, which promotes cooling and thereby prevents rise in temperature. If cell temperatures are allowed to increase, positive feedback occurs in low-energy nuclear reaction experiments, leading to even higher cell temperatures. Seebeck calorimeter. The enclosure type of calorimeter is a thermally insulated container in which an experiment is placed (Fig. 1c). Many thermocouples are embedded within the walls of the enclosure, and they measure temperature within the container and outside the container. These data are collected and used to determine the heat generated within the container. An advantage of the enclosure calorimeter is that it is relatively simple to use and can thereby provide more error-free and rigorous results. It can also be used in parallel with a mass-flow calorimeter, fully enclosed within it, for redundant calorimetry.

2.3

Calorimetry critique

Improper stirring. The first critique of M. Fleischmann and S. Pons’ excess heat claims came from N. Lewis and W. E. Meyerhof (Stanford) at the May 1989 American Physical Society meeting. Both scientists publicly stated the claimed temperature gradient was invalid because a mechanical stirrer was neglected. Fleischmann and Pons put this speculations to rest a week later in Los Angeles at the Electrochemical Society meeting with a videotape that showed that the natural geometry of the cell they designed, in conjunction with the bubbling action of the electrolyte, accomplished the task of stirring in seconds, without the need for a mechanical stirring device. Mathematical error. In 1990, W. Hansen, with the physics and chemistry departments at Utah State University, performed an independent analysis of the Fleischmann–Pons data. He reported to the Utah State Fusion/Energy Council that the quantity of excess energy found in the Fleischmann–Pons cells was ‘over a thousand times the energy required to vaporize the electrode.’ The Wilson group from General Electric, in July 1992, also performed an independent analysis of the Fleischmann–Pons calorimetry

442

Electrochemical Devices | Electrochemical Low-Energy Nuclear Systems

data but they were unable to dismiss the excess-heat claim. The Wilson group concluded that the M. Fleischmann and S. Pons cell generated 40% excess heat, amounting to 0.736 W, more than 10 times the error levels associated with the data. Recombination error. As heavy water (D2O) dissociates within the electrolytic cell, deuterium and oxygen gases leave the cell, taking with them some chemical energy. In order to perform a full accounting of potential excess heat, researchers must account for this energy loss. Some critics speculated that errors in such a recalculation could lead to an overcorrection and could be responsible for the claims of apparent excess heat. The criticism became moot when most researchers switched to thermodynamically closed cells, wherein the devolving deuterium and oxygen gases were recombined, thus keeping the chemical energy fully within the cell and eliminating the need for corrective calculations. Low magnitude of heat effect. Calorimeters must be able to measure temperature changes precisely and maintain such precision over long periods. M. Fleischmann and S. Pons designed a calorimeter that was accurate to 70.1 mW for an 800-mW input. Many of their experiments reported excess heat in the hundreds of milliwatts. R. Garwin, at that time the director of applied research at the IBM Thomas J. Watson Research Center, in October 1993, along with N. Lewis of Caltech, visited the SRI International laboratory of M. McKubre,19 director of energy research, and conducted an extensive, 2-day evaluation of the LENR research performed at SRI International. In their report to L. M. Hammarstrom at the Pentagon, R. Garwin and N. Lewis reported that the ‘uncertainty in excess power measurement is about 50 mW, but the excess power appears to be on the order of 500 mW or even 1 Watt peak. However, excess power is still a deduced quantity and depends upon the calibration of the calorimeter.’ Conservatively, this provides a signal 10 times that of the uncertainty. However, calibration was a minor issue, according to M. McKubre, who designed and built this first-principles mass-flow calorimeter. It had a 99.3% efficiency, and only 0.7% of the remaining conductive and radiative heat depended on calibration. Furthermore, R. Garwin wrote, “on cells L3 and L4, we note that a chemical reaction involving the Pd at perhaps 1.5 eV per atom would correspond to about 3.5 kJ of heat; this is to be compared with the 3 MJ of ‘excess heat’ observed, so such an excess could not possibly be of chemical origin.” Garwin also wrote in his report: “We have found no specific experimental artifact [that is, error] responsible for the finding of ‘excess heat’ in McKubre’s laboratory.”

2.4

Energy reactions

The primary reaction, which converts matter and releases energy, in the form of heat, with helium-4 as a dominant by-product, contrasts with conventional electrochemical cells, which typically do not produce nuclear-level energy or convert elements but instead store and transfer energy. Low-energy nuclear reaction cells have a distant relationship to fuel cells: both continually consume fuel, and both continually produce by-products. The similarity ends there. The most significant difference is that the net reaction in a fuel cell is an ordinary chemical reaction and the reaction in an LENR cell is a nuclear reaction. Theoretical calculations conservatively estimate that the potential energy release from hydrogen in a LENR cell (based on an assumed 24 MeV fusion reaction) would be 8 million times more per reaction than that of a fuel cell or any chemical reaction using hydrogen, if the reaction is, in fact, a fusion reaction. If the LENR mechanism is the result of a non-fusion weak interaction process as proposed in the Widom-Larsen model,33 it would still release 4 million times more energy than a chemical reaction. The original Fleischmann–Pons experiment and the SRI experiment audited by R. Garwin indicate that the reactions are at least 1000 times more energetic than any known chemical reaction. Low-energy nuclear reactions also contrast with thermonuclear fusion. One of the thermonuclear fusion reaction channels releases heat and helium-4, but this reaction channel is extremely rare and is always accompanied by deadly gamma radiation. Low-energy nuclear reactions do not produce such radiation.

3 3.1

Nuclear evidence Helium-4

The palladium–deuterium experiment, based on the original Fleischmann–Pons electrolysis setup, can produce excess energy, in the form of heat, and helium-4 as a dominant nuclear by-product. Half a dozen independent reports show a close temporal correlation between the excess heat and the evolution of helium-4 (Fig. 2a). However, helium-4 can be difficult to distinguish from deuterium without the use of a high-resolution mass spectrometer. Another issue makes helium a difficult nuclear ash to identify and measure: it readily permeates many materials, including glass. In the early 1990s, M. Miles,21 at the US Navy China Lake laboratory, identified helium-4 as a dominant nuclear product of LENR experiments. M. Miles and J. O’M. Bockris27 at Texas A&M University subsequently performed experiments in stainless steel and detected significant amounts of helium-4. One of the best helium-4 results was obtained in an experiment at SRI International when researchers repeated an experiment developed by L. Case, a researcher in New Hampshire, using a palladium catalyst. Air from the active cell as well as a control cell was evacuated before starting the experiment. Both cells showed a starting value of helium-4 close to zero. The helium-4 value for the control cell, using hydrogen, stayed relatively flat, never greater than 1 ppm. The helium-4 signal for the active cell increased steadily over 15 days, at which point it surpassed the level of helium-4 in the environment (5.22 ppm), and then continued steadily upward to near 11 ppm (Fig. 2b).

Electrochemical Devices | Electrochemical Low-Energy Nuclear Systems

443

Fig. 2 (a) Temporal correlation of excess energy (heat) and helium production in deuterium gas cell. Two heat measurements (differential and gradient) are used and displayed. The figure shows heat and helium-4 increase in sealed cells containing palladium on carbon catalyst (SRI Case replication). The experiment is named after a man named Lester Case. (b) Increase of helium-4 in palladium/deuterium gas cell. (c) Comparison of proton recoil from known neutron source to emissions from Pd/D co-deposition experiment. Reproduced from 1st edition Krivit, S.B., Cold Fusion—Precursor to Low-Energy Nuclear Reactions, Encyclopedia of Electrochemical Power Sources, 1st edn.; Elsevier: Amsterdam, 2009; Vol. 2, pp. 255–270.

3.2

Tritium and helium-3

On very rare occasions and in low but statistically significant proportions, tritium and helium-3 (thought to be the product of tritium decay) have been observed. Tritium has been measured both in the gas phase and in the electrode. When tritium is found, excess heat is not observed. P. K. Iyengar and M. Srinivasan at the Bhabha Atomic Research Centre (BARC), in Trombay, India, witnessed a burst of tritium evolution on 21 April 1989. At Texas A&M University, a group led by J. O’M. Bockris noticed extremely high concentrations of tritium in their experiments on 24 April 1989. Eleven 11 of a set of 24 cells produced tritium at levels “100 to 1015 times above that expected from the normal isotopic enrichment of electrolysis.” E. Storms,28,29 with Los Alamos National Laboratory at the time, found that the signature of spiked tritium in LENR in no way matched the signature observed in the J. O’M. Bockris group’s cells. The J. O’M. Bockris group sought and obtained extensive confirmations of the tritium after the initial analysis at Texas A&M. Additional analyses were performed by Los Alamos National Laboratory (National Tritium Center), Argonne National Laboratory, Battelle Pacific Northwest Laboratory, and General Motors Research Laboratory. An experiment performed at SRI International, repeating one designed and originally performed at Osaka University by Y. Arata and Y. C. Zhang,6 produced tritium and helium-3. The key part of the cell design comprised a hollowed-out 14 mm palladium cathode containing finely divided palladium, also called palladium-black, which was welded close before the experiment began. SRI reported total production of tritium between 21015 and 51015 atoms. After the experiment, the cylindrical cathode was sectioned horizontally, and an assay was performed on the radial section for helium-3 on the inside of the chamber, on the outer wall of the chamber, and at five intermediate locations between the two. A signal of helium-3 was found at all points, the highest from inside the chamber, then sequentially lower at each of the remaining six locations. These data support the claim that the helium-3 was created inside the chamber. Decay rates from the tritium indicate that the causative event occurred during cathodic electrolysis (Fig. 3).

Fig. 3 Helium-3 and tritium found in higher concentrations on the inside of the cathode. The figure corresponds to a double structure cathode: radial distribution of helium-3 and tritium produced inside electron beam-welded cathode (SRI Arata–Zhang replication). Reproduced from 1st edition Krivit, S.B., Cold fusion—Precursor to Low-Energy Nuclear Reactions, Encyclopedia of Electrochemical Power Sources, 1st edn.; Elsevier: Amsterdam, 2009; Vol. 2, pp. 255–270.

444 3.3

Electrochemical Devices | Electrochemical Low-Energy Nuclear Systems Radiation and particles

Prompt radiation is produced and emitted from its source immediately. When the reaction stops, so does the prompt radiation. In contrast, emissions that continue after the reaction stops are called radioactive decay. LENR experiments produce various forms of nuclear radiation. The types, energies, and intensities of these various emissions are unlike those from conventional nuclear reactions, such as those occurring in nuclear power plants. In LENR research, radioactive decay has been reported only in rare situations. Of the prompt forms of radiation, the intensity levels, or flux, have been unusually low in proportion to the energy released. According to what was known of thermonuclear fusion, M. Fleischmann and S. Pons should have been killed by the intense commensurate neutron emissions from the 4 W of heat they claimed to produce. Of course, they were not. The low energy of the charged particles emitted in LENRs means that they are stopped, e.g., by a thin piece of paper. Notably missing are highly energetic gamma radiation and high fluxes of neutron emissions. If such emissions were present, thick layers of lead or concrete would be required to shield bystanders from the radiation. Types of prompt radiations detected include X-ray, gamma ray, and energetic particles (ions and electrons) and neutrons. All of these radiations are emitted at very low intensities so they are difficult to measure in LENR experiments. Furthermore, most X-rays and energetic charged particles rarely travel outside of an LENR experiment. Neutrons, at low levels, have been reported for many years in LENR experiments, particularly by T. Mizuno, assistant professor of nuclear engineering at Hokkaido National University. Detectors. The use of solid-state nuclear track detectors, commonly known as CR-39 plastic track detectors, has been refined by S. Szpak and P. A. Mosier-Boss30,31 at the US Navy’s Space and Naval Warfare Systems (SPAWAR) Center in San Diego, CA. The detectors are placed just outside the cell, separated by a thin membrane, and sometimes directly within the cell. They observed evidence of charged and neutral particles. The use of solid-state nuclear track detectors is convenient and produces permanent, constantly integrated recordings of nuclear emissions. The detectors are not designed to be immersed in an electrolyte; they are designed for use in air or vacuum. When the detectors are used in the electrolyte, careful analysis is required. In March 2007, SPAWAR researchers reported evidence of neutron emission through proton recoil effects in the detectors. Tracks were observed on the front as well as the back side of the detectors. The back side of the detectors faced away from and had no contact with the cathodes. The sets of tracks on the front and back were spatially correlated with each other and with the cathode. The researchers reported a very strong signal-to-noise ratio. Many experimental results from the SPAWAR group detectors display most of the expected characteristics of alpha particle emissions: consistent and smooth shapes of individual tracks, consistent patterns and groups of tracks, significant track depth relative to the diameter of the track, and sharp edges and pinpoint centers. However, an optical, two-dimensional analysis is not ideal for an absolute confirmation that the apparent tracks clearly demonstrate the mathematically predictable geometry that is expected from the ionization path created by a charged particle. Researchers at SRI International, using a BF3 detector, observed apparent neutron counts above background in at least three experiments. F. Tanzella reported a measurement of a neutron burst 14 times higher than background, which ran for 14 h. Concurrently with the onset of the neutron signal, he observed a distinct drop in cell potential, suggesting a heating effect in the electrolyte. A CR-39 detector configured with this experiment also showed the neutron signal. It was independently analyzed in Russia by yet a third method, serial etching at the Russian Academy of Sciences. A. Roussetski and A. Lipson confirmed that the detector showed “real nuclear (proton recoil) tracks.” Additionally, they compared the signals to a known neutron source and concluded that the “experimental evidence can be considered as a strong, unambiguous proof of fast neutron (2.5 MeV) exposure” (Fig. 2c).

3.4

Ultrasound experiments

In 1989, R. Stringham used acoustic cavitation with ultrasonic waves for loading deuterium into materials in closed cells. Acoustic inertially confined fusion does not use condensed matter inside the apparatus. R. Stringham has produced evidence of melted and even vaporized 100-mm target foils as a result of his acoustic cavitation experiments. The acoustic energy inputs varied from 5 to 16 W with exposures of 5 min to several weeks in duration. Indeed, a vaporized section in the center of a palladium foil was presented. Scanning electron microscope (SEM) photomicrographs of the foils after the experiments show a palladium surface that looks identical to melted metal, as well as nanometer-size-diameter eruptions. Other changes to the cathodes include unusual morphological deformations, craters, and ‘hot spots.’ Exploding wires have also been reported experimentally by a number of researchers. None of these effects can be caused by Joule heating or by arcing because this type of experiment involves no electrolysis; only acoustic energy is input to these experiments. However, ultrasound is known for steep temperature gradients and deformations in the material.

3.5

Heavy element transmutation

Transmutation refers to atomic changes, addition or subtraction of protons to a nucleus through the use of high-energy particle accelerators. The change of deuterium into helium is such a transmutation. Transmutation of heavy elements were reported in the early 1990s largely through the work of J. O’M. Bockris at Texas A&M University. G. Miley,22,23 director of the Fusion Studies Laboratory at the University of Illinois, Urbana, noted LENR heavy element transmutation evidence reported by 15 independent laboratories.

Electrochemical Devices | Electrochemical Low-Energy Nuclear Systems

445

Y. Iwamura13 and coworkers at Mitsubishi Heavy Industries in Japan forced deuterium gas to pass through a multilayered substrate containing palladium and calcium oxide. A positive correlation between deuterium gas permeation and the elemental conversion rate was found on the front side of the substrate, where atoms from the new element are found in place of the element initially deposited there: cesium into praseodymium, barium into samarium, and strontium into molybdenum. Great care was taken to isolate possible contamination in the work. Some of the confirmatory measurements were independently confirmed in situ, using X-ray fluorescence, at the Japanese SPring-8 Synchrotron Radiation Research Institute. The resultant elements often show anomalous isotopic ratios, supporting the hypothesis that such elements are created by LENRs and are not natural artifacts. F. Celani reported a replication of the strontium-to-molybdenum reaction using electrolysis.

4 4.1

Other aspects of low-energy nuclear reaction Deuterium loading

Even though the specific loading ratio varies by cathode size and geometry, a minimum deuterium-to-palladium loading ratio of 0.90 is required to observe the excess-heat effect. Loading ratios lower than 0.90 sometimes produce the excess-heat effect, but it becomes increasingly unlikely below this threshold. In the case of the early Fleischmann–Pons work with bulk palladium, a wait time of 10 weeks was not unusual to achieve a sufficiently high loading ratio. For example, N. Lewis’ team at Caltech reported a failure to replicate after only 39 days. The maximum loading ratio they obtained, 0.80, was insufficient. Later, researchers have found ways to obtain the required loading quite early, without the long wait. The second parametric requirement is a relatively high current density through the cathode surface, a minimum of 250 mA cm−2. Again, this parameter varies based on cathode size and geometry. The third parametric requirement calls for a dynamic trigger, a stimulus that will cause the electrochemical cell to enter a state of disequilibrium. Fleischmann and Pons’ dramatic 1985 reaction occurred after the current was changed abruptly. Other known triggers are the application of temperature changes, low-power (30 mW) laser excitation, external electrical fields, and external magnetic fields.

4.2

Power and energy quantified

One of the largest amounts of excess heat as a percentage of input power was reported in 2004 by the Energetics Technologies Ltd. laboratory in Omer, Israel. A. El-Boher’s experiment demonstrated 2500% excess heat, and 1.1 MJ integrated over 17 h. A key aspect of the Energetics work is the use of a proprietary triggering waveform that is delivered to the experiment by the applied current. In 2007, replication of the Energetics experiments was reported from SRI International and the ENEA laboratory in Italy, including multiple excess-heat results greater than 1 W with a calorimetric uncertainty of three sigma. A significantly different group of experiments was performed in the 1990s by researchers at Siena University in Italy. S. Focardi and F. Piantelli12 used hydrogen gas in combination with nickel rods and reported 18 W and 72 W of excess heat—600 and 900 MJ of energy integrated—over 319 and 278 days, respectively. They also reported evidence of neutrons, gamma rays, charged particles, and the presence of anomalous elements. Their 1994 II Nuovo Cimento paper was challenged by E. Cerron-Zeballos and coworkers from CERN in 1996, and the Focardi–Piantelli group published a successful rebuttal in 1998. A much greater storage capacity is required for hydrogen than to transport an amount of gasoline that delivers the same amount of energy. Some attempts have been made to quantify the volumetric energy density relative to the volume of the palladium cathode. These estimates indicate that LENRs might have a very high energy density, even higher than that of the uranium fuel rods used in fission reactors. M. Fleischmann and S. Pons, and G. Preparata published energy densities in the range of 10–100 kW cm−3 based on single and nonreproduced experiments. Estimates from M. McKubre suggest that the maximum rates being observed are in the range of 1–10 kW cm−3 if the fuel consumed is believed to be deuterium, not palladium.

4.3

Excess heat boil-offs

Several rare reactions have been reported, many of them anecdotal and none of them repeatable at will. In 1992, M. Fleischmann and S. Pons did not replenish the electrolyte in a set of four cells and allowed them to run dry. A videotape of this event in a conference in Nagoya, Japan, displayed four cells, each of them initially producing small, fine bubbles from the hydrogen evolution during the loading process. The current in the first cell was 0.500 A. The initial current in each of the other three cells was 0.200 A, which was increased to 0.500 A at the beginning of days three, six, and nine of the experiment, respectively. In the course of the experiment, each cell, one by one, reaches the boiling point. Based on the volume of vaporized heavy water, Fleischmann and Pons calculated 144 W of excess power, 86.7 MJ of excess energy for a 10-min period and an energy density (based on the volume of the lead cathode) of 3.7 kW cm−3. When the electrolytic circuit was broken as a result of the absence of the electrolyte, the cell continued to give off excess heat for 3 h. A Kel-F plastic support melted, indicating temperatures above 300  C. In the early 1990s, P. A. Mosier-Boss (SPAWAR Center) observed a boil-off. The cathode also vaporized. A metallurgist noted that the silver streaks seen on the sides and top of the cell appeared similar to formations he had only previously seen when the metal had melted under water.

446

Electrochemical Devices | Electrochemical Low-Energy Nuclear Systems

At an MIT symposium in the early 1990s, L. Forsley of JWK Technologies Inc. reported on a cell that had been running at 80  C, at equilibrium, for 1 day. After the abrupt power interruption, the cell temperature shot up to 125  C, cracked a plastic insulator, and rapidly boiled off all the electrolyte—at a power input far below that required for Joule heating. In the early 1990s, T. Mizuno24,25 of Hokkaido University reported the boil-off of a cell initially running 24 W of input power that, in its last 8 days with current turned off, boiled more than 15 L of water. Mizuno had placed the cell in a bucket of water after disconnecting it from the power supply. Other researchers reporting excess-heat boil-offs are G. Mengoli20 of the Instituto di Polarografia in Italy and M. Miles of the US Navy’s China Lake Weapons Center.

4.4

Repeatability and materials science issues

How to obtain and sustain that loading ratio of at least 0.90 is one of the greatest mysteries, The characteristics of the palladium used as cathodes in LENR experiments, and consequently the results, vary widely based on its source, manufacture, and preparation. Palladium that develops significant cracks is not able to hold the required amount of deuterium within its lattice. Increasing the surface area of the palladium directly increases the excess-heat effect, which supports the hypothesis of a surface effect. The reactions do not appear to be uniformly distributed across the entire sample of palladium-black cathodes but appear in randomly distributed, tiny spots. Three notable exceptions exist to the irrepeatability of these experiments. In 1989, S. Szpak and P. A. Mosier-Boss used a neutral substrate for the cathode, typically a thin wire or mesh of nickel, silver, or gold. Palladium is introduced into the cell through a solution of palladium chloride and lithium chloride. Electrolysis simultaneously deposits deuterium and palladium, in particles 60 nm in diameter, onto the cathode. The required high loading ratio is achieved almost instantly as palladium and deuterium atoms deposit concurrently on the cathode. In 1994, A. M. Imam at the Naval Research Laboratory made cathodes from a unique palladium–boron alloy. M. Miles, who was with the US Naval Air Warfare Center Weapons Division at China Lake, CA, saw excess heat in seven of eight experiments with the material. On inspection, an obvious crack was apparent in the nonfunctional cathode.

4.5

Role of lithium

S. Pons and B. Huggins of Stanford University stated in 1989 that lithium might play an important role. A. Widom and L. Larsen state that lithium is a key factor in their theoretical explanation of LENRs.

4.6

Role of normal water

Light water excess heat might disprove the fusion hypothesis and make way for some other as yet unexplained mechanism. In heavy water experiments, introduction of normal water to a heavy water cell halts any excess-heat effect. Normal water has trace amounts of deuterium, one deuterium atom for every 6000 hydrogen atoms. Stanley Pons revealed to reporters at a press conference on 12 April 1989, at the American Chemical Society meeting in Dallas that his group’s normal water cells were also showing a slight but significant signal of excess heat. Later Fleischmann, as well as M. Srinivasan, dismissed the possibility of excess heat from normal water experiments. Research by F. Piantelli, G. Miley, F. Celani, T. Mizuno, J. Patterson, and R. Mills also adds support for the possibility that normal hydrogen may produce excess heat.

4.7

Environmental issues

All of the observed reactions appear to lack significant high-flux neutron and gamma ray emissions. Low levels of radiation found in at least some of these reactions is usually absorbed directly and promptly within the experiments. In addition, the experiments do not appear to produce any greenhouse gases or long-lived radioactive decay emissions. A wide variety of conditions have been reported to produce both excess heat and anomalous nuclear products. These include other variations of electrolysis, pressurized deuterium gas, gas-electric field discharge, gas diffusion, plasma electrolysis, ion bombardment, acoustic and mechanically induced cavitation, nanostructured or finely divided palladium, and even biological organisms.

5 5.1

Theories Fusion and fission

Low-energy nuclear reaction theories that propose a fusion or fission mechanism try to explain three key experimental observations: (1) the mystery of how the Coulomb barrier is penetrated, (2) the lack of high-flux neutron emissions, and (3) the absence of hard radiation in the form of lethal emissions of gamma or X-rays.

Electrochemical Devices | Electrochemical Low-Energy Nuclear Systems

447

Most LENR theories invoke the strong interaction as the underlying physical mechanism—that is, that some form of a Coulomb barrier-penetrating fusion process accounts for the observed phenomena. Some theorists propose a fusion mechanism, e.g., P. Hagelstein, S. Chubb, T. Chubb, A. Takahashi, and X. Z. Li.

5.2

Neutron-catalyzed reactions

Neutron-catalyzed reactions have been proposed by H. Kozima,15 J. Fisher, G. Anderman, L. Chatterjee and others as the underlying mechanism. T. Mizuno, Y. Iwamura, S. Szpak, A. Widom, and L. Larsen refer to weak interactions of ultracold, ultralong-wavelength neutrons that then facilitate subsequent neutron-catalyzed reactions. An electron and a proton combine into a slow neutron and a neutrino in a fluctuating electric field. The slow neutron recombines with other elements in the experiment through a chain of nuclear reactions. Nonlinear collective coupling effects associated with the neutron creation step provide a two-way mechanism able to bridge a huge energetic gulf normally separating the chemical and nuclear energy realms.   Q 36 Li + 2n ! 2 42 He + e + ne ¼ 26:9 MeV Given that the fundamental basis for the Widom–Larsen theory is weak interaction neutron creation and subsequent neutron-catalyzed nuclear reactions, rather than the fusing of deuterons, the Coulomb barrier problem is irrelevant. Ultracold neutrons are created on solid metallic surfaces that involves a direct reaction between ‘heavy’ surface electrons and protons or deuterons to produce neutrons and neutrinos. Such collectively formed ultracold neutrons are almost always absorbed locally by nearby nuclei, producing new stable or unstable isotopes and elements through transmutation and related beta decay processes that can also generate heat. Some of the products of such reactions are helium-4, helium-3, and tritium. The absence of lethal photon radiation is explained by a mechanism in which high-energy gamma photons are internally converted into more benign infrared (heat) radiation by electromagnetic interactions with heavy electrons. Among others, D. Bushnell (NASA Langley Research Center) has expressed strong enthusiasm for the Widom–Larsen: “There is now a very successful theory for LENRs; weak interactions on surfaces fully explicable by the Quantum Standard Model,” and “Neutron formation allows subversion of the Coulomb barrier. It is not thermonuclear fusion at all, but it is very useful, with virtually no radioactivity concerns.” The Widom–Larsen model states that the expected energy produced per helium-4 atom should be roughly 13 MeV. The excess energy measured in LENR experiments have ranged from approximately 12–48 MeV per helium-4 atom produced. Their model also proposes an explanation for excess heat from hydrogen as well as deuterium experiments. G. Greenman (Lawrence Livermore National Laboratory) was skeptical that LENRs would be able to produce significant amounts of energy at high rates, as observed in experiments. By definition, weak interactions are weak energetically. However, some weak interactions with certain neutron-rich isotopes have beta decays with energies as high as 20 MeV. According to Widom and Larsen, LENRs produce such neutron-rich isotopes. In some unusual circumstances, reaction rates can be very high. The theory also shows how energy is produced by alpha decay, which produces helium-4. Another process is gamma-shielded neutron captures that occur on substrates such as transition metals and certain noble metals. Prompt gamma rays resulting from neutron captures are converted directly into infrared energy by heavy electrons by extraordinarily high localized surface electric fields that occur in LENR systems.

6

Conclusion

The ‘cold fusion’ episode is an instructive lesson on controversial discoveries. After more than 30 years, the unexplained experimental anomalies as the result of heavy water electrolysis remain a phantom. Low-energy nuclear reactions (LENRs) have been discussed that might generate large energy gains. Scientific rigor provides the method and confidence to distinguish fact from artifact, observation from illusion. The claims of cold fusion, however, are unusual in that even the strongest proponents assert that the experiments, for unknown reasons, are not consistent and reproducible at the present time. The history of science was enriched by a complex and instructive chapter.

References 1. 2. 3. 4. 5. 6. 7. 8. 9.

Krivit, S. B. Cold fusion—Precursor to low-energy nuclear reactions. In Encyclopedia of Electrochemical Power Sources, 1st edn; Elsevier: Amsterdam, 2009; vol. 2; pp 255–270. Krivit, S. B. Cold Fusion: History. In Encyclopedia of Electrochemical Power Sources, 1st edn.; Elsevier: Amsterdam, 2009; vol. 2; pp 271–276. Fleischmann, M.; Pons, S. Calorimetry of the Pd–D2O System: From Simplicity Via Complications to Simplicity. Phys. Lett. A 1993, 176, 118–129. Fleischmann, M.; Pons, S.; Anderson, M. W.; Li, L. J.; Hawkins, M. Calorimetry of the Palladium–Deuterium–Heavy Water System. J. Electroanal. Chem. 1990, 287, 293–351. Fleischmann, M.; Pons, S.; Anderson, M. W.; Li, L. J.; Hawkins, M. Background to cold fusion: The genesis of a concept. In Proceedings of the Tenth International Conference on Cold Fusion; Hagelstein, P. L., Chubb, S. R., Eds.; World Scientific: London, 2003; pp 1–11. Arata, Y.; Zhang, Y. C. A New Energy Generated in DS-Cathode with ‘Pd-Black’. Koon Gakkai Shi, J. High Temp. Soc. 1994, 20 (4), 148–155. Beaudette, C. Excess Heat & Why Cold Fusion Research Prevailed, 2nd edn; Oak Grove Press: South Bristol, ME, 2002. Bridgman, P. W. The Physics of High Pressure; International Textbooks of Exact Science: London, 1947. Bush, B. F.; Lagowski, J. J.; Miles, M. H.; Ostrom, G. S. Helium Production during the Electrolysis of D2O in Cold Fusion Experiments. J. Electroanal. Chem. 1991, 304, 271–278.

448

Electrochemical Devices | Electrochemical Low-Energy Nuclear Systems

10. Czerski, K.; Huke, A.; Heide, P.; Ruprecht, G. Experimental and Theoretical Screening Energies for the 2H(d, P)3H Reaction in Metallic Environments. Eur. Phys. J. A 2006, 27 (S1), 83–88. 11. De Ninno, A.; Frattolillo, A.; Rizzo, A.; Del Giudice, E.; Preparata, G. Experimental Evidence of 4He Production in a Cold Fusion Experiment, Report Prepared and Distributed by Servizio Edizioni Scientifiche—ENEA; Centro Ricerche Frascati: Rome, Italy, 2002. 12. Focardi, S.; Gabbani, V.; Montalbano, V.; Piantelli, F.; Veronesi, S. Large Excess Heat Production in Ni–H Systems. Il Nuovo Cimento A 1998, 111, 1233–1242. 13. Iwamura, Y.; Sakano, M.; Itoh, T. Elemental Analysis of Pd Complexes: Effects of D2 Gas Permeation. Jpn. J. Appl. Phys. 2002, 41, 4642–4650. 14. Koldamasov, A. I.; Yang, H. I.; Mcconnell, D. B.; Kornilova, A. A.; Vysotskii, V. I.; Desyatov, A. V. Observation and Investigation of Nuclear Fusion and Self-Induced Electric Discharges in Turbulent Liquids. In Proceedings of the Twelfth International Conference on Condensed Matter; Takahashi, A., Ota, K., Iwamura, Y., Eds.; World Scientific: London, 2005; pp 97–107. 15. Kozima, H. Discovery of the Cold Fusion Phenomenon; Tokyo: Ohotake Shuppan, 1998. 16. Krivit, S. B.; Winocur, N. The Rebirth of Cold Fusion: Real Science, Real Hope, Real Energy; Pacific Oaks Press: Los Angeles, CA, 2005. 17. Li, X. Z.; Liu, B.; Tian, J.; Wei, Q. M.; Zhou, R.; Yu, Z. W. Correlation between Abnormal Deuterium Flux and Heat Flow in a D/Pd System. J. Phys. D Appl. Phys. 2003, 36, 3095–3097. 18. Marwan, J., Krivit, S. B., Eds.; In Low-Energy Nuclear Reactions Sourcebook; American Chemical Society/Oxford University Press: Washington DC, 2008. 19. McKubre, M.; Tanzella, F.; Tripodi, P.; Hagelstein, P. The Emergence of a Coherent Explanation for Anomalies Observed in D/ Pd and H/Pd system: Evidence for 4He and 3He Production. In Proceedings of the Eighth International Conference on Cold Fusion; Scaramuzzi, F., Ed.; Bologna, Italian Physical Society: Lerici La Spezia, Italy, 2000; pp 3–16. 20. Mengoli, G.; Bernardini, M.; Manduchi, C.; Zannoni, G. Calorimetry Close to the Boiling Temperature of the Pd–D2O Electrolytic System. J. Electroanal. Chem. 1998, 444, 155–167. 21. Miles, M. Correlation of Excess Enthalpy and Helium-4 Production: A Review. In Proceedings of the Tenth International Conference on Cold Fusion; Hagelstein, P. L., Chubb, S. R., Eds.; World Science: London, 2003; pp 123–131. 22. Miley, G. H.; Shrestha, P. Review of Transmutation Reactions in Solids. In Proceedings of the Tenth International Conference on Cold Fusion; Hagelstein, P. L., Chubb, S. R., Eds.; World Science: London, 2003; pp 361–378. 23. Miley, G. H.; Name, G.; Woo, T. Use of Combined NAA and SIMS Analyses for Impurity Level Isotope Detection. J. Radioanal. Nucl. Chem. 2005, 263 (3), 691–696. 24. Mizuno, T. Nuclear Transmutation: The Reality of Cold Fusion; Infinite Energy Press: Bow NH, 1998. 25. Mizuno, T.; Ohmori, T.; Akimoto, T.; Takahashi, A. Production of Heat during Plasma Electrolysis in Liquid. Jpn. J. Appl. Phys. 2000, 39, 6055–6061. 26. Oriani, R. A.; Nelson, J. C.; Lee, S. K.; Broadhurst, J. H. Calorimetric Measurements of Excess Power Output during the Cathodic Charging of Deuterium into Palladium. Fusion Technol. 1990, 18, 652–662. 27. Packham, N. J. C.; Wolf, K. L.; Wass, J. C.; Kainthla, R. C.; Bockris, J. O. M. Production of Tritium from D2O Electrolysis at a Palladium Cathode. J. Electroanal. Chem. 1989, 289, 451–458. 28. Storms, E. The Science of Low Energy Nuclear Reaction: A Comprehensive Compilation of Evidence and Explanations about Cold Fusion; World Scientific: London, 2007. 29. Storms, E.; Talcott, C. L. Electrolytic Tritium Production. Fusion Technol. 1990, 17, 680–706. 30. Szpak, S.; Mosier-Boss, P. A.; Gordon, F. E. Further Evidence of Nuclear Reactions in the Pd/D Lattice: Emission of Charged Particles. Naturwissenschaften 2007, 94 (6), 511–514. 31. Szpak, S.; Mosier-Boss, P. A.; Young, C.; Gordon, F. E. Evidence of Nuclear Reactions in the Pd Lattice. Naturwissenschaften 2005, 92 (8), 394–397. 32. Violante, V.; Castagna, E.; Sibilia, C.; Paoloni, S.; Sarto, F. Analysis of Ni-Hydride Thin Film After Surface Plasmons Generation by Laser Technique. In Proceedings of the Tenth International Conference on Cold Fusion; Hagelstein, P. L., Chubb, S. R., Eds.; World Scientific: London, 2003; pp 421–434. 33. Widom, A.; Larsen, L. Ultra Low Momentum Neutron Catalyzed Nuclear Reactions on Metallic Hydride Surfaces. Eur. Phys. J. C 2006, 46 (1), 107–111.

Electrochemical Terminology | Electrochemical Cell Nomenclature Peter Kurzweil, Technical University of Applied Sciences (OTH), Amberg-Weiden, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of S P. Kurzweil, BATTERIES | Nomenclature, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 381–394, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00042-3.

1 2 3 3.1 3.2 3.2.1 3.2.2 3.2.3 3.2.4 3.2.5 3.3 3.3.1 3.3.2 3.3.3 3.3.4 3.4 3.4.1 3.4.2 3.5 4 4.1 4.2 4.3 4.4 4.5 4.6 5 6 6.1 6.2 6.3 6.4 7 References

Introduction Sign conventions and cell polarity Definitions of potentials and cell voltage Electrochemical half-cells Electrode potential Open-circuit potential Equilibrium potential Overpotential Reference potential Absolute potential Standard electrode potential Equilibrium potential under standard conditions Standard hydrogen electrode (SHE) Electrode potential under ambient conditions Formal potential Cell voltage Terminal voltage Open-circuit voltage Resistance Current direction and power rating Current and short-circuit current Electric charge and capacity Electric energy Electric power Capacitance Impedance Self-discharge and cycling conditions Constants, units, and conversions for power sources International system of units (SI) Common units Electrochemical constants Quantities and units for electrochemical power sources Conclusion

450 450 451 451 451 451 452 452 452 452 453 453 453 454 454 454 454 455 456 456 456 457 457 458 458 458 458 459 459 460 461 461 464 464

Abstract This article explains the correct use of electrochemical terms based on IUPAC recommendations: (1) Sign conventions and definitions for electrochemical cells: conductors, reactants, cell polarity, potentials and cell voltage. (2) Electrical and electrochemical definitions for quantities and rated data describing the charge–discharge behavior and the performance of power sources. (3) Terminology of physical quantities, list of SI units for power sources, fundamental constants, and conversion factors for common units.

Key points

• • •

Fundamentals of electrochemical terminology Overview of definitions and conventions in electrochemistry Correct usage of symbols, signs and units

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00035-9

449

450

1

Electrochemical Terminology | Electrochemical Cell Nomenclature

Introduction

In the history of electrochemistry, from the early experiments on electrical discharges to modern battery technology and photo electrochemistry, many introduced technical terms have undergone a change of meaning or have been clarified. Outdated terms like ‘electromotive force’, ‘overvoltage,” and ‘polarization’ require a meaningful re-evaluation. Recently, the base units of the international system of units were redefined on the basis of established natural constants. This article is based on IUPAC definitions and practical conventions.1–10

2

Sign conventions and cell polarity

Every electrochemical cell – primary cell, secondary cell, fuel cell, electrochemical capacitor, electrolysis cell, or corrosion element – consists basically of two electrodes dipping into an electrolyte, also called the two half-cells. At the electrodes, electrochemical reactions take place that involve electrons as reactants. Electrode: An electronic conductor, e.g. of metal,carbon, graphite, metal oxide (such as PbO2, MnO2), or semiconducting materials. The polarity of the electrode refers to the charges residing at its surface and its terminal to which the external electric circuit is connected. Electrolyte: An ionic conductor, liquid, solid, or fused, which transports the electric current by the mobility of ions, for example, water-based or organic salt solutions, diluted acids or bases, salt melts, oxide ion conducting ceramics (such as ZrO2), or conducting polymers. An ionic liquid is composed of a salt that is liquid below 100
C. In a solid electrolyte such as NASICON (Na1+xZr2P3−xSixO12, 0 < x < 3), the predominant charge carriers are ions. Positive charge carriers are referred to as cations; negative charge carriers as anions. The intrinsic conductivity of pure water is too low to be an appropriate electrolyte: k ¼ 0.0635 mS cm−1 at 25
C owing to autoprotolysis (2H2O Ð H3O+ + OH−). At the electrode–electrolyte interface change occurs from conduction by electrons to conduction by ions. Anode: The electrode at which electrochemical oxidation takes place, a chemical reaction that results in the release of electrons by the active material, so that the oxidation number of the active species is increased (Table 1). M Ð Mz+ + z e Connected with the positive pole of a power source, nearby every metal M is dissolved and metal cations Mz+ are formed. The number of electrons z transferred in the electrochemical reaction is determined by the valency (oxidation number, charge number) of the species M. When the electrooxidation is accompanied by chemical processes or a side reaction, the apparent value of z will differ from that in the redox equation. At an anode, electrons are produced in a galvanic cell or extracted in an electrolytic cell. The concepts of anode and cathode are related only to the direction of electron flow, not to the polarity of the electrodes.8 (1) The anode is the negative terminal of a primary or secondary battery cell, as the electrode material gives up electrons to the external load circuit and dissolves into the electrolyte or is precipitated on the electrode surface in a reaction with the electrolyte. In a proton exchange membrane (PEM) fuel cell, hydrogen molecules are oxidized to protons, whereby two electrons per H2 are released. (2) In an electrolysis cell, the anode is connected to the positive pole of a power source. Anions are discharged and oxygen is produced from aqueous solutions. Metals may dissolve into the electrolyte. Cathode: The electrode at which electrochemical reduction takes place, i.e. electron acceptance and decrease of oxidation number. Mz+ + z e Ð M (1) The positive terminal of a primary or secondary battery cell, toward which electrons flow through the external circuit when the cell discharges, for example, a noble metal that absorbs electrons. In a PEM fuel cell, the reduction of every oxygen molecule consumes four electrons and four protons to form two water molecules. (2) In an electrolysis cell, the cathode is connected to the negative pole of a power source. Cations are discharged and hydrogen is evolved from aqueous solutions. Anodic currents: They have a positive sign (I > 0) and denote the flow of electrons from the electrolyte into the electrode. Cathodic currents: They have a negative sign (I < 0) and denote the flow of electrons from the electrode into the electrolyte. Table 1

Definitions of anode and cathode. Cathode

Electrochemical cell (battery) Electrolytic cell (electrolyzer)

Anode

Sign

Function

Sign

Function

Positive Negative

Electron-accepting Cation-attracting Cation-reducing

Negative Positive

Electron-supplying Anion-attracting Anion-oxidizing

Electrochemical Terminology | Electrochemical Cell Nomenclature

451

Oxidation number: Assigned to a specific atom in a specific ion or molecule, the oxidation number is the electric charge (not in coulomb, but in electron charge units) actually carried by this atom, just like an ion in an ionic bond. In a covalent bond, all binding electrons are entirely displaced to the most electronegative partner. Equal electronegative partners share the binding electrons. The oxidation number is obtained by the number of valency electrons, i.e., the atom’s group number in the periodic table of elements. Since oxidation number does not necessarily correspond with a real ionic charge, it is indicated by a Roman numeral according to IUPAC. For example, Mn2+ carries the oxidation number +II (plus two) and the electric charge 2+ (two plus). Manganese in MnO2 has the oxidation number +IV, oxygen in oxides always has the oxidation number −II. Hydrogen has the oxidation number +I, and in metal hydrides it is −I. Free elements (H2, N2, O2, Cl2, and metals) have the oxidation number 0 (zero). The sum of the oxidation numbers yields the electric charge of the particle. For example [MnO4]−, +VII + 4(−II) ¼ −1. The equation means that the negatively charged MnO4− ion contains a Mn(VII) cation and four oxide anions. Oxidizing agents: When the oxidation number of an atom in a particle – in an element, a molecule, or an ion – decreases due to the absorption of electrons, it is said that this atom is being reduced and that the particle as a whole acts as an oxidant: Ox + z e Ð Red Examples: F2, S2O82−, H2O2, MnO4−, Cr2O72−, O2, Cl2, Ag+, Cu2+. E0 > 0, for all; see the Section 3.3. Reducing agents: When the oxidation number of an atom in a particle – in a chemical element, a molecule, or an ion – increases due to the release of valency electrons, it is said that this atom is being oxidized and the particle as a whole acts as a reductant. Red Ð Ox + z e Examples: K, Na, Al, Zn, Fe, Sn2+, hydroquinone, and Fe2+. E0 < 0, for all. Redox couple: A pair of redox forms, linked by a given half-cell reaction, is present at each electrode. Example: The reaction Ag Ð Ag+ + e− is based on the Ag/Ag+ redox couple, and takes place at a silver electrode in an aqueous solution. A strong oxidizing agent (Ox) corresponds to a weak reducing agent (Red) and vice versa, which is called a conjugated oxidant–reductant couple. x Ox 1 + y Red2 Ð p Red1 + q Ox2 By analogy with acid–base reactions, this redox reaction, whether self-sustaining or not, mainly progresses from the direction of relatively strong reductants and oxidants toward weaker conjugated oxidants and reductants. Electronegativity (EN): It denotes the degree to which an atom will reduce and function as an oxidizing agent or a cathode. In the Pauling electronegativity scale, the ability of an element to attract the electrons of a covalent bond is allotted relative to fluorine (EN 4.0). Nonmetals have the highest electronegative values, H (2.1), C and S (2.5), N and Cl (3.0), and O (3.5); followed by metalloids, Si (1.8) and B (2.9). Metals range below 1.7: K, Na, Li (0.8–1), and Al (1.5). Their behavior is termed electropositive as they will oxidize readily and function as a reducing agent or an anode.

3 3.1

Definitions of potentials and cell voltage Electrochemical half-cells

According to International Union of Pure and Applied Chemistry (IUPAC), a half-reaction equation is written in the form of a reduction: x Ox + z e– Ð y Red. This represents a redox couple of an oxidized (higher oxidation state) and a reduced form (lower oxidation state) of the same substance in equilibrium. An electrode dipping in an electrolyte is termed a half-cell. Two half-cells form a galvanic cell that is capable of producing an electric voltage by redox reactions, or an electrolytic cell that consumes electric power for the decomposition of the electrolyte. In a divided cell, the electrodes reside in different electrolyte chambers separated by a porous diaphragm or a membrane, which prevents mixing but allows the passage of electric current.

3.2

Electrode potential

The term potential is reserved for the case when an electrode is considered, whereas ‘voltage’ or ‘applied potential (difference)’ describes a complete cell.

3.2.1 Open-circuit potential The open circuit potential (OCP) is the electrode potential of a working electrode relative to the reference electrode when no potential or electric current is applied to the electrochemical cell. In the case of a reversible electrode system, the OCP is also referred to as the equilibrium electrode potential. Otherwise, it is called the rest potential, or the corrosion potential, depending on the system being studied.8 The electric potential j of an electrode, more precisely, of an electrode reaction at the electrode–electrolyte interface, is measured as the potential difference against a reference potential jref, whereby there is no current flowing through the reference electrode.

452

Electrochemical Terminology | Electrochemical Cell Nomenclature

The cell current I flows between the working electrode (indicator electrode) and a counter electrode (auxiliary electrode). In a three-electrode cell, the electrode potential is measured between the working and reference electrode, therefore written as E versus reference. EðIÞ ¼ jðIÞ  jref The type of reference electrode is written behind the symbol of the measured quantity. For example: E vs. SCE ¼ 0.45 V denotes an electrode potential which was determined against (versus) a saturated calomel electrode (SCE). ‘Corrosion potential’ denotes the electrode potential spontaneously acquired by a corroding material in a particular environment.

3.2.2 Equilibrium potential The equilibrium electrode potential, Nernst potential or reversible potential of an electrode is defined when no electric current flows through the cell and all local charge transfer equilibria across phase boundaries that are represented in the cell diagram (except at possible electrolyte-electrolyte junctions) and local chemical equilibria are established.8 Temperature and activities of relevant species must be specified. The equilibrium potential is related to the standard electrode potential by the Nernst equation. It is the open-circuit potential, formerly ‘electromotive force’ of a cell, in which the electrode on the left is a standard hydrogen electrode (SHE) and the electrode on the right is the electrode in question (IUPAC): E ¼ j  jSHE ðat I ¼ 0Þ The electrode potential is thermodynamically related to Gibbs free enthalpy of the potential-determining electrode reaction, DG ¼ −z F E; see Section 3.4. An electrode or an electrochemical cell is in equilibrium when there is no net current flowing (besides the exchange current owing to the Fermi level adjustment). In practice, without any externally applied potential, parasitic anodic and cathodic partial currents may flow because of secondary corrosion processes; then the measured open-circuit potential (OCP), mixed potential, or rest potential does not exactly equal the equilibrium potential (related to Gibbs free energy) of the desired electrode reaction. For example, at a zinc rod in diluted sulfuric acid, two kinetically inhibited and potential-determining electrode reactions take place simultaneously and independently from each other: Zn Ð Zn2+ + 2e− and H3O + e− Ð H2O + ½H2. The mixed potential measured at I ¼ 0 comes rather close to the equilibrium potential E of pure zinc, which can be studied better in neutral solution, where the hydrogen evolution reaction plays no role (because of its high overpotential).

3.2.3 Overpotential The overpotential  is the electrode potential minus the equilibrium electrode potential of an electrochemical reaction. The operating potential E(I) of an anode is always more positive, and that of a cathode is more negative, than its equilibrium potential.
0:

anode

The overpotential  is the extra potential, in relation to the equilibrium value (I ¼ 0), required to cause a given electric current I to flow through the electrode.8 Overpotential is positive for oxidation reactions (anode) and negative for reduction reactions (cathode). The term ‘overvoltage’ is deprecated. A polarizable electrode is an electrode whose potential changes with an applied potential. A reference electrode should be a non-polarizable electrode that holds its potential essentially constant by efficiently allowing electric current to pass. The historical term ‘polarization’ should no longer be used. The modern term overpotential measures the extent of polarization, i.e., the change of potential of an electrode from its open-circuit potential (corrosion potential) upon the application of a current. The most important overpotential is caused by the kinetically inhibited charge transfer process (activation overpotential ct) through the electrode–electrolyte interface. Other overpotentials arise from diffusion processes d, a rate-determining chemical reaction step r or a slow crystallization of metal atoms.

3.2.4 Reference potential A reference electrode – according to IUPAC – maintains a virtually invariant potential under the conditions prevailing in an electrochemical measurement, and serves to permit the observation, measurement, or control of the potential of the indicator or working electrode: EðSHEÞ ¼ EðRef Þ + Eref For example, to obtain a value versus standard hydrogen electrode (SHE), 0.241 V has to be added to the measured values obtained againsta saturated calomel electrode (here: Ref ¼ SCE).

3.2.5 Absolute potential The electric potential j of any conducting phase is not measurable by thermodynamically rigorous means. Experimentally, the potential difference between two electronic conductors is available only at a metal/metal-ion electrode, e.g., silver immersed in a

Electrochemical Terminology | Electrochemical Cell Nomenclature

453

silver nitrate solution. The potential difference between the metal electrode (M) and the solution phases (S) cannot be measured directly, but requires an electrical connection to the solution phase with setting up another electrode potential:   E ¼ ðjM  jS Þ  jref  jS’ ¼ EM  Eref The recommended absolute electrode potential of the hydrogen electrode under standard conditions is. E0 ðH+ jH2 Þ ¼ ð4:44  0:02Þ V at 298:15 K The absolute electrode potential describes the difference in electronic energy between a point inside the metal electrode (Fermi level) and a point outside the electrolyte (an electron at rest in vacuum). In practice, the universal hydrogen reference system is best realized by an ideally polarizable mercury electrode at the point of zero-charge: E0 ðH+ jH2 Þ ¼ F + c −EðH+ jH2 jHgÞ, where F is electron work function of mercury, c is the contact (Volta) potential difference at the mercury-solution interface.

3.3

Standard electrode potential

Synonyms: standard potential, standard electrode potential of an electrochemical reaction E0; standard reduction potential. Deprecated: standard electromotive force (emf ).

3.3.1 Equilibrium potential under standard conditions The equilibrium potential of a half-cell reaction is measured as electric potential difference against the SHE at standard conditions: ambient pressure of 105 Pa, all activities a ¼ 1, and all solids at 298.15 K ¼ 25
C in their most stable modifications. Molecular hydrogen under standard pressure is oxidized to solvated protons at the left-hand electrode (IUPAC). For example, the standard potential, e.g., of the Ag|Ag+ couple is E0(Ag|Ag+) ¼ +0.799 V, which is the standard electrode potential of the reduction Ag+ + e− Ð Ag; but that of the silver oxidation is −0.799 V versus SHE: Pt | H2 ða ¼ 1Þ | H+ ða ¼ 1Þ ¦¦ Ag+ ða ¼ 1Þ|Ag   E0 ¼ lim ’0right − ’0left I!0

A solid vertical bar | represents a phase boundary, a vertical dashed bar ¦ represents a junction between miscible liquids, and double dashed vertical bars ¦¦ represent a liquid junction in which the liquid junction potential is assumed to be eliminated, e.g. by a salt bridge. Noble metals (oxidants) exhibit a positive standard potential (E0 > 0), as they prefer the reduced state and can thus be deposited from metal salt solutions by reduction. Base metals are negative (E0 < 0, reductants), because they tend to dissolve spontaneously in aqueous solution. The couple with more negative standard potential is able to reduce couples with more positive standard potential in the table of electrochemical series. E01 > E02 : Species 1 oxidizes species 2: The standard potential E0 is defined by the standard Gibbs energy of the cell reaction in the direction in which reduction occurs at the right-hand electrode in the diagram representing the cell: RT ln K zF The subscript r (reaction) is usually omitted. The standard equilibrium constant is denoted by K (also called thermodynamic equilibrium constant) for the cell reaction, z its charge number (1, 2, 3, . . .), F the Faraday constant, R the molar gas constant, and T the thermodynamic temperature. Dr G0 ¼ −zFE0 ¼

3.3.2 Standard hydrogen electrode (SHE) For solutions in protic solvents, the SHE is the universal reference electrode for which, under standard conditions, the standard electrode potential E0(H+|H2) is zero at all temperatures (IUPAC). The SHE, by convention, is always placed on the left and connected through a salt bridge to the half-cell under test. The ideal hydrogen electrode consists of a platinized platinum sheet, immersed in 1-active hydrochloric acid (pH 0, and molality b(H+) ¼ 1.184 mol kg−1 at 25
C), over which hydrogen gas is bubbled. In the standard state of the hydrogen electrode the standard state pressure equals p(H2) ¼ p0 ¼ 105 Pa ¼ 1 bar, and the mean activity of the hydrochloric acid solution reads a(HCl) ¼ 1. The former use of the ‘standard atmosphere’ 1 atm ¼ 1.01325 bar should be discontinued (IUPAC recommendation since 1982):   E0 ð101325 Pa Þ ¼ E0 105 Pa + 0:17 mV The error is about 5 mV in 1-molar acid, taking into account the partial pressure of water vapor in the hydrogen stream at ambient pressure (1 bar). A true SHE can hardly be realized in practice; therefore, the standard potential is extrapolated for increasing activity.

454

Electrochemical Terminology | Electrochemical Cell Nomenclature

3.3.3 Electrode potential under ambient conditions The Nernst equation supplies the equilibrium potential E of the half-cell reaction as a function of temperature and the activities of the oxidized and reduced species: x Ox + z e Ð y Red z Ð z H+ðaqÞ + z e H 2 2ðgÞ EðT, c, . . .Þ ¼ E0 −

aðRedÞy aðH+ Þz RT : ln zF aðOxÞx aðH2 Þz=2

Herein, a(H+) ¼ a(H2)  1 is omitted for the standard hydrogen electrode. In neutral aqueous solution, the reduction potential of the half-reaction H+|H2 equals −0.41 V, and for the half-reaction O2|H2O this is +0.81 V. Hydrogen is a stronger reducing agent and oxygen a stronger oxidant in water at pH 7 than at pH 0. At pH 7, hydrogen ions are less likely to oxidize zinc metal; and oxygen gas is less likely to oxidize iron:   1 pðH Þ EðH+ jH2 , 25
CÞ ¼ −0:05916  pH + lg 02 2 p   1 pðO Þ EðO2 jOH − , 25
CÞ ¼ 1:229 − 0:05916  pH − lg 02 4 p The reversible hydrogen electrode is immersed directly into the working electrolyte of the electrochemical cell and operated with hydrogen gas under atmospheric pressure. The equilibrium potential depends on the hydrogen ion activity, hence on the pH of the electrolyte. In a dynamic hydrogen electrode, a small amount of hydrogen is evolved electrolytically by a current of 1 mA cm−2 flowing between two internal platinum electrodes in a glass tube, immersed in the electrolyte solution of the working cell.

3.3.4 Formal potential As it is usually inconvenient to deal with activities, because ion activity coefficients g are almost always unknown, measured formal standard potentials, E00 versus SHE, are tabulated for designated solutions and concentrations. For the measurement of E00 , the concentration ratio of the oxidized and reduced species c(Ox)x/c(Red)y must be unity: gðRedÞy RT cðRedÞy RT − E ¼ E0 − ln ln zF gðOx Þx zF cðOxÞx |fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl} E00

where E is the equilibrium electrode potential, R the gas constant, T the thermodynamic temperature, F the Faraday constant, z the electron number of the electrochemical reaction; x, y the stoichiometric coefficients in the equation of the electrode reaction; c the amount concentrations of the species involved. The formal standard potential E00 is used when the concentrations of the various species are known, but their activities are not, e.g. in analytical chemistry and biochemistry. E00 can differ from the standard electrode potential E0 owing to the influence of real conditions (pH, ionic strength, complex forming substances). At 25
C, the Nernst equation is often written as: 0

E ¼ E0 −

y

0:0592 cðRedÞ lg z cðOxÞx

where the factor 0.0592  (RT/F)  ln 10 at 298.15 K.

3.4

Cell voltage

The term ‘voltage’ refers to the applied potential difference between the two electrodes of an electrochemical cell when current is flowing through the cell.

3.4.1 Terminal voltage Generally, the cell voltage U(I) means the applied potential difference or terminal voltage between the two electrodes of an electrochemical cell at any temperature, when current is flowing through the cell. U ðIÞ ¼ U 0 + Za ðIÞ + |Zc ðIÞ| + U S Herein U(I) is the cell voltage at the current I. U0 is the open-circuit voltage,  is the overpotential at the anode (a) and the cathode (c). When there is no current flowing through the cell, the overpotential is zero per definition. US is the ohmic voltage drop in the electrolyte between anode and cathode. In a three-electrode arrangement, US is the ohmic drop between working electrode and reference electrode. For sign convention, see Table 2.

Electrochemical Terminology | Electrochemical Cell Nomenclature Table 2

455

Sign convention in electrochemical cells. Galvanic cell (battery in discharge mode) n

Electrolyzer cell (or battery in charge mode) n

Electrode potential versus a nonpolarizable reference electrode

E ¼ ’ − ’ref

Open-circuit voltage (cell voltage at I ¼ 0) Terminal voltage including overpotentials

DE ¼ Ec – Ea ¼ U0 > 0 U(I) ¼ U0 − a − |c| − IRS

U0 ¼ Ea − Ec > 0 (decomposition voltage) U(I) ¼ U0 + a + |c| + IRS

U(I) ¼ U0 − IRi

U(I) ¼ U0 + IRi

+ −

anode cathode

Electric power

P ¼ UI ¼ U2 Ri R+aRa > 0 consumed by the load Ra

Maximum power

U (for Ri ¼ Ra) P ¼ 4R i

2

E¼

+ −

cathode anode

P ¼ UI ¼ I 2Ri > 0 consumed by the cell resistance Ri

E0 thermodynamic equilibrium potential at I ¼ 0 (V), I electric current in the direction +! – (A), Q electric charge (C ¼ As), Ri internal resistance of the cell (O), RS electrolyte resistance (O), U voltage (V ¼ J C−1), W electric energy (J ¼ Ws), j0 electrode potential at I ¼ 0 (V), Z overpotential (V), a ¼ anode, c ¼ cathode, 0 ¼ open-circuit

The terminal voltage consists of the sum of (1) equilibrium potential difference of the cell reaction, i.e., the reversible cell voltage or open-circuit voltage; (2) anodic and cathodic overpotentials, and (3) ohmic voltage drop between anode and cathode, caused by the electrolyte resistance. In the case of a three-electrode cell, the cell voltage equals the potential difference between the working electrode and the counter-electrode. The current I flowing through the cell is proportional to the rate of conversion pertaining to the cell reaction.

3.4.2 Open-circuit voltage Reversible cell voltage, the equilibrium electrode potential difference, U0 ¼ DE (in V), between the two electrodes of a galvanic cell, for zero current through the cell, all local charge transfer equilibria and chemical equilibria being established after a waiting period under constant ambient conditions (temperature and concentrations): ð− Þ PtjOx 1 , Red2 jjRed1 , Ox2 jPt ð + Þ ð + Þ Cathode : x Ox1 + z e− > y Red1 ð− Þ Anode : p Red2 > q Ox 2 + z e− Cell : x Ox1 + y Red2 > p Red1 + q Ox 2 The reaction quotient K (based on mean ionic activities) of the cell reaction determines the change of Gibbs free energy and thus the cell voltage, where K tends to be the equilibrium constant for DG ¼ DE ¼ 0: K¼

aðRed1 Þp  aðOx 2 Þq aðOx2 Þx  aðRed1 Þy

DG ¼ DG0 + RT ln K ¼ −zF DE DE ¼ DE0 −

RT ln K zF

A galvanic cell (battery, fuel cell) transforms chemical energy into electric energy as long as the change of Gibbs free enthalpy DG is negative for the total cell reaction; it delivers the reversible cell voltage or emf, DE, which is given by the difference of the reduction potentials of the two half-cell reactions. The cell voltage can be compensated by applying a counter voltage −DE to the cell: DE0 ¼ E0cathode − E0anode

456

Electrochemical Terminology | Electrochemical Cell Nomenclature DEðT, c, . . .Þ ¼ Ecathode − Eanode
DE ¼

>0:

spontaneous from left to right

0: anodic current dQ I¼ ¼ dt 80%) of total rated capacity; drains all electric energy by withdrawal of current until an end voltage is reached. The cell is fully discharged using low current, so that the voltage falls below the final discharging voltage. Shallow discharge: Discharge using only a small portion of total rated capacity; drains up to 50% of the energy partially on each discharge. Cutoff voltage or final voltage: The voltage of a rechargeable battery at which the charging or discharging is terminated. It is the prescribed lower-charging limit voltage at which useful capacity is considered complete. End-point voltage: Cell voltage below which the connected equipment will not operate or below which operation is not recommended. Float charging: Maintaining a rechargeable battery in a fully charged condition by continuous, long-term, and constant-voltage charging, at a level sufficient to balance self-discharge. Trickle charging: It is also known as boost charging; continuous, long-term, and slow-rate charging, at a level sufficient to balance self-discharge and occasional discharge. Taper charging: Charge at moderately high current when the battery is at a low state of charge and tapering the current to lower rates as the battery becomes fully charged.

6 6.1

Constants, units, and conversions for power sources International system of units (SI)

A coherent system of units, such as the SI, defines the base quantities and the base units independently of any other fundamental quantities or units, i.e., there are no numerical factors between two quantities or two units. SI quantities are organized in a dimensional system built upon seven base quantities, each of which is regarded as having its own dimension (Table 3). All other physical quantities are called derived quantities and are regarded as having dimensions derived algebraically from the seven base quantities by multiplication and division. The value of a physical quantity is expressed as the product of a numerical value and a unit; e.g., electric charge:



Q ¼ fQg ½Q ¼ 100 C ¼ 100 As





dim Q ¼ dim ðI t Þ ¼ I t The letters of quantities are printed in italic (sloping) style, units in roman (upright), and dimensions in serif-less letters. Neither the name of the physical quantity nor the symbol used to denote it should imply a particular choice of unit. Prefixes allow one to signify decimal multiples and submultiples of SI units (Table 4). Quantities, numerical values, and units may all be manipulated by the rules of algebra. Thus, it is correct to write, for example,



ð12 mmÞ2 ¼ 144 mm2 ¼ 144



2 106 m ¼ 1:44 1010 m2

Useful conversions: 1 cm2 ¼ 10 −4 m2 1 cm−1 ¼ 100 m−1 1 cm−2 ¼ 1000 m−1



460

Electrochemical Terminology | Electrochemical Cell Nomenclature Table 3 Base quantity

Symbol

Base unit

Dimension

Length Mass Time Electric current Thermodynamic temperature Amount of substance Luminous intensity

l m t I T n Iv

m kg s A K mol cd

L M T I Y N J

Table 4

6.2

Base physical quantities and units.

Decimal multiples and submultiples of SI units.

Multiple

Prefix

Factor

Multiple

Prefix

Factor

deci centi milli micro nano pico femto atto

d c m m n p f a

0.1 0.01 0.001 10−6 10−9 10−12 10−15 10−18

deca hecto kilo Mega Giga Tera Peta Exa

da h k M G T P E

10 100 1000 106 109 1012 1015 1018

Common units

Some traditional units are allowed for use together with the SI in appropriate contexts.

• • • • • • • •

minute: 1 min ¼ 60 s, hour: 1 h ¼ 3600 s, day: 1 d ¼ 86400 s, and year: 1 a ¼ 365 d plane angle degree (1
¼ p/180 rad), minute (0 ), and second (00 ) kilowatt-hour: 1 kWh ¼ 3.6106 Ws ¼ 3.6 MJ liter: 1 L]1 dm3 ¼ 1000 cm3 ¼ 0.001 m3 tonne, metric ton: 1 t ¼ 1000 kg bar: 1 bar ¼ 105 Pa ¼ 1000 hPa ¼ 105 Nm−2 ¼ 10 N cm−2 electronvolt: 1 eV  1.602210−19 J atomic mass unit (u)

Non-metric units are not allowed in the SI system. Conversion factors are needed for selected US units and obsolete values with importance for power sources:

• • • • • • •

1 in ¼2.54 cm 1 ft. ¼ 12 in ¼ 0.3048 m 1 yd. ¼ 3 ft. ¼ 0.9144 m 1 mi ¼1760 yd. ¼1609.344 m 1 barrel ¼ 158.987 dm3 1 gal ¼ 3.785 41 dm3 1 lb. ¼ 0.453592 37 kg 1 oz. ¼ 28.349523 g 1 atm ¼ 101325 Pa ¼ 1.01325 bar 1 mmHg ¼ 1 Torr ¼ 133.322 Pa 1 mWS ¼ 9806.65 Pa ¼ 98.0665 mbar 1 psi ¼ 68.9476 mbar 1 cal ¼ 4.184 J 1 PS ¼ 735.49875 W 1 hp. ¼ 745.7 W 1 Ci ¼ 3.7  1010 Bq 1 G ¼ 10−4 T 1 D ¼ 3.33564  1030 Cm

Electrochemical Terminology | Electrochemical Cell Nomenclature Table 5

461

Fundamental physical constants and definitions.

Speed of light in vacuum Elementary charge FARADAY constant Newtonian constant of gravitation Standard acceleration of gravity PLANCK constant BOLTZMANN constant AVOGADRO constant Electron rest mass Neutron mass Proton mass Standard atmosphere Molar gas constant Standard temperature (ideal gas) NERNST factor (25
C ¼ 298.15 K) Molar volume (0
C ¼ 273.15 K) LOSCHMIDT constant Atomic mass constant Permittivity of vacuum Permeability of vacuum Magnetic flux quantum Stefan-Boltzmann constant

m s−1 C C/mol m3kg−1 s−2 ms−2 Js J/K mol−1 u u u Pa J mol−1 K−1 K V m3 mol−1 m−3 kg F/m H/m Wb W m−2 K−4

¼ 299,792,458 ¼ 1.602,176,634 10−19 ¼ 96 485.332,12  6.674,30  10−11 ¼ 9.806,65 ¼ 6.626,070,15 10−34 ¼ 1.380,649  10−23 ¼ 6.022,140,76  1023 ¼ 5.485,799,090,65  10−4  1.008,664,915, 95  1.007,276,466,621 ¼ 101 325 ¼ 8.314,462,618 ¼ 273.15 ¼ 0.059,159,3. . . ¼ 22.413,969,54. . .  10−3 ¼ 2.686,780,111. . .  1025 ¼ 1.660,539,066,60  10−27 ¼ 8.854,187,817. . .10−12 ¼ 12.566,370,614. . .  10−7 ¼ 2.067,833,484. . . 10−15 ¼ 5.670374419. . .  10−8

c e F ¼ NA e G g h k ¼ R/NA NA me mn mp p0 R ¼ k NA T0 UN ¼ ln 10  RT/F Vm ¼ R T/p0 NL ¼ NA/Vm u ¼ 1/12 m(12C) e0 ¼ 1/(m0c2) m0 ¼ 4p10−7 F0 ¼ h/(2e) s

Exact value (¼) or uncertain last digits printed in italics. Conversion of number concentration N/V and molar concentration: N/V¼NA c.

6.3

Electrochemical constants

In May 2019, the numerical values of fundamental constants were fixed, with no uncertainty, thus defining the SI units. Of importance to electrochemistry are the values of the Boltzmann constant, the elementary charge, and the Avogadro constant. Faraday constant and gas constant are calculated values. Table 5 compiles fundamental constants and fixed values used in electrochemistry, battery technology, and all fields of energy conversion.

6.4

Quantities and units for electrochemical power sources

Table 6 lists the terminology, recommended symbols, and SI units for physical and electrochemical quantities. Conventional units such as A/cm shall be written with exponential numbers (A cm−1). Table 6

Quantities and units.

Quantity

Symbol

SI units

Dimension

Area Absorbance Magnetic vector potential Acceleration Activity (of substance) Mean ionic activity Molality (of a solute) Magnetic flux density, magnetic induction Capacitance Concentration Specific heat capacity (at constant pressure) Diffusion coefficient Electric displacement Thickness, distance Energy Potential energy Kinetic energy

A A Am a a a b B C c cp D D d E Ep, V Ek, T

m2 –– Wb m−1 ¼ Vs m−1 m s−2 –– –– mol kg−1 T ¼ V s m−2 F ¼ C V−1 ¼ As V−1 mol L−1 J kg−1 K−1 m2s−1 C m−2 m J J J

¼ m2 ¼1 ¼1 ¼1 ¼ kg s−2A−1 ¼ m−2kg−1s4A2 ¼ m−3kmol ¼ m2kg−2 K−1 ¼ m−2A s ¼m ¼ m2kg s−2 ¼ m2kg s−2 ¼ m2kg s−2

Definition A(l) ¼ −lg I/I0 ¼ e c d B ¼ rot Am a ¼ dv/dt ai ¼gi ci /(mol L−1) a ¼ gc/(mol L−1) bi ¼ni /mLm F ¼Q v B C ¼ Q/U ci ¼ni / V cp ¼ (dH/dT )p/m n_ ¼ –DA dc/dx D ¼ eE R Ep ¼ − F ds Ek ¼ 12mv2 (Continued )

462 Table 6

Electrochemical Terminology | Electrochemical Cell Nomenclature (Continued)

Quantity

Symbol

SI units

Dimension −2

−1

Definition −2

Modulus of elasticity, Young’s modulus Electric field strength (vector) Electrode potential versus a reference Potential of the electrochemical cell reaction Standard potential of the cell reaction Activation energy Fermi level Half-wave potential in voltammetry Force Frequency Shear modulus GIBBS free energy GIBBS free energy change in a chemical process Standard GIBBS free energy change Conductance Enthalpy

E E E E E0 EA EF E1/2 F f, n G G DG DG0 G H

Pa ¼ Nm V m−1 V ¼ J C−1 V V J mol−1 J, eV J N Hz Pa ¼ Nm−2 J J J S ¼ O−1 J ¼ Nm ¼ Ws

¼ m kg s ¼ m kg s−3A−1 ¼ m2kg s−3A−1 ¼ m2kg s−3A−1 ¼ m2kg s−3A−1 ¼ m2kg s−2 mol−1 ¼ m2kg s−2 ¼ m2kg s−2 ¼ m kg s−2 ¼ s−1 ¼ m−1kg s−2 ¼ m2kg s−2 ¼ m2kg s−2 ¼ m2kg s−2 ¼ m−2 kg−1 s3A2 ¼ m2kg s−2

Standard enthalpy change in a chemical process Magnetic field strength Electric current Ionic strength Radiant intensity Imaginary part of impedance Moment of inertia Flux of species j at location x at time t Electric current density Exchange current density Equilibrium constant Cell constant Force constant Heat transition coefficient Rate constant Mass transfer coefficient, diffusion rate constant Electrochemical equivalent Angular momentum, action Self-inductance Length Torque, moment of a force Molar mass Magnetization, magnetic dipole moment per volume Mass Mass flow rate Number of particles Number concentration, electron density in a semiconductor Amount of substance Number of electrons Refractive index Power Electric power Dielectric polarization Momentum Pressure, partial pressure Heat Radiant energy Electric charge, quantity of electricity Electric resistance Charge transfer resistance Electrolyte resistance Polarization resistance Reynolds number Real part of a complex quantity

DH0 H I Ic I Im Z J Ji(x,t) j j0 K K k k k kd, b k L L l M, T M M m _ m N n n, C n n P P P p p Q Q, W Q R Rct Re RP Re Re Z

J A m−1 A mol m−3 W sr−1 O ¼ V A−1 kg m2 mol m−2 s−1 A m−2 A m−2

¼ m2kg s−2 ¼ m−1A

m−1 N m−1 ¼ J m−2 W m−2 K−1 (mol−1 m3)n–1 s−1 m s−1 kg C−1 Nms¼Js H ¼ V s A−1 ¼ O s m Nm kg mol−1 A m−1 kg kg s−1 –– m−3 mol –– –– W ¼ J s−1 W C m−2 Ns Pa ¼ N m−2 J ¼ Ws J C O ¼ V A−1

¼ m−1 ¼ kg s−2 ¼ kg s−3 K−1

–– O

¼ m−3 mol ¼ m2kg s−3 sr−1 ¼ m2kg s−3A−2 ¼ m2kg ¼ m−2A ¼ m−2A

¼ kg A−1 s−1 ¼ m2kg s−1 ¼ m2kg s−2A−2 ¼m ¼ m2kg s−2 ¼ m−1A ¼1 ¼ m−3 ¼1 ¼1 ¼ m2kg s−3 ¼ m2kg s−3 ¼ m−2 A s ¼ m kg s−1 ¼ m−1kg s−2 ¼ m2kg s−2 ¼ m2kg s−2 ¼ As ¼ m2kg s−3A−2

¼1

E ¼ ds/de E ¼F/Q ¼ −grad ’ E ¼ ’ – ’ref E ¼ ’c – ’a (I ! 0) E0 ¼ −DrG0/(zF) EA ¼RT2d ln k/dT F ¼ dp/dt ¼ m a f ¼ c/l ¼ T −1 G ¼ dt/dg G¼H–TS DG ¼ −z F DE DG0 ¼ −z F E 0 G ¼ 1/R ¼ Re Y dH ¼ m cp dT H ¼ U+ p V B ¼m H I ¼ dQ/dt P 2 Ic ¼ 12 ci z i I ¼ dF/dO X ¼ RIm Z J ¼ r2 dm j ¼ I/A K ¼ k1/k−1 K ¼ k R ¼ d/A k ¼ –dF/ds dQ/dt ¼ k A DT k ¼ M/zF L ¼ |r p| ¼ mr2o U ¼ −L dI/dt M ¼ r F Mi ]mi /ni M ¼B/m0 – H

n ¼ N/V ni ¼ Ni /NA see z P ¼ dW/dt P¼UI P ¼D – e0 E p¼mv p ¼ F/A ¼S pi Q ¼ m cp DT Q¼It R ¼ U/I

Re ¼ gvd/

Electrochemical Terminology | Electrochemical Cell Nomenclature Table 6

463

(Continued)

Quantity

Symbol

SI units

Dimension

Radius, radial distance Heterogeneous reaction rate Homogeneous reaction rate Path length, length of arc Entropy Standard entropy change Poynting vector Period Thermodynamic temperature Time Transport number of species i Internal energy Electric voltage, cell voltage Electric mobility of ion i Volume Molar volume Linear velocity (of solution flow), speed Linear potential scan rate Work, energy Mass fraction Reactance Mole fraction of species i Admittance Impedance Charge number of an ion, electron number of an electrochemical reaction Angular acceleration Degree of dissociation Transfer coefficient Mass concentration Surface (excess) concentration of species i Activity coefficient for species i Mean ionic activity coefficient Shear strain Thickness of diffusion layer Loss angle Linear strain, relative elongation Porosity Permittivity Relative permittivity, dielectric constant Molar (decadic) absorption coefficient Electrokinetic potential, zeta potential Overpotential Dynamic viscosity Surface coverage (of an interface by species i) Conductivity (of a solution) Molar conductivity (of an electrolyte i) Wavelength (of light) Ionic conductivity, molar conductivity of an of ion i Reduced mass Permeability Electric dipole moment Chemical potential of species i in phase (a) Wave number Kinematic viscosity Stoichiometric number of species i Density, mass density Resistivity (Space) charge density Normal stress Surface tension

r r, v r, v s S DS0 S T T t ij U U ui V Vm v v W w X xi Y Z z a a a b Gi gi g g d d e, e e e er e z   yi k Li l li m m m, p m(a) i ~n n ni r r r s s, g

m mol m−2 s−1 mol m−3 s−1 m J/K

¼m

W m−2 s−1 K s –– J V ¼ J C−1 m2V−1 s−1 m3 m3 mol−1 m s−1 V s−1 J ¼ Nm ¼ Ws –– O ¼ V A−1 –– S ¼ O−1 O ¼ V A−1 –– rad s−2 –– –– kg m−3 mol m−2 –– –– –– m rad –– –– F/m –– –– V V Pa s –– S m−1 ¼ O−1 m−1 S m2 mol−1 m S m−1 kg H m−1 Cm J mol−1 m−1 m2s−1 –– kg m−3 Om C m−3 Pa ¼ N m−2 N m−1 ¼ J m−2

¼ kg s−3 ¼ s−1 base unit base unit ¼1 ¼ m2kg s−2 ¼ m2kg s−3A−1 ¼ kg−1s2A

¼m ¼ m2kg s−2 K−1

¼ m2kg s−4A−1 ¼ m2kg s−2 ¼1 ¼ m2kg s−3A−2 ¼1 ¼ m−2kg−1s3A2 ¼ m2kg s−3A−2 ¼1 ¼ s−2 ¼1 ¼1 ¼ kg m−3 ¼ m−2mol ¼1 ¼1 ¼1 ¼m ¼1 ¼1 ¼1 ¼ m−3kg−1s4A2 ¼1 ¼1 ¼ m2kg s−3A−1 ¼ m2kg s−3A−1 ¼ m−1kg s−1 ¼1 ¼ m−3kg−1s3A2 ¼ kg−1s3A2mol−1 ¼ m−3kg−1s3A2 ¼ kg ¼ m kg s−2A−2 ¼msA ¼ m2kg s−2 mol−1 ¼ m−1 ¼ m2s−1 ¼1 ¼ m−3kg ¼ m3kg s−3A−2 ¼ m−3s A ¼ m−1kg s−2 ¼ kg s−2

Definition

r ¼ (dc/dt ) / ni dS  dQ/ T S ¼ E H T ¼ 1/f ti ¼Ii /I DU ¼ Q + W U ¼ E ¼ D’ ui ¼vi /E Vm ¼ V/ni v ¼ dr/dt v ¼ dU/dt R W ¼ − F ds wi ¼mi /S mi X ¼ Im Z xi ¼ni /ntot Y ¼ 1/Z ¼G+ j B Z ¼R+ j X zi ¼Qi /e a ¼ do/dt (in Tafel equation) bi ¼mi /V ¼ c M ¼ r w Gi ¼ ni /A ai ¼ gi ci /c⁎ g¼ (g+)z+(g−)z_ d ¼ D/kd d ¼ (p/2) + ’I – ’U e ¼ Dl/l e ¼ er e0 A(l) ¼e c d  ¼ E – E0 – I Rel tx,z ¼ (dvx /dz) yi ¼ Ni /Nm,i j ¼ k E ¼ 1/r Li ¼ k/ci l ¼ c/ f li ¼zi F ui m ¼ m1m2/(m1 + m2) B ¼m0mrH p ¼Q r mi ¼ (dG/dni)T,p ~n¼ 1/l n ¼ /r r ¼ m/V r ¼ k−1 r ¼ Q/V s ¼ F/A s ¼ dW/dA (Continued )

464

Electrochemical Terminology | Electrochemical Cell Nomenclature

Table 6

(Continued)

Quantity Surface charge density Time constant, relaxation time Shear stress Radiant power Phase angle Electric potential Potential drop, Galvani voltage Magnetic susceptibility Electric susceptibility Electric flux Solid angle Circular frequency, pulsatance Angular velocity

Symbol s t t F ’ ’, V D’ w we C O o o

SI units −2

Cm s Pa W rad V V m3 mol−1 –– C sr rad s−1 rad s−1

Dimension −2

¼m sA ¼s ¼ m−1kg s−2 ¼ m2kg s−3 ¼1 ¼ m2kg s−3A−1 ¼ m2kg s−3A−1 ¼ m3mol−1 ¼1 ¼As ¼1 ¼ s−1 ¼ s−1

Definition s ¼ Q/A t ¼ |dt/dlnx| ¼ RC t ¼ F/A F ¼ P ¼ dQ/dt ’ ¼ dW/dQ wm ¼ Vmw we ¼ Rer – 1 C ¼ D dA O ¼ A/r2 o ¼ 2pf o ¼ dj/dt

Vectors are printed in bold.

7

Conclusion

Some quantities used in electrochemistry have undergone a change of meaning or a sharper definition over the decades. Examples are the terms ‘voltage,’ ‘electromotive force,’ and ‘polarization.’ The recent redefinition of the base units of the international system of units has affected some constants, such as elementary charge, Faraday constant, Avogadro constant, and Boltzmann constant.

References 1. Bard, A. J.; Memming, R.; Miller, B. Terminology in Semiconductor Electrochemistry and Photoelectrochemical Energy Conversion (IUPAC Recommendations 1991). Pure Appl. Chem. 1991, 63, 56–596. 2. Cohen, E. R.; Cvitaš, T.; Frey, J. G. Quantities, Units and Symbols in Physical Chemistry – The Green Book, 3rd ed; RSC Publishing: Cambridge, 2007. 3. Gritzner, G.; Kreysa, G. Nomenclature, Symbols, and Definitions in Electrochemical Engineering (IUPAC Recommendations 1993). Pure Appl. Chem. 1992, 65, 1009–1020. 4. Heusler, K. E.; Landolt, D.; Trasatti, S. Electrochemical Corrosion Nomenclature (IUPAC Recommendations 1988). Pure Appl. Chem. 1989, 61, 19–22. 5. IUPAC, Compendium of Chemical Terminology. In The “Gold Book”, 2nd ed.; McNaught, A. D., Wilkinson, A., Eds.; Blackwell Scientific Publications: Oxford, 1997. https://doi. org/10.1351/goldbook. 6. Leigh, G. J.; Favre, H. A.; Metanomski, W. V. Principles of Chemical Nomenclature: A Guide to IUPAC Recommendations; Blackwell Science: Oxford, 1998. 7. Parsons, R. Manual of Symbols and Terminology for Physicochemical Quantities and Units, Appendix III, Electrochemical Nomenclature (IUPAC Recommendations 1973). Pure Appl. Chem. 1974, 37, 499–516. 8. Pingarrón, J. M.; Labuda, J.; Barek, J.; Brett, C. M. A.; Camões, M. F.; Fojta, M.; Hibbert, D. B. Terminology of Electrochemical Methods of Analysis (IUPAC Recommendations 2019). Pure Appl. Chem. 2020, 92 (4), 641–694. 9. Rubenbauer, H.; Henninger, S. Definitions and Reference Values for Battery Systems in Electrical Power Grids. J. Energy Storage 2017, 12, 87–107. 10. Rumble, J., Ed.; In CRC Handbook of Chemistry and Physics, CRC Press: Boca Raton, 2022.

Electrochemical Terminology | Capacity H Wenzla, R Bengerb, and I Hauerc, aConsulting for Batteries and Power Engineering, Osterode, Germany; bResearch Center Energy Storage Technologies, Clausthal University of Technology, Clausthal-Zellerfeld, Lower Saxony, Germany; cChair of Electrical Energy Storage Technology, Institute of Electrical Power Engineering, Clausthal University of Technology, Clausthal-Zellerfeld, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of H. Wenzl, BATTERIES | Capacity, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 395–400, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00043-5.

1 1.1 1.2 2 2.1 2.2 2.3 2.4 2.5 2.6 3 3.1 3.2 3.3 3.4 3.5 3.6 4 5 5.1 5.2 5.3 6 References

Introduction Definition of battery capacity Mechanisms for storage of charge Necessary differentiation of the term capacity Theoretical capacity Nominal capacity Operating capacity Useable capacity Actual capacity Residual capacity Cell capacity Electrical conductivity of active material (lead acid batteries) Consumption of electrolyte Irreversible changes of active material and aging processes caused by deep discharging Rechargeability Current inhomogeneity Diffusion processes Capacity of cells connected in series Dependence of capacity on design and operating conditions, temperature and rate of discharge Effect of high discharge current Effect of temperature on capacity Effects of previous use on capacity Conclusion

466 466 466 466 467 468 468 468 468 469 469 469 470 470 470 470 470 471 471 471 471 472 472 472

Abstract This article defines the term capacity for batteries (in Ah). Theoretical capacity of the active materials can be calculated from fundamental physical and chemical data, but the actual capacity of an individual cell and of cells connected in series is considerably lower. The main reasons for the difference between theoretical capacity and the capacity of a power system are explained. In addition, the term capacitance (in F ¼ As/V) for supercapacitors is explained.

Key points

• • •

Overview of electrochemical terminology Definition of the term capacity Correct usage of symbols and units

Nomenclature

C F I Q U t

Capacitance (F), not recommended: battery capacity (Ah) Faraday constant: 96485C/mol ¼ 26.8 Ah/mol Electric current (A) Electric charge (C ¼ A s), capacity (1 Ah ¼ 3600C) Voltage (V) Time (s)

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00056-6

465

466

1

Electrochemical Terminology | Capacity

Introduction

The term capacity describes the amount of electric charge that can be drawn from a battery until a limiting value, usually a voltage threshold, is reached. The value is given in ampere hours (Ah) instead of coulombs (C) or ampere seconds (As) for practical reasons. It is important to note that the capacity (electric charge) of a battery and the capacitance of a capacitor (in F ¼ As/V) are different quantities. The difference is due to the fact, that charge in capacitors is stored in an electric field. The term capacity if used as well for fuel cells and redox-flow cells with respect to the size and the total amount of charge that can be extracted from a full tank. The active materials are stored in a tank and pumped into the electrochemical cell for converting chemically stored energy into electrical energy more or less continuously. The duration of operation and the amount of charge that can be drawn from a fuel cell is therefore a function of the size of the storage tank.

1.1

Definition of battery capacity

The capacity Q (the symbol C is not recommendeda) of a cell is defined as the integral of the discharge current I over time, until the end of discharge voltage has been reached. The voltage threshold is usually defined as the value beyond which either the voltage during discharging begins to fall very rapidly, loads can no longer be operated, or severe aging processes may start to occur. The upper (Umax) and lower (Umin) voltage thresholds are set by the manufacturer for their specific product and may have application-specific values. ð Umin Q¼ − Iðt Þ dt U max

or, for a constant current discharge, Q ¼ I  t D ðfor U > Umin Þ where tD is the discharge time and I the discharge current. A minus sign is introduced because the discharge current is usually given as a negative value. For determining the capacity, only discharge currents may be used. Small cells for portable products have a capacity in the range of ca. 0.1 Ah (100 mAh) to ca. 3 Ah (3000 mAh) whereas the largest cells used in submarines have a capacity of a few thousand Ah. The capacity of a battery consisting of a number of cells connected in series is the same as the capacity of the individual cells when it can be assumed that all cells have an identical capacity. Otherwise, the weakest cell determines the maximum capacity which can be drawn from the system (without over-discharge). When capacitors are connected in series, the total capacitance, C ¼ Q/U, is less than any one of the series capacitors’ individual capacitances because the voltage increases proportional to the number of capacitors connected in series whereas the total electric charge which can be supplied to a load remains the same. C¼

1.2

Q ¼ U 1 + U 2 + . . . +U n C1 + 1

1 C2

1 + . . . + C1n

Mechanisms for storage of charge

In batteries, charge is not stored but supplied from two coupled electrochemical reactions which supply electrons on the negative electrode and consume electrons on the positive electrode during discharging. Current as a result of electrochemical reactions is sometimes referred to as Faraday current. The electrochemical reactions between the electrolyte and the electrodes lead to a spontaneous separation of charges as soon as electrodes and electrolyte are in contact with each other. There is a surplus of electrons on the negative electrode and a surplus of positive ions in a layer of the electrolyte on the interface between electrolyte and electrode, and a lack of electrons and a lack of positive ions on the positive electrode. The two adjacent layers of charge with opposite polarity are called a double layer. Charges which are removed from the double layer during discharging are immediately replaced by the electrochemical reaction.

2

Necessary differentiation of the term capacity

The definition of capacity needs to be differentiated into a number of concepts because the amount of charge that can be drawn from a battery depends on the operating conditions (e.g., constant or variable discharge current and temperature), lifetime expectations, age and its current state of charge, e.g., what amount of charge has been removed already. In addition, specific capacity (in Ah/kg) is an important value. a The International Union of Pure and Applied Chemistry IUPAC proposes to use the symbol Q for charge. In this article therefore capacity is denoted by Q. This reduces the confusion of capacity with capacitance and the unit Coulomb (C). However, in data sheets and many publications capacity is denoted as C, and current values are usually given as C-rate.

Electrochemical Terminology | Capacity 2.1

467

Theoretical capacity

The theoretical capacity of a battery can be calculated from the number of electrons which are transferred during reaction and the available amount of active materials, assuming 100% conversion of the active materials. Using lead acid batteries as an example because of their clearly defined discharge reactions: Positive electrode : PbO2 + H2 SO4 + 2H+ + 2e − Ð PbSO4 + 2H2 O Negative electrode : Pb + H2 SO4 Ð PbSO4 + 2H+ + 2e − Per conversion of one lead dioxide molecule, one lead atom and two molecules of sulfuric acid, two electrons become available, which can be used in an external load. One mole of active mass contains NA ¼ 6.021023 molecules and will therefore deliver 12.041023 electrons of charge. The number N of electrons per Coulomb (¼1 As) is 6.241018 (the elementary charge of an electron e is 1.60210−19 As). The ratio of number of molecules per mole to number of electrons per As is called Faraday’s constant F, the amount of charge (As) delivered per mole of reactands by: F ¼ e N A ¼ 96485 As=mol: The electric charge of 96,485C is equivalent to 1 mol electrons. For practical reasons, F is usually given in Ah and has the value of 26.8 Ah/mol. In electrochemical systems where one electron is transferred (NiCd, NiMH, Li-Ion, etc.), the theoretical capacity is therefore 26.8 Ah/mol, in electrochemical systems where two electrons are transferred, the theoretical capacity is 53.6 Ah/mol. In order to compare different electrochemical systems, the specific theoretical capacity (in Ah/kg) is used: Q PzF ¼ m Mi where m is the total mass, M the molar mass of reactand i, and z ¼ electrochemical valency (number exchanged electrons in the redox equation). The weight of all reactands of a lead acid battery—comprising 1 mol of lead (207.2 g/mol), 1 mol lead dioxide (239.2 g/mol) and 2 mol sulfuric acid (98.08 g/mol)—is 642.5 g and thus the theoretical specific capacity is 296,485C mol−1/ (0.6425 kg) ¼ 300,342C/kg ¼ 83.4 Ah/kg. Table 1 gives an overview of the specific theoretical capacity of some electrochemical systems. The data show that lithium-ion cells have by no means the highest specific theoretical capacity. However, as their voltage is high in comparison to most other systems, their specific energy is very high. Please note: For electrochemical systems where oxygen from the air is one of the reactants, the theoretical specific capacity is sometimes given without the weight of oxygen. Electrode materials under development such as silicon nanowires with a theoretical capacity of 4200 mAh/g and the polymer PEDOT (poly(3,4-ethylenedioxythiophene))1 with a specific capacity of 691 mAh/g which has been achieved experimentally are not yet stable for commercial application but show the potential of further developments.

Table 1

Specific theoretical capacity of selected electrochemical couples. Specific theoretical capacity (Ah/kg) for cell reaction

Secondary cells (rechargeable) Lead acid Ni–Cd Ni–H2 NiMH (dependent of the hydrogen storing alloys) Na–S Na–NiCl2 LiNiO2/graphite Primary cells (non rechargeable) Li/SO2 Zn/MnO2 (Leclanché) Ag/Zn Zn/Air Fuel cell data for comparison Hydrogen fuel cell (oxygen from the air is not considered) Hydrogen fuel cell including oxygen Data mostly from Linden D, Handbook of batteries, 2nd edn., McCraw Hill.

83 180 289 209 377 305 158 379 224 283 658 26,800 2978

468 2.2

Electrochemical Terminology | Capacity Nominal capacity

The nominal capacity (Symbol QN, not recommended: CN) of a cell is given by the manufacturer as part of the product specification. The nominal capacity is defined as the amount of charge which can be discharged from a fully charged cell under clearly defined constant current and temperature conditions, the so-called nominal conditions. The charging characteristics which leads to a fully charged battery is defined by the manufacturer, and the current and temperature to be used are often application specific and usually agreed upon by the industry. The index N denotes the hours required for discharging. There are two different ways to describe capacity and the nominal current associated with the capacity:

• •

For lithium-ion batteries the capacity value is given for a one-hour discharge time. The discharge current associated with a given discharge time is given as I ¼ QN/t (in A) and is referred to as C-rate. Example: A 20-Ah battery can be discharged at 100 A at the rate 5C for t ¼ (1/5) h, and at the rate 0.5C (10 A) for 2 h. “5C” is not a physical quantity, but a traditional designation of current amplitude in battery technology. For lead acid batteries and most other battery systems, the capacity has a strong dependence on the discharge time (see below: Peukert law). The capacity is therefore given as nominal capacity QN and product specifications give details of the capacity for different discharge times (Table 2). The nominal current IN associated with CN is defined as IN ¼ QN/tN.

The nominal capacity and discharge current are a design value of a cell. A newly delivered cell may have an initial capacity below or above the nominal capacity but after a few cycles, the capacity, measured by multiplying the nominal discharge current with the number of hours until the end of discharge voltage assigned to the discharge current is reached, is at or above the nominal capacity.

2.3

Operating capacity

In cases where the operating conditions differ greatly from the nominal conditions, the capacity for the temperature and average discharge current during operations must be known for planning reasons and use of the battery. For most cells, low temperatures and currents of a multiple of the nominal current will reduce the capacity which can be drawn from the cell considerably.

2.4

Useable capacity

Due to restrictions from the application, safety, or lifetime reserves the really useable capacity can be smaller than the actual available capacity. This means that, for example, the theoretically available capacity cannot be used because the application requires the observance of voltage limits (e.g., DC link of an inverter).

2.5

Actual capacity

The term “actual capacity” describes the amount of charge that can be drawn from a fully charged cell under nominal discharge conditions. During the lifetime of a battery, the actual capacity may increase at the beginning until it has a stable value before falling

Table 2

Overview of values used for the definition of nominal capacity in different batteries.

Lead acid, SLI application Lead acid, motive power application Lead acid, stationary applicationsa

Lithium-ion, portable communication equipment Lithium-Ion, hybrid vehicle application a

Nominal capacity (in Ah)

End of discharge voltage (per cell)

Nominal temperature

Capacity use considering lifetime aspects

Q20 ¼ 20 h  I20 Q5 ¼ 5 h  I5 Q10 ¼ 10 h  I10 Q5 ¼ 5 h  I5 Q1 ¼ 1 h  I1 Q1/6 ¼ 1/6 h  I1/6 Q1 ¼ 1 h  I1

1.75 V 1.7 V 1.8 V 1.79 V 1.74 V 1.6 V 2.5 V dependent on materials used 2.5 V dependent on materials used

25  C 30  C 20  C

80% (100–20% SOC) 80% (100–20% SOC) 80% (100–20% SOC)

20  C

100%

20  C

65% (90–25% SOC)

Q1 ¼ 1 h  I1

End of discharge voltage values are from the data sheet of BAE Berliner Akkumulatorenwerke.

Electrochemical Terminology | Capacity

469

(e.g., lead acid batteries) or may start to decline immediately after production or commissioning (e.g., lithium ion batteries). In many applications, the end of life is considered to be reached, once the actual capacity falls below a given percentage of the nominal capacity, usually 80% of nominal capacity especially in the mobile sector, for stationary uncritical applications sometimes 60%.

2.6

Residual capacity

The term residual capacity describes the amount of charge which can be drawn from a battery at any given time during operation. The residual capacity is for instance used to calculate the remaining run time of an autonomous energy system or the remaining mileage of a car. For practical reasons, the residual capacity is often calculated under the existing operating conditions (average discharge current, temperature) and measured using application specific currents and temperatures.

3

Cell capacity

The capacity (available electric charge) of a battery cell is considerably below its theoretical value because the assumption of discharging the active materials completely (100%) at a technologically useful voltage is unrealistic and damages the cell more than the gain in energy would justify. The main reason for the difference between cell capacity and theoretical capacity is the limited utilization of active materials. The following overview describes a number of factors for limited mass utilization which apply to most electrochemical systems, however, in differing degrees. Examples are given for lead acid and lithium-ion batteries. In addition, the specific capacity of a cell includes a number of passive components of a cell. As a general rule, the capacity of a cell is in the range of ca. 20–40% of its theoretical value. The quantity of active material which electrodes contain per Ah of capacity (g/Ah) is an important value for comparisons. If some cells are connected in series or in parallel to battery systems, both the specific capacity and energy will be significantly reduced further.

3.1

Electrical conductivity of active material (lead acid batteries)

The active materials of a cell must be conducting at all times during charging and discharging. As the chemical composition and crystal structure of the charged and discharged active materials are different, changes in their electrical conductivity can be expected. This is particularly pronounced in lead acid batteries. The charged active materials lead and lead dioxide have a conductivity of ca. 5.3105 S/m and 1.35104 S/m. In contrast, the conductivity of lead sulfate, the discharged active material has a conductivity of ca. 10−6 S/m. As the active material is discharged, more and more lead sulfate accumulates and decreases the conductivity of the electrode. The conductivity of a mixture of two materials with different conductivities can be described, for instance, by the effective medium theory. It predicts that once approx. 50% of conducting material has been replaced by non-conducting material (depending on the porosity of the electrode), the electrodes become non-conductive and it is no longer possible to discharge the remaining charged active material at a useful discharge rate. Fig. 1 shows calculations for the conductivity of lead dioxide electrodes. The change of conductivity in electrode materials of other battery systems is smaller and sometimes of little or no effect, e.g., in lithium-ion batteries.

Fig. 1 Conductivity of lead dioxide electrode calculated using the effective medium theory. Courtesy Dirk Uwe Sauer (From Sauer DU, Optimierung des Einsatzes von Blei-Säure-Akkumulatoren in Photovoltaik-Hybrid-Systemen unter spezieller Berücksichtigung der Batteriealterung, PhD Thesis, University of Ulm, 2000).

470 3.2

Electrochemical Terminology | Capacity Consumption of electrolyte

In a number of electrochemical systems (lead-acid, NiMH) with aqueous electrolytes, the electrolyte is consumed and its concentration and conductivity changes during discharging. A surplus of electrolyte has to be available in the cell. At low temperatures and high discharge currents local depletion of electrolyte respectively ions from the electrolyte can occur and limit the capacity of a cell.

3.3

Irreversible changes of active material and aging processes caused by deep discharging

A very high rate of active mass utilization leads to a reduction in the number of cycles and total energy throughput of the cell during its lifetime in virtually all electrochemical systems. The conversion processes from charged to discharged active material or the removal of ions from their crystal lattice site during discharging and the subsequent reversals of these processes during charging are in most electrochemical systems processes which lead to volume changes and mechanical stress in the active materials of the electrodes. In addition, the voltage at the end of discharging may lead to corrosion or dissolution effects in passive components of the cell and other undesirable side reactions. The capacity of a cell, from its fully charged state to its fully discharged state, is therefore often not available if a very long lifetime is required and the useable capacity of a battery considering lifetime requirements is lower than the nominal capacity. For many lead acid battery applications, the fully charged battery can only be discharged to 20% SOC and the capacity is limited to 80% of the capacity under nominal conditions. As only ca. 50% of the active material can be used before the conductivity falls below acceptable values, only ca. 35–40% of the active material of a lead acid battery can be discharged. For some Li-Ion battery applications, the battery capacity is also only used to 70% (because intercalation and deintercalation in the electrodes of lithium is not 100%, and charging and discharging of often limited to between 90% and 10% SoC for lifetime reasons). and for NiMH batteries for hybrid vehicle applications the capacity is maintained between ca. 75% and 55% SOC during most of the operation. Ideally, the capacity of a cell or battery therefore should refer to the capacity that can be removed without shortening its lifetime unduly. Table 2 gives some examples of how lifetime considerations reduce the nominal capacity of a cell. The mass utilization of a cell is therefore always a trade off against lifetime and the amount of active material per Ah capacity is, therefore, an important design value with an impact on the number of cycles until the end of lifetime is reached.

3.4

Rechargeability

The need for easy rechargeability of the cell after a complete discharge may also limit the utilization of active mass. The individual effects are often linked to those which limit the use of the capacity for lifetime reasons, i.e., delithiation of active mass in lithium-ion batteries or the formation of large, non-conductive lead sulfate crystals in lead acid batteries.

3.5

Current inhomogeneity

The current homogeneity in a battery depends on the conductivity of the materials used and the design, particularly in cells with thick electrodes and large dimensions. Under high-rate discharge conditions the current inhomogeneity increases. As a consequence, the local utilization of the active materials is reduced. This effect is particularly pronounced if the specific conductivity of the various cell components (active and passive materials) differ greatly. Both measurements and simulations in lead acid batteries show that the discharge occurs mainly on the outside of the electrode making the discharge of remaining active material difficult or impossible within the voltage constraints that have been set. Even in systems with very thin electrodes, e.g., lithium-ion-systems, this effect is not negligible. According to the requirements for rather high-energy cells, the electrodes are thicker and less porous than cells for high-power applications with thin electrodes, high porosity and thicker current collectors. In addition, the thickness and arrangement of the current collectors have a decisive influence on the performance of the cell, which is why high-power cells often have current collectors on opposite sides. This ensures better current homogeneity.

3.6

Diffusion processes

In electrochemical systems, mass transfer phenomena often limit the availability of reactants. Diffusion processes can cause a pronounced decrease of the available capacity at high discharge rates and/or low temperature. In NiMH batteries, hydrogen cannot reach the electrode surface sufficiently fast at very low temperatures and therefore the capacity at high rates is diminished considerably. In lead acid batteries, diffusion of the electrolyte from the space between the electrodes into the pores of the electrodes limits the discharge at very high discharge currents and low temperatures. It is well-known that in LIB, the diffusion processes of lithium ions in the graphite lattice, which are limited at low temperatures, reduce the charging current. Even if this happens indirectly, it also affects the discharge capacity of a lithium-ion battery. Diffusion processes are the main reason why an interruption of a discharge can lead to an increase of capacity. During the interruption, diffusion and mass transport phenomena influence the availability of reactants at the electrode/electrolyte interface and the discharge voltage increases again. Such an artificial or forced increase of capacity may, however, induce more irreversible changes than a normal discharge and reduces lifetime.

Electrochemical Terminology | Capacity

4

471

Capacity of cells connected in series

In principle, cell capacity does not change by connecting them in series. However, cells even when new have slight variations in capacity due to variations in production. Over the lifetime of the battery these differences tend to become larger. The capacity of a battery consisting of cells in series will then be limited by the cell with the lowest capacity. For some electrochemical power sources, e.g., lithium-ion batteries, it is of great importance to end the discharge of the battery when the capacity of the weakest cell has been spent and its voltage reaches its lowest acceptable value. Safety devices that limit the discharge of all cells once the weakest cell reaches its end of discharge voltage are an integral part of the battery in such cases. In other electrochemical systems, e.g., lead-acid, nickel-cadmium, or nickel metal hydride, the damage to the overall battery remains limited if the discharge is not ended. Special safety precautions are not usually taken in such systems. However, it is still advisable to limit the discharge of such battery once the voltage of an individual cell has fallen below a limiting voltage threshold.

5

Dependence of capacity on design and operating conditions, temperature and rate of discharge

All of the effects described above together lead to a significant difference between the theoretical cell capacity per mole of active material, the useable capacity of the cell and the capacity of a battery, and to a dependence of the available capacity on design, discharge rate, temperature and application. Design aspects such as the thickness of electrodes, the resistance of the passive components, homogeneous current distribution and mass utilization, and heat dissipation are product specific features and cannot be dealt with here in detail.

5.1

Effect of high discharge current

An empirical formula, the Peukert law, is often used to estimate the capacity of cells with a large dependence of their capacity on the discharge rate. Peukert’s law combines all the effects which reduce capacity with increasing discharge current in one formula.  n I t In  t ¼ constant , 1 ¼ 2 I2 t1 where I is a constant discharge current and t the discharge time associated with that current. n is an empirical value. Typical values of the Peukert coefficient are in the range of 1.05 for lithium-ion batteries and 1.1 to 1.3 for lead acid batteries. Peukert’s law is highly useful when a rough estimate of the capacity at a certain discharge rate is required, but not very accurate and measured data should be used when sizing batteries for a specific application.

5.2

Effect of temperature on capacity

The capacity of a battery depends also strongly on temperature. The rate of the chemical reactions taking place in a cell, the equilibrium voltage and the resistance of the active and passive components in the cell are all temperature dependent and the end of discharge voltage is reached earlier at low temperatures. The most important effects are however those linked to mass transport phenomena and diffusion. The limiting factor at low temperatures is different for different electrochemical systems:

▪ Lead acid batteries: Electrolyte diffusion and conductivity, ice formation at low temperatures and low state-of-charge ▪ NiMH: hydrogen diffusion in the negative electrode ▪ Lithium batteries: Electrolyte resistance and mass transport in the active material, esp. lithium-plating conditions. NiCd-batteries have the best low temperature performance of all commercially successful battery systems and are only affected little by the above factors. An increase of temperature above the nominal temperature does not lead to a significant further increase in capacity because the capacity is then limited by non-temperature dependent factors, such as the availability of active materials. Increasing the temperature only causes a significant increase of corrosion and in some cases safety risks. An example of capacity versus discharge current and temperature is given in Fig. 2 for lead acid batteries. Over a wide range of temperature, a constant percentage loss of capacity per degree can be used to estimate the capacity at lower temperature. When more detailed calculations are required, it is important to measure the capacity versus temperature curve for the range of discharge currents which are of interest.

472

Electrochemical Terminology | Capacity

Fig. 2 Dependence of capacity on temperature and discharge rate for a lead acid battery.

5.3

Effects of previous use on capacity

In addition to temperature and discharge rate, the capacity also depends on the microstructure of the active material and passivating layers which restrict the current flow. The microstructure of the active material which exist at full charge prior to discharging is the result of the history of use, particularly the last discharge, the last charge and the conditions at float charge. The microstructure changes have a reversible aspect and an irreversible aspect which contributes to aging and loss of performance. The irreversible changes are part of a discussion of aging processes. However, the reversible changes have to be considered when discussing the effect of operating conditions on capacity. The effect of the previous discharge on the capacity of lead acid batteries has been reported more than 100 years ago and was investigated in more detail by Hullmeine et al.2 leading to the concept of reversible insufficient mass utilization (RIMU). The widely known “memory effect” of certain types of nickel cadmium batteries is another example of the effect of previous use on the capacity. It is not clear if comparable effects are relevant for other battery systems.

6

Conclusion

The capacity (available electric charge) of a cell and battery, its ability to provide electrical current to a load, is one of the most important characteristics of a battery. However, it is important to stress that the amount of charge that can be provided depends on many different parameters, e.g., current amplitude, temperature, state of charge and age. The capacity of a cell even when new and fully charged is considerably below its theoretical value.

References 1. Zhan, L.; Song, Z.; Zhan, J.; Tang, J.; Zhan, H.; Zhou, Y.; Zhan, C. PEDOT: Cathode Active Material With High Specific Capacity in Novel Electrolyte System. Electrochim. Acta 2008, 53, 8319–8323. 2. Hullmeine, U.; Winsel, A.; Voss, E. Effect of Previous Charging and Discharging History on the Capacity of the PbO2/PbSO4 Electrode: The Hysteresis or Memory Effect. J. Power Sources 1988, 25, 27–47.

Electrochemical Terminology | Capacitance H Wenzla, R Bengerb, and I Hauerc, aConsulting for Batteries and Power Engineering, Osterode, Germany; bResearch Center Energy Storage Technologies, Clausthal University of Technology, Clausthal-Zellerfeld, Lower Saxony, Germany; cChair of Electrical Energy Storage Technology, Institute of Electrical Power Engineering, Clausthal University of Technology, Clausthal-Zellerfeld, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 2.1 2.2 2.3 3 3.1 4 5 6 References

Introduction Differences between conventional capacitors and supercapacitors Capacitance of capacitors connected in series Energy content of capacitors and supercapacitors Equivalent circuit diagram of capacitors and supercapacitors Capacitance of electrochemical double layer capacitors Effect of porous electrode structure on charging and discharging behavior Materials for supercapacitors and batteries Capacitance of batteries Conclusion

473 474 474 474 474 475 475 475 476 478 478

Abstract This article describes the term capacitance, the ability to store electrical charge directly in an electric field and not, as is the case in batteries, indirectly in chemical form which provides charge as a result of electrochemical reactions. As a result, capacitance has the unit Farad (As/V) in contrast to batteries where the capacity has the unit Ah. The discussion here focusses on electrochemical double layer capacitors (EDLC) commonly referred to as supercapacitors, in particular their pseudocapacitance, but also includes a discussion of the double layer capacitance properties of batteries.

Key points

• •

1

Definition of capacitance and pseudocapacitance Correct usage of symbols and units related to capacitance

Introduction

Capacitance describes the ratio of charge (in Coulomb ¼ Ampere second) which can be stored in the electric field between two electrodes in relation to the voltage difference between the electrodes. Capacitance and the unit Farad (F ¼ As/V) must not be confused with capacity (in Ah) used for batteries and Faraday’s constant (As/mol). The energy content which can be stored in a capacitor can be calculated using standard electrical engineering principles as 1 W ¼ CU 2 2 where C is the capacitance and U the nominal voltage of the capacitor. For very large supercapacitors, the energy content is in the range of small battery cells. However, the power output of capacitors is much higher compared to battery cells of similar energy content, and the application of capacitors therefore focusses on application where high power at a low weight and size are more important than energy content. Capacitors are usually only discharged to 50% of the initial voltage to limit the input voltage range for loads and thus only 75% of the energy content can be used practically. In a more generalized form, the energy content of a capacitor therefore is:  1
W ¼ C U 20 − U 21 2 with U0 as the voltage at the beginning of discharge and U1 as the voltage at the end of discharging. When charging a capacitor at constant current, the voltage curve is proportional to time as the amount of charge stored is directly proportional to the voltage. This article focusses on the capacitance of supercapacitors also called electrochemical double layer capacitors which have some similarities to conventional capacitors but are able to store considerably more charge albeit at comparatively low voltages. The main process for storing such large amount of charge is electrochemical in nature and differs from the storage of charge in an electric field

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00121-3

473

474

Electrochemical Terminology | Capacitance

in conventional capacitors. Additionally, in this article, the capacitive properties of batteries are discussed which are important under certain conditions. The information given in this article is restricted to the concept of capacitance and their underlying processes.

2

Differences between conventional capacitors and supercapacitors

It is worthwhile to point out similarities and differences between the capacitance of conventional capacitors and supercapacitors. The energy of conventional capacitors is stored exclusively in an electric field between two electrodes by applying an external voltage source. In conventional capacitors, the electric field strength depends on the dielectric material between the electrodes and can be very high. The dielectric is always an insulating layer between the electrodes. Electrolytic aluminum-oxide capacitors with a rated voltage of 6000 V are commercially available.1 In supercapacitors, the electric field is the result of a charge separation between the electrodes and ions in the electrolytic double-layer, which, however, is conductive due to electrochemical processes. As a result of the direct contact between the electrode material and the ions in the electrolyte, the nominal voltage of supercapacitors is very small compared to conventional capacitors, usually 2.7 V in organic electrolytes. The nanostructure of the double layer leads to a significantly higher charge density on the electrodes and therefore a much higher capacitance by a factor of about 104 between the largest commercially available capacitors and supercapacitors. The capacitance of an ideal plate capacitor can easily be calculated as area of the electrodes A multiplied by the permittivity er of the electrically non conducting material (dielectric) between the electrodes divided by the distance d between the electrodes. C¼

e0 er A d

The effective surface area of the electrodes is therefore important for both conventional capacitors and supercapacitors.

2.1

Capacitance of capacitors connected in series

When n individual capacitors (either conventional capacitors or supercapacitors) of the same capacitance Ci are connected in series, their capacitance decreases according to standard electrical engineering principles according to: 1 X C C ¼ i ¼ n C i¼1 i n

It must be noted that, when discharging a string of capacitors connected in series, the available capacitance can only be used if all capacitors have very similar capacitances in order to avoid polarity reversal and damage to the smallest capacitors. This is in contrast to batteries, where the capacity (electric charge) of a string of cells is identical to the capacity of the individual cells respectively the capacity of the cell with the lowest capacity.

2.2

Energy content of capacitors and supercapacitors

Conventional capacitors used for electronics and electrical engineering have a capacitance in the range of pF (10−9 F) up to a few hundred mF and maximum operating voltages as high as a few thousand Volts.1 Despite high operating voltage, the energy content is usually much below about 0.01 Wh, a fraction of the energy of even the smallest batteries used in consumer electronics. When comparing the energy content of conventional capacitors and supercapacitors, the voltage of the capacitor respectively the series connection of supercapacitors need to be considered. For stabilizing the DC-voltage of an DC/AC inverter (e.g., 350 V), one conventional capacitor is sufficient whereas about 130 supercapacitors with a nominal voltage of 2.7 V each would be required. One of the largest supercapacitor which is commercially available, 3 V, 3600 F, has an energy content of used of 3.375 Wh when discharged to 50% of its nominal voltage. Conventional capacitors and supercapacitors do not compete in any given application as conventional capacitors have a dominant advantage whenever their energy content is sufficient for the application. Supercapacitors in contrast have a dominant advantage when a higher energy content is required. Depending on application specific requirements lithium-ion batteries compete with supercapacitors.

2.3

Equivalent circuit diagram of capacitors and supercapacitors

Fig. 1 shows the equivalent circuit diagram which is generally used both for conventional capacitors and supercapacitors: an inductance, a resistor and a capacitor in series with an additional resistor parallel to the capacitor to represent self-discharge processes.

Electrochemical Terminology | Capacitance

475

Rself discharge

RSeries Fig. 1 Equivalent circuit diagram of a capacitor or supercapacitor.

3

Capacitance of electrochemical double layer capacitors

In supercapacitors, the externally applied voltage creates an electric field between each electron conducting electrode and the ions of an electrolyte align themselves according to the potential. As there is no electrically insulating layer on the electrode surfaces, ions of the electrolyte can be adsorbed directly without their solvent shell on the electrode surface and considerably greater amounts of charge can be stored depending on the size of the ions and their availability in the solvent and the properties of the electrode surface. The two charged layers, a layer of electrons in the electrode surface and a layer of positively charged ions in the electrolyte (negative electrode) and a lack of electrons in the electrode and a surplus of negatively charged ions in the electrolyte (positive electrode) are called a double layer,2 and are separated by a distance similar to the size of solvent molecules and desolvated ions, e.g., 0.3–1 nm. The selection of materials is crucial and the maximum applied voltage is limited by electrolysis of the electrolyte—in commercial supercapacitors usually to 2.7 V. The amount of charge which can be stored depends on the surface area of the electrodes. A preferred material is activated highly porous activated carbon with an effective surface area of up to 1000 m2/g.3 The process described so far is purely electrostatic in nature. However, to maximize the storage of charges on the electrode surface, the electrode materials are treated with surface active materials which can transfer charges between the electron conducting electrode material and the ions of the double layer which is a fully reversible electrochemical charge transfer reaction. This process is faradaic in nature as there is a physical movement respectively flow of charges albeit only within the dimensions of the double layer. The process of transferring charges from the electrode (electrons) to the surface active layer and ions of the electrolyte was described by Conway4 and termed pseudocapacitance. There is no formation of chemical bonds and thus no aging processes caused by conversion processes must be expected. As a result, supercapacitors usually have an extremely high cycle life in the range of 106 cycles.a Pseudocapacitance can increase the capacitance by a factor of 100 in relation to the storage of charges only in the electric field between the electrodes in the double layers. There is a whole range of materials with which the surface of carbon electrodes can be doped (e.g., metal oxides) or coated (e.g., conducting polymers such as polyaniline). As a result, the capacitance of supercapacitors is a complex function of the activation process to create an extremely large effective surface while maintaining good electronic conductivity, the choice of surface active materials, the production processes, and the choice of electrolyte.

3.1

Effect of porous electrode structure on charging and discharging behavior

The equivalent circuit diagram shown in Fig. 1 does not explain the relaxation processes as a result of the porous electrode structure. In order to create a double layer in the pores of the electrode material, the pores must be filled with electrolyte. When charging a supercapacitor, the maximum charging voltage is reached relatively quickly. However, it takes some time for the surface of inner pores to develop a double layer with the maximum charge density that can be created based on the electrolyte properties and surface active materials. Therefore, once the end of charge voltage is reached, current will continue to flow if the voltage is maintained or, if there is no more current flow, the voltage will decrease because of redistribution processes of electrolyte ions in the pores. The equivalent circuit diagram to represent this behavior is a “transmission line” equivalent circuit.5 Fig. 2 shows the equivalent circuit diagram for the charging and discharging process in the pores of an electrode. When analyzing the voltage behavior of a supercapacitor it is therefore important to use both equivalent circuit diagrams. To achieve a full charge of a supercapacitor requires some time.

4

Materials for supercapacitors and batteries

Materials for supercapacitors which lead to the existence of pseudocapacitance are characterized by a chemical affinity with ions of the electrolyte. Desolvated ions are adsorbed at the surface and then a charge transfer process occurs in the double-layer. The process a It is worth noting that lithium ion cells of similar weight and size to supercapacitors also have a very high cycle lifetime when cycles of the same energy content are used—from a systems point of view with the advantage of only small changes of voltage.

476

Electrochemical Terminology | Capacitance

Bulk electrolyte between electrodes Electrolyte in pores of electrode Fig. 2 Equivalent circuit diagram of the charging process of a supercapacitor in the pores of the electrode. The individual capacitor elements are separated by resistors which represent the ohmic voltage drop along the electrolyte during charging or discharging.

is called pseudocapacitance because the electrochemical signature, cyclic voltammetry (CV) and galvanostatic charge discharge curves (GCD), are those of capacitors. In cyclic voltammetry, the current flowing through a sample during a linear increase of voltage is measured. In a galvanostatic charge discharge test, the voltage is measured while there is a constant current flowing through the sample. Fig. 3. shows schematically the electrochemical signature of conventional capacitors, supercapacitors and batteries.6 Fig. 3 implies a clear distinction between materials used for batteries and supercapacitors. However, it has been shown that this distinction becomes blurred for certain materials. LiCoO2, for instance, is a well-known battery material but begins to exhibit pseudocapacitance like properties when its crystal size is reduced, e.g., to 6 nm.6 As with many nanomaterials, the surface properties of the material and not the bulk properties of the material begin to dominate. Nano sized LiCoO2 is believed to exhibit pseudocapacitance properties because the surface offers a significant number of ion adsorption sites for Li-ions containing electrolytes and the diffusion lengths are greatly reduced. Similar properties have been identified for other materials as well. It can therefore be expected that further developments in material science will increase the capacitance of supercapacitors further and extend their range of applications.

5

Capacitance of batteries

The electrochemical reaction between the active electrode material and the ions in the electrolyte leads to a charge separation on the electrode-electrolyte surface without applying an external voltage. A layer of positive ions and a surplus of electrons on the negative electrode and a layer of surplus negative ions and a lack of electrons on the positive electrode are created as soon as electrolyte and electrode materials are in contact with each other. These layers of opposite charged particles are usually called a double layer capacitor and its structure is referred to as Helmholtz layer as for supercapacitors.3 The capacitance of the double layer has to be considered when characterizing a battery using electrochemical impedance spectroscopy or when analyzing the voltage behavior after a fast current change, i.e., when switching high power loads on or off. During operation, however, the amount of energy supplied by small or relatively low changes in voltage can be ignored as the

Fig. 3 Electrochemical signature of ideal double-layer capacitors, supercapacitors with pseudocapacitance, and batteries: (a) cyclic voltammetry, (b) galvanostatic charge discharge curves, and (c) voltage change as a function of current. Modified after Jiang, Y.; Liu, J. Energy Environ. Mater. 2, 2019, 30–37.

Electrochemical Terminology | Capacitance

477

Fig. 4 Voltage and current of a 65 Ah/12 V lead acid starter battery when discharging the battery with a current of 50 mA above the thermodynamic equilibrium voltage of the battery between 13.5 V (2.25 V/cell) and 13.0 V (ca. 2.17 V/cell).

electrochemical reactions provide or take up most of the energy. However, the discharge behavior of cells and batteries above their thermodynamic equilibrium voltage can only be explained when the capacitance of the double layer capacitor is taken into account. The following discussion is based on lead-acid batteries because the thermodynamic equilibrium voltage of the system is precisely known. The concept applies equally to all other battery systems with porous electrode structure and after a prolonged period of charging at a voltage above the thermodynamic equilibrium voltage, i.e., also for lithium ion batteries. Fig. 4 shows the voltage behavior of a 12 V/65 Ah lead-acid battery after a long period of overcharging. Once the charging current is stopped, the voltage falls and after many hours, the thermodynamic equilibrium voltage of the battery is reached. The thermodynamic equilibrium voltage of the battery can be calculated exactly from the known concentration of the diluted sulfuric acid assuming that the long-time of overcharging has led to a chemically full charge (no more lead sulfate contained in the active materials, homogeneous acid concentration throughout the cell). At an acid concentration of 1.28 g/cm3 the equilibrium voltage of a cell is 2.13 V per cell, and 12.78 V for a 12 V battery. Above this equilibrium voltage, it is not possible to discharge the active materials of the battery. However, the battery can sustain a discharge current of 50 mA for minutes at a voltage above 13 V and deliver 0.6 Wh of energy. The origin of this energy cannot be an electrochemical discharge reaction and must come from another energy storage property of the battery. The only possible explanation is the discharge of the capacitor formed by the double layer capacitor on the negative and positive electrodes of the cells and their voltage value as a result of an externally applied charging voltage. The voltage during this discharge process above the thermodynamic equilibrium voltage falls proportional to the discharge time respectively the amount of charge which is removed. This is of course the property of a capacitor. The capacitor is characterized by electrodes which

• • • •

have a highly porous structure, can form a double layer without electrochemical reactions as the active materials of the electrodes are fully charged and are above the thermodynamic equilibrium potential, are electrically connected via an ion conducting electrolyte, and do not have an electrically insulating layer which is in contrast to conventional electrolytic capacitors.

The discharge current between 13.5 and 13.0 V is provided by the double layer capacitor, and when the discharging current stops, there is an equalization process of the double layer in the electrodes. From the voltage difference and the discharge current, the magnitude of the capacitance can be estimated to be about 1000 F per cell. These are obvious similarities to the processes which lead to the energy storage in supercapacitors. Fig. 5 is an equivalent circuit diagram of the porous electrode of a cell in analogy to Fig. 2. The only electrochemical reaction which can take place above the thermodynamic equilibrium voltage of the active materials is the hydrolysis of water. This reaction consumes energy and it is thus a self-discharge reaction. The hydrolysis reaction is driven

• • •

during overcharging by the current from a charger, once the external current has been switched off, by the double layer capacitor which was originally built up by the electrochemical reaction of the main reaction (conversion of lead sulfate to lead or lead dioxide) and subsequently maintained by continuously applying a charging voltage despite fully charged electrodes, and once the voltage has fallen below the thermodynamic equilibrium voltage, by discharging the active material of the cell.

478

Electrochemical Terminology | Capacitance

Bulk electrolyte between electrodes

Electrolyte in pores of electrode Fig. 5 Equivalent circuit diagram of the electrode of a cell in analogy to the equivalent circuit diagram for charging and discharging the pores of a supercapacitor.

In comparison to the equivalent circuit diagram of a supercapacitor, it is therefore necessary to add for each capacitor element representing pores the ideal voltage source for the hydrolysis of water and the resistor to represent its charge transfer overpotential. During charging, current flows through the main reaction paths of the electrodes and charges the double layer at the electrodes, and the ‘supercapacitor’ characteristics of the cell. The more the cell becomes fully charged, the lower becomes the main reaction current and, once fully charged at constant voltage, the current only drives the side reactions and maintains the capacitance of the double layer capacitor. When the charging voltage is terminated, the capacitance of the supercapacitor drives the hydrolysis of water and the voltage decreases. The practical relevance of the capacitance which exists in electrochemical cells may be small as the energy content is small compared to the capacity of a battery. However, there are a number of applications, where batteries are operated under float charging conditions above their thermodynamic equilibrium voltage and due to small changes in the loads or properties of chargers small discharge currents occur for a short time without the voltage falling below the thermodynamic equilibrium voltage. These are not currents associated with conversion of active material (no Faradaic current), and the aging effects if caused at all are different from aging processes associated with cycling and active mass conversion or intercalation processes.

6

Conclusion

Capacitance, the storage of charge in the electric field between electrodes of conventional capacitors and double layer capacitors (supercapacitors), is characterized by a linear relationship between voltage and the amount of charge which is stored (C ¼ Q/U). Electrochemical double layer capacitors have a much higher capability to store charges because the electric field is created directly between the electrodes and ions of the electrolyte without their solvent shell which reduces their size. However, the maximum voltage is in comparison very small, i.e., usually less than 3 V. If the surface of electrodes is suitably treated, electrons from the electrode can be directly transferred to ions of the electrolyte in an electrochemical reaction which is fully reversible and causes no chemical bonds. In theory, the electrochemical reactions therefore do not lead to any aging processes. This property is called pseudocapacitance and its magnitude depends on the materials used to activate the electrodes, the electrolyte, the design and production processes. Pseudocapacitance can be the dominant factor compared to the storage of charges in the electric field by a factor of about 100. Double layer capacitance effects can also be observed in batteries and have to be considered when discussing the charge and discharge properties of a battery above their thermodynamic equilibrium voltage.

References 1. Maxwell Inc (n.d.) Data sheet of Maxwell Inc.: 3V-3400F-Cell-BCAP3400-P300. 2. Sabo, L.; Jacob, T. Grundlagen Elektrochemischer Doppelschichten. Bunsen-Magazin 2014, 16, 260–266. 3. Bharti, A.; Kumar, G.; Ahmed, M.; Gupta, P.; Bocchetta, R.; Adalati, R.; Chandra, Y. Kumar: Theories and Models of Supercapacitors with Recent Advancements: Impact and Interpretations. Nano Express 2021, 2. https://doi.org/10.1088/2632-959X/abf8c2. 4. Conway, B. E.; Pell, W.; Liu, T. Diagnostic Analyses for Mechanisms of Self-Discharge of Electrochemical Capacitors and Batteries. J. Power Sources 1997, 65, 53–59. 5. Pean, C.; Rotenberg, B.; Simon, P.; Salanne, M. Multi-Scale Modelling of Supercapacitors: From Molecular Simulations to a Transmission Line Model. J. Power Sources 2016, 326, 680–685. 6. Jiang, Y.; Liu, J. Definitions of Pseudocapacitive Materials: A Brief Review. Energy Environ. Mater. 2019, 2, 30–37.

Electrochemical Terminology | Energy H Wenzla, R Bengerb, and I Hauerc, aConsulting for Batteries and Power Engineering, Osterode, Germany; bResearch Center Energy Storage Technologies, Clausthal University of Technology, Clausthal-Zellerfeld, Lower Saxony, Germany; cChair of Electrical Energy Storage Technology, Institute of Electrical Power Engineering, Clausthal Universtiy of Technology, Clausthal-Zellerfeld, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of H. Wenzl, BATTERIES | Energy, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 401–406, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00044-7.

1 2 2.1 2.2 2.3 2.4 2.5 2.6 3 4 5 6 References

Introduction Thermodynamics of batteries, fuel cells, and redox flow systems Calculation of equilibrium potentials Theoretical energy of cells Energy and specific energy of a cell Mass utilization Voltage und energy content Dependence of energy content on operating conditions and design Supercapacitors Specific energy of cells connected in series Influence of application on energy content and specific energy Conclusion

480 480 480 481 481 482 482 483 484 484 486 486 486

Abstract This article describes the energy content of batteries and supercapacitors and the factors that determine the discrepancy between the theoretical energy content to the actual and useable energy content. The concepts of specific energy and energy density are defined, and values for different types of systems are given.

Key points

• • •

Overview of electrochemical terminology Definition of the terms related to energy Correct usage of symbols and units

Nomenclature

E E0 F G H I m P Q R R S T t U W z h

Electrode potential (V) Electrode potential at standard conditions (V) Faraday constant: 96485C/mol ¼ 26.8 Ah/mol Gibbs free enthalpy (J mol−1) Enthalpy of the cell reaction (J mol−1) Electric current (A) Mass (kg) Electric power (W) Electric charge (C ¼ A s) Molar gas constant: 8.3144 J mol−1 K−1 Resistance (O) Entropy of the cell reaction (J K−1) Thermodynamic temperature (K) Time (s) Cell voltage (V) Electric energy (J) Electrochemical valency Overpotential (V)

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00068-2

479

480

1

Electrochemical Terminology | Energy

Introduction

The term energy of a battery, redox-flow battery and supercapacitor only refers to the energy available during discharging and not to the energy required for charging. The energy in watt-hors (Wh) describes the electrical energy which is taken up by an external load and does not include internal losses in the battery during discharging, e.g. in the form of heat. For fuel cells, the term energy, if used at all, might refer to the total energy which can be supplied from a hydrogen tank. In contrast to capacity (useful electric charge), there is no equivalent differentiation of the term energy despite the fact that for most applications the energy that can be drawn from the battery is much more important than the amount of charge. Equally, the most relevant concept for controlling the operation of a battery is the state of charge (SOC), and the ‘state of energy’ is only a seldomly used concept despite the fact that the energy and/or power that can be delivered is often much more important, e.g. in grid applications.

2 2.1

Thermodynamics of batteries, fuel cells, and redox flow systems Calculation of equilibrium potentials

The useful energy (in kJ/mol) of all chemical and electrochemical reactions is given by DG ¼ DG − TDS where DH is the total enthalpy of reaction (including effects of pressure and volume changes), DG the Gibbs free enthalpy, T the absolute temperature (in K) and DS the change of entropy. The electrical energy which an electrochemical reaction can supply corresponds to Gibbs free energy DG. The changes in entropy DS cause heating or cooling of the system, depending on the direction of the reaction and is referred to as reversible heat. In some batteries, most notably Ni/Cd batteries, a reduction of temperature can readily be observed during charging and is sometimes used for controlling the charging process. For most other battery systems, however, the effect of reversible heat can be neglected in operation in most cases. For reactions in aqueous solutions values of DH and DG at standard conditions (1013 mbar, 25  C ¼ 298 K) are available in tables for most substances, so that the energy of reaction can easily be calculated (see Table 1). For non-aqueous solutions, e.g. all reactions in lithium-ion batteries, values are often not available. The thermodynamic values, DH, DG and DS refer to a system in equilibrium. Assuming that the total amount of active material in a cell can be converted under equilibrium conditions, then DG is the theoretical electrical energy which the system can supply. Using DG, the equilibrium voltage of an electrochemical system can be calculated using DE ¼ −

DG ¼ Ecathode − Eanode zF

Table 1 Overview of the theoretical specific energy of some important electrochemical power sources by using the nominal voltage of the electrochemical systems.

Secondary cells (rechargeable) Lead-acid LiNiO2/graphite FePO4/graphite CoO2/LTO FePO4/LTO Na/S Na/NiCl2 Ni/Cd Ni/H2 Ni/MH Primary cells (non rechargeable) Li/SO2 Zn/MnO2 (Leclanché) Ag/Zn Zn/air

Theoretical capacity Q (Ah/kg) for cell reaction

Nominal voltage U (V)

Theoretical specific energy W ¼ QU (Wh/kg)

83.5 158 117 151 113 377 305 180 289 209

2 3.6 3.2 2.5 2.1 2 2.5 1.2 1.2 1.2

167 568.8 373 376 236 754 762.5 216 346.8 250.8

379 224 283 658

3.1 1.2 1.85 1.2

1174.9 268.8 523.55 789.6

The data have been compiled from data sheets and various other sources, most notably the handbook of batteries by David Linden.1

Electrochemical Terminology | Energy

481

where E is the electrode potential, if only one electrode reaction is considered, and the cell voltage DE, if both electrodes of a cell are considered, both under standard conditions. z is the number of electrons which are transferred per reaction, and F is Faraday’s constant (96,485 As/mol). E and DE were formerly referred to as the electromotive force (EMF). This term should no longer be used. Please note, that the open circuit voltage (OCV) is only identical to the equilibrium voltage E once all diffusion and relaxation processes have ended and if there are no side reactions, i.e. no other electrochemical reactions parallel to the main cell reaction. The above values for the thermodynamic data are given for one molar concentration (aqueous solutions), a pressure of 1013 mbar and a temperature of 293 K. In order to calculate the thermodynamic values for other concentrations of solutions, the Nernst equation is used. It is here given here for calculating the equilibrium voltage under application specific concentrations of the solution: Y i am DG ¼ DG0 + RT ln i E ¼ E0 −

RT Y mi ln ai zF

where E is the equilibrium potential of the electrochemical reaction at the existing concentrations of the reactants and ai are the activities of the reactants and not their concentrations due to effects of interaction of reactants and the effect of solvent molecules on the reactivity of the reactants. The values of activities are a function of the concentrations. However, in many cases it is possible to approximate the activity by the concentration c  a. Similar adjustments can be made for variations of partial pressure in gas electrodes.

2.2

Theoretical energy of cells

The change of free energy of a reaction DG has been widely tabulated for aqueous solutions of many materials based on standard concentrations of reactants. These values can be used to calculate the theoretical energy content of battery systems and, using Nernst’s equation, the theoretical energy content at a given concentration. W ¼QU where W is the theoretical energy content of the system, U ¼ DE the cell voltage (equilibrium Nernst voltage) and Q the theoretical capacity of the electrodes assuming complete conversion from charged to discharged active material. The specific theoretical energy is calculated by dividing the theoretical energy W by the mass m of all reactants required for the theoretical capacity: Wm ¼ W/m. It is worth noting that

• •

For non-aqueous solutions, where the thermodynamic values are not available, the term theoretical energy or specific theoretical energy are used despite not being able to calculate the equilibrium voltage from thermodynamic data. In general, the theoretical capacity is then multiplied by the nominal voltage of the system to calculate a theoretical energy, W ¼ QU. For aqueous solutions it should be straightforward to calculate the theoretical energy of a system. Using lead acid batteries as an example shows the difficulty. The theoretical specific energy based on the Gibbs energy for one molar concentration of sulfuric acid is 161.2 Wh/kg. For batteries with sulfuric acid with a density of 1.24 kg/L (e.g. stationary batteries), the theoretical specific energy is 174.5 Wh/kg and for a density of 1.32 kg/L (e.g. VRLA batteries), the theoretical specific energy is 180.8 Wh/kg, in both cases assuming constant conditions, i.e. no concentration changes during discharging and disregarding the weight of water which is required in the electrolyte to provide diluted sulfuric acid. Such a concentration dependent value is impractical and usually the theoretical specific energy is simply calculated from the theoretical capacity (83.5 Ah/kg) and the nominal voltage of the galvanic cell (see Table 1). The value of the theoretical specific energy given usually for lead-acid batteries is therefore 167 Wh/kg.

In addition, the above calculations assume that the reactants and products are well defined. For instance, the crystal structure and stoichiometry and phase composition of the active material of the positive electrodes of Ni/Cd and Ni/MH cells are complex, and the thermodynamic data are not available.2–4 A similar situation exists for battery systems in which the stoichiometry varies continuously over a wide composition range, as in intercalation processes. For supercapacitors, the concept of theoretical energy is meaningless as the capacitance is a characteristic of design, materials and manufacturing processes.

2.3

Energy and specific energy of a cell

It is obvious that conversion of the active material under conditions of a thermodynamic equilibrium is not possible and even a very slow discharge rate at near equilibrium conditions does not allow a complete discharge of the active material at the equilibrium voltage.

482

Electrochemical Terminology | Energy

Under application specific or nominal discharge conditions, two aspects cause a major reduction of the energy content of a cell in comparison to its theoretical value:

• •

Reduction of capacity from the theoretical capacity because of mass utilization restrictions Reduction of voltage from the equilibrium voltage due to overpotentials and the ohmic voltage drop in the electrolyte: U ðIÞ ¼ U 0 − cathode − anode − IRe

2.4

Mass utilization

Only a limited amount of the active material can be converted during discharging to the end of discharge voltage. The main reasons for limited mass utilization are:

• • • •

Poor electrical conductivity of active material (especially the discharged), particularly for lead acid batteries Irreversible structural changes associated with deep discharge Diffusion and mass transport limitations of reactants Current inhomogeneity.

The reduction of mass utilization caused by these factors is technology dependent and the overall effect differs greatly.

2.5

Voltage und energy content

Batteries cannot be discharged close to their thermodynamic state, i.e. with extremely small currents and therefore have a voltage during discharging which is 5–20% lower than the theoretical equilibrium voltage which can be derived from Gibbs free energy and Nernst equation. The main reasons why the battery voltage during discharging differs from its thermodynamic values are: (1) ohmic losses due to substantive electrical resistances inside cells, (2) kinetic overpotentials due to slow electrochemical reactions (Butler-Volmer equation), (3) concentration overpotentials due to mass transport limitations or reaction steps that determine the reaction rate, (4) changes of the equilibrium voltage due to the reduction of concentrations of reactants or potential of lattices spaces during de-intercalation and (5) poor current distribution. The cell voltage U is therefore often given as: U ðIÞ ¼ U 0 − ct − D − C − U R where U0 is the equilibrium voltage of a cell, ct is the charge transfer overpotential, D is the voltage drop because of diffusion effects, and C is the voltage drop caused by concentration changes. UR is the ohmic voltage drop in the electrolyte, which does not belong to the overpotentials. Experimentally, it is sometimes difficult to distinguish between these voltage drops, and it needs some experience to distinguish between these effects properly. For calculating the specific energy of a cell, the weight of electrochemically inactive components needs to be considered which are required and ensure mechanical stability.5

•

• • •

Electrolyte solution and the separator. The electrolyte solution is required to provide ionic conductivity between the positive and negative electrodes. It might be thought that in systems where the electrolyte does not take part in the electrochemical reactions, the layer of electrolyte between the electrodes might be extremely thin. However, as the positive and negative electrodes must not touch each other, a separator has to be used which has to be soaked in the electrolyte solution to ensure ionic conductivity. Thus, the electrolyte solution is always present in excess. Future cells with solid electrolytes will probably have a much lower electrolyte weight. In lead acid batteries, sulfuric acid is an active component of the electrochemical reaction and the amount of water required to dilute sulfuric acid to the required range of concentration is the single biggest cause of reducing the theoretical energy. Grids and electron collectors. The porous active mass of the electrodes is, in most cases, structurally weak and poorly conducting. A grid (as in lead-acid batteries), nickel-coated polypropylene fibers (as in some Ni/Cd cells), or thin metal foils (as in some lithium-ion cells) are therefore required. To ensure a good connection between the active material and the electron-conducting foil, binders are used that further reduce the energy density. Other passive components. The most important passive components are cell containers, which must be designed to be chemically inert and mechanically strong. Auxiliary components and materials. Cells often need to be equipped with additional components for safety or maintenance reasons. Such items include safety vents, water ports, high current or high temperature fuses, sensors, and other such items. All these features must be taken into account when calculating the specific energy of cells.

There are usually additives to improve the properties of active materials. However, their additional weight can be neglected. For these reasons, the energy content of a cell must be determined empirically and is typically expressed in Watt-hours (Wh). For measuring the energy content of an electrochemical cell, a fully charged cell is usually discharged at a constant current. As the discharge voltage falls during discharge, the following equation must be used:

Electrochemical Terminology | Energy Z

Dt

W¼I

483

U ðt Þ dt

0

where U(t) denotes the voltage during discharging, I the constant discharge current and t the discharge time until the end of discharge voltage has been reached. Only very few product data sheets provide information on the average discharge voltage during discharging and thus it is usually not possible to calculate the energy content of a battery using W ¼ I U Dt where U denotes the average discharge voltage. As a good approximation, the nominal voltage of the cell can be substituted for the average discharge voltage. Values for the energy content of a cell, let alone values for the energy content at different discharge rates and temperatures, are usually not given as part of the product specification. In cells used for stationary power applications the duration of providing a constant power rather than a constant current is often given and thus, the energy content of a cell under different discharge rates can be calculated. The specific energy (or gravimetric energy) in Watt-hours per kilogram (Wh/kg) is an important value. And for some applications the energy density (Wh/L) is used. The energy density is often also referred to as volumetric energy. The units Wh/kg or Wh/L are not used for fuel cells or redox flow systems because the energy is stored externally in those cases. Fig. 1 provides a diagram of the mass distribution of different types of lithium-ion cells.

2.6

Dependence of energy content on operating conditions and design

Temperature and discharge rate not only have an impact on capacity but also on the energy content:

• •

Temperature. Diffusion and other types of mass transport (convection, migration) are very sensitive to temperature. As a result, there are diffusion overpotentials and the voltage is lowered at low temperatures. For lithium-ion batteries the maximum energy content could be delimited at low temperatures as the charge current and the useable state of charge can be reduced to avoid lithium plating. Design restrictions. The design of a cell has a strong influence on the electrical resistance of grids and electrolyte solutions, and thus directly influences the current distribution. This has important repercussions in the design of batteries for low power and high-power applications. In particular, the specific energy of cells with the same electrochemistry can differ greatly.

A detailed comparison between the theoretical specific energy of an electrochemical power source and the specific energy of commercial cells and complete battery systems is complex. For lithium-ion cells the details of the active mass composition are often not given, size of cell, type of cell (cylindrical, prismatic, pouch, energy or power cell) lead to differences and the number of auxiliary components which are included (converters, protective class of housing, etc.) differ in data sheets. Comparisons therefore must be restricted to cells of similar size (see e.g.7) and type or great care must be taken to compare like with like.

Fig. 1 Mass distribution of commercial 2.1 Ah lithium-ion cells.6

484

3

Electrochemical Terminology | Energy

Supercapacitors

For capacitors, the energy content W is determined empirically using the well-known formula from electrical engineering 1 W ¼ CU 2 ¼ QU 2 where C is the capacitance of the supercapacitor (in F ¼ As/V) and U the voltage of the supercapacitor (in V), and Q the stored electric charge (in C ¼ As). If supercapacitors are discharged to 50% of their rated voltage, 75% of their total energy stored is extracted.  1  W ¼ C U 2N − U 2E 2 Here, UN the nominal voltage, and UE the voltage at end of discharge. It should also be noted that, in data sheets, the value for the energy content is mostly given for a complete discharge to 0 V, implying UE ! 0. A further impact on available energy is associated with the internal resistance of supercapacitors. When discharging supercapacitors at high rates, an ohmic loss may also be significant for the discharged power P ¼ I2R. Finally, when discharging a supercapacitor into a matched load (external load resistance equals internal resistance) to achieve the highest possible power transfer, the energy output to a load will be decreased by 50% of the above value. For both batteries and supercapacitors, the specific energy can be calculated by dividing the energy with the mass of the complete device. For a large commercially available supercapacitor, Maxwell 3.4kF, BCAP3400 P300, with a maximum voltage of 3.0 V and a rated capacitance of 4080 F at that voltage,8 the maximum energy content is given as 4.4 Wh and the specific energy as 9.1 Wh/kg. Compared with any battery chemistry, these are very low values. The advantage of a supercapacitor lies in the power output which is rated for the above given supercapacitor as 17 kW/kg or, if a matched load is used, even higher.

4

Specific energy of cells connected in series

It is not possible to use the energy content of a cell and its specific energy to exactly calculate the energy content or specific energy of a string of cells. The charge capacity of batteries or supercapacitors in series cannot be utilized in full. Voltage balancing is required to prevent individual cells from experiencing extreme voltages. Additional components may also be required for safety or lifetime reasons. As a result of the increased mass and volume, the specific energy and energy density of a series combination is always lower than that of a component cell. Care must be taken that cells connected in series have uniform properties. For critical applications, cell capacities are measured individually during production and only well-matched cells are placed in series. In such cases, the differences of cell voltages during discharging are negligible and the energy content of the string is simply the sum of the energy content of the individual cells. However, in most cases small but relevant differences in cell capacity can be expected and during aging and under operational constraints, these differences become bigger. As the current through all cells is the same, the amount of charge removed is also the same. Cells with a below average capacity will therefore suffer deep discharge and be irreversibly damaged. Analogously, these cells are overcharged, which can have an even more disastrous effect in the case of lithium-ion batteries. Where matched cells cannot be assembled (or over-discharge leads to severe damage) the discharge and charge must be terminated early. This precaution is particularly necessary in the case of lithium-ion batteries, where over-discharging and subsequent charging can lead to catastrophic failures. Ending the discharge or charge of a battery early because of a weak cell further decreases the specific energy of the overall system. Balancing-circuits can be used to equalize differences in cell voltage; due to unavoidable losses the energy content is lowered by this measure. In some battery systems, such as sealed Ni/Cd and Ni/MH, the electrodes contain excess active material to minimize damage if they are over-charged or over-discharged. Additives with a similar function are also incorporated in some lithium-ion cells. The additional weight is undesirable but necessary. Lithium-ion batteries, especially cells used for vehicle applications, are also fitted with various monitoring devices and safety features that decrease the specific energy considerably. Fig. 2 shows a lithium-ion battery including the frame and the electronics for monitoring cell voltages. The resulting specific energy for this battery is 105 Wh/kg. For most batteries the weight of the cells makes up between 50% and 80% of the total weight and volume of the battery. To increase the specific and volumetric energy of battery systems, new concepts try to do without modules and housings. Instead, the battery cells are connected to each other and integrated directly into the overall battery housing (Cell-to-pack technology). In a further step, the battery cells are integrated directly into the chassis of a vehicle (cell-to-chassis technology).9 Fig. 3 shows schematically the factors that lead from the theoretical specific energy of an electrochemical couple to the actual specific energy of a complete battery. The specific energy tends to decrease with cell capacity, nominal voltage and a long lifetime requirement.

Electrochemical Terminology | Energy

485

Fig. 2 High energy battery with a 360 V nominal voltage and 120 Ah capacity (ca. 43 kWh) consisting of 200 cells (two parallel strings) with 60 Ah each (Li(Ni0,5Co0,5)O2/graphite). The mass of the complete battery is 410 kg, the specific energy 105 Wh/kg (Courtesy of Gaia Akkumulatorenwerke GmbH, Nordhausen, Germany).

TheoreƟcal specific energy Ca. 25% due to intercalaƟon levels are kept between 0.1 – 0.15 and 0.85 – 0,9 for fully charged and discharged cells.

Ca. 15% due to restricƟons of SoC usually in the range of 95 to 10%. Between 0% and ca. 25% for applicaƟon specific restricƟons of SoCrange, e.g. for hybrid vehicles. Ca. 15% for copper and aluminium current collectors Ca. 4% for separators, electrolyte acƟve mass addiƟves (binders etc.) Ca. 2% for cell container and poles

PracƟcal specific energy

Ca. 5 – 20% for further components for cells in series (connectors, BMS, frame, sensors, cooling or heaƟng systems)

Fig. 3 Schematic diagram of the losses of a Li-ion battery that prevent the attainment of the theoretical specific energy of cells and batteries. The components making up the battery have roughly the same weight as the cells in this application (power assist and hybrid vehicle application) and the SOC range is restricted to achieve the required lifetime and safe operation.

Calculating and quoting the weight or specific energy of a battery without components which are required for safety reasons (BMS, fuses, switches, venting systems, module housings to contain fire risks) or for lifetime reasons (e.g. cooling and heating systems) is misleading. Dependent on the application also the weight and volume of charging equipment must be added to give a realistic picture and to allow a fair comparison with competing battery systems or alternatives. The concept of theoretical specific energy is meaningless for fuel cells as the reactants are stored in an external tank. However, a specific energy value can be calculated for a closed fuel cell system complete with fuel tank. A portable fuel cell systems, e.g. EMILY 3000 by SFC Energy AG, has a specific energy of 880 Wh/kg, higher than any commercially available battery system.10

486

5

Electrochemical Terminology | Energy

Influence of application on energy content and specific energy

The depth of discharge (DOD) of a battery (or the allowed range of the state-of-charge SOC in which it is operated) is often restricted for lifetime reasons. When talking of the energy content of batteries, the maximum SOC range for a particular application should always be mentioned. Measurements of a Toyota Prius battery have shown that the maximum SOC is ca. 75% and the minimum SOC is ca. 30%. As a further example, a lithium-ion battery used as a “power assist” battery for hybrid vehicle applications is limited to an SOC range between 90% and 25%. Therefore only 65% of the energy content of the cell based on a 100% discharge is used. The energy available from a battery in many applications is therefore considerably diminished.

6

Conclusion

The specific energy of a cell is considerably less than the theoretical value due to limited utilization of active material and the mass of passive components. When connecting cells in series, a further decrease often occurs due to the need for balancing systems and other components needed for safety reasons. Great care must be taken when comparing different systems, in particular when cells are connected in series, as the number, mass and volume of auxiliary components may vary widely. In general, the specific energy should be quoted only in the context of a given application. The temperature and lifetime considerations strongly impact specific energy values. The specific energy of a cell or battery is an important parameter for automotive power applications. However, mass and volume are of less importance for other applications, such as forklift trucks requiring counterbalance weights, and for stationary applications where weight and volume are often of no practical concern. In yet other applications, such as aviation and portable electronics, the specific energy is the main criterion for choosing the most suitable energy storage technology.

References 1. Linden, D., Ed. Handbook of Batteries, 2nd ed.; McGraw Hill, 1995. 2. Bode, H. Zur Kenntnis der Nickelhydroxidelektrode. II. Über die Oxydationsprodukte von Nickel(II)-hydroxiden. Zeitschrift für anorganische und allgemeine Chemie 1969, 366, 1–21. 3. Berndt, D. Maintenance-Free Batteries, a Handbook of Battery Technology; Research Studies Press Ltd.: Taunteon, Somerset, 1993. 4. Delahay-Vidal, A.; Portemer, F.; Beaudoin, V.; Tekaia-Elhsissen, K. In D.A. Corrigan: Nickel Hydroxide Electrodes, Proceedings Vol. 90–4; Genin, P., Figlarz, M., Eds.; The Electrochemical Society Inc.: Pennington USA, 1990; p. 44. 5. Cao, Y.; Li, M.; Lu, J.; Liu, J.; Amine, K. Bridging the Academic and Industrial Metrics for Next-Generation Practical Batteries. Nature Nanotechnology 2019, 14, 200–207. 6. Marshal, J.; Gastol, D.; Sommerville, R.; Middleton, B.; Goodship, V.; Kendrick, E. Disassembly of Li ion cells, characterization and safety considerations of a recycling scheme. Metals 2020, 10 (6), 773. 7. Quinn, J. B.; Waldmann, T.; Richter, K.; Kasper, M.; Wohlfahrt-Mehrens, M. Energy density of cylindrical Li-ion cells: A comparison of commercial 18650 to the 21700 cells. J. Electrochem. Soc. 2018, 165, A3284. 8. Maxwell Datasheet BCAP3400 P300. 9. Fichtner, M. Recent Research and Progress in Batteries for Electric Vehicles. Batterie&Supercaps 2021. https://doi.org/10.1002/batt.202100224. 10. SMC Energy AG, Datasheet EMILY 3000.

Electrochemical Terminology | Power H Wenzla, R Bengerb, and I Hauerc, aConsulting for Batteries and Power Engineering, Osterode, Germany; bResearch Center Energy Storage Technologies, Clausthal University of Technology, Clausthal-Zellerfeld, Lower Saxony, Germany; cChair of Electrical Energy Storage Technology, Institute of Electrical Power Engineering, Clausthal University of Technology, Clausthal-Zellerfeld, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is a update of H. Wenzl, BATTERIES AND FUEL CELLS | Power, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 559-565, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00045-9.

1 2 2.1 2.2 2.2.1 2.2.2 2.2.3 2.3 2.4 2.5 3 4 5 6 7 References

Introduction Pulse power Measuring internal resistance Factors influencing internal resistance Dependence on pulse duration Dependence on current amplitude Dependence on state of charge and temperature Power of supercapacitors Short-circuit current Pulse power of fuel cells and redox flow cells Constant high power Ragone diagram Power of fuel cells and redox flow batteries Design aspects to improve the power capability Conclusion

488 488 488 489 489 490 491 492 493 493 493 494 496 497 497 497

Abstract This article describes the power capability and internal resistance of batteries, fuel cells, and supercapacitors. It is shown that the “Ragone diagram” is a convenient means of comparing the specific energy and specific power of these technologies.

Key points

• • •

Overview of electrochemical terminology Definition of the terms related to power Correct usage of symbols and units

Nomenclature E I m OCV P Q R t U W Wh

Electrode potential (V) Electric current (A) Mass (kg) Open-circuit voltage (V) Electric power (W) Electric charge (C ¼ A s) Resistance (O) Time (s) Cell voltage (V) Electric energy (J) Watt-hour (Wh ¼ 3600 J)

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00082-7

487

488

1

Electrochemical Terminology | Power

Introduction

In a number of applications, the ability of supercapacitors and batteries to provide power P (in W) over a short period of time and without time lag (pulse power) is an important property. Familiar examples are the starting of internal combustion engines, providing additional power for acceleration and absorbing braking energy in electric vehicles, the starting of electrical motors with high inrush currents, and the triggering of fuses. The batteries in hybrid vehicles are also designed for higher power rather than high energy content (P/W > 10). In other applications, e.g. the protection of computers and telecommunications equipment against brief interruptions of mains power supply, very high discharge rates are required for a short time (e.g. 5–15 min). For grid ancillary services such as the provision of virtual inertia or voltage and frequency control the ratio between power and energy can become very large (P/W  10).1 The energy which is supplied to the load in high power applications is considerably smaller than the energy which can be supplied at a lower discharge rate. The different applications lead to so called power batteries and energy batteries. A useful method of comparing the specific power and specific energy of different power sources in these circumstances is the Ragone diagram. Power is the rate at which energy is used. This chapter focuses on the high-power properties of electrochemical power sources and methods to measure them. Standard electrical engineering principles show that high-power output capability corresponds to low internal resistance: P ¼ I2R ¼ U2/R. Batteries and supercapacitors may exceed their rated power considerably for a few seconds. In contrast, it has been found that the ability of a fuel cell or redox flow system to provide power beyond the maximum rated power, even for a very short period of time, is very limited. In the following, a distinction is made between discharging a battery at high power, e.g. within 5–30 min, and pulse power requirements for a few seconds only.

2

Pulse power

In applications that have a very high-power demand for a short period of time, there is usually also the requirement of a very fast response time, i.e. no time lag between the time of the power requirement and only a few milliseconds until the power is made available. Supercapacitors are sometimes considered to be a better choice than batteries. However, due to their inductance and sometimes the need for a DC/DC-converter to adjust the voltage of the supercapacitor to the voltage requirements of the load, supercapacitors may have in special cases a slower response time and a lower power gradient than batteries. Furthermore, it is also known that the capacitance and thus the energy content of supercapacitors decreases sharply with frequent changes in current direction (at high frequencies), as may be required e.g. in applications to improve power quality. Batteries have the disadvantage of a sudden and sharp voltage drop when a power pulse has to be provided but then have a relatively constant voltage compared to supercapacitors. The key is in both cases their low internal resistance. As in all electrical systems, the maximum power is delivered when the internal resistance and the external resistance are equal. However, this point is at 50% of the cell voltage and therefore, for most battery chemistries, outside the voltage range which is specified by the manufacturer. Some cell manufacturers do allow voltage levels below the normal end of discharge voltage for a very short time (seconds). Please note that at the point of maximum power, the heating power generated within a cell and the power delivered to a load are equal.

2.1

Measuring internal resistance

There are basically two methods for measuring the internal resistance of an electrochemical power source.

• •

A current or voltage step is applied to the cell and the voltage or current curve is evaluated in the time domain: current step method (chronopotentiometry) or voltage step (chronoamperometry). Fig. 1 shows a simple, but widely used equivalent circuit diagram for a current step method consisting of an ideal voltage source variable voltage and a variable internal resistance of the battery connected to a load. Using the impedance method (EIS), a small AC voltage or current is applied to the battery by the help of a frequency response analyzer and an electronic load, and the phase shift between excitation signal and cell answer is measured. Frequency by frequency, usually between 100 kHz to 0.01 Hz, the impedance spectrum is recorded, which allows to determine the model parameters of the equivalent circuit of the cell. Fig. 2 is often used for electrochemical impedance spectroscopy, but can also be used for the evaluation of current step measurements. By appropriate representation of measurements using a Nyquist or Bode diagram, the DC-resistance can be readily determined. If only the electrolyte resistance (equivalent series resistance) of is wanted, this can be determined more easily with the aid of an AC ohmmeter at a frequency of 1 kHz. The resistance at 1 kHz is simply the ratio of the peak values of AC voltage and current.

Electrochemical Terminology | Power

489

Fig. 1 Equivalent circuit diagram of a battery; E0 is the open circuit voltage of the cell at rest and Ri is the lumped internal resistance (combining all effects which lead to a change in voltage during current flow).

UR

UCT

UD

Load/Charger

UL

E0

Fig. 2 Equivalent circuit diagram of a battery including ideal voltage source E0, its inductance, the ohmic resistance of its components and two RC-couples with different time constants (charge transfer CT, diffusion and other equalization processes D).

Ri,1kHz

   b U 1kHz   ¼   I1kHz  b

In datasheets, the internal resistance is often given as DC-resistance or resistance at 1 kHz. It is usually not disclosed whether the DC-resistance is measured using the current step method or EIS. The methods described above do not provide any information on the changes of voltage and current during a power pulse. But this information is often not required.

2.2

Factors influencing internal resistance

The internal resistance and thus power depends for most electrochemical systems strongly on temperature, pulse duration, state of charge and current amplitude.

2.2.1 Dependence on pulse duration Fig. 3 shows the measurement of voltage and current during a pulsed discharge through a resistor. Resistors rather than a constant current load are often used for practical reasons as the current for such tests using commercial batteries can easily be in the range of a few hundred amperes. In Fig. 3 the current and voltage values at the start of the pulse, immediately after start of discharge, immediately before the end of the pulse and immediately after the end of the pulse can be used to calculate the internal resistance. But usually, the values before the pulse and immediately before ending the pulse are used. Ri ¼

DU U 0 −U 2 ¼ I2 −I0 DI

where Ri is the internal resistance (in O). From Fig. 3. it is immediately obvious that the internal resistance depends on the duration of the pulse. Often, a pulse duration of 10 s is used to determine the power capability as this reflects the time for accelerating or breaking in-vehicle applications. The generalized formula for a time-dependent resistance is

490

Electrochemical Terminology | Power

Current

Fig. 3 Voltage and current during a short high-power discharge using a resistor as a load.

Ri ð t Þ ¼

U OCV  U ðt Þ Iðt Þ

where UOCV is the open circuit voltage prior to discharge (in V), and U(t) is the voltage at the time of measurement t. I is the discharge current (in A).

2.2.2 Dependence on current amplitude In addition to the pulse duration, the internal resistance of a cell depends on the current because the charge transfer overpotential described by the Butler-Volmer equation has a strong dependence on current. The increase of the charge transfer overpotential is lower than the increase of the current, so that the resistance becomes smaller, as long as diffusion effects are not considered. Fig. 4 shows the internal resistance of a 6.5 Ah Ni/MH module, having a nominal voltage of 7.2 V, measured 1 ms and 1 s after the start of discharge. It can be seen that the internal resistance is both a function of discharge current and time, and that the

Internal resistance (mOhm)

14 12 10 8 6 4

Ri (1 sec)

2

Ri (1 msec)

0 0

100

200

300

400

500

600

700

800

Discharge current (A) Fig. 4 Internal resistance of a 6.5 Ah, 7.2 V Ni/MH module 1 millisecond and 1 s after start of discharge at full state-of-charge. Room temperature measurement, polynomial fit.

Electrochemical Terminology | Power

491

dependence on either parameter is complex. The small increase of internal resistance for a one second pulse at higher currents is possibly the result of the depletion of the concentration of hydrogen at the electrode/electrolyte interface. It is important to note, that the determination of the internal resistance by impedance spectroscopy does not show these current dependences because the excitation with small AC currents is usually in the linear range. As a consequence, one should select the measurement method according to the application (e.g. for high-power application ! current step method (time domain)). For determining the power capability of a cell, an application specific power or current curve followed by a rest period is often used, for instance, the EUCAR power assist cycle for hybrid vehicles. The power requirement is fulfilled as long as the voltage during the test does not fall below a set voltage limit.

2.2.3 Dependence on state of charge and temperature Short, very high rate, power requirements and short-circuit power are not only important when the battery is fully charged, but may also be required at any point during the discharge process and at any temperature that can be expected in the application. To understand the voltage response during such brief discharges, the effects of mass transport and inhomogeneous current distribution must be investigated. Fig. 5 shows the measured internal resistances of a lithium-ion cell at different temperatures one second after a 5C current pulse. The increase of internal resistance at low SOC and temperature is not unexpected. This general behavior is typical for batteries, although there are differences in detail. Lead-acid batteries, for instance, show a marked increase at low SOC, while supercapacitors have a much smaller dependence of internal resistance on temperature. Some battery systems, most notably NiMH and some lithium-ion systems cannot be used for power applications below 0  C. Fig. 5 also shows that a pulse power requirement can change the capacity of a cell considerably. Below 40% SOC, the voltage during the discharge pulse falls below the threshold voltage of the cell and the measurement has to be stopped. In applications with short, high-rate power requirements this decreases the capacity or SOC range that can be used. In batteries, where the voltage threshold is lower, or is not of critical concern, the loss of capacity is less severe. Electrochemical power sources such as batteries and supercapacitors not only deliver power but also take up power in some applications. For example, during braking in different types of electric vehicles, a problem is that, at high states-of-charge, the voltage during the charging pulse can exceed the normal threshold value if there is no means to divert or dissipate the power in some other way. This limits the useful range of operation as shown in Fig. 5. Note that fuel cells are usually protected against charging currents. In grid applications, where symmetrical power provision is required, the ideal state of charge is the point at which discharge and charge power are equal. This point of maximum charging and discharging power is not necessarily at 50% SOC (see Fig. 6). Fig. 7 shows the power that can be delivered (and taken up) during short, 10-s pulses at different depth of discharge (here: DoD ¼ 1 – SoC) for another electrochemical power source. Knowledge of the power capability of a cell under operating conditions requires a wide range of parameter variations. Practically, a cell is often discharged in steps, i.e. after drawing 10% of capacity, a power pulse is administered and the internal resistance respectively the power capability measured over the whole range of state of charge. This must be repeated during charging as the power capability differs for providing power and taking up power in particular at very high and very low states of charge. Fig. 7 shows such measurements.2 In addition the measurement has to be repeated for the range of temperatures which can be expected for the application in mind. 0,06 0°C RT 40°C

Internalresistance(Ohm)

0,05 0,04 0,03 0,02 0,01 0,00 1

0,8

0,6

0,4

0,2

0

SOC Fig. 5 Internal resistance of a 2.1 Ah lithium-ion cell measured at different SOC and temperatures by means of 5C current pulses one second after start of pulse. Due to the voltage threshold, measurements at 0  C were not possible below 40% SOC.

492

Electrochemical Terminology | Power

Fig. 6 SOC-range of maximum power in charge and discharge direction for a 5 Ah LiNiMnCoO2 cell.

Fig. 7 Ability of a hybrid lead/acid-supercapacitor cell (Ultrabattery) to supply and take up power for 10 s at different states of charge. From Lam LT, Louey R, Haigh NP, Lim OV, Vella DG, Phyland CG, Vu LH, Furukawa J, Takada T, Monma, D, and Kano T. VRLA Ultra Battery for High-Rate Partial-State-of-Charge Operation. J. Power Sources 174, 2007, 16–29.

2.3

Power of supercapacitors

Supercapacitors have an extremely high-power capability when fully charged but a low energy content compared to batteries of similar weight and size. However, a given power requirement leads to a rapidly increasing current during discharging because the voltage decreases proportionally to the amount of charge which has been removed. In IEC 62391-2:2006 “Fixed electric double-layer capacitors for use in electronic equipment – Part 2: Sectional specification – Electric double-layer capacitors for power applications” there is a description of how to measure the power density Pd of a supercapacitor: Pd ¼

I  ðU −U 6 + U e Þ 2m

Here U is the charging voltage, U6 is a voltage drop caused by the high rate of discharge (corresponding to 20% of the charging voltage in most applications), Ue is the end-of-discharge voltage (40% of the charging voltage), I is the discharge current, and m is the mass of the supercapacitor. For typical values of U6 and Ue, the average voltage is 60% of the charging voltage. For larger supercapacitors, other definitions are used. Maxwell technologies, for instance, use the following definitions: Pd ¼

0:12 U 2 m RDC

P max ¼

U2 4mR

where Pd is the average discharge power, Pmax is the theoretical maximum power for matched impedance (load resistance ¼ internal resistance), U is the charging voltage, RDC is the equivalent series resistance measured at direct current, and R is the equivalent series resistance measured at 1 kHz.3 Table 1 provides some data for supercapacitors. Their power capability is only little affected by low temperatures.

493

Electrochemical Terminology | Power Table 1

Power values for supercapacitors (Maxwell technologies data sheet3).

Capacitance (F)

Pmax (W)

Pd (W)

Pmax (W/kg)

Pd (W/kg)

Mass (kg)

650 1200 1500 2000 3000

3020 4140 5184 7000 7590

1080 1500 1856 2480 2970

15,100 13,800 16,200 17,500 13,800

5400 5000 5800 6200 5400

0.20 0.30 0.32 0.40 0.55

2.4

Short-circuit current

As a special, safety-relevant aspect, batteries and supercapacitors are required in most applications to provide short-circuit currents for triggering fuses in the case of electrical malfunction. In this case, the requirement is for a very high current, not for very high power. Depending on the installed fuse, the current has to be sustained for up to about 0.5 s. If the fuse cannot be triggered by the battery, then the energy of the battery will be discharged at the point of the electrical malfunction, and this may cause fires or other hazards. IEC 896-2 provides a measuring method to determine the short-circuit power for stationary lead acid batteries. The short circuit current or the resistance values measured by this method are often included in data sheets. Values exceeding 20,000 A are not uncommon for large stationary batteries. If only the internal resistance of a cell is given, then the short circuit current of a cell can be calculated by Isc ¼

U0 Ri

where U0 is the nominal voltage of the cell. Matching fuses to batteries is difficult, because the short circuit power of the battery as calculated above is a rough estimate only and does not provide information for the short circuit power of an aged battery at low SoC and low temperature.

2.5

Pulse power of fuel cells and redox flow cells

In contrast to batteries and supercapacitors, the power output of fuel cells and redox flow cells during a pulse power requirement is limited. A reasonably fast increase in power output is possible as there is always a surplus of reactants, but the supply of reactants must be increased sharply to maintain the power output. The maximum power output is limited to the maximum rated power level (respectively the power level which can be maintained indefinitely provided that the supply of fuel is maintained, and the output power depends only on the rate of fuel supply and the rate of fuel utilization) as the capacitance values of fuel cells are very small.

3

Constant high power

Batteries are sometimes referred to as high power cells or energy cells. There are no agreed upon criteria concerning a high-power cell with a more or less “constant” high power discharge and an energy cell. Examples for the application of high-power cells are:

• • •

hybrid vehicles in a power assist mode, e.g. where the battery is used to support acceleration or to absorb breaking power and the average currents are considerably higher than 1C or uninterruptible power supply systems, where batteries are required to protect loads for 10–15 min. grid ancillary services like providing virtual inertia, voltage and frequency regulation, primary reserve power.

Energy cells in contrast would be used for electric vehicles where the average discharge current is in the range of ¼ C or less or emergency power supplies for lighting, where loads have to be supplied for e.g. up to 10 h. Batteries that temporarily store solar energy in households also have the characteristics of an energy storage where the average discharge current is lower 1C. Please note that pulse power requirements may also exist and it is always necessary to provide sufficiently high short-circuit power for safety reasons. A low internal resistance is important for all power sources, but design and materials to reduce the internal resistance are of greater importance in power cells. Table 2 shows the key differences between the design of power and energy cells – current collectors, porosity of electrodes, effective internal electrode surface area. To determine the power capability of a battery, a constant power discharge instead of a constant current discharge is often used as this corresponds better to the application specific requirements. Product data sheets do not always provide this information and

494

Electrochemical Terminology | Power Overview of design differences between energy and power cells.4

Table 2 Component

Energy cell

Power cell

Electrodes

High coat weights Low coating porosity Medium + large particle sizes Low conducting carbon content Minimum possible binder content Thinner Coated to improve adhesion Thin High conductivity Thin/narrow tags Single tag on each electrode

Low coat weights High coating porosity Small + medium particle sizes High conducting carbon content

Current collectors Separator Electrolyte Connection tags

Thicker Coated to reduce resistance Thin High conductivity Thick/wide tags Multiple tags

therefore the average power which is required in a given application for the whole discharge time must be measured or estimated using average discharge voltages where given. An interesting aspect of measurements at high discharge currents is the increase in cell temperature that occurs when the power capability is measured at low temperatures. It is then possible that the discharging voltage increases during the discharge and the term constant power output at constant temperature is no longer straightforward.

4

Ragone diagram

Both capacity (electric charge) and energy decrease with increasing discharge current. A useful method to visualize the relationship between the energy that can be drawn from a cell, and the rate of energy removal, i.e. the power output, is a Ragone diagram. In a Ragone diagram, either specific energy (Wh/kg) is plotted against specific power (W/kg), or energy density (Wh/L) is plotted against the power density (W/L). In order to make fair comparison of different products and technologies, the following approach has been used for the Ragone diagrams shown in Figs. 8 and 9:

▪ Only commercially available products are included. ▪ Datasheet information is used as far as possible. ▪ It is assumed that the capacity of the cells is used fully, i.e. they can be discharged to their application-specific end-of-discharge voltage without considering lifetime aspects.

▪ Only cells are compared. 10000

W/kg(W/kg) Specific power

1000

100

12V OGiV 100 FNC 20 Ah energy cell

10

FNC 10 Ah power cell 12V 50 Ah NiCd, 2.5 AH S-SCR Li-Ion 2.1 Ah high rate Li-Ion 4.0 Ah energy cell

1 0

20

40

60

80

100

120

140

Wh/kg Specific energy (Wh/kg) Fig. 8 Ragone diagram of specific energy and power of different electrochemical couples and types of cells (data from data sheets and recalculations of low temperature data).

Electrochemical Terminology | Power

495

100000

W/l(W/l) Power density

10000

1000

100 12V OGiV 100 FNC 20 Ah energy cell FNC 10 Ah power cell

10

12V 50 Ah NiCd, 2.5 AH S-SCR Li-Ion 2.1 Ah high rate Li-Ion 4.0 Ah energy cell

1 0

50

100

150

200

250

300

350

400

Energy density Wh/l(Wh/l)

Fig. 9 Ragone diagram of energy density and power density of different electrochemical couples and types of cells (data from data sheets and recalculations of low temperature data).

A complication is that most manufacturers only give constant current discharge data, while constant power discharge data are only given for some applications. However, in most cases, the average power may be calculated from the average discharge voltage or discharge time. The following cells are included in Fig. 9:

▪ ▪ ▪ ▪ ▪ ▪

Nickel/cadmium cell, round cell, 2.4 Ah Nickel/cadmium cell, prismatic cell, FNC technology, 10 Ah, high rate and 20 Ah energy cell Lead-acid cell used for stationary applications OGiV (12 V, 100 Ah) Lead-acid starter battery, 12 V, 50 Ah Lithium-Ion cell 4.0 Ah energy cell and Lithium-Ion cell 2.1 Ah high-rate cell

The diagrams show that all cells have a similar dependence of Wh/kg on W/kg. At low power levels, the specific energy changes only a little, but at high power levels, the specific energy decreases considerably. Somewhat surprisingly, there are significant differences between different designs based on the same electrochemical couples. In particular, the differences between the Ni/Cd batteries, all three of which being commercial products, are significant and show that a high specific energy and power are only one of many requirements that have to be fulfilled by an electrochemical power source. For lead-acid SLI batteries, the available data sheets provide data only on the 20-h capacity C20, and the cold cranking capacity CCC, which describes the maximum short-term power requirement during starting. Correcting the values for the temperature used for capacity testing, values for a high rate specific energy and power can be calculated. These values are included in the Ragone diagram and show the high rate capability of lead/acid SLI batteries (12 V, 50 Ah). Ragone diagrams are often used as schematic diagrams without reference to particular products and product data to highlight differences between different technologies, and their use is perfectly legitimate. However, when drawing conclusions and selecting technologies, great care must be taken when evaluating the data. For example, when adjusting the energy and power values of the Ragone diagram for the weights and volumes of external components, the differences between cell values and battery values become very significant indeed. Sarpal and co-workers5 have determined experimentally the relationship between specific energy and energy density as a tool for application specific sizing of batteries. For applications that require both high-rate discharge and charge power, e.g. for ancillary services in the electrical grid6 it is useful to introduce Ragone diagrams for both current directions, see Fig. 10. This highlights, in particular, the limited charging power for many technologies (the maximum charging current is in some cases only one-tenth or less of the maximum discharging current). Another issue to consider is the cell size. Due to the delayed heat transport in large cells, the maximum power is limited in comparison to smaller cells. This is also the reason why most “ultra high-power cells” only have capacities of less than 10 Ah. Therefore, it is sometimes difficult to compare cell technologies in a fair way, especially for immature technologies for which only small cells are available.

496

Electrochemical Terminology | Power

Fig. 10 Ragone diagram showing the range of values of energy density versus specific energy for different battery technologies used for ancillary services for the electricity grid under highly dynamic load conditions for both charging and discharging currents in the range of 50–100C. From Deblon F, Klink J, Benger, R. Enhanced RAGONE-Diagram for Visual GAP-Analysis and Benchmark of Stationary Storage Systems for the Provision of Dynamic Grid Power. In: Advanced Battery Power Conference, Aachen, 2023. Poster P4-012.

5

Power of fuel cells and redox flow batteries

In contrast to batteries and supercapacitors, the power output of fuel cells and redox flow batteries can be maintained indefinitely provided that the supply of fuel is maintained. The output power depends only on the rate of fuel supply and the rate of fuel utilization. An example of output power as a function of current and fuel utilization is given in Fig. 11 for a SOFC operated at constant gas composition and temperature. As the fuel gas is consumed during its passage through the cell, the concentration of the fuel decreases and ultimately limits the power. The power of a fuel cell and its fuel utilization is therefore strongly coupled. In addition, it is often desirable to maintain a minimum fuel concentration at any point in the cell for lifetime reasons.

Fig. 11 Power density of a SOFC operated at constant temperature and fuel gas concentration as a function of current density and fuel gas flow rate.

Electrochemical Terminology | Power

497

Fuel cells and redox flow batteries are usually maintained within a narrow temperature range during operation, so that the temperature dependence of the power is not an important consideration.

6

Design aspects to improve the power capability

There are a number of options to improve the power capability of an electrochemical power source, some based on modifications of materials, others based on the design. Improvements in materials concentrate on improving the electrical conductivity of the components, e.g. by

• • •

additives and the addition of conducting materials or by coating of the active material to lower what could broadly be called contact resistances or by improved transfer from the electrolyte respectively the SEI into the host crystal structure.

Interesting examples are, for instance, the coating of LiFePO4 particles with a glass-like structure to enhance the fast exchange of lithium ions between the active material and the electrolyte and the addition of titanium wire with a ratio of ca. 1:100 of diameter to length to the active material of lead-acid batteries. More important however are design solutions. The use of thin electrodes, thin separators and thin electrolyte layers in systems, where the electrolyte is not consumed during the reaction are obvious solutions. An area where there still seems to be ample scope for further improvement is the design of the electrodes, the position of the current collectors and their relative positions to another. Both simulations and measurements show that the current homogeneity can be improved significantly. This reduces ohmic voltage drops and electrode overpotential. In addition, the bipolar design of batteries, similar to the standard design of fuel cells, offers very good current homogeneity and low internal resistance in the electrode due to the very short length of material which the current has to pass through. Batteries with bipolar design are under development but have to overcome the difficulty of creating a permanent, electrolyte-proof seal between the cells connected in series.

7

Conclusion

The ability of an electrochemical power source to deliver high power for a short period of time, and in some cases to absorb high power for a short period of time, is sometimes more important than its ability to store energy. In this regard, Ragone diagrams provide a convenient tool for comparing different electrochemical systems. However, when choosing a system for a given application, the data must be carefully analyzed, and the weight and volume of components for the battery assembly must always be taken into account.

References 1. Deblon, F.; Reinecke, S. S.; Werther, B.; Turschner, D.; Benger, R.; H.-B. Peck: Implementation and Evaluation of a High-Performance Beattery Converter System for Providing Synthetic Inertua at Distriubtion Network Level. In Proceedings the 3rd European Conference on Power Electronics and Applications (EPE21), Gent; 2021. 2. Lam, L. T.; Louey, R.; Haigh, N. P.; Lim, O. V.; Vella, D. G.; Phyland, C. G.; Vu, L. H.; Furukawa, J.; Takada, T.; Monma, D.; Kano, T. VRLA Ultra Battery for High-Rate Partial-Stateof-Charge Operation. J. Power Sources 2007, 174, 16–29. 3. Maxwell Technical Documentation: Standard Capacitance- and ESR-Test Procedures. 4. Lain, M. J.; Brandon, J.; E. Kondrick: Design Strategies for High Power vs. High Energy Lithium Ion Cells. Batteries 2019, 5, 64. 5. Sarpal, I.; Bensmann, A.; Mähliß, J.; Hennefeld, D.; Hanke-Rauschenbach, R. Characterisation of Batteries with E-P Curves: Quantifying the Impact of Operating Conditions on Battery Performance. Electr. Power Energy Syst. 2018, 99, 722–732. 6. Deblon, F.; Klink, J.; R. Benger: Enhanced RAGONE-Diagram for Visual GAP-Analysis and Benchmark of Stationary Storage Systems for the Provision of Dynamic Grid Power. In Advanced Battery Power Conference, Aachen; 2023. Poster P4-012.

Electrochemical Terminology | Efficiency H Wenzla, R Bengerb, and I Hauerc, aConsulting for Batteries and Power Engineering, Osterode, Germany; bResearch Center Energy Storage Technologies, Clausthal University of Technology, Clausthal-Zellerfeld, Lower Saxony, Germany; cChair of Electrical Energy Storage Technology, Institute of Electrical Power Engineering, Clausthal University of Technology, Clausthal-Zellerfeld, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of H. Wenzl, BATTERIES AND FUEL CELLS | Efficiency, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 544–551, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00047-2.

1 2 2.1 2.2 2.3 2.4 3 3.1 3.2 3.3 4 5 References

Introduction Efficiency of batteries Efficiency and applications Coulombic efficiency and charge factor Voltage Factors influencing energy efficiency Efficiency of supercapacitors Internal losses Self discharge processes and standby losses System components Efficiency of fuel cells Conclusion

499 499 499 499 501 501 502 502 503 503 503 505 506

Abstract This article discusses the different efficiencies of electrochemical power sources and their dependence on operating conditions. Under optimal conditions, the most interesting “electrical efficiency” may reach values above 95% for galvanic cells and supercapacitors; however, under the operating conditions of most applications and considering the total power system, the electrical efficiency is much lower.

Key points

• • •

Overview of electrochemical terminology Definition of the term efficiency Correct usage of symbols and units

Nomenclature

E F I Q R T t U z h

498

Electrode potential (V) Faraday constant: 96485C/mol ¼ 26.8 Ah/mol Electric current (A) Electric charge (C ¼ A s), capacity (1 Ah ¼ 3600C) Molar gas constant: 8.3144 J mol−1 K−1 Thermodynamic temperature (K) Time (s) Voltage (V) Electrochemical valency Efficiency (1 ¼ 100%)

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00085-2

Electrochemical Terminology | Efficiency

1

499

Introduction

The term efficiency has a number of aspects:

• • • •

energy efficiency referring only to electrical energy coulombic efficiency of galvanic cells such as batteries, redox-flow-systems, and supercapacitors thermal efficiency of fuel cells overall or fuel efficiency of fuel cells.

The term voltage efficiency is sometimes also used, but its interpretation requires detailed knowledge of the charging characteristics and has little practical use. It is possible to define “theoretical efficiency” for fuel cells but not for batteries and supercapacitors. Please note that the concept “efficiency” is not generally used for primary batteries.

2

Efficiency of batteries

The energy efficiency Z (dimension: J/J ¼ 1) of a battery cell, redox-flow battery or supercapacitor is defined as the ratio of the electrical energy that can be drawn from it, to the electrical energy required for recharging it to the same state of charge. The requirement of “same state of charge” at the beginning and at the end of a time period for which the efficiency is calculated causes difficulties in determining the efficiency experimentally. As side reactions and self-discharge processes can be reduced or enhanced depending on the state of charge during the charging or discharging process, the SOC-range has to be considered carefully.
¼

Electrical energy removed during discharge  100% Electrical energy supplied during charge

As the heat generation during discharging and charging is never a useful by-product of the operation of galvanic cells or supercapacitors, it is important to note that “energy efficiency” is here always synonymous “electrical energy efficiency.” In any discussion of efficiency, the coulombic efficiency and energy efficiency which includes the effects of the changing voltage levels should be considered separately for a clearer understanding. In addition, it is important to consider application specific aspects.

2.1

Efficiency and applications

The efficiency of an electrochemical power source depends on the details of its use. The efficiency of operation with small depthof-discharge (DoD) and 50% state-of-charge (SoC) will differ from the efficiency of operation with 100% DoD and 50% SoC. Current amplitude and temperature will also affect efficiency as internal losses will change. There are similar issues with the concept efficiency of converters: What are the current amplitudes and what is the input voltage? In some cases, there is very little power drawn from the battery but there is a constant energy demand that does not contribute to the power output at all. A good example is a battery that is kept fully charged for backup power. This requires continual float charging to counteract side reactions on processes, or—as in the case of lead-acid batteries—continual float charging to maintain the system at a voltage where the corrosion rate is minimal. Such charging never contributes to the discharge capacity. In such applications, the values for coulombic and electrical energy efficiency can be calculated, but have little practical value. For example, a lead-acid battery with 220 V nominal voltage and 100 Ah capacity on float charging requires a charging current of about 50 mA when new and, including the losses in the charging equipment, requires ca. 12 W of power. Over a year, the energy requirement is more than 100 kWh which is about 5 times the energy content of the battery! Furthermore, the older the battery, the higher the energy requirement of the battery becomes. The energy demand during float charging is often referred to as the “standby losses.” If the battery in an emergency power system is not discharged for a year, then the efficiency would be 0%. In contrast, the efficiency of charge/discharge cycling is often referred to as the roundtrip or cycling efficiency and depends greatly on application specific aspect of the cycles (depth of discharge, average state of charge), temperature, the charging characteristics and the discharge and charge rate. In general, it is advisable to evaluate and compare the efficiency of different electrochemical power sources at the systems level and for the application in mind. Power requirements of components necessary for the safe and reliable operation of the system must be included.

2.2

Coulombic efficiency and charge factor

The coulombic efficiency (dimension: C/C ¼ 1) refers to the ratio of the amount of electric charge that can be removed from a battery during the discharge cycle, to the amount of charge supplied during the charge cycle. Alternatively, the charge factor (reciprocal of coulombic efficiency) is used. The coulombic efficiency is 100% for electrochemical power sources where there are no side reactions, i.e. the discharging and charging current only convert active materials. Sodium-sulfur batteries are an example for this. As soon as

500

Electrochemical Terminology | Efficiency

side reactions occur, the coulombic efficiency falls below 100%. In systems where side reactions, i.e. corrosion reactions or electrolyte decomposition, are very small in comparison to the main charging reaction, the coulombic efficiency is close to 100% (e.g. for many types of lithium-ion cells). However, in systems with aqueous electrolyte, there are side reactions which increase considerably when the charging voltage increases. The battery current I is then divided into the main reaction current IMR for reconverting discharged active material to charged active material and the side reaction current ISR: I ¼ IMR + ISR Both the main and the side reaction current increase with charging voltage and temperature. At a high SoC and with increasing age the side reaction currents increase disproportionally. The side reaction current ISR can be estimated for a fully charged cell using the Butler-Volmer equation. Only the part of the Butler-Volmer equation which describes the discharge process needs to be considered because the electrode potential is considerably above the equilibrium potential for the side reaction (SR):
 azF ISR ¼ I0,SR exp ðE − E0,SR Þ RT Here I0,SR is the exchange current of the side reaction of the cell, a is the transfer coefficient, z is the total number of electrons transferred, F is the Faraday constant, R is the molar gas constant, T is the absolute temperature, E is the electrode potential and E0,SR is the equilibrium potential of the side reaction. As a “rule of thumb,” it is generally considered that reaction currents double for each 10 K rise in temperature. The side reaction current becomes most evident when the battery is nearly fully charged, i.e. when the main reaction current is small and approaching zero. Not surprisingly, an increase of voltage and/or temperature causes an exponential increase in the side reaction current and thereby decreases the coulombic efficiency significantly. The time integral of the current flowing into the side reaction constitutes the coulombic loss, and, in batteries with aqueous electrolytes, allows the calculation of the amount of water that needs to be refilled. The coulombic efficiency is the ratio of the integral of the main reaction (MR) current to the integral of the battery current (B). ZDt Coulombic loss ¼

ISR dt 0

ZDt IMR ðt Þ dt Coulombic efficiency ¼

0

ZDt IB ðt Þ dt 0

where the integral extends over the period of observation. In flooded lead-acid batteries, some gas evolution as a result of water electrolysis is actually useful to counteract acid stratification. A minimum charge factor is necessary for a long lifetime. Some typical charge factors for motive power batteries after a discharge with nominal conditions are listed in Table 1 showing the impact of design features on efficiency. Battery manufacturers often demand higher charge factors for laboratory testing to ensure full charge. For instance, the charging requirement for lead-acid batteries for SLI applications according to EN 50342 is a 24-h charge at 16 V with a current limit of 5 IN (nominal current). The charge factor is therefore influenced by the criterion used for determining the end of the charging process. It is not simple to determine the end of the charging process, as only the battery current can be measured and it is not possible to measure the current for the main reaction directly. A suitable end-of-charge criterion is, for instance, that there is no change in the charging current for a defined period of time at a constant charging voltage and temperature. If the charging process is terminated too late, then the charging factor is higher than it needs to be and the coulombic efficiency and consequently the electrical efficiency falls as well. Although batteries with aqueous electrolytes have significant side reactions at high charging voltages, their coulombic efficiency can be still high, if their state-of-charge and thus charging voltage is limited during operation. Measurements in a Toyota Prius have shown that the state-of-charge of the NiMH battery is maintained in a narrow band around 60% SOC and the side reaction therefore should be very low and the coulombic efficiency high, respectively. Table 1 Typical charge factors and coulomb efficiency for motive power applications of lead acid batteries after a discharge with nominal conditions and standard charging characteristics. Battery design

Charge factor

Coulombic efficiency

Flooded lead-acid battery (new) Flooded lead-acid battery (old) Flooded lead-acid battery fitted with an electrolyte circulation system Valve regulated lead-acid battery (gel or AGM)

1.15–1.18 1.25–1.4 1.07 1.05–1.07

85–87% 71–80% 93.5% 93.5–95%

Electrochemical Terminology | Efficiency 2.3

501

Voltage

As the current of a galvanic cell changes depending on the voltage and state of charge, the concept of average voltage during charging is difficult to use, in particular as soon as the charging current is reduced at the end of the charge. An important factor is the internal resistance of the cell which is, however, strongly dependent on current direction and state of charge, mainly as a result of the influence of the Butler-Volmer equation. It is possible to calculate the ratio of average voltage during discharging and average voltage during charging, but the value is difficult to interpret.

2.4

Factors influencing energy efficiency

To determine the energy efficiency, it is necessary to measure (i) current and voltage and (ii) to have detailed knowledge of the state of charge, so that the value of SoC at the beginning and end of measurement are the same. Obviously, measurements starting with a fully charged battery and ending with a fully charged battery can easily be evaluated. When data for other SoC ranges are required, in particular where side reactions exist or self-discharge processes are relevant, the uncertainty of a reaching a different SoC at the end of measurements needs to be evaluated and an average over many measurements should be considered. In addition, energy efficiency must be measured taking all necessary systems components into account. The efficiency of a single cell and of a battery consisting of many cells connected in series and parallel will be different. The main factors can be summarized as follows:

•

Temperature: An increase of 10 K typically doubles the rate of all chemical reactions (Arrhenius law). Under otherwise constant charging conditions, therefore, the rate of side reactions is also doubled and the time for charging is increased. At very high temperatures, the current due to the side reactions in Ni/Cd batteries, for instance, is so large in comparison with the main reaction that full charge cannot be achieved.

In contrast, the resistance values, in particularly of the electrolyte and the charge-transfer resistance associated with the ButlerVolmer-equation decrease with increasing current. Fig. 1 shows the electrical efficiency of a 2.1 Ah lithium-ion cell at different temperatures during cycling.

•

•

Age of the battery. Side reaction currents tend to increase with the age of batteries. For example, in the case of a lead-acid battery, inhibitors lose their effectiveness while impurities from corrosion processes or topping-up water catalyze side reactions. IEC 21/455/CD:1998 gives the side reaction rates for new and used lead-acid batteries. For most lead-acid battery types, an increase of the current (flowing due to the side reaction) during float operation by a factor of 5 can be expected by the end of life. In addition, in most battery chemistries, the internal resistance increases with age. Charging characteristics. The time spent at high charging voltage has a strong influence on the charge factor. In lead-acid batteries, the charge factor needed to achieve full charge and elimination of electrolyte stratification is claimed to be in the range of 1.05–1.07 for IR-free charging, pulse charging and electrolyte circulation, whereas it is ca. 1.15–1.17 for standard CC/CV/CC charging. 1 0,99 0,98

Energy efficiency

0,97 0,96 0,95 0,94 0,93 0,92 0,91 0,9 0

5

10

15

20

25

30

35

40

Temperature for charging and discharging Fig. 1 Measurement of (electrical) energy efficiency of a 2.1 Ah high-rate lithium-ion cell (LiCoNiMnO2/Carbon) at different temperatures (identical ambient temperature during discharging and charging) but otherwise identical conditions during charging and discharging. The discharge was interrupted after a discharge by 10% SOC to measure the open circuit voltage after a rest period, and the charge was also interrupted at the same SOC values.

502

Electrochemical Terminology | Efficiency

1 0,99 0,98

Efficiency (%)

0,97 0,96 0,95 0,94 0,93

0,92 0,91 0,9 0

1

2

3

4

5

Current (A)

6

7

8

9

10

Fig. 2 Measurement of (electrical) energy efficiency of a 2.1 Ah high-rate lithium-ion cell (LiCoNiMnO2/Carbon) during constant current discharging and charging at different current rates. The current rate during charging and discharging was identical.

•

Application. Factors influencing the charge factor and electrical efficiency are depth of discharge, current, time on float charge, and time during discharge. Fig. 2 shows the electrical efficiency of a 2.1 Ah lithium-ion cell at room temperature, which has been charged and discharged at 1C, 2C and 4C.

Standby losses of batteries during float charge operation are generally low. However, in lead-acid batteries, where are significant side reactions, standby losses are higher than for many other batteries. Nevertheless, for new lead-acid batteries, the standby losses are below 0.1% of the energy content of the cell per day.

3

Efficiency of supercapacitors

The coulombic efficiency of supercapacitors is close to 100%. Important factors affecting the energy efficiency of a supercapacitor are internal losses, self-discharge processes, leakage currents, and systems components such as DC/DC converters and balancing systems.

3.1

Internal losses

The main losses when cycling a supercapacitor are caused by the internal resistance, also known as the equivalent series resistance (ESR). The ESR is usually given in data sheets or may be measured using impedance spectroscopy at 1 kHz or constant-current discharge methods. Typical values are 0.3 milliohm for large supercapacitors (3000 F) and 0.8 milliohm for intermediate supercapacitors (650 F). The voltage drop across the equivalent series resistance decreases the voltage efficiency and it becomes clear that the electrical efficiency depends on the current amplitude. Despite the small resistances, thermal losses (I2R) may still be high due to the massive currents involved (several hundred amperes) when using the full power capability of supercapacitors. Ashtiani1 shows that under very high rate constant current charge and discharge conditions, the efficiency of asymmetric supercapacitors can be as low as 35%. In principle, the electrical efficiency of supercapacitors depends on the same parameters as the electrical efficiency of batteries, but in practice the dependence on temperature is low. Fig. 3 shows the efficiency of a supercapacitor discharged and charged with a constant current.6 When the charging and discharging currents are low (discharge time of 1000 s), the electrical efficiency is close to 100% because standby losses are not yet relevant. For very high charging and discharging currents, however, the electrical efficiency falls to 85% because the voltage rise is faster than the capacitance-forming electrode processes (by resistive losses due to electrolyte, charge-transfer reactions and pore diffusion). In hybrid vehicle applications, Faggioli et al.2 have reported electrical efficiency values of 92% and 85% for two different fuel cell hybrid vehicles.

Electrochemical Terminology | Efficiency

503

1 0.95 0.9 0.85

K

0.8 0.75 0.7 0.65 0.6 0.55 0.5 100

101

t (s)

102

103

Fig. 3 (Electrical) energy efficiency during constant current charging and discharging with different current rates (cycle durations) of a 2600 F supercapacitor with an equivalent series resistance of 0.7 milliohm. The supercapacitor was discharged to 50% of its maximum voltage to yield 75% of its stored energy. Courtesy of P. Barrade and A. Rufer. From Barrade, P.; Rufer, A. Current Capability and Power Density of Supercapacitors: Considerations on Energy Efficiency; EPE 2003, 10th European Conference on Power Electronics and Applications. Toulouse.

3.2

Self discharge processes and standby losses

The standby losses of supercapacitors are a few micro to milliamperes, e.g. 12 mA leakage current at nominal voltage for a large supercapacitor (Maxwell 3.4kF BCAP3400 P3003), and are sufficient for keeping them fully charged. Comparing the leakage current of this capacitor to the capacitance of 4080 F at nominal voltage, shows that the leakage current leads to ca. 8.5% loss of charge and a proportional reduction of voltage per day. Self-discharge is therefore an important performance factor when using supercapacitors and the energy efficiency is low when there are long periods of standby operation.4 Compared to batteries, the standby losses of supercapacitors are high particularly if the comparison with batteries is based on the energy content of the two systems.

3.3

System components

Supercapacitors connected in series for supplying high voltages need to be kept balanced to achieve a long lifetime. For this reason, supercapacitors connected in series are often discharged via the balancing system. The drop in energy efficiency caused by this must be taken into account when comparing the self-discharge of supercapacitors with batteries. When using DC/DC converters to supply a stable load voltage and a charging system, the electrical systems efficiency is reduced further.

4

Efficiency of fuel cells

The definition of energy efficiency used for batteries and supercapacitors cannot be used for fuel cells as there is no input of electrical energy to charge it. In order to calculate “electrical energy efficiency,” the enthalpy DH of the fuel cell reaction must be used as energy input term. It is thus possible to define the theoretical electrical efficiency of a fuel cell thermodynamically. This is given by the formula
¼

DG T DS ¼1− DH DH

where DH is the enthalpy, DG the Gibbs free enthalpy and DS the entropy change (and T DS the reversible heat) of the reaction. For hydrogen and oxygen as fuels, the theoretical electrical efficiency at room temperature is 83%. The thermodynamic values, and

504

Electrochemical Terminology | Efficiency

depend strongly on temperature. At the operating temperature of solid oxygen fuel cells, e.g. at ca. 800  C, the theoretical electrical efficiency is only 74%. Interestingly enough, the above calculation of efficiency shows that the theoretical thermodynamic efficiency of a fuel cell can be above 100%, for instance the reaction of formic acid with oxygen to water and carbon dioxide. HCOOH + ½ O2 ! CO2 + H2 O has a theoretical efficiency of 105.6% calculated from thermodynamic values. Here the number of gaseous particles increases, and thus the reaction entropy is DS > 0, but heat is consumed from the environment, and the cell cools down, which explains Z > 1 for this exothermic reaction (DH < 0). As long as the number of particles decreases, the cell reaction produces waste heat (T DS < 0, and |DG| < |DH|), which is true for the hydrogen-oxygen fuel cell. Obviously, the theoretical electrical efficiency cannot be obtained in real systems, because the losses due to reaction kinetics must not be neglected. A fuel cell is not a combustion engine, but battery! The practically achievable electrical energy efficiency therefore is defined by replacing DG with the output of electrical energy per mole of H2 (E ¼ cell voltage).
¼

electrical output per mol H2 zFE combustion heat of hydrogen DH

Heat is often a useful by-product of the operation of fuel cells, and therefore the term thermal efficiency is also important. In analogy to the electrical efficiency defined above, the thermal efficiency is:
¼

thermal energy supplied energy content of fuel

The thermal efficiency may differ significantly from the electrical efficiency. In fuel cells used for combined heat and power (CHP) applications, the ratio of electrical and thermal output power is an important design parameter for CHP plants and influences the electrical, thermal and overall efficiency of the plant. The total efficiency of a fuel cell or any other power generator for supplying heat and electricity to loads is
tot ¼

Electrical energy supplied + thermal energy supplied Energy content of fuel

The total efficiency of a fuel cell used for CHP applications, sometimes also called fuel utilization or fuel efficiency, can be as high as ca. 90%. In stationary applications where heat and electricity are the products, the electrical efficiency and the thermal efficiency are both normally given. Viessmann7 provides the following data for its combined heat and power fuel cell system Vitavalor 300-P for domestic applications: Output power: 750 W (electric): 1 kW (thermal), total energy efficiency: 90% (lower calorific value, LCV), electrical efficiency: 37%. The efficiency of fuel cells is of course lower than the theoretical value. Some of the well-known reasons for lower practical efficiency are:

• • • •

High charge transfer overpotentials as given by the Butler-Volmer equation Diffusion overpotentials (because the concentration of reactants decreases along the flow field and through the gas diffusion layer). Electrical resistance of components Crossover of fuel from the cathode to the anode, due to diffusion of hydrogen through the membrane

Most of these factors are also influenced by temperature. With increasing temperature, the practical electrical efficiency increases due to lower losses and overpotentials whereas the theoretical electrical efficiency decreases (see Fig. 4 for an SOFC-system as example). Please note: the efficiency values given in this diagram are approximate values for complete systems. In contrast to batteries and supercapacitors, fuel cells always require auxiliary components such as pumps, compressors, heating and cooling devices. In addition, the fuel utilization rate for most fuel cell systems is typically in the range of 60–80%. This is because fuel leaves the fuel cell unused, and must therefore be returned or burnt in an afterburner. As a result, the efficiency of a fuel cell must be determined on the systems level, taking into account the need to re-circulate (or dispose of ) unused fuel. This implies that efficiency data should be given for a stack or for an entire system. Corbo and coworkers5 give an interesting assessment as a function of output power. In practical systems, the auxiliary components continue to consume energy at low power and even during standby operation. As a result, the efficiency is very low. With increasing power, however, the energy consumption of the auxiliary components remains constant (or increases only a little) and the efficiency increases. Finally, as the power output increases further, the internal losses become significant and the voltage and thus efficiency falls. The resulting efficiency curve has the characteristic shape shown in Fig. 5. The efficiency of a fuel cell stack without auxiliary components and of an internal combustion engine is shown for

Electrochemical Terminology | Efficiency

505

Fig. 4 Theoretical (thermodynamic) and real electrical efficiency (H2-air and natural gas - air), approximate values only, of a SOFC fuel cell systems as a function of temperature.

Average power of NEDC

E ff ic ie n c y η

Fuel cell stack Fuel cells ystem

Internal combustion engine

Power City driving

Motorway driving Acceleration

Pmax

Fig. 5 Schematic diagram of the theoretical and practical electrical efficiency curve of a fuel cell and an internal combustion engine.

comparison. The advantages of fuel cells for vehicle operation where the average power is close to the maximum efficiency point of a fuel cell become clear. The electrical efficiency at maximum power is very similar, but the fuel cell has a higher efficiency at lower power, in particular, the average power required by a vehicle for the new European drive cycle (NEDC) is close to the maximum electrical efficiency of the fuel cell.

5

Conclusion

There is very little published information on the efficiency of electrochemical power sources as a function of operating conditions, e.g. for different current rates and temperatures as shown in Figs. 1 and 2. A number of analyses do, however, exist in the field of automotive applications. These provide average efficiency values or quantify changes of overall efficiency for different types of energy supply systems.

506

Electrochemical Terminology | Efficiency

References Ashtiani, C.; Wroght, R.; Hunt, G. Ultracapacitors for Automotive Applications. J. Power Sources 2006, 154, 561–566. Faggioli, E.; Piergioargio, R.; Danel, V.; Andrieu, X.; Mallant, R.; Kahlen, H. Supercapacitors for the Energy Management of Electric Vehicles. J. Power Sources 1999, 84, 261–269. Maxwell, Data sheet for 3.4kF BCAP3000 P300. Liu, K.; Yu, C.; Guo, W.; Ni, L.; Yu, J.; Xie, Y.; Wang, Z. Recent Research Advances of Self-Discharge in Supercapacitors: Mechanisms and Suppressing Strategies. J. Energy Chem. 2021, 58, 94–109. 5. Corbo, P.; Corcione, F. E.; Migliardini, F.; Veneri, O. Experimental Assessment of Energy-Management Strategies in Fuel Cell Propulsion Systems. J. Power Sources 2006, 157, 799–808. 6. Barrade, P.; Rufer, A. Current Capability and Power Density of Supercapacitors: Considerations on Energy Efficiency; EPE. In 10th European Conference on Power Electronics and Applications. Toulouse; 2003. 7. Viesmann, Data sheet for Vitavalor 300P CHP plant.

1. 2. 3. 4.

Electrochemical Terminology | Self-Discharge H Wenzla, R Bengerb, and I Hauerc, aConsulting for Batteries and Power Engineering, Osterode, Germany; bResearch Center Energy Storage Technologies, Clausthal University of Technology, Clausthal-Zellerfeld, Lower Saxony, Germany; cChair of Electrical Energy Storage Technology, Institute of Electrical Power Engineering, Clausthal University of Technology, Clausthal-Zellerfeld, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of H. Wenzl, BATTERIES | Self-Discharge, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 407-412, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00046-0.

1 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 3 4 5 5.1 5.2 5.3 5.4 5.5 6 7 8 9 References

Introduction Self-discharge processes Coupled anodic and cathodic reactions Redox shuttle in batteries Leakage current and short circuits Cross over Diffusion constraints Oxidation processes Energy consumption of auxiliary components Benefits of self-discharge and side reactions Self-discharge of batteries Self-discharge of lead–acid batteries Coupled anodic and cathodic reactions of the active materials Redox shuttle process Reactions between positive active material and separator Leakage currents and short circuits Reduction of self-discharge processes Self-discharge of supercapacitors Self-discharge of fuel cells Self-discharge of redox flow batteries Conclusion

508 508 508 509 509 509 510 510 510 510 511 512 512 513 513 513 514 514 515 515 516 516

Abstract This article describes the processes causing a reduction of the amount of charge or electrical energy which is stored in an electrochemical power source, and their relevance under operating conditions. Electrochemical systems that have to be kept at a higher temperatures level than the ambient temperature (e.g., NaS or NaNiCl batteries), also suffer of a “thermal self-discharge” due to heat losses.

Key points

• • •

Overview of electrochemical terminology Definition of the concept self-discharge Reversible and irreversible self-discharge processes

Nomenclature

I IMR ISR Q U t

Electric current (A) Main reaction current (A), which leads to providing or storing electric charge Side reaction current (A), which does not lead to providing or storing electric charge Electric charge (C ¼ A s), capacity (1 Ah ¼ 3600 C) Voltage (V) Time (s)

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00093-1

507

508

1

Electrochemical Terminology | Self-Discharge

Introduction

Self-discharge processes decrease the amount of available electric charge (capacity) of a battery. The energy content of any electrochemical power source drops in open-circuit operation without current flow through an external circuit, i.e., no load is connected between the positive and negative poles. Some self-discharge processes are reversible and the amount of charge or energy can be restored by charging. Other processes are irreversible and therefore are part of aging processes. Some self-discharge processes not only lead to a loss of capacity or energy but also to a loss of performance characteristics, in particular an increase of the internal resistance, e.g., growth of SEI layers in lithium-ion batteries. A high self-discharge rate is a disadvantage in all applications. Batteries with a high self-discharge rate have shorter service life and require monitoring and recharging to avoid deep discharge. Self-discharge also decreases the energy efficiency of the electrochemical power sources, particularly if the power source has to be maintained in its fully charged state for a long period of time, e.g., in emergency power supply systems. The relevance of self-discharge reactions is therefore specific for the application and should be evaluated at a systems level considering auxiliary components and their energy consumption. Self-discharge processes reduce the energy efficiency of all electrochemical power sources. In high power applications with many charge and discharge cycles, the capability to provide or consume energy with high rates can be more important than the loss of energy by self-discharge. Some self-discharge processes make it possible to equalize cells connected in series and have therefore a beneficial effect on monitoring and operation. For instance, cell balancing systems are not required for lead acid batteries and other electrochemical systems with aqueous electrolyte due to such self-discharge reactions, but are required for lithium-ion-batteries.

2

Self-discharge processes

The charged active materials of batteries have high Gibbs free energy. Not surprisingly, therefore, such materials tend to react with other materials of lower Gibbs energy (electrolyte solution, grid, separator, additives, and impurities). Self-discharge reactions may be either chemical or electrochemical, and may also be reversible or irreversible in nature. In this article, aging processes which lead to an irreversible loss of active materials and thus loss of capacity are also considered to be self-discharge reactions. There are different self-discharge processes with different characteristics and they should be clearly distinguished.1 Their relevance for different electrochemical power sources differs, and some of the processes discussed below may have little technical relevance in most applications. Important types of self-discharge reactions are as follows:

2.1

Coupled anodic and cathodic reactions

Coupled anodic and cathodic reactions take place simultaneously at the same electrode in batteries and fuel cells. They convert charged active material to discharged active material and lead to a mixed potential. Fig. 1 shows an equivalent circuit diagram of a cell with coupled reactions on both electrodes. Each reaction is represented by an ideal voltage source (E0 is the thermodynamic equilibrium voltage of the reaction) and a current dependent resistor R, which represents the charge transfer overvoltage according to the Butler-Volmer equation. The main reaction (MR) on both electrodes is the normal discharge reaction of the cell, and has a higher potential E0,MR and a significantly lower internal resistance than the side reactions (SR). The main reactions provide the energy via their discharge reactions and drive the side reactions which consume energy. Despite there is no external current flow, the current through the main reaction leads to an electrode potential different from the equilibrium voltage of the main reactions. -

+ R

Eo,MR,-

Eo,SR,-

R RCT,MR,+

RCT,MR,RElectrolyte

RCT,SR,-

RCT,SR,+

Eo,MR,+

E0.SR,+

R – ohmic resistance of all components RCT – charge transfer resistance Fig. 1 Equivalent circuit of a cell with self-discharge reactions on both electrodes. E0 is the thermodynamic equilibrium voltage of the reactions. RCT is the charge transfer resistance, which can be calculated using the Butler-Volmer equation. The indices MR and SR denote the main reaction and side reaction and (+) und (−) denote the positive and negative electrode. R is the ohmic resistance of all electron conducting components in contrast to the electrolyte resistance. The circular current flow as a result of the different equilibrium potentials of main and side reactions is indicated. For simplicity reasons, the double layer capacitance generated by the electrochemical reactions is not shown. The type of reactions depends on the battery chemistry.

Electrochemical Terminology | Self-Discharge

509

The electrode potential depends on both reactions and is called a mixed potential which depends on the kinetic behavior of the reactions described by the Butler-Volmer equation. As in all electrochemical reactions, the current is a function of (i) the surface area, (ii) the exchange current density, which depends on the activity of the reactands, a function of their concentration, and impurities acting as catalysts, and (iii) the voltage difference between the equilibrium potential and electrode potential. As self-discharge reactions take place without an electrical connection between the poles (assuming the electrical conductivity of the electrolyte is very low) and there are no other electron conducting connections like internal shorts, it is clear that the self-discharge reactions on the anode and the cathode are independent of each other and may lead to different discharge rates for the two electrodes. The reactions may differ profoundly depending on the type of reaction and rate of reaction. As a result, the state of charge of the electrodes may become unbalanced and recharging may become problematic. The higher the cell voltage, i.e., during charging or at a high state of charge, the higher the self-discharge rate will be. This self-discharge process is usually higher for power-oriented batteries compared to energy-oriented batteries owing to their larger electrochemically active surface area. In addition, aging will tend to increase the self-discharge rate because of catalytic impurities, degradation of inhibitors and micro-cracks on anode or cathode passive layers between the active materials of the electrodes and the electrolyte.

2.2

Redox shuttle in batteries

Redox shuttle reactions transfer charge from one electrode to the other (see Fig. 2). Ions in the electrolyte are oxidized at one electrode, move to the other electrode by means of diffusion and migration processes, where they are reduced and return via diffusion and migration processes. In this way, both electrodes are discharged by the same rate. In lithium-ion batteries redox-shuttles aim at preventing overcharging. Here, the redox shuttle is an electrolyte additive that can be reversibly oxidized/reduced at a characteristic potential and provides an intrinsic overcharge protection for the battery. During normal operation, the redox potential of the redox shuttle is not reached and the molecules stay inactive. When the cell is overcharged, the potential of the positive electrode increases, and the redox cycle of the redox shuttle molecules is then activated.2

2.3

Leakage current and short circuits

Leakage currents and internal short circuits with a high resistance cause slow discharging of a cell without external current flow across a load. Leakage currents in new batteries and supercapacitors are the result of faulty designs or material defects. Poor maintenance or adverse environmental conditions can lead to the formation of a conductive path on the outside of the cell container and thus to current flow between the poles. A more severe problem are internal short circuits, the formation of an electron conducting path between the electrodes of a cell inside the cell container either by defects in the separator or by dendrites. Dendrites developing on one electrode, usually, the negative electrode, may also cause short circuits across the separator. In lithium-ion batteries, short circuits by dendrites as a result of metallic lithium deposition at the anode (so-called plating) or the creation of copper dendrites at the cathode by deep discharge are severely relevant to safety and lifetime. To detect float or leakage currents, shorts and micro-shorts, multiple approaches have been discussed in recent literature.2–4

2.4

Cross over

Diffusion +

+

R

R

e—

e—

R

R Diffusion

Negative electrode

Positive electrode

In redox-flow batteries and, fuel cells, charged active material (either as an ion or in the case of fuel cells as a molecule) may migrate or diffuse through the separator and reach the other electrode and react with the charged active material there, however without making electrons available to the external load.

Fig. 2 Schematic diagram of a redox shuttle process. R denotes a molecule or ion that is oxidized at the positive electrode and reduced at the negative electrode.

510 2.5

Electrochemical Terminology | Self-Discharge Diffusion constraints

The pseudocapacitance of supercapacitors cause by reversible redox reactions between electrons from the electrode and surface-active molecules adsorbed on the electrode surface with ions from the electrolyte is a diffusion-limited process and the cause of a number of different self-discharge processes.

2.6

Oxidation processes

In batteries, supercapacitors, fuel cells, and redox-flow batteries—the positive active material of all electrochemical power sources has a high oxidation potential. Chemical reactions between components of the cell and active materials lead to the loss of active material or increased internal resistance, and therefore reduce the energy content of the cell.

2.7

Energy consumption of auxiliary components

Energy consumption occurs in batteries, redox-flow batteries, supercapacitors, and fuel cells, A number of electrochemical power sources require auxiliary components for their operation (e.g., pumps for fuel cells and redox-flow batteries), monitoring systems, and cell balancing systems (e.g., lithium-ion batteries and supercapacitors). If these systems are operated without an external power supply and the energy demand has to be provided by the energy content of the electrochemical power source, then this energy loss is a self-discharge process which is completely reversible and has a direct impact on the overall energy efficiency. Typically, the self-discharge rate is proportional to the amount of active material remaining, and, as in any process where the rate of change depends on the amount of substance, exhibits an exponential decay curve. Quantitative descriptions of the self-discharge rate as a percentage per time reflect this. But in electrochemical systems where the energy of the charged active material is distributed over a wide range of values (as in intercalation compounds or mixed valence salts), the self-discharge rate may vary in a complicated way with time. Yazami and Ozawa5 showed this complex behavior for both the open-circuit voltage (OCV) of spinel electrodes in Li/LixMn2O4 cells and the self-discharge currents as a function of voltage. In all cases, however, the overall self-discharge rate falls with time. When investigating the self-discharge rate of individual electrodes, it is sometimes possible to measure the rate at which the concentration of a product of self-discharge increases, e.g., the volume of oxygen that has been generated as a result of a self-discharge reaction or the float current that is required for maintaining a given voltage. All batteries which are not connected to loads or chargers exhibit a slow decrease in cell voltage as a result of self-discharge reactions and the difference from the desired voltage level can then be used to recharge cells to maintain them at an appropriate level of state of charge. Self-discharge is typically assessed by keeping a battery or supercapacitor in a load-free state for a fixed period of time at a specified temperature (e.g., 6 months and 20–30  C for stationary lead–acid batteries according to EN 60896, or 21 days at 40  C for lead–acid starter batteries according to EN 50342), and then subsequently measuring the remaining amount of charge.

3

Benefits of self-discharge and side reactions

As a rule, self-discharge processes are undesirable as they lead to a reduction of capacity and energy content during a standstill, and usually cause also some irreversible changes. However, overcharge reactions, i.e., side reactions, which play an important role at the end of charging and cause relatively little damage over the lifetime of a battery, contribute to self-discharge but can also be useful for charging cells connected in series and equalize cell voltages. All batteries with aqueous electrolyte have such side reactions, i.e., the hydrolysis of water. Where such side reactions are not an integral part of the electrochemical system, additives are sometimes used to encourage a low rate of redox shuttle reactions for overpotential protection. At the end of charging and at a high state of charge, the cell voltage of batteries increases quickly. The rate of self-discharge reactions increases also with voltage. The battery current is the sum of the current for the main charging reaction and the current for the self-discharge reaction. I ¼ IMR + ISR where I is the battery current, IMR the current that leads to the conversion of discharged active materials to charged active materials (main reaction), and ISR is the self-discharge or side reaction current. All cells with low-rate side reactions, e.g., electrolysis of water in aqueous electrolytes, can be charged fully even if they are connected in series. As the current through all cells of a battery is identical, small changes in capacity lead to variations in cell voltage. The charging current can be maintained at a low level until the last cell has been fully charged. The surplus current for cells (respectively electrodes), which are already fully charged, is simply used up by the side reaction. In contrast, the voltage of electrochemical systems without appropriate side reactions increases at the end of charging. This can destroy a cell if the cell voltage is not monitored and not limited. The charging process ends as soon as the first cell or electrode is fully charged (respectively its voltage reaches a threshold value). All other cells of the battery cannot be charged further. As overpotential protection for such batteries, additives that act as redox shuttles have been introduced. Once the potential at which they can be oxidized or reduced is reached, they transport charge from one electrode to the other. A constant current charging regime

Electrochemical Terminology | Self-Discharge

511

can then be introduced, which limits the cell voltage and can also lead to an equalization between all electrodes and cells. These additives will contribute to the overall self-discharge as long as the cell voltage is above the redox potential range of these additives. In an overview of safety mechanisms by Balakrishnan and coworkers,6 the importance of this process for over-potential protection is described. The operation is not affected by the introduction of such redox shuttles as long as the operating voltage range remains outside the redox potentials of the additives. As an alternative to intrinsic overcharge protection, some battery system provide an electronically controlled bypass for each cell to allow full charge of all cells.

4

Self-discharge of batteries

Self-discharge reactions in batteries sometimes are reversible, i.e., the products of the self-discharge reactions can be restored to their charged states without any aging effects. In other cases, the self-discharge reactions cause permanent changes of the active or passive components of a cell and reduce both calendar life and cycle life. A good example of the irreversible effects is the growth of a solid electrolyte interface (SEI) on the negative electrode of lithium-ion cells. This process consumes lithium, and thus diminishes the available capacity. Interesting data on the relationship between reversible and irreversible self-discharge effects are, for instance, given in a paper by Yazami and Ozawa.5 Self-discharge processes that do not correspond to the normal discharging reactions may lead to passivating layers and other effects, which impact performance. The capability of a cell to respond to power requirements after a long period of self-discharge in the same way as a cell that has been discharged to the same state of charge may be lower. Details are available in data sheets. Self-discharge processes always reduce the capacity. If the results of self-discharge reactions are, for instance, passivating layers, the energy content, and in particular the power of the cell, decreases much more than the capacity. Usually, the self-discharge rate of a battery is reported as a percentage loss of capacity per month or per year. Some typical values of self-discharge rates are given in Table 1. As the rates of all reactions are temperature dependent, storage at low temperature decreases self-discharge considerably. Fig. 3 shows the self-discharge rate for a number of primary systems. For primary batteries that are used to supply power to electronic devices at a low rate for many years, the self-discharge rate must be very low, cf. the lithium-manganese dioxide cell in Table 1 as an example. These batteries tend to have a very long shelf life and, even after a few years of storage without recharging, the capacity of the cell is still very high and all other performance criteria are within acceptable limits. Secondary batteries tend to have higher self-discharge rates and lower shelf life. Batteries that are stored without electrolyte between the electrodes, for instance, reserve batteries, or batteries that require a molten electrolyte at elevated operating temperature, and are stored with the electrolyte as a solid at low temperatures, have virtually no self-discharge. When high-temperature batteries (Na–S, ZEBRA) are maintained at their operating temperature, heat losses are important and can be called thermal self-discharge processes. The elevated operating temperature causes heat losses on the order of 90 W for the battery type Z5 (21.2 kWh) by MES-DEA. This is despite the excellent thermal isolation of such battery types. Depending on ambient temperature, the thermal loss has to be covered in remote autonomous power systems by discharging of the battery for operating a heater. In order to minimize thermal self-discharge in high-temperature cells, the operating strategy and thermal management have to be coordinated with each other. In the case of Na-S batteries, the heat released in the exothermic discharge process is used to maintain the high temperature. The operating strategy can then be designed in such a way that sufficient but also not too much heat is released. Most high-capacity lithium-ion batteries with many cells connected in series have built-in electronic control circuits that run constantly. These necessary control systems increase the discharge rate by as much as an additional 3% per month as they should be activated for safety reasons even when not connected to an external power supply. Modern battery management systems therefore try to reduce their idle current. This can be achieved by switching off most of the BMS-functions and repeated wake-up routines. Self-discharge reactions must be significantly slower than normal charging reactions; otherwise, it is not possible to charge a battery efficiently. For example, the inability to recharge Ni–Cd batteries fully above ca. 60  C is due to very high self-discharge rates at high temperatures.

Table 1

Typical values of self-discharge rates at room temperature.

Battery type

Self-discharge rate

Lithium-ion Ni–MH Ni-MH (low self-discharge type) Ni-Cd Lead-acid batteries Lithium-manganese dioxide (non-rechargeable)

discharge

Liz C6 , 0  z  1

(II)

Here, z describes the mol fraction of Li+ ions inside the negative electrode. As a result of these electrochemical charge-transfer reactions, Li+ ions must cross the electrolyte under current-flowing conditions (see Fig. 2). The electrolyte in lithium-ion batteries is based on a dissociated lithium-containing salt, for example, lithium hexafluorophosphate (LiPF6) or lithium perchlorate (LiClO4), which is an ionic-conductive medium. Various mixtures of ethylene carbonate (EC), diethyl carbonate (DEC), and dimethyl carbonate (DMC) are used as nonaqueous solvents. The ions in the electrolyte are transported by both diffusion and migration, the latter process being induced by the electric field between the electrodes across the electrolyte. The overall main electrochemical storage reaction can be represented as LiCoO2 + Liz C6

charge

>

discharge

Li1 −x CoO2 + Liz+x C6

(III)

An important battery characteristic used in SoC indication is the OCV when the battery is in equilibrium. This is defined as the difference between the equilibrium potentials of the positive and negative electrodes, according to U 0 ¼ DE ¼ E+ − E−

(2)

During current flow the battery voltage is not equal to the equilibrium value. Fig. 3 illustrates the development of the discharge voltages of a lithium-ion battery (lithium–cobalt oxide/graphite chemistry) at various discharge rates. As can be seen in the figure, the higher the discharge rate, the lower the voltage will be. Note that another important dependence but nonetheless relevant for SoC indication based on voltage measurement is that on temperature. The difference between the observed battery voltage during (dis)-charging and the open-circuit voltage is the summation of ohmic-drop voltage UO and overpotential , as defined by U O +  ¼ U ðIÞ − U 0

(3)

where UO and  have equal signs that depend on the direction of the current that flows through the battery (positive for charging, negative for discharging). Part of the observed overpotential  can be attributed to kinetic limitations in both electrodes (k), whereas the remaining part represents mass transfer limitations (diffusion, d):  ¼ k + d

(4)

Fig. 4 illustrates the kinetic limitations in the negative and positive electrodes of a general rechargeable battery accordingly in assumed absence of mass transport limitations. Anodic currents Ia are shown in the upper half and cathodic currents Ic are shown in the bottom half. These currents are equal in value, defined as exchange current I0, and opposite in direction in a state of equilibrium,

Voltage (V)

4.0

1C

0.5C

0.2C

2C

3.5

4C 3.0

0

200

400

600

800

Capacity (mAh) Fig. 3 Development of the discharge voltage (V) as a function of discharged capacity (mAh) for a cylindrical Li-ion battery (lithium–cobalt oxide/graphite chemistry) at 25  C at various discharge rates.

540

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

Ia,−

I Ia,+ I o+

I o− Ic,− I o−

I

U0

k −

V

Ic,+ I o+

eq

E−

I

U(I)

eq E+ k +

E+

E−

Fig. 4 Current–voltage relation for the positive (+) and negative (−) electrode of a rechargeable battery. Mass transport limitations have been left out for simplicity.

resulting in a zero net current. In a state of nonequilibrium, however, one current is larger than the other. For charging, a solid line represents the net electrode reaction current and currents Ia and Ic are represented by dashed curves at each electrode. The battery voltage U(I) equals E+–E− and is larger than the battery equilibrium potential, U0. The direction of current at both electrodes is reversed during discharging. In that case, U(I) will be smaller than U0. A complicating factor for SoC-indication systems is the fact that the overpotentials depend on many factors, including current, SoC, temperature, and SoH of the battery. Therefore, many SoC-indication systems based on voltage measurement rely on the open-circuit voltage to get rid of the influence of current on the SoC-determination accuracy. One of the major problems of the lithium-ion battery chemistry is that a thorough (electro)chemical protection against both overcharging and overdischarging is not yet available. Recombination cycles, as employed in Ni–Cd and Ni–MH batteries, do not exist at the moment for lithium-ion batteries. However, it should be emphasized that the organic electrolyte can also be decomposed irreversibly at high voltages occurring during charging. Another dangerous process occurs at the positive electrode during overcharging. Charging the lithium-ion battery means a decrease in the mol fraction of Li+ ions in the positive electrode. When the lithium content becomes too low, the following decomposition reaction starts: Lix CoO2 !

ð1 − xÞ½Co3 O4 + O2  + xLiCoO2 3

(IV)

The active electrode material decomposes into inactive cobalt oxide (Co3O4) material, which will be formed at the surface of the lithium cobalt oxide electrode and will contribute to the increase of the battery impedance and to a decrease of the maximum storage capacity. These effects need to be taken into account when designing an accurate SoC-indication system. The occurrence of decomposition reactions implies that overcharging and overdischarging should be avoided under all circumstances in lithium-ion batteries and care should be taken that the battery voltage is never outside the range within which the basic electrochemical reactions occur. During charging, the constant-current, constant-voltage (CCCV) charging profile achieves this. It is shown in Fig. 5 and reveals the development of the cell voltage U(I), the applied current (I), and the stored capacity as a function of charging time. In the initial stage (CC mode), the battery is charged rapidly with a moderate constant current. The battery manufacturer generally prescribes the value of this current. The cell voltage gradually increases up to a value of 4.2 V. This value is considered to be the upper allowable limit and ensures a proper functioning of the battery. As soon as this limit is reached, the charging regime changes from amperostatic (CC mode) to potentiostatic (CV mode). As a result, the current decreases rapidly to lower values. Evidently, charging proceeds more slowly which can be recognized by the slower increase in stored capacity. At the end of the charging process, the current diminishes to very low values when the battery is fully charged. In total, charging may take up to 2 h. Recently, the concept of ‘boost charging’ was introduced by the present authors in order to cope with the customer request to charge lithium-ion batteries more quickly without introducing any negative cycle-life effects. In this way, for example, one-third of its rated capacity can be charged within 5 min. The specific charging profile for lithium-ion batteries is sometimes used as an advantage in SoC-indication systems. Other side reactions may occur in lithium-ion batteries as well. One of the most important is that leading to the formation of the solid electrolyte interface (SEI). The SEI films form on the surface of the negative carbon electrode due to reduction of the electrolyte solvent by tunneling electrons. The SEI has good Li+-ionic conductance but poor electronic conductance. Good ionic conductance is vital for the main storage battery reaction. Poor electronic conductance will prevent tunneling of electrons, which would induce reduction of the solvent and thereby increase the SEI layer thickness. The SEI is formed by insoluble lithium-based salts; thus some electrochemically active lithium becomes captured in SEI. The gradual capacity fade that lithium-ion batteries suffer can be partially attributed to this SEI formation. As a result of aging processes, the capacity of lithium-ion battery declines and the impedance grows, as described above. This also holds for other battery chemistries. Moreover, even the open-circuit voltage reveals some changes in behavior as a function of battery age, as shown in Fig. 6 (lithium–cobalt oxide/graphite chemistry). This figure shows that at 25.4% capacity loss by aging, the curve looks substantially different. All these effects need to be taken into account in the SoC algorithm to ensure accurate SoC indication, even when the battery ages.

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

5.0

541

150

2.0 Vbat

Capacity 1.0

100

2.0

Capacity (%)

1.5 3.0

I (C rate)

Voltage (V)

4.0

50 1.0 0

0.5

I

0

1

0

2 3 Charge time (h)

0

Fig. 5 Constant-current, constant-voltage (CCCV) charging regime typical for Li-ion batteries: voltage (V), capacity (%), and current (C rate) versus charge time (h).

4.2

Voltage (V)

4 3.8 3.6 3.4 3.2 3 100

Fresh Aged 5.4% Aged 25.4% 80

60

40

20 SoC (%)

0

Fig. 6 Open-circuit voltage in equilibrium (rest voltage U0) (V) obtained by voltage relaxation after charge steps versus normalized state-of-charge (SoC) (%) at 25  C obtained for fresh and aged batteries at 5.4 and 25.4% capacity loss (lithium–cobalt oxide/graphite chemistry).

3

State-of-charge-indication methods

A proper definition of the term SoC is important. A first distinction that must be made is between the charge inside the battery and the charge that is actually available to the user under the actual discharge conditions. Depending on these conditions, the difference between these two may be substantial, for example, when trying to discharge the battery at very low temperatures.

• • • • •

State-of-charge (SoC): The charge (in Ah) that is present inside the battery reflects an estimated value in most cases. The SoC can be expressed in percentage of the maximum possible charge; here 100% reflects a full battery and 0% reflects an empty battery. Remaining capacity Qr(t): The charge (in Ah) that is available to the user under the actual discharge conditions reflects an estimated value in most cases, which is equal to or smaller than the maximum possible charge. Remaining time of use (tr): The estimated time that the battery can supply charge to a load under the actual discharge conditions before it will stop functioning when the battery voltage will drop below the end-of-discharge voltage. Charge supplied to the battery (Qch): The charge (in Ah) supplied to the battery during charging, that is, the integral of charge current Ich over time. The value of Qch starts at zero at the start of each subsequent charge step. Charge obtained from the battery (Qd): The charge (in Ah) obtained from a battery during discharging, that is, the integral of discharge current Id over time. The value of Qd starts at zero at the start of each subsequent discharge step.

An SoC-indication system may estimate the battery’s SoC and/or capacity and/or tr; Variables Qch and Qd may be used in some form in the estimations. In the remainder of this section, the term SoC will be used as a collective noun for simplicity. The distinction between SoC and capacity will be made when applicable. The remaining time of use will be most interesting for the user of a battery-operated device. This may imply remaining talk time where a mobile phone is concerned, but alternatively may also be applied in electrical vehicles, in which case a more useful

542

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

definition would be remaining distance as defined by velocity multiplied by remaining time of use. The remaining time tr (in s) can be inferred from residual capacity in two ways, depending on the type of the load: tr ¼

Qr I

(5)

for a current-type load with value I and Z tr ¼

U’

U EOD

Qr ðU ÞdU (6)

P

for a power-type load with value P, where U0 expresses the battery voltage at the moment tr is estimated. The integral of residual capacity over the voltage range applicable during the subsequent discharge expresses the energy (in Wh) obtained from the battery. It is impossible to exactly anticipate future load conditions. Therefore, tr can only be inferred from the actual current or power consumption values at the moment it is calculated, assuming that this consumption will remain constant until the end-of-discharge voltage (EOD) is reached. As an alternative, a worst- and best-case value of tr can be displayed. The best case applies to the minimum expected load, while the worst case applies to the maximum expected load. For example, a cellular phone may indicate talk time left and standby time left. The current will increase for a power-type load when the battery voltage decreases during discharge. It is important for the user of a battery-operated device to know what the state of the battery is and whether it needs to be replaced or not. Therefore, some SoC-indication systems take SoH into account to compensate the SoC estimations for aged batteries.

3.1

Direct SoC measurements

The direct measurement method is based on a reproducible and pronounced relation between a measured battery variable and the SoC. Examples of such battery variables are battery voltage U, battery impedance Z, and voltage relaxation time (t) after application of a current step. All relations between battery variables and SoC depend on the battery temperature T, which should also be measured. The relation fTd between the directly measured battery variable at a given temperature and the SoC, can be stored in the system, for example, in the form of a lookup table. The basic principle of an SoC-indication system based on direct measurement is shown in Fig. 7A.

U, Z, τ, T

SoC = f Td (U, Z, τ)

(A) Coulomb counting I

∫Idt

U, T, I

bk

SoC = fU, T, I (∫Idt )

(B)

Update of information (Specific battery behavior) Basic set of information (Standard battery behavior)

Estimated battery behavior Ym + −

Adaptive control unit

Battery model description

Charge/discharge conditions: I, T, U

Battery: I, T, U

Observed battery behavior Yb

(C) Fig. 7 Basic principle of (A) a state-of-charge (SoC)-indication system based on direct measurement, (B) an SoC-indication system based on bookkeeping, and (C) an adaptive SoC-indication system.

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

543

The main advantage of a system based on direct measurement is that it does not have to be continuously connected to the battery. The measurements can be performed as soon as the battery has been connected, after which the SoC can be directly inferred from the function f T d . The main problem is determining the function f T d itself, which should describe the relation between the measured battery variable and the SoC under all applicable conditions, including spread in battery behavior. Conditions include the discharge current, which may vary quite a lot depending on the application, temperatures (e.g., outdoor use involves a substantially wider temperature range than indoor use), and storage times. Battery behavior depends strongly on these conditions, for example, for current dependence of the discharge voltage. In general, the greater the amount of variation in conditions in practical use, the less accurate a system based on direct measurement will be. The reason for this is that it is difficult to derive f T d for all conceivable conditions, including aging. Adaptive systems may be able to cope with battery spread and aging.

3.2

Bookkeeping systems

Bookkeeping systems are based on current measurement and integration. This can be denoted as coulomb counting, which literally means ‘counting the charge units flowing into (defined as Qch) or out of the battery (defined as Qd).’ This yields an accurate system when all the charge applied to the battery can be retrieved under any condition, at any time. However, this is not the case in practice. Therefore, other battery variables such as voltage and temperature also need to be measured. The basic principle of an SoC-indication system based on bookkeeping is shown in Fig. 7B. bk where ‘bk’ means bookkeeping and V, I, and T are parameters in the function, is based on the content of the The function fV,T,I coulomb counter, which is the integral of current I over time. Other battery variables (voltage V and temperature T ) are measured to compensate the content of the coulomb counter for the battery behavior. The value of current I itself is also used to compensate the counter. Examples of specific battery behavior for which the coulomb counter needs to be compensated in a bookkeeping system are the following:

• •

• • •

3.3

Charging efficiency: Side reactions occur at the end of charging in, for example, lead-acid, Ni–Cd, and Ni–MH batteries. As a result, not all charge applied to the battery is effectively stored. The charging efficiency depends on the SoC, I, and T. Discharging ‘efficiency’: Depending mainly on the SoC, T, and I, only part of the available charge inside a battery can be retrieved under the actual conditions, but may be retrievable under other, less demanding, conditions. In general, less charge can be obtained from a battery at low temperatures and/or large discharge currents; see, for example, Fig. 3. The battery age also influences the discharging efficiency, for example, due to an increased internal resistance. The compensated SoC value equals residual capacity under the actual discharge conditions. Self-discharge: Any battery will gradually lose charge, which becomes apparent when the battery is left unused for some time. A coulomb counter cannot measure this, as no net current flows through the battery terminals. The self-discharge rate of a battery depends strongly on temperature as well as on the SoC. Capacity loss: The maximum possible battery capacity (in Ah) decreases when a battery ages. In general, the more the battery is misused, for example, overcharged and overdischarged on a regular basis, the larger the loss will be. Voltage measurement is used to deal with capacity loss in most commercial bookkeeping systems. Storage effects/memory effect: Some types of batteries will temporarily show deviating behavior after long times of storage. The behavior will depend not only on the storage time, but also on temperature. Some types of batteries, such as Ni–Cd batteries, show a memory effect. This will lead to a temporal and reversible capacity loss, as opposed to the irreversible capacity loss due to aging.

Adaptive systems

The main problem in designing an accurate SoC-indication system is the unpredictability of both battery and user behavior. Battery behavior depends strongly on conditions, including age, some of which may be unanticipated. Moreover, spread in behavior of batteries of the same type makes life more difficult. A possible solution is to add adaptivity to a system based on direct measurement, bookkeeping, or a combination of the two. The basic principle of adding adaptivity to an SoC-indication system in Fig. 7C is the battery model description. The measured battery variables I, T, U are the inputs of this model, which estimates battery behavior in the form of output vector Ym on the basis of these inputs. Vector Ym contains at least the SoC, but could also contain additional battery variables, such as an estimated value of the battery series resistance. Another possibility would be to estimate the battery voltage on the basis of current and temperature measurements, and to compare this estimated value with the measured battery voltage. The model may contain the function f T d of ‘Direct Measurements’ or the function fU; T; I bk ‘Bookkeeping Systems’, or a combination of the two. The system starts with a basic set of information, which describes standard battery behavior for the type of battery concerned. Adaptivity of the model is based on a comparison of Ym with observed battery behavior in the form of vector Yb. This comparison is made whenever possible. It results in an error signal e, which is input to an adaptive control unit. The unit updates the information in the model by updating parameter values or even by changing the model description. As a result, the model is adapted on the basis of behavior specific to the battery to which the system is connected and the error between estimation and observation is minimized.

544

4 4.1

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

State-of-charge-indication systems Direct voltage measurement in practice

Many existing direct measurement SoC-indication systems are based on voltage measurement, for example, those in cellular phones, camcorders, and portable audio equipment. The battery voltage is then measured and translated into a number of bars shown on a bar graph. The accuracy and reliability of these systems is generally poor, especially when temperature dependence has not or marginally been taken into account and when the battery current varies a lot. Ohmic voltage drops and overpotentials depend on the battery type and chemistry, charge/discharge current, temperature, SoH, and variations in contact resistance in case the battery can be detached from the portable device. An example of the dependency of the battery discharge voltage on the discharge current is given in Fig. 3. For a given SoC, the battery voltage varies substantially with current. When this is not properly taken into account in the function fTd, the accuracy of the system will be compromised. As a general rule, the various ohmic drop and overpotential contributions to the battery voltage will increase when the battery ages. This also needs to be taken into account properly for a maintained accuracy over the lifetime of the battery. Another issue influencing the accuracy of a direct measurement system based on voltage measurement is the occurrence of a limited voltage variation for most of the discharged capacity. For example, the voltage variation from a full to empty Ni–Cd or Ni–MH battery is limited to 200 mV. When used in a battery pack with, for example, three Ni–Cd or Ni–MH cells in series, the voltage variation over the full discharge curve also increases, which is beneficial for accuracy. The discharge curve depicted in Fig. 8 illustrates why SoC systems based on voltage measurement will usually have a poor accuracy for a battery with a relatively flat discharge voltage curve. The voltage variation △U around voltage U can result from variations in current, resistance, or temperature. Due to the relatively flat discharge curve, this voltage variation will lead to a considerable variation △SoC around the estimated SoC value. As illustrated in Fig. 6, the rest voltage of a lithium-ion battery shows a gradual decrease of roughly 500 mV when discharged from 100% full to 10% full, after which a more steep shape occurs when the battery approaches the empty stage. Although the U(SoC) curve depends on cycle life and temperature, these dependencies are relatively small. This makes the voltage curve attractive for use as a measured battery variable in an SoC-indication system. Another important advantage is the absence of current and impedance influence on accuracy. For the same reasons rest voltage is often used for SoC indication for lead-acid batteries too, in which case the U0(SoC) is even more or less linear. A complicating factor in all SoC-indication systems based on U0(SoC) is how to determine the open-circuit voltage when no external current flows and all overpotentials have fully relaxed to zero. The first method is that of voltage relaxation. The battery voltage will eventually relax to the equilibrium voltage after current interruption. A complicating factor is how to determine that the battery voltage has fully relaxed, considering that the relaxation time depends on SoC, SoH, and temperature. The other two methods include linear interpolation, where the average battery voltage is determined from the battery voltages during charging and discharging with the same currents with opposite signs, and linear extrapolation, where the battery voltages obtained with different currents with the same sign and at the same SoC are linearly extrapolated to a current of zero value.

4.2

Impedance and chronopotentiometry

U

The battery impedance is a frequency-dependent complex quantity describing the relation between battery voltage and current. It can be measured in the frequency domain and in the time domain. In the former case, the alternating current (AC) impedance valid at a certain bias point is measured. This is referred to as impedance spectroscopy. In the latter case the complete battery voltage relaxation curve is measured and used as a criterion for the SoC. This technique is referred to as chronopotentiometry, which is the equivalent of impedance spectroscopy in the time domain.

ΔU U

SoC

SoC

ΔSoC Fig. 8 Schematic discharge curve of voltage versus state-of-charge (SoC) illustrating poor accuracy of voltage measurement–based SoC-indication systems for a relatively flat voltage curve.

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

545

Measurement of battery impedance as a function of frequency is not practical for SoC indication in a portable device, because a signal with a frequency sweep has to be applied. The impedance dependence on SoC is generally limited. For example, consider the battery discharge voltage curves for various discharge currents in Fig. 3. The voltage difference between two curves for different discharge currents is more or less constant over most of the discharge range. For this reason, this method is not applied in practice. For the same reason, chronopotentiometry is also not used. However, both impedance spectroscopy and chronopotentiometry are very useful for studying battery behavior in detail in a laboratory setup when it is used to characterize separate electrodes. The measurement of the resistive part of the battery impedance is sometimes used for indicating the SoH, since an increased series resistance indicates battery wear-out. For example, this is used in industrial applications such as uninterruptible power supplies (UPSs), in which a large number of batteries are placed in series and parallel.

4.3

Bookkeeping-system implementations

A practical example of a bookkeeping system is shown in Fig. 9. In addition to an analog measuring function, involving A/D conversion and preprocessing, a microcontroller or microprocessor is used to perform the actual bookkeeping calculations in the bk . Two types of memory are needed. Basic battery data, such as the self-discharge rate as a function of form of function fV,T,I temperature and the discharging efficiency as a function of current and temperature, is read from the read only memory (ROM). The random access memory (RAM) is used to store the history of use, such as the number of charge/discharge cycles, which can be used to update the maximum battery capacity. bk . An important difference with respect to a direct measurement system is the presence of the variable ‘time’ in the function fV,T,I The time reference t is obtained from a crystal oscillator (Xtal). As a result, a memory at least has to remain connected to the battery at all times. A microcontroller inside the device can be used to calculate the SoC and store regular SoC updates in the memory inside the battery pack with a time stamp when a battery pack is connected to a portable device. When the pack is detached, the microcontroller can read the latest time stamp upon renewed attachment and estimate the time for which the battery has been unused. Another alternative is to integrate the entire system within the battery pack. Errors will accumulate over time in a bookkeeping system due to measurement inaccuracy and limited accuracy of the bk . Therefore, the system must be calibrated from time to time. The estimated SoC is reset to an assumed compensations in fV;T;I value in a calibration point. For example, the charger can signal to the SoC system when an end-of-charge trigger occurs. This serves as a ‘battery full’ calibration point, as indicated in Fig. 9 by the arrow labeled ‘Charger information.’ The effectiveness of this calibration of course depends on the accuracy of the ‘battery full’ trigger. Moreover, the system can be calibrated to ‘empty’ when the battery voltage drops below the end-of-discharge voltage defined as ‘battery empty.’ However, this is rather tricky, because when a battery is discharged with a large current and/or at a low temperature, the voltage will drop below UEoD while the battery still contains a considerable amount of charge. The reason for this is the discharging ‘efficiency’ phenomenon described in ‘Bookkeeping Systems.’ Bookkeeping systems are used predominantly in notebook computers. Communication between the SoC-indication IC and the operating system allows for the implementation of smart battery management, where for example the battery performance is shown to the user as a function of a chosen power management scheme. In the mid-1990 s Intel and Duracell set up the Smart Battery System (SBS) and System Management Bus (SMBus) specifications. The SBS was developed as a high-level battery chemistryindependent battery management specification. It uses the SMBus, based on the I2C (inter-integrated circuit) protocol, as physical layer, and the Smart Battery Data (SBD) as command language. I2C is a multimaster serial computer bus developed by Philips, which is used to attach low-speed peripherals to, for example, a motherboard, embedded system, or cell phone. The SBS and SMBus have become widely adopted standards in notebook computers, and Microsoft has included SMBus software drivers with Windows software since Windows 2000. Many IC vendors offer bookkeeping SoC-indication ICs, most of which are SBS compliant. The main vendors include Texas Instruments, Maxim, Microchip, and Linear Technology.

ROM

RAM

U

T

Analog preprocessing: • Amplification • Filtering • A/D conversion

Microcontroller/ Microprocessor

t Xtal

I I

Charger information

Fig. 9 Practical setup of a bookkeeping system. A/D, analog to digital; ROM, read only memory; RAM, random access memory.

546 4.4

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination Adding adaptivity to a state-of-charge-indication system

The most common example of adaptivity can be found in bookkeeping systems, where the maximum battery capacity, starting at the rated capacity corresponding to a new battery, is updated from time to time, taking into account capacity loss. Update algorithms are usually rather simple, but prove to be effective in many situations. For example, by starting a coulomb counter at zero at the moment a full battery is discharged, the amount of charge being extracted from the battery can be tracked. When the battery voltage subsequently drops below UEoD, depending on the actual conditions such as discharge current and temperature, the contents of this discharge coulomb counter can then be adopted as the new maximum-battery-capacity reference. A different and more complex approach in adding adaptivity is using a system as shown previously in Fig. 7C. For example, a so-called Kalman filter can be used to identify the proper model parameters based on battery variable measurements, such as voltage, current, and temperature. The Kalman filter is an efficient recursive filter that estimates the state of a dynamic system from a series of noisy measurements. The dynamic system, that is, the battery in this case, is described in the form of a model. The term ‘recursive’ implies that the estimated state from the system in the previous time step and a new set of measurements are used to estimate the new state of the system. Making the model more complex by adding, for example, dynamic contributions to the OCV and temperature effects, helps to create an SoC-indication system that adapts to the battery while maintaining an estimation error that is smaller than the quantization error related to the chosen implementation. Alternatively, use can also be made of neural networks or fuzzy logic. The effectiveness of these methods depends largely on the application, for example, the magnitude and amount of changes in behavior that can be expected over time.

4.5

Hybrid state-of-charge-indication systems

Hybrid SoC-indication systems have started to appear in an attempt to increase the accuracy. In a hybrid SoC-indication system, the principles of direct measurement and bookkeeping are combined, as well as adding forms of adaptivity. For example, Texas Instruments introduced the Impedance Track™ technology in 2006.

5

An adaptive state-of-charge-indication system

In the hybrid SoC-indication system described in this section, bookkeeping, direct voltage measurement based on the rest voltage U0(SoC) and adaptivity of the maximum battery capacity (Qmax), and the open-circuit voltage as function of SoH are combined. The general principle of the SoC-indication algorithm is illustrated in the form of a state diagram in Fig. 10. The basis of the SoC algorithm is the rest voltage measurement during the equilibrium state. The algorithm resides in this state when no or only a small current (smaller than defined minimum current Ilim) flows and the battery voltage has fully relaxed from previous (dis)charge currents. The measured battery voltage can be translated into SoC by means of a U0(SoC) curve stored in the system. During current flow larger than Ilim into or out of the battery, SoC indication is based on coulomb counting in either the charge or the discharge state, respectively. In the discharge state, the influence of the overpotentials on the available capacity is taken into account. After current interruption, a voltage prediction method is used in the transitional state to predict the equilibrium voltage based on the first part of the voltage relaxation curve. As will be described in the next section, the predicted open-circuit voltage is |I| ≤ I lim and voltage not stable |I| ≤ I lim and voltage stable

Initial state I < −I lim I > I lim

I > I lim I > I lim

−I lim < I < I lim Equilibrium state

I < −I lim

|I| < I lim and voltage stable

Charge state I > I lim

Transitional state

|I| < I lim and voltage not stable

|I| ≤ I lim

I > I lim

I < −I lim

|I| ≤ I lim

Discharge state

I < −I lim I < −I lim

Fig. 10 State diagram of a hybrid state-of-charge (SoC)-indication system based on direct measurement and bookkeeping developed by Bergveld, Danilov, Notten, Pop, and Regtien.

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

547

close to the actual value and can be used to calibrate the system. This implies that the influence of errors that integrate over time while integrating charge or discharge current can be compensated for. Moreover, using a voltage prediction algorithm allows the system to proceed to the equilibrium state without having to wait for the battery voltage to stabilize. The predicted open-circuit voltage can simply be transferred into an SoC value using the stored U0(SoC) curve. Since the SoC is estimated on a relative scale using the voltage curve in equilibrium and the coulomb counter yields an absolute SoC value in the charge and discharge state, the maximum battery capacity Qmax is needed to translate the absolute values into relative values. An update mechanism for Qmax has therefore been implemented to deal with aging. Moreover, it is shown that updating the stored U0(SoC) curve to deal with aging, as revealed in Fig. 6, is possible.

5.1

Voltage prediction

The battery voltage is built up from the open-circuit voltage, ohmic voltage drops, and overpotentials. After current interruption the overpotentials will relax to zero. This is a process that depends on many parameters, including the SoC, temperature, the value and duration of the previously applied (dis)charge current, and the age of the battery. This large number of dependencies makes it difficult to predict the equilibrium cell voltage at the very beginning of the relaxation period. For this reason, the relaxation process should be observed during a certain time interval to arrive at an accurate prediction of the final U0 value. The voltage prediction method employs a mathematical model for the lithium-ion battery voltage relaxation process. The voltage relaxation end value (U0) is determined on the basis of the measured first part of a voltage relaxation curve in the form of a set of samples (ti, Ui), where ti is the time value at which battery voltage value Ui has been measured. Then, fitting of the mathematical model to this measured part of the relaxation curve is performed. In addition to the unknown voltage relaxation end value, the mathematical model also contains three more parameters that are again found by fitting. This means that these parameters are updated for each individual situation, without the need to store values beforehand. The advantages are (1) that the equilibrium voltage U0 can be predicted with enough accuracy in the first few minutes after current interruption and (2) that no previously stored parameters are used. The first advantage improves the SoC determination by offering more calibration opportunities and solving the problem of the inability to determine a battery’s exact equilibrium state. It will also improve any SoC-indication system based on the rest voltage method. The second feature makes the method more robust with respect to battery aging. Fig. 11A illustrates the voltage relaxation process after application of a charge step at a 0.5C rate and 25  C (lithium–cobalt oxide/ graphite chemistry). The voltage eventually relaxes to a stable value of 3.71 V, that is, the equilibrium voltage corresponding to this SoC value. After 5 min of relaxation, the OCV, of which points Um denote the measured samples, still differs by 15 mV from the end value. This means that if the algorithm were to return to the equilibrium state after these 5 min, the battery voltage would still differ by 15 mV from the actual open-circuit voltage and a 2.5% SoC error would result. Returning to the equilibrium state after 60 min would lead to an acceptable SoC error of less than 1%. This also means that in cases like the one illustrated in Fig. 11A, a system cannot be calibrated until it has returned to the equilibrium state after 60 min. If a user starts (dis)charging a battery before these 60 min have elapsed, the calibration opportunity will be lost. However, the SoC calculated using the predicted open-circuit voltage (Up) based on voltage samples Um in the first 5 min is very close to the final SoC value calculated on the basis of the real equilibrium voltage (U0) voltage. In this example the calculated SoC inaccuracy is −0.2% SoC. The SoC errors obtained when using either the voltage relaxation model (SoCe(Up)) or the instantaneous OCV value (SoCe(OCV)) obtained after a discharge step at 0.25C rate and 25  C are shown in Fig. 11B for the same battery. The error in the SoC based on the voltage prediction model is about 0.05% after 5 min of relaxation, whereas the SoC error obtained when using the instantaneous OCV value is about 0.8% at that time. An error SoCe(OCV) of 0.05% is obtained only after a relaxation period of 55 min. It can be concluded that voltage prediction results in accurate equilibrium voltage predictions only after 5 min of voltage relaxation.

5.2

Remaining-time-of-use calculation in the discharge state

In order to calculate the remaining run time tr during discharge, the effect of the discharge conditions needs to be taken into account due to the discharge efficiency phenomenon. To this purpose, the system described in this section uses a mathematical function that calculates the SoCl value, that is, the SoC still remaining in the battery after the battery voltage has dropped below UEoD under the actual discharge conditions (current, temperature). Based on (1) the difference between the SoC at the moment tr is calculated and SoCl (which difference expresses residual capacity on a relative scale), (2) the maximum capacity Qmax, and (3) the value of the discharge current (to calculate tr from Q as indicated in Eq. (5)), the algorithm can calculate the remaining run time tr. Fig. 12 illustrates the accuracy of the implemented model for a lithium–M oxide/graphite battery chemistry, where M is a combination of manganese, nickel, and cobalt. The measured SoCl values (SoClm) have been used to fit the model and to obtain the fitted SoClf curves. The difference between measured and fitted curves over a range of discharge conditions varies between 0 and 2% SoC. In case a battery ages, the SoCl model parameters can also be updated. This is performed each time the battery has been discharged until UEoD. After current interruption, the voltage prediction model is used in this case to find the equilibrium value and then the corresponding SoCl value is found via the stored U0(SoC) curve.

548

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

Voltage (V)

3.75 Um OCV Up

3.74

3.73

3.72

3.71

3.70

0

100

200

300

400

500

600

700

Time (min)

(A) 1.5

SoCe (%)

SoCe (OCV) SoCe (Up) 1.0

0.5

0.0

0

10

20

30

40

50

60

Time (min)

(B)

Fig. 11 (A) Battery relaxation voltage open-circuit voltage (OCV) (V) after a charge step at 0.5C rate at 25  C versus time (min). The data points in the first 5 min of the relaxation process (Um) are used for the voltage prediction curve (Up). (B) Error in SoC (SoCe (%)) calculated based on the predicted voltage (SoCe(Vp)) using the voltage prediction model and SoC calculated based on instantaneous OCV values (SoCe(OCV)) after a discharge step at 0.25C rate and 25  C versus time (min) (lithium–cobalt oxide/graphite chemistry).

15

SoCI (%)

SoClf SoClm

5 °C

10 25 °C

5

0

45 °C

0

0.2

0.4

0.6

0.8

1.0

C rate Fig. 12 Measured (SoClm (%)) and predicted (SoClf) (%)) SoCl values versus C rate for various temperatures (lithium–M oxide/graphite chemistry, with M a combination of manganese, nickel, and cobalt). SoC, state-of-charge.

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination 5.3

549

Adaptation of maximum battery capacity

Fig. 13 illustrates how the SoC algorithm can adapt to changing values of Qmax. For updating the maximum capacity it is necessary for the system to run through a sequence of states: equilibrium state, (dis)charge state, transitional state, and again equilibrium state. The new value of Qmax is simply calculated by relating the charge (Qch in the charging case, Qd in the discharging case) added to (subtracted from in case of discharging) the battery during (dis)charging to the difference in SoC (SoCef – SoCes) before and after (dis)-charging. This is reflected in the equation below: Qmax ¼

100%  Qch SoCef − SoCes

(7)

The SoCes and SoCef values are inferred from a stored U0(SoC) curve. The value of SoCef may be determined using the voltage relaxation model. Updating Qmax during the final stages of charging has the advantage that the overpotentials are very small at the end of the CV region; see Fig. 5. In that case, the voltage relaxation model does not have to be used. This has a positive impact on the accuracy.

5.4

Update mechanism for the rest voltage curve

Storing an updated version of the U0(SoC) curve starts with gathering a number of equilibrium open-circuit voltage values at various SoC values. Every time the current is interrupted, the above-described voltage relaxation model can be used to predict U0 and the corresponding (SoC, U0) pair can be stored. Alternatively, during overnight charging, the system can force the battery through a number of charge or discharge steps and a number of U0 points can be found from the voltage relaxation model and stored for a number of selected SoC values between 100 and 0%. After a number of (SoC, U0) points have been gathered, a mathematical SoC ¼ f (U0) function can be fitted to these points to yield a new curve. Fig. 14 illustrates this process for a lithium–cobalt oxide/graphite battery showing a 5.4% capacity loss due to aging. A total of 12 voltage points have been obtained after current interruption using the voltage relaxation model and the mathematical function

Equilibrium Voltage

Charge

Equilibrium

Transitional SoCch

SoCef

SoCes Qch = ∫Ichdt Time Fig. 13 Schematic representation of maximum-capacity adaptation method based on voltage measurement and coulomb counting developed by Bergveld, Danilov, Notten, Pop, and Regtien, the case for battery charging is shown.

Voltage (V)

4.2 4 3.8 3.6 3.4 3.2 3 100

GITT Model fit Predicted 80

60

40

20 SoC (%)

0

Fig. 14 Equilibrium open-circuit voltage U0 (V) vs state-of-charge (SoC) (%), obtained by means of fitting 12 predicted points ( ) in mathematical model compared with U0 obtained by means of galvanostatic intermittent titration technique (GITT) at 25  C using a battery that suffered 5.4% capacity loss due to aging (lithium–cobalt oxide/graphite chemistry).

550

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

fitted to these points yields the least-squares fit curve. This curve is compared to the U0(SoC) curve obtained with the galvanostatic intermittent titration technique (GITT). In the latter case, the battery voltage is allowed to relax for a long period after current interruption. The comparison in Fig. 14 shows that the fitted curve obtained using the gathered U0 points fits very well with the GITT curve. Since the voltage relaxation model has been used to obtain the U0 points, long resting times are not needed. This increases the practical usefulness of this method.

5.5

Results obtained with the considered state-of-charge-indication system

Fig. 15A illustrates the obtained accuracy in tr prediction in case aged batteries (lithium–cobalt-oxide/graphite chemistry) are connected to the described SoC-indication system. In this case, after connection the described update mechanisms for the EMF and SoCl functions as well as the Qmax value have been performed. The solid gray lines indicate the limits of 1-min error for remaining run times smaller than 100 min and  1% for times larger than 100 min. The figure illustrates that most predicted tr values are within the indicated limits. Note that the predictions have not only been obtained for various discharge rates, as indicated in the figure, but in fact the depicted points have also been obtained at various temperatures, including 5, 25, and 45  C. Moreover, various starting SoC values have been applied at the moment when discharging at the indicated discharge rates was started. Fig. 15B indicates similar results when a fresh lithium-ion battery of a type different from the type for which the system has originally been developed is connected to the system (lithium–cobalt oxide/graphite). This type, lithium–M oxide/graphite, was also used to obtain Fig. 12, as described above. Again, the U0(SoC) and SoCl functions have been fitted to this new battery type, as well as the Qmax value. In this case the results are even better.

t r (min)

20

10

0

−10

0.1 C rate 0.25 C rate 0.5 C rate 0.75 C rate 1 C rate

−20 1

10

(A)

100

1000

Time (min)

t r (min)

20

10

0

−10

0.1 C rate 0.25 C rate 0.5 C rate 0.75 C rate 1 C rate

−20 1 (B)

10

100

1000

Time (min)

Fig. 15 (A) Error in predicted remaining run time (tr (min)) vs actual remaining run time (min) for discharging an aged Li-ion battery at various C rates; the stateof-charge (SoC)-indication system has adapted the EMF curve and SoCl parameters to fit the aged battery behavior (lithium–cobalt oxide/graphite chemistry); (B) Error in predicted remaining run time (tr (min)) versus actual remaining run time (min) for discharging a Li-ion battery of a different type at various C rates; the SoC-indication system has adapted the EMF curve and SoCl parameters to fit the behavior of the different battery types (lithium–M oxide/graphite chemistry, with M a combination of manganese, nickel, and cobalt).

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

551

Fig. 16 Artist impression of a mobile phone with indicated talk time, standby time, and state-of-charge (SoC).

Finally, Fig. 16 illustrates what a practical user interface may look like for a mobile phone in which the described system has been implemented. In addition to the SoC (in % and on a bar graph), the remaining talk time under the actual discharge conditions as well as the remaining standby time based on an assumed discharge current in standby mode are shown.

6

Conclusions

There is an increasing demand for rechargeable batteries in many applications, ranging from low-power applications such as autonomous devices to high-power applications such as hybrid electrical and future plug-in vehicles. In many applications an accurate prediction of the SoC and even more important the remaining time of use is crucial. Accurate and reliable SoC indication will prompt the user to use the battery more effectively, resulting in a decrease in the number of necessary recharges and thereby increasing user convenience. For any SoC-indication system to be accurate, it is important to take battery behavior into account in a proper way. In recent years, hybrid SoC-indication systems have emerged that combine the advantages of several methods, mostly combining direct voltage measurement through assessment of the equilibrium situations with no or only limited current flow with current measurement and integration during current flow. In any practical implementation of an SoC-indication system, taking into account the effects of battery aging leading to decreased capacity and increased series impedance is important. For newer lithium-based battery technologies such as lithium iron phosphate (LiFePO4) systems, the OCV curve is flatter versus SoC than for conventional lithium-based systems and may mean that, for example, a bookkeeping system will be a better alternative in this case. A hybrid system that includes adaptivity means to deal with battery aging demonstrates the effectiveness of the applied method. As promising as the described results may be, it still remains a big challenge to come up with an SoC-indication system that adapts itself to any battery type or chemistry even when it ages and still remains accurate under all conceivable conditions. Such a system will remain the subject of choice in many research projects in the years to come.

See also: Batteries – Battery Types – Lead-Acid Battery: Overview; Batteries – Battery Types – Nickel Batteries: Overview; History of Electrochemistry: Primary and Secondary Batteries; Lithium Batteries – Lithium Secondary Batteries – Li-ion battery: Overview

552

Electrochemical Terminology | Adaptive State-of-Charge (SoC) Determination

References Alzieu, J.; Smimite, H. Glaize C, Improvement of Intelligent Battery Controller: State-of-Charge indicator and Associated Functions. J. Power Sources 1997, 67, 157–161. Aylor, J. H.; Thieme, A.; Johnson, B. W. A Battery State-of-Charge indicator for Electric Wheelchairs. IEEE Trans. Ind. Electron. 1992, 39, 398–409. Bergveld, H. J.; Kruijt, W. S.; Notten, P. H. L. Battery Management Systems: Design by Modelling; Kluwer Academic: Boston, 2002. Broussely, M.; Herreyre, S.; Biensan, P.; Kasztejna, P.; Nechev, K.; Staniewitz, R. J. Aging Mechanism in li-Ion Cells and Calendar Life Predictions. J. Power Sources 2001, 98, 13–21. 5. Chiasson, J.; Vairamohan, B. Estimating the State-of-Charge of a Battery. IEEE Trans. Control Syst. Technol. 2005, 13, 465–470. 6. Coleman, M.; Lee, C. K.; Zhu, C.; Hurley, W. G. State-of-Charge Determination from EMF Voltage Estimation: Using Impedance, Terminal Voltage, and Current for Lead-Acid and lithium-Ion Batteries. IEEE Trans. Ind. Electron. 2007, 54, 2550–2557. 7. Gerard, O.; Patillon, J. N.; d’Alche-Buc, F. Neural Network Adaptive Modeling of Battery Discharge Behavior. Lect. Notes Comput. Sci 1997, 1327, 1095–1100. 8. Huet, F. A Review of Impedance Measurements for Determination of the State-of-Charge or State-of-Health of Secondary Batteries. J. Power Sources 1998, 70, 56–69. 9. Kim, I. S. Non-linear State-of-Charge Estimator for Hybrid Electric Vehicle Battery. IEEE Trans. Power Electron. 2008, 23, 2027–2034. 10. Kim, M.; Hwang, E. Monitoring the Battery Status for Photovoltaic Systems. J. Power Sources 1997, 64, 193–196. 11. Mundra, T.; Kumar, A. Micro-Power Battery State-of-Charge Monitor. IEEE Trans. Consum. Electron. 2008, 54, 623–630. 12. Notten, P. H. L. Rechargeable batteries: Efficient energy storage devices for wireless electronics. In Amlware: Hardware Technology Drivers of Ambient Intelligence; Mukherjee, S., Aarts, E., Roovers, R., Widdershoven, F., Ouwerkerk, M., Eds.; Dordrecht: Springer, 2006; pp 315–345. 13. Notten, P. H. L.; van Beek, J. R.; Op Het Veld, J. H. G. Boostcharging li-Ion Batteries: A Challenging New Charging Concept. J. Power Sources 2005, 145, 89–94. 14. Piller, S.; Perrin, M.; Jossen, A. Methods for State-of-Charge Determination and their Applications. J. Power Sources 2001, 96, 113–120. 15. Plett, G. Extended Kalman Filtering for Battery Management Systems of LiPB-Based HEV Battery Packs: Part 1, 2, 3. J. Power Sources 2004, 134, 252–292. 16. Pop, V.; Bergveld, H. J.; Danilov, D.; Regtien, P. P. L.; Notten, P. H. L. Battery Management Systems: Accurate State-of-Charge Indication for Battery-Powered Applications; Springer: Dordrecht, 2008. 17. Pop, V.; Bergveld, H. J.; Notten, P. H. L.; Regtien, P. P. L.; Op Het Veld, J. H. G.; Danilov, D. Battery Aging and its Influence on the Electromotive Force. J. Electrochem. Soc. 2007, 154, A744–A750. 18. Rodrigues, S.; Munichandraiah, N.; Shukla, A. K. A Review of State-of-Charge Indication of Batteries by Means of a.c. Impedance Measurements. J. Power Sources 1999, 87, 12–20. 19. Rong, P.; Pedram, M. An Analytical Model for Predicting the Remaining Battery Capacity of lithium-Ion Batteries. IEEE Trans. Very Large Scale Integr. VLSI Syst. 2006, 14, 441–451. 20. Salkind, A. J.; Fennie, C.; Singh, P.; Atwater, T.; Reisner, D. E. Determination of State-of-Charge and State-of-Health of Batteries by Fuzzy Logic Methodology. J. Power Sources 1999, 80, 293–300. 21. Shen, W. X.; Chan, C. C.; Lo, E. W. C.; Chau, K. T. Estimation of Battery Available Capacity under Variable Discharge Currents. J. Power Sources 2002, 103, 180–187. 22. Spotnitz, R. Simulation of Capacity Fade in Lithium-Ion Batteries. J. Power Sources 2003, 113, 72–80.

1. 2. 3. 4.

Further reading 1. Brodd, R. J. Secondary Batteries | Overview. In Encyclopedia of Electrochemical Power Sources; Garche, J., Ed., 2009, pp 254–261, ISBN 9780444527455, https://doi.org/10. 1016/B978-044452745-5.00125-8. 2. Shukla, A. K.; Venugopalan, S.; Hariprakash, B. Secondary Batteries – Nickel Systems | Nickel–Cadmium: Overview. In Encyclopedia of Electrochemical Power Sources; Garche, J., Ed.; Elsevier, 2009, pp 452–458, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00153-2.

Electrochemical Terminology | State-of-Health (SoH) H Wenzla, R Bengerb, and I Hauerc, aConsulting for Batteries and Power Engineering, Osterode, Germany; bResearch Center Energy Storage Technologies, Clausthal University of Technology, Clausthal-Zellerfeld, Lower Saxony, Germany; cChair of Electrical Energy Storage Technology, Institute of Electrical Power Engineering, Clausthal University of Technology, Clausthal-Zellerfeld, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 2.1 2.2 2.3 3 4 4.1 4.2 5 6 References

Introduction State of health Definitions of SoH Relation of state of charge and state of health Inclusion of other performance characteristics State of power State of function Definition State of function for end of lifetime determination as a multidimensional vector Measuring and estimating state of health, state of power and state of function Conclusion

553 554 554 554 555 555 556 556 557 558 558 558

Abstract This article describes concepts which are used to quantify how much a battery has aged or how much of its lifetime has already been used up. There are a number of “state of X” terms used in batteries, the most important being state of charge, state of energy, state of power, state of health and state of function. The distinction is blurred between the states used to describe the operating conditions and thus a fast change of the state variable during a discharge cycle and states associated with lifetime and aging which vary slowly and are independent of the operating conditions. State of health, state of function and state of power are discussed.

Key points

• • • • •

1

Definition of the terms SoH, SoF and SoP Correct usage of lifetime states Lifetime states as slowly varying parameters vs state variables reflecting current operating conditions Lifetime states need more precise definitions Capacity loss and increase in internal resistance should be distinguished when using State of health

Introduction

The end of life is reached, when a battery is no longer capable to fulfil a required function. Lifetime may be considered as a multidimensional vector, with each dimension denoting a function of importance for the application under consideration. For practical purposes, it is of interest to quantify how an aged battery is still capable to perform the required functions. The most important terms, so-called lifetime states, are state of health (SoH) and state of function (SoF), where state of health refers to the performance characteristics of a fully charged battery under reference or nominal conditions and state of function to the capability of the battery to fulfil required functions under current operating conditions, e.g. at low temperature and low state of charge. Terms such as state of fitness and state of balance have also been introduced to describe the state of a battery. The definitions of the state of power (SoP) often include factors associated with aging of batteries. It is interesting to note that battery “state of X” terms appeared in the peer reviewed scientific journals some 25 years ago with the onset of the commercialization of lithium-ion batteries whereas state of charge (SoC) is a term which has been used for batteries for a considerably longer period of time. Lifetime states change very slowly over the whole lifetime whereas state of charge and state of energy vary over their whole range of values during cycling and do not take aging factors into account as a result of their definition. The term state of safety (SoS) requires a definition of safety of batteries. In the following the terms SoH, SoC, SoP and SoF are explained independently of the kind of electrochemical power source as different uses of the terms for different systems should be avoided.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00107-9

553

554

2

Electrochemical Terminology | State-of-Health (SoH)

State of health

The following discussion focusses on batteries to allow a more concise and clearer description.

2.1

Definitions of SoH

The term state of health (SOH)a is mostly used only in reference to loss of capacity assuming either that no other performance characteristics defines end of life or that loss of capacity is the lead factor for all other performance characteristics. These are in most cases valid assumptions. An overview by Meissner and Richter1 and the article on lifetime in the previous edition of the Encyclopedia of Electrochemical Power Sources2 proposed the following definition: SoH ¼

Q − Qmin Q0 − Qmin

where Q is the capacity (electric charge in Ah, the symbol C is not recommended) of the battery under reference discharge and charge conditions, i.e. nominal discharge current and nominal temperature after a full charge. In some battery technologies, i.e. NiCd or lead-acid, there are operating conditions leading to a reversible capacity loss. Therefore, before determining Q, procedures to remove a reversible capacity loss should be carried out. Qmin is the minimum capacity (threshold value) acceptable for the application, and Q0 is the capacity of the new battery under reference discharge and charge conditions. As the capacity of lead acid batteries can increase after cycling beyond the value of Q0, this definition of SOH may lead to a value above 1. If the reference charge conditions have been chosen well, the state of health in this definition describes a battery where all reversible effects which lead to a loss of performance have been removed and the integral effect of all aging effects on capacity is measured in one parameter. When state of health as described in1 is zero or has a negative value, the battery capacity is below the threshold value and the battery is no longer considered capable of fulfilling required functions. The battery has reached its end of life. This definition, however, is no longer widely used. Instead, the ratio of actual capacity Q and nominal capacity QN is used: SoH ¼

Q QN

In this definition, it is not necessary to measure the capacity of the new battery. As the capacity of most batteries when new is slightly above the nominal capacity, this definition will lead in most cases to a SoH value slightly above 1 for new batteries. For practical reasons this may be ignored. End of life is reached either when the SoH is 0,8 or 0,7 depending on the definition of what the lowest acceptable capacity is. It is important to note that the value of SoH, whichever of the definitions above is used, does not allow a straightforward prediction of the remaining lifetime of a battery.

• •

Lithium-ion batteries have in general an exponential loss of capacity, so that the time during which SoH is reduced by 10% gets longer, the lower the initial SoH is. For a rough estimate a linearization of the loss of capacity over time may be acceptable. For lead acid batteries in many applications, the SoH remains at a value of 100% or above during most of the lifetime and decreases only at the very end of lifetime. A SoH of, e.g.. 90% after some years therefore does not mean that the battery can be used for a few more years but that the end of life is immanent, and a SoH of 100% may describe a new battery or a battery which has spent 80% or 85% of its lifetime.

It should be pointed out that even in peer reviewed scientific journals it is not always clearly stated which of the definitions of SoH above is used and the likely definition must be based on the context of the results presented.3

2.2

Relation of state of charge and state of health

The state of charge (SoC) of a battery refers to the amount of charge which has been drawn from a battery in relation to a reference value, usually the nominal capacity, despite the dependence of the capacity on current rate and temperature. As the battery ages and its capacity decreases, the battery voltage will fall below its end of voltage value earlier. A further discharge is then not advisable for safety reasons, risk of damaging the battery or because the loads cannot be operated due to an unacceptable low input voltage level. SoH 0.8 and reaching the end of discharge voltage at a SoC of 20% are therefore related and, in the case of a discharge at nominal current and temperature, mathematically identical. Ideally, state of charge and state of health should both be used to control the operating conditions of a battery where this is possible, e.g. in hybrid power systems or in situations where the charging process can easily be initiated. a It is wrong to draw an analogy to human beings, which can recover from poor health or can have good health despite of being old and not capable of competing with young healthy individuals. The battery equivalent to a human state of health would be a parameter describing the battery as having the maximum possible performance characteristics at any time during its lifetime after removing all reversible effects which reduce the performance characteristics.

Electrochemical Terminology | State-of-Health (SoH) 2.3

555

Inclusion of other performance characteristics

The definitions above use only capacity as a parameter to quantify the ability of a battery to perform the required functions. In hybrid vehicles, however, the battery is kept in a narrow range of state of charge so that the battery can both take up or provide high power for a short period of time. Loss of capacity is therefore of less importance than increase of internal resistance. In line with lifetime as a multidimensional vector, SoH can also be considered as a vector with a number of dimensions, the definitions above describing SoHC, the loss of capacity, and the increase of internal resistance to be described by SoHR. Lando4 used the following definition in analogy to the SOH definition: SoHR ¼

R R0

where R is the internal resistance of the aged battery, and R0 the internal resistance of the new battery, both measured under identical conditions, nominal temperature, SoC of 50% and in discharge direction. It is important to define the conditions under which internal resistance is measured, because the internal resistance of any battery does not only depend on SoC, temperature and age, but also on the current direction. For lead acid batteries, for instance, internal resistance values according to DIN EN 60896-11:2003 are given at 100% SoC at room temperature in discharging direction. End of life is considered to be reached by Lando when the internal resistance HAS reached 200% of its initial value resulting in SoHR of 2 at the end of lifetime. A low Internal resistance and the ability to provide or absorb a high power pulse are equivalent and therefore SoHR and state of power are related terms.

3

State of power

The power capability of a battery is, in some applications, the most important performance characteristics and may determine the end of lifetime. The power capability is a direct function of its internal resistance. Not surprisingly, therefore, efforts to define a state of power (SoP) and estimate a numerical value have recently gained great importance. Electrically, power is defined as P ¼ U I, and is thus an instantaneous value. In technical applications, however, it is important to know how long a given power level can be sustained, e.g. for 0.1 s for triggering a fuse or 10 s to determine the capability of a vehicle with an electric drive system to accelerate or store energy during breaking. In all electrochemical systems, the voltage is a function of the discharge current, current direction and time. Knowledge of the internal resistance at a given SoC or temperature is therefore not the same as knowing its power capability, unless the internal resistance is measured in the same way as its power capability. An appropriate equation for power of a battery depends on how it is measured, using either a constant current discharge, using a resistor with a constant resistance during the discharge time or using an electronic load which delivers a constant power level. Technically, average power levels for the time period of time Dt or the power level at the end of Dt are required. The average power for a time dependent power output is: P¼

1 Dt

ZDt P ðt Þ dt ¼ 0

1 Dt

ZDt U ðt Þ  Iðt Þ dt 0

where Dt is the time interval for which the average power level is required. The voltage U must remain within the allowed voltage range: an upper limit for taking up power and a lower limit for providing power. Measuring power capability is most easily carried out with a resistor as a load and using the power level at the end of the time period Dt, e.g. 10 s, while observing the voltage restrictions of the cell. Fig. 1 shows data for a 5 Ah NMC lithium-ion cell which shows, how much the power capability in the charging and discharging direction differ, in particular at very high and very low SoC. In view of the above it is clear, that the conditions under which power levels are measured or estimated and the term state of power is defined must be very clearly described in order to be practically useful. Definitions of state variables, generally referred to as SoX, are not usually precisely defined and not used unambiguously in the scientific and technical community. In,5 state of power is defined as “the maximum power that can be released or absorbed steadily by the power battery within a pre-determined time interval.” Frequently, however, state variables are defined as a ratio of the value which currently exists and a reference value. The following definitions may be appropriate:

•

•

State of power as a parameter to control operation. Because of the dependence on state of charge, current amplitude and current direction as the dominant factors, state of power, whatever is its definition, varies more strongly than any other state variable during operation. Algorithms for estimating state of power are therefore closely linked to algorithms to estimate state of charge (see e.g.5). State of power as a variable to control the operation of a power system is not considered further in this article as it is not a lifetime state. State of power as a lifetime state. Here, state of power is a variable which varies only slowly over time and can be defined by choosing a fixed reference point (e.g. SoC and temperature) relevant for the application or the battery chemistry, ideally the same

556

Electrochemical Terminology | State-of-Health (SoH)

Fig. 1 SoC-range of maximum power in charge and discharge direction for a 5 Ah LiNiMnCoO2 cell.

as that used for determining the state of health for internal resistance SoHR. Under this definition, the SoP is defined similar to zje state of health above: P P0 where P refers to the actually available power level under reference conditions for an aged battery and P0 refers to the power level under reference conditions for a new battery. All aging processes which impact on the internal resistance and mass transport phenomena, as the most important factor, then lead to a single parameter which describes the power capability of the battery. The underlying assumption is that a reduction of the power level at the reference conditions can be used to estimate the reduction of the power levels for any other operating conditions. It is widely assumed that the loss of capacity of a battery as quantified by an unacceptable low value of SoHC also indicates that all other relevant performance characteristics have reached unacceptable values – neither beforehand nor afterwards. In addition, it is assumed that there are no applications where loss of capacity is not of great importance as long as other performance characteristics, in particular the power capability of a battery still meets all functional requirements. Obviously, these assumption can no longer be used for battery based energy systems with sophisticated battery and energy management systems. SoP ¼

4

State of function

The lifetime states discussed above describe the decrease of the performance under reference conditions, e.g. a given state of charge and temperature. However, in many applications it is important to know the performance under considerably less favorable conditions within the permitted range of operating conditions. Failure to fulfil a required function under such conditions is an obvious end of lifetime criterion, regardless of the state of health.

4.1

Definition

The state of function (SoF) describes the ability of an electrochemical power source to fulfil a required function at the current operating conditions, e.g., low state of charge and low temperature. This is the difference to the state of health where nominal conditions (fully charged, nominal temperature) are used to quantify the performance characteristics of an aged battery compared to a new one. The ability of a battery to provide high power pulses, e.g. for acceleration in electric or hybrid vehicles or for starting of an internal combustion engine, are good examples to explain the concept. Meissner1 defined the state of function for starting an internal combustion engine. SoF ¼

U min − U 1 U0 − U1

where Umin is the minimum voltage during the starting process, U1 is the lowest acceptable voltage during the starting process, and U0 is the minimum voltage of a battery in new and good condition during the starting process. This definition was used for evaluating the capability of a battery to start an engine from online measurements during operation. The value of SoF as defined above is likely to be above 1 under optimal operating conditions and zero or negative under extremely poor operating conditions, e.g. very low state of charge and temperatures below −18  C, outside the expected range of operating conditions. Under this definition, a value of SoF below zero under extremely harsh conditions would not mean the end of lifetime.

Electrochemical Terminology | State-of-Health (SoH)

557

It is therefore necessary to generalize this concept and formulate the conditions more precisely. The example for the following description refers to starting an internal combustion engine or the internal resistance of a battery. 1. State of function, a characteristics of the battery, must not depend on other systems components. The power requirement of a battery for successfully starting an internal combustion engine depends on many engine specific factors (oil viscosity, age of oil, engine temperature, time since last operation, etc.). It is therefore more appropriate to use the internal resistance of the battery instead of the minimum voltage during starting. 2. The concept of state of function must be restricted to a range of operating conditions. Even new batteries cannot fulfil a required function below certain temperature limits and outside a realistic range of state of charge. State of function must therefore be restricted to include only those operating conditions which are within the design limits or the expected range of operating conditions. It may be appropriate to include some abuse conditions if these cannot be avoided with certainty. 3. The function must reflect realistic requirements. Acceptance tests for starter batteries require a power profile (e.g. 30 s constant current at −18  C within a required voltage range) which exceeds by far what is actually required for starting an engine. It is therefore necessary to define an appropriate power profile which will enable the start of an engine under the worst conditions which may exist and which can be tested easily in the laboratory. 4. An estimate of state of function should be possible from online measurements of current, voltage and temperature and it should be possible to calculate state of function using battery models. EN 62 for diesel generators used for emergency power applications requires that the battery must be capable to enable 3 tries of starting the engine, each with a duration of 10 s and a rest period of a few seconds between each try. Batteries for such applications are kept always fully charged und are kept in buildings where there are always moderate temperatures at the location of installation. Based on measurements of the current drawn during a successful engine start (e.g. 1000 A with a voltage above 9 V), a functional requirement could therefore be to require that the internal resistance of a fully charged battery at, e.g. 10  C must be below a threshold value to allow delivery of such a 30 s discharge. The state of function can then be defined as: SOFR ¼

RðSoC, T Þ − Rm ðSoC, T Þ R0 ðSoC, T Þ − Rm ðSoC, T Þ

where R is the internal resistance of the aged battery, R0 is the resistance of the new battery. The indices refer to SoC and temperature T: the most unfavorable conditions under which the battery must be able to fulfil the required function, e.g. for the case described above 10  C, fully charged, Rm is the value that the internal resistance must not exceed to fulfil the required function, e.g. providing a current pulse of 1000 A for 30 s with the battery voltage remaining above 9 V. The value of SoFR ¼ 1 for a new battery and becomes negative if the internal resistance has increased beyond its maximum acceptable value. Not being able to fulfil a required function, defined here as a current profile under well-defined conditions, means that the end of lifetime has been reached and the battery must be replaced.b Alternatively, in SOFR ¼

Rðt Þ − Rm ðSoC, T Þ R0 ðSoC, T Þ − Rm ðSoC, T Þ

R(t) refers to the internal resistance at the time of measurement. In this case, the SoF can have values considerably above 1. It may be appropriate to define the state of function not as a scalar value but as either SoFR > 0 or SoFR  0 in line with the concept that a function can be fulfilled or not.

4.2

State of function for end of lifetime determination as a multidimensional vector

In most applications batteries must be able to fulfil a number of different requirements, some of which may reflect non-technical and economic constraints. A battery with adequate capacity and internal resistance but a very high self discharge rate may, for instance, require more frequent recharging and maintenance and has to be replaced for cost reasons. A state of function can and should therefore be defined for every requirement which the battery has to fulfil. The state of function for other required performance characteristics has to be defined specific to the application and in analogy to the example given above for internal resistance and the ability to start an internal combustion engine. It is important to point out that the life of a battery, where at least one dimension of state of function has become zero, can be extended in systems with appropriate energy management systems, particularly hybrid systems with additional energy conversion components. In these systems it may be possible to prevent a very low temperature of the battery and very low state of charge. As a result, the battery will be able to fulfil the required functions under the changed, less severe operating conditions. The concept of second life is a practical implementation of such modified functional requirements which extend lifetime.

b Diesel engines for uninterruptible power supply systems (often preheated, well maintained state) tend to have a starting time of ca. 3 s. As a result, a battery with an internal resistance value as defined here will still be able to start a diesel engine, but does not fulfil the requirements set out in the standard and thus, for safety reasons, should be replaced.

558

5

Electrochemical Terminology | State-of-Health (SoH)

Measuring and estimating state of health, state of power and state of function

It is straightforward to measure lifetime states under laboratory conditions. Using the state of function definitions given above, it is also possible to measure the state of function for starting an internal combustion engine from voltage measurements during a starting process. However, for most applications, measurements of voltage, current and temperature during operation of the battery do not allow the direct determination of above battery states. Even in uninterruptible power supply systems, where the battery is kept fully charged (e.g. lead acid batteries), or at a high state of charge (e.g. lithium-ion batteries) and where there is an annual discharge test, the test procedure is often carried out at the application specific currents and limited to the backup time which the power system has to fulfil. As a result determining state of health SoH as the most straightforward state variable requires a battery model which can, however be very simple. It is therefore necessary to use sophisticated battery models for any lifetime state estimation which must include state of charge estimation. Not surprisingly, any such sophisticated battery model is able to calculate values for any of the above state variables, in the case of state of function limited to capacity and internal resistance but not state of functions which refer to maintenance and cost aspects. Such battery models can, for instance be found in.6,7,8

6

Conclusion

Lifetime states quantify changes of performance characteristics due to aging of a battery. The most commonly used state variable is state of health based on capacity loss and its widely spread use is based on the assumption that all other functional requirements which a battery must fulfil lie within acceptable levels as long as the capacity under nominal conditions is sufficiently high and lie outside acceptable levels as soon as the capacity is no longer sufficient. The more sophisticated an energy system is, in particular when used in hybrid systems, the more inappropriate these assumptions are. End of lifetime can easily be determined by an appropriate definition of state of function for each function which a battery must be capable of fulfilling.

References 1. Meissner, F.; Richter, G. The Challenge to the Automotive Battery Industry; the Battery Has to Become an Increasingly Integrated Component within the Vehicle Electric Power System. Journal of Power Sources 2005, 144, 438–460. 2. Wenzl, H. Lifetime. In Encyclopedia of Electrochemical Power Sources; Garche, J., Dyer, C., Moseley, P., Ogumi, Z., Rand, D., Scrosati, B., Eds.; vol. 1; Elsevier: Amsterdam, 2009; pp. 552–558. 3. Aksakal, C.; Sisman, A. On the Compatibility of Electric Equivalent Circuit Models for Enhanced Flooded Lead Acid Batteries Based on Electrochemical Impedance Spectroscopy. Energies 2018, 11, 118. 4. Tchoupou-Lando, E. Entwicklung eines ereignisbasierten Lebensdauerprognosemodells und Validierung der linearen Schadensakkumulationshypothese für NMC/Graphit Lithium-Ionen Zellen; PhD thesis, Technical University of Clausthal, 2021. 5. Wu, M.; Qin, L.; Wu, G. State of Power Estimation of Power lithium-Ion Battery Based on an Equivalent Circuit Model. Journal of Energy Storage 2022, 51, 104538. 6. Xiao, Y.; Wen, J.; Shen, Y. State of Health, a Comprehensive Review of the lithium-Ion Battery State of Health Prognosis Methods Combining Aging Mechanism Analysis. Journal of Energy Storage 2023, 15, 107347. 7. von Bülow, F.; Meisen, T. A Review on Methods for State of Health Forecasting of lithium-Ion Batteries Applicable in Real-World Operational Conditions. Journal of Energy Storage 2022, 57, 105978. 8. Pradhan, S. K.; Chakraborty, B. Battery Management Strategies: An Essential Review for Battery State of Health Monitoring Techniques. Journal of Energy Storage 2022, 51, 104427.

Electrochemical Terminology | Safety and State-of-Safety (SoS) H Wenzla, R Bengerb, and I Hauerc, aConsulting for Batteries and Power Engineering, Osterode, Germany; bResearch Center Energy Storage Technologies, Clausthal University of Technology, Clausthal-Zellerfeld, Lower Saxony, Germany; cChair of Electrical Energy Storage Technology, Institute of Electrical Power Engineering, Clausthal University of Technology, Clausthal-Zellerfeld, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 2.1 2.1.1 2.1.2 2.2 2.2.1 2.2.2 3 4 4.1 4.2 4.3 4.4 5 References

Introduction Overview of risks Risks arising from normal operating conditions Electrical risks Gas emissions risks Risks associated with abnormal operating conditions Electrical risks Risks from overcharging and overdischarging Ensuring safety State of safety Interpretation of ASIL and EUCAR hazard levels as state of safety values Effects on safety as a result of operation within the permitted range of operating condition Effects of operating conditions outside the permitted range or under abuse conditions Proposed descriptions and algorithms for calculating state of safety Conclusion

559 560 560 560 560 560 560 560 561 562 562 563 563 563 564 564

Abstract Safety is the absence of unacceptable risks for causing damage and health hazards outside the physical boundaries of a system (see ISO/IEC Guide 78:2012 for further details). Safety requirements therefore go beyond reliability issues of failed products, and must not cause unacceptable risks. A failed product should not reduce required safety levels. This article describes the concept of safety, procedures to ensure and assess safety, and the efforts to quantify safety as a state of safety.

Key points

• •

1

Overview of safety risks Definition of the terms related to safety

Introduction

Safety is the absence of unacceptable risks for causing damage and health hazards outside the physical boundaries of a system (ISO/IEC Guide 78:20121). Safety requirements therefore go beyond product reliability issues of failed batteries, and must not cause unacceptable risks. A failed product should not reduce required safety levels. Environmental protection aspects (ISO/IEC Guide 64:20082) go further and require minimizing the impact on soil, air, water and biodiversity. All electrochemical power sources pose considerably risks due to their high energy content if not used properly, as a result of product defects, inadequate control of operating conditions or malfunctioning of systems components. They may emit toxic, flammable or explosive gases either under normal operating conditions or during abuse. The damage caused by an electrochemical power source can be high. When discussing safety risks of electrochemical power sources it is important to note that a number of safety risks can be managed by good industrial practices, for example electrical hazards. All electrochemical power sources if connected in series and with sufficiently low internal resistance (e.g. all lead acid, nickel based alkaline batteries and lithium-ion batteries) can deliver very high short circuit currents for a prolonged period of time. This can cause arcing, fires and electrocution which may lead to death. All of these risks can be handled by standard protection procedures used in electrical engineering. In the case of electrical short circuits in external loads, the choice of fuses and switches, their location and the control systems to trigger the switches may be more challenging than in other fields of electrical engineering, but the risk of a fire caused by an external short circuit can be minimized to the same degree.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00120-1

559

560

Electrochemical Terminology | Safety and State-of-Safety (SoS)

Safety risks which can be largely excluded by adherence to good industrial practices and the application of appropriate engineering principles such as fail safe battery management systems are not discussed further in the following.

2

Overview of risks

There are risks arising from the normal mode of operation and risks from abnormal operating conditions.

2.1

Risks arising from normal operating conditions

2.1.1 Electrical risks The kinetic energy of a small vehicle (1000 kg) driving at 100 km/h is 107 Wh. This compares to an electrical energy content of a 30 Ah lithium-ion cell. The sudden release of this amount of energy may obviously cause catastrophic results. Depending on the design and chemistry, the short circuit current of a battery can easily be in the range of a few thousand amperes even for small cells. Batteries deliver short-circuit currents for many seconds, much longer than supercapacitors. In contrast to that, conventional capacitors can provide very high currents for a few milliseconds. For this reason it is always necessary to protect the system from extremely high discharging currents over a prolonged period of time unless these are application specific. An example for an application specific high current is the starting process of an internal combustion engine. The battery is more or less short circuited in the initial phase of starting the engine and it is therefore unusual and expensive to provide fuses as a protection. Cells connected in series have a high voltage and once the DC voltage exceeds 60 V, standard electrical engineering principles require a number of protection mechanisms which go beyond those at lower voltage levels. The choice of a nominal voltage of 48 V for many applications is the result of this safety limit, as 60 V is not reached even under charging conditions. Above 1000 V DC the technical requirements are even higher and thus very few battery systems exist with a voltage above 1000 V DC.

2.1.2 Gas emissions risks All batteries with aqueous electrolytes generate hydrogen and oxygen gas during operation, even if not connected to chargers. As a result, the atmosphere inside the cell and the gases emitted from the cell vents must be considered explosive at all times. Good industrial practice minimizes the risks, such as prevention of electrostatic discharges and venting of the area in which the batteries are kept. There are mandatory guidelines to remove gases which are emitted from the cell so that gas explosions are prevented (see EN 60896). However, corrosion processes may lead to arcing inside the cell and may lead to explosions particularly when there is a sudden high current pulse delivered by the battery. There are no statistics available as to the frequency of their occurrence, but the consequences can be very severe. It should be noted that lithium-ion cells generate small quantities of toxic gases during their lifetime and the internal pressure increases. When opening a cell, special precautions should therefore be taken.

2.2

Risks associated with abnormal operating conditions

To ensure safety of operation, a number of abuse tests have to be carried out. These reflect abuse conditions, e.g., short circuiting a cell with a resistance of 5 mO until the voltage has dropped to zero volt.

2.2.1 Electrical risks Faulty loads, connections and damaged insulation cause risks and therefore monitoring systems are in place at least for larger battery systems, e.g., fuses and monitoring of the insulation resistance. The biggest problem are faulty loads which draw an excessive current, sufficiently high to create high temperatures and fire but too low to trigger a fuse. This problem is not unique to batteries but is a general electrical engineering problem.

2.2.2 Risks from overcharging and overdischarging The use of a fail-safe battery management system is able to avoid the associated risks. Risks from mechanical damage, external heat sources and uncontrolled short circuits. These risks are not unique to lithium-ion batteries but the damage is much more severe than for other battery systems. If, for whatever reason, a lithium-ion cell heats up beyond its critical temperature it may suffer degradation processes which affect its safety and, at a temperature usually referred as the onset temperature, suffer a thermal runaway process which cannot be stopped. Details of the process and concepts to reduce their occurrence and severity have been discussed in numerous articles (e.g.3,4). The process can be characterized by:

• • • •

exothermic processes from the decomposition of the solid electrolyte interface and the decomposition of the cathode material, melding and shrinking of the separator thus enabling an internal short circuit, increase of internal pressure, and ultimately a fast rise in temperature (>1 K/s) with cell and gas temperatures in the range of 600 to 800
C and local temperature hot spots which may exceed 1250
C.

Electrochemical Terminology | Safety and State-of-Safety (SoS)

561

This process is referred to as a thermal runaway5 and, once the initial cell which undergoes a thermal runaway causes the thermal runaway of neighboring cells, thermal propagation. It is generally accepted that it is not yet possible to design, manufacture and use cells and battery systems in such a way that thermal runaway cannot occur. As a result, cells are usually equipped with vents (which rupture before there is a risk of the cell exploding) and gas ducts to ensure a controlled transport of gases to the atmosphere. Thermal runaway processes are also a phenomenon in lead acid batteries and nickel based alkaline batteries, however do not lead to significant safety risks. The battery, however, is destroyed. The amount of heat generated by exothermic reactions is much higher than the electrical energy which can be drawn from the battery during discharging. If the gases emitted from a cell start to burn, the exothermic reactions with oxygen increase the amount of released energy by a factor of about 4, and the environment of the cell, module or battery catch fire. Due to the presence of fluorine and phosphorous in the electrolyte, highly toxic substances (HF and organic phosphorous compounds similar to insecticides and pesticides) can be formed. Cleaning up after a thermal runaway event therefore requires protective clothing.

3

Ensuring safety

Lithium-ion cells are safe and reliable products. Safety concerns consider the high number of cells produced annually and the high potential damage associated with a failure of a cell at any point between production and safe disposal. In all areas where products and systems may cause major damage and loss of life (mining, aircrafts, railway systems) there is a process to reduce risks, generally referred to as “documented safety.” The questions raised are: 1. What might go wrong and how will a fault propagate through the system (failure tree and failure mitigation analysis)? 2. What is the likelihood of something going wrong? 3. What will be the consequences regarding functional requirements, health and safety risks and damage to other objects and the environment? 4. Is it possible to control the course of events in the case of a failure arising? The questions raised include not only cell design but also production, systems integration and maintenance. It is required to document this process in full. ISO 26262 (road vehicles – functional safety) provides a detailed framework to assist the development process and is applicable to batteries. It is interesting to note that the role of the overall safety management (chapter 2-5 of ISO 26262) and safety culture (chapter 12-6 of ISO 26262) are explicitly highlighted in standards. Ashtiani6 proposed to assign 10 likelihood levels (see Table 1). The European Car Association (EUCAR) introduced 8 hazard levels, level 0 being effects which do not even lead to loss of functionality (Table 2). Similar hazard levels have been proposed by the Society of Automotive Engineers (SAE). Test procedures have been defined to show what the safety level of a cell or module is under the test conditions which have been agreed upon and manufacturers do provide information on the test results. In addition EUCAR has introduced so-called Automotive Safety Integrity Levels (ASIL) to combine severity of a hazard, exposure (impact on health and the environment) and controllability (options to mitigate faults as they occur). Each category has three levels. Depending on the combination of these levels, actions are required. In Table 3, QM denotes that this safety level is only a quality management issue, whereas A to D are various risk categories requiring actions. ASIL level D (severe, high exposure, limited or no controllability) are considered unacceptable and the product must be reengineered. ASIL levels can also be called safety categories. Obviously, the quality of this approach depends on the people who carry it out and the management systems which got sufficient training to have the appropriate know-how, allow time and resources and take concerns seriously. It may be an interesting exercise for the reader when studying accident reports to consider also the role of the company’s management and safety culture. A more fundamental approach to ensuring product safety has been suggested in.5,7 In analogy to standard procedures in chemical engineering, it is proposed that cells are designed, manufactured and monitored in such a way that they always remain within a “safe zone” (Semonov theory) under conditions where the rate of heat removal is always greater than the maximum rate of Table 1

Categories of likelihood levels of a fault arising (after2).

Category

Likelihood

Description

10 9 8 7 6 5 4 3 2 1

10% 5% 2% 1% 0.5% 0.2% 0.1% 0.05% 0.01% 0.001%

Extremely high Very high High Above average Average Below average High low Average low Low Very low

562

Electrochemical Terminology | Safety and State-of-Safety (SoS)

Table 2

Eucar hazard levels.

Hazard Level

Description

Criteria for severity classification and effects

0 1

No Effect Reversible loss of function

2

Irreversible defect/damage

3

Leakage, mass change Mozambique (23000), and the total is 1,200,000 metric tons. Natural graphite is highly crystalline, and its XRD pattern shows a sharp peak at around 2y ¼ 26.5
corresponding to the 002 line. A scanning electron microscope (SEM) image of flake graphite is shown in Fig. 4.2 Graphite has two different crystal structures as shown in Fig. 5, namely, (1) hexagonal and (2) rhombohedral.3–5 The normal form is hexagonal but the rhombohedral structure tends to be formed, for example, when shearing stress is applied to the crystal during milling. The molecular-level structure of graphite is very characteristic due to the stacking of the two-dimensional graphene

a c a

rp*

c0

c

0.760 nm

p

= 60°

a = 0.1379 nm c = 0.1207 nm

0.501 nm

rp* = 0.069 nm

0.501 nm

b a b 0.233 nm

rc*

0.476 nm 0.416 nm (a)

(b)

c

c0

= 30°

a = 0.1379 nm b = 0.1282 nm rc*= 0.1194 nm

(c)

Fig. 1 Structure of carbyne. Reproduced with permission from Inagaki, M. (ed.) Carbon Family, Tokyo, Japan: Agune Shofu Sha Publishing Co. 2001; p 63.

602

Chemistry and Electrochemistry | Carbon

Counts

(c)

(b)

(a)

10

20

30

40 2

50

60

(q)

Fig. 2 X-ray diffraction (XRD) pattern of carbynes and graphite. (a) Cu-terminated carbyne formed from Cu-acetylene; (b) phenyl-terminated carbyne formed from phenyl, and Cu-acetylene; (c) graphite. Reproduced by permission of John Wiley & Sons Inc. Cited from Cataldo, F. A Study on the Structure and Electrical Properties of the Fourth Carbon Allotrope: Carbyne. Polymer Int. 1997, 44, 191–200.

Fig. 3 Photograph of natural graphite ore. Courtesy of Kyushu University Museum.

planes, which are comprised of double bond-conjugated six-member carbon rings as shown in Fig. 5. The plane parallel to the graphene layer is called as the “basal plane,” and the perpendicular plane is termed as the “edge plane.” The edge plane has openings between the graphene layers. The graphene layer is mechanically very strong as the average bond order is larger than that of diamond, whereas the interlayer bond, along the c-axis direction (interlayer binding), is merely due to van der Waals forces, and accordingly, is very weak, allowing easy cleavage along the graphene layer direction. This is the reason why graphite is used as a solid lubricant. The main constituent of the lead of pencil is graphite.

Chemistry and Electrochemistry | Carbon

603

Fig. 4 Scanning electron microscope (SEM) image of flake graphite (KS6, Timcal). Reproduced with permission from Aurbach, D.; Teller, H.; Levi, E. Morphology/ Behavior Relationship in Reversible Electrochemical Lithium Insertion into Graphitic Materials. J. Electrochem. Soc. 2002. 149, A1258. Hexagonal Z

000

A

1

1

B 00 2

00 2 1 2 1 3 3 2

000

A X

000 2 1 3 3

0

000

000

Y

(a)

Rhombohedral 2 1 3 3

A 000

0

2

00 3

C 1 2 2 3 3 3

B 1 2 1 3 3 3

2 1 1 3 3 3

A 2 1 3 3

0

000

(b)

Fig. 5 Crystal structure of graphite. Reproduced with permission from Inagaki, M. (ed.) Carbon Family, Tokyo, Japan: Agune Shofu Sha Publishing Co. 2001; p 63.

604

Chemistry and Electrochemistry | Carbon Table 1

The crystallographic parameters and physical-properties of graphite.

Item

Characteristics

Crystal structure Lattice parameters (hexagonal structure)

Hexagonal structure, Rhombohedral structure ao ¼ 0.24612 nm co ¼ 0.6708 nm d002 ¼ 0.3354 nm 2.26 Direction of a-axis 4  7  10−5 O cm Direction of c-axis 1  5  10−1 O cm Coefficient of static friction: 0.3 Coefficient of dynamic friction: 0.1 or less 700
C in air. 2.05 cal/gatm
C (at 300 K) 0.2  0.4 cal/cm sec
C Chemically stable except in strong acids at room temperature

Mean distance between layers Specific gravity Volume resistivity Coefficient of friction Heat-resisting property Specific heat Heat conductivity Chemical stability

The crystallographic parameters and the physical properties are shown in Table 1. As seen in Table 1, the interlayer distance between graphene layers is as long as 0.3354 nm, and many kinds of chemical entities can be intercalated in the interlayer space. The powder XRD pattern of graphite is shown in Fig. 6, where the 002 diffraction intensity is very strong. In Fig. 7 the transmission electron microscope (TEM) image of natural graphite is shown, where the dozens of 002 fringe are stacked in parallel. Such an image is typical of a well-crystallized layered structure. In spite of the strong anisotropy in the crystallographic structure, there is a product named “isotropic graphite” whose bulk electric and mechanical properties are isotropic. This is due to the random orientation of small graphite powder particles. The product is formed by binding a finely pulverized graphite powder with a pitch-based binder, kneading well, and heating at high temperature to pyrolyze the binder. The synthesized material is highly anisotropic in the microscopic view but highly isotropic based in the bulk sense.

1.2.2.2 Crystal formation Crystalline graphite is formed by heating a precursor such as petroleum pitch at high temperature. Such a precursor is referred to as “soft carbon,” “graphitizable carbon,” or “mesophase carbon.” Another category of carbon raw material is named “hard carbon” or “non-graphitizable carbon” (which is better called “hardly graphitizable carbon”). Hard carbon is quite difficult to transform into a well-developed single crystal of graphite even though local crystallization takes place. The reason why hard carbon is very difficult to transform into graphite is that the embryos of graphite (crystallites), having random orientation, are intrinsically joined together by strong carbon–carbon bonds, which are difficult to break. Schematic representations of the two types of carbon raw materials and the crystal growth processes are illustrated in Fig. 8.6 The process temperature should be as high as 3000
C. However, the presence of an appropriate catalyst, such as cesium metal, assists the growing of graphite crystals at low temperature, where the raw material and catalyst should be contained in a high-pressure vessel. In the total graphite structure, shown in Fig. 5, the planes comprised of six-membered carbon atoms are stacked in layers so that each carbon atom lies above another in every second layer. In the case of a “turbostratic structure” which is often encountered in graphite’, however, the interlayer arrangement is irregular even though the interlayer parallelism is maintained. In Fig. 9, the top

004

002 101

110

102 104

20

30

40

50

60

70

105

100

006 112

103

80

90

2 (Cu KD)(°)

Fig. 6 X-ray diffraction (XRD) pattern of natural graphite. Reproduced with permission from Inagaki, M. (ed.) Carbon Family, Tokyo, Japan: Agune Shofu Sha Publishing Co. 2001; p 63.

Chemistry and Electrochemistry | Carbon

605

5 nm

Fig. 7 High-resolution transmission electron microscopy image of natural graphite. Courtesy of Miss Lijun Fu, Department of Chemistry, Fudan University, Shanghai, China.

Graphitizable carbon

Hardly graphitizable carbon

Nonorientation structure

Nonorientation structure

Heating temperature

Local orientation Basic structural unit

Crystallite

Local orientation Basic structural unit

Crystallite

Fig. 8 Structure models of graphitizable carbon and less graphitizable carbon. Reproduced with permission from Inagaki, M. (ed.) Carbon Family, Tokyo, Japan: Agune Shofu Sha Publishing Co. 2001; p 63.

606

Chemistry and Electrochemistry | Carbon

Fig. 9 Top view of turbostratic structure of two graphene layers oriented in parallel. Mutual orientation differs between (a and b), accommodated between the two layers.

: Li particles can be

views for two examples are shown for a turbostratic structure between two parallel layers where the in-layer orientation differs. There are only a restricted number of coincidences in position for a six-member ring between the two layers. 1.2.2.3 Manufacturing of graphite One of the important commercial graphite products is electric furnace electrodes. An example of the production process is illustrated in Figs. 10–12. For the graphitization, the most popular electric furnace is the Acheson furnace but now other furnaces (Fig. 13) are also utilized. 1.2.2.4 Applications There are a number of application fields for graphite and the main application fields will be shown by classification as follows:

• • •

Electric power applications (1) Furnace wall material for nuclear fusion test equipment. (2) Nuclear reactor materials. Electric furnace applications (1) Electrodes of the arc furnace for making steel. Power source applications (1) Battery applications Negative electrode active material of lithium-ion batteries. Positive active materials for zinc–air batteries.

Silo Raw material treatement

Crusher

Coke

Sieving Binder pitch Kneading

Mixing tank Kneader

Extrusion molding Fig. 10 Production process of graphite electrodes-1 crushing and kneading. Courtesy of Tokai Carbon Co., Ltd.

Chemistry and Electrochemistry | Carbon

607

Pitch

Pitch impregnation

Pitch impregnater

Primary heating at 1000 °C

Primary furnace

Fig. 11 Production process of graphite electrodes-2 pitch impregnation and preheating. Courtesy of Tokai Carbon Co., Ltd.

Secondary heating at 700 °C

Furnace

Graphitization at 3000 °C

Shaping

Fig. 12 Production process of graphite electrodes-3 secondary heating and graphitization. Courtesy of Tokai Carbon Co., Ltd.

Fig. 13 Continuous graphitization process machine. Courtesy of SEC Carbon Co., Ltd.

Conductive materials for lithium-ion batteries, lithium–manganese dioxide batteries, lithium–thionylchloride batteries, zinc–carbon batteries, alkaline manganese oxide batteries, zinc–silver oxide batteries. (2) Fuel cell (FC) applications Separator materials. Positive electrode materials. (3) Capacitor applications Negative and positive electrode materials for double-layer capacitors.

608

•

•

• •

• •

Chemistry and Electrochemistry | Carbon

Electric equipment applications (1) Direct motor brush for steel and other heavy industries and electric trains. (2) Brushes for car electric appliances. (3) Brushes for the motor of home appliances. (4) Moving electric contact parts: Widely used as clutches and brakes for construction, agricultural, transportation, and other industrial machinery. Electronic equipment applications (1) Key board and other touching switches. (2) Button switches of cellular phones and others. (3) Touch panel application. (4) Conductive paint for screen printing, cathode-ray tube black matrix, and others. (5) Conductive additives for tantalum solid capacitors. (6) Light shadowing for black matrix of electronic monitors. (7) Conductive slurries of coating on the inner surface of the cathode can of alkaline manganese oxide batteries. (8) Conductive pastes. Semiconductor industry applications (1) Heater rods for polycrystalline silicon production. (2) Czochralski (CZ)-crystal pulling equipment. (3) Silicon carbide-coated graphite is used for silicon wafer annealing furnace. High-temperature equipment (1) Heater and heat shield for manufacturing optical fibers. (2) Heat shield and heater for diamond sintering for making diamond blade. (3) Die-cast production of high-quality nonferrous cast product. (4) Crucible as a container of the treating material. Moving contact equipment applications (1) Composite of graphite powder and metal is used for clutches and brakes. (2) Lubricant for machining process (applied to extrusion and engine oil additives). (3) Lubricant for high-temperature-extrusion machining. Stationary applications (1) Pencil lead. (2) Ink for dot printer.

1.2.3 Fullerenes and nanotubes 1.2.3.1 Introduction The existence of C60 was predicted by E. Osawa of Toyohashi University of Technology in 1970, who suggested that the structure was just like a soccer ball. During the course of molecular beam experiments on black carbon soot prepared by irradiating graphite with a laser beam, H. Kroto, R. Curl, and R. Smalley from Rice University, Houston, TX, USA, discovered Buckminsterfullerene (C60) in 1985 (Fig. 14) and were awarded the 1996 Nobel Prize in Chemistry. A micrograph of the C60 crystals is shown in Fig. 15. Buckminsterfullerene (C60) was named after Richard Buckminster Fuller, a noted architect who popularized the geodesic dome, as buckminsterfullerenes have a similar shape to that sort of dome. As the discovery of the fullerene family came after buckminsterfullerene, the name was shortened. The molecular structures of fullerenes and nanotubes are shown in Fig. 16 which illustrates the relationship among the fullerenes families. Carbon nanotubes, CNTs were discovered by S. Iijima.7 After the discoveries of fullerenes and CNTs, a number of studies have been conducted; 177 papers on fullerenes and 178 papers on CNTs, have been published upto 2007 April. The research fields have covered fundamental analysis including quantum mechanical calculations, physical measurements, structure analysis, and organic synthesis, various applications in the electronics industry, medical field, and cosmetics. Fullerenes have a cage structure in which one can insert atoms and simple molecules. Cage-in compounds such as those shown in Fig. 17 have been synthesized. The compounds scandium dicarbide and scandium nitride are very reactive in an ambient atmosphere but if they are caged-in as Sc2C2@C84 or Sc3N@C80 where @ means the state of being caged in the fullerene molecule as shown in Fig. 17 they become chemically stable. Iijima showed TEM images of a single wall nanotube and a multiwall nanotube (Fig. 18). The discoveries of Iijima pioneered the research and development of CNTs extensively and now a wide array of information on CNTs can be obtained from internet homepages. The physical properties of nanotubes are very attractive. They are mechanically extremely strong, the electrical conductivity can be managed from highly conductive to semiconductive, and the products enable fine processing technology. The construction of sophisticated molecular structure of nanotubes is shown in Fig. 19, where many C60 molecules are caged into a nanotube. When we heat this tube, the C60 begins to polymerize to form a nanotube inside the mother nanotube. Such a tendency is useful to provide a new type of tube material. 1.2.3.2 Production Now there are a number of research groups including MIT group and venture companies beginning to prepare nanotube and fullerene samples. In Fig. 20 nanotube products produced by NanoLab are shown. The National Institute of Advanced Industrial

Chemistry and Electrochemistry | Carbon

609

Fig. 14 Molecular model of fullerene C60. Courtesy of Dr. Kentaro Satoh.

Fig. 15 Photograph of fullerene C60 crystals formed in soot (250). Courtesy of Wikipedia.

and Science Technology (AIST), Japan announced the highly efficient synthesis of impurity-free single-walled carbon nanotubes (SWCNTs) in Science. Macrostructure arrays of SWCNTs are shown in Fig. 21. These were prepared by gas flow chemical vapor deposition (CVD) methods using no substrate. The reactors capable of large-scale production of SWCNTs and was developed by the joint research of the Research Center for Advanced Carbon Materials, Japan and Nikkiso Co., Ltd., Japan (Fig. 22). The Nano-C company located in Westwood, MA, USA is aiming to make a low-cost large-scale production of fullerenes and nanotubes by controlled premixed combustion of hydrocarbons.

1.2.3.3 Applications 1.2.3.3.1 Fullerenes According to Satoh, when the complex of cyclodextrin and fullerene sodium hyposulfate were mixed with a gas mixture source of nitrogen and hydrogen and irradiated with light, ammonia was formed with a maximum yield of 45%. This finding is epoch

610

Chemistry and Electrochemistry | Carbon

C240 C540

C70

C60

C82

C108

Fig. 16 Polymorphism of fullerenes. Reproduced with permission from Inagaki, M. (ed.) Carbon Family, Tokyo, Japan: Agune Shofu Sha Publishing Co. 2001; p 63.

Fig. 17 Molecular model of fullerene C80 and C84 caged with Sc3N and Sc2C2. Courtesy of Dr. Kentaro Satoh.

making, because until now the production of ammonia is known to be impossible in the absence of a metallic catalyst. Fullerene requires no metallic catalyst for formation of ammonia. This example suggests that fullerenes do have a great latent potential of usefulness for applications for developing new medicines effective for recovering against diseases such as AIDS, cancer, and so on. Fullerenes, as precursors of the microcrystal of artificial diamond, are effective in providing highly pure diamond, facilitating superhard-cutting blades. When fullerenes are coated on both sides of a piezoelectric crystal of a chemical sensor, the detection sensitivity can be improved remarkably. If C60 is incorporated in the photoresist for micropatterning, the mechanical strength becomes much enhanced allowing more precise patterning.

1.2.3.3.2 Nanotubes Carbon nanotubes have a versatile application field and have now been put into practice, although it is still at a very early stage of development. Conductive nanotubes of several nanometer diameters can be used as the tips of detecting electrodes of the scanning tunneling microscope (STM) or the atomic force microscope (AFM), where the precision will be dramatically enhanced. It is estimated that a rope of nanotubes having no defect and having a diameter of 1 cm can bear the weight of 1200 tons. If nanotubes applied to reinforce tall buildings, then they will be very durable against major earthquakes. Because of their low density, one can make any kind of vehicles, such as Shinkansen automobiles, to be super lightweight and very strong, leading dramatic energy saving. It has been reported that the nanotube-reinforced tennis rackets are already out in the market.

Chemistry and Electrochemistry | Carbon

611

Fig. 18 Carbon nanotubes. Electron micrographs of microtubules of graphitic carbon. Parallel dark lines correspond to the (002) lattice images of graphite. A cross section of each tubule is illustrated. (a) Tube consisting of five graphitic sheets, diameter 6.7 nm; (b) two-sheet tube, diameter 5.5 nm; (c) seven-sheet tube, diameter 6.5 nm, which has the smallest hollow diameter (2.2 nm). Reproduced with permission from Iijima, S. Helical Microtubules of Graphitic Carbon. Nature 1991, 354, 56–58.

Fig. 19 Model of a nanotube in which four C60s are caged in; the situation is captured using transmission electron microscope (TEM) in soot. Reproduced with permission of Dr. Satoh cited from http://www.org-chem.org/yuuki/nanotube/nanotube_en.html.

Fig. 20 Scanning electron microscope (SEM) images of multiwall carbon nanotubes grown on the nickel metal substrate and silicon solid substrate. Courtesy of Microphase Co., Ltd., Tsukuba, Japan.

Our social life owes much to the computer however, silicon-based technology will tend to become limited in miniaturization. A nanotube semiconductor system is capable of providing a smaller semiconductor device by one order of magnitude. In addition, the processing speed will be 10 times faster. As it is known that a very tiny electron beam can be available from the tip of a CNT, combination with a fluorescent plate will give us very high precision display devices. Another important application field is for power sources. CNT can absorb large amounts of hydrogen. The amount is estimated to be up to 13–14%, which enables us to use it as a lightweight hydrogen tank. A Fuel cell car installed with a CNT hydrogen tank is

612

Chemistry and Electrochemistry | Carbon Each rod diameter: 10 µm

2.5 mm

Top view

Side view

High density alignment of SWCNT bundle

Cylindrically aligning of SWCNT bundle

Plate-like aligning of SWCNT

Fig. 21 Several types of the bundle of single-walled carbon nanotubes (SWCN). Courtesy of AIST, Japan.

Electric furnace High frequency electrode Source gas + carrier gas

Substrate

Plasma

Vacuum pump

High frequency electrode

Fig. 22 Plasma chemical vapor deposition (CVD) reactor for carbon nanotube synthesis. Schematic diagram of a plasma CVD system in Wikipedia appearing on September 20, 2008 was modified. (http://ja.wikipedia.org/wiki/%E3%83%95%E3%82%A1%E3%82%A4%E3%83%AB:PlasmaCVD-JP.PNG).

now under development. Further, the specific high surface area of CNTs is attractive for developing double-layer capacitors working with CNTs as the active electrode material. CNTs are now being examined for the negative active mass of lithium ion batteries.

1.2.4 Diamond 1.2.4.1 Introduction Diamond is the sp3 carbon allotrope. Diamond is one of the most precious jewels as it is highly stable, hard, colorless, transparent, and the refractive index is very high (nD ¼ 2.417). It is very unique since it is an element jewel. The density is 3.6 g cm−3, which is the highest among the carbon allotropes. The Debye temperature is very high (2240 K). Basically it is an electric insulator, whereas the thermal conductivity is high (0.33 cal cm−1 s−1 deg−1). The crystal has a cubic lattice. Diamond appears to have no power source application, but once it is doped with boron it becomes electrically conductive. It is well-known that there are two crystal directions of cubic and hexagonal structure in diamond whose crystal relationship is shown in Fig. 23. In Fig. 24 a photograph of a natural diamond sample is shown.

1.2.4.2 Production and applications Once it was believed that diamond would be impossible to synthesize, however, later its preparation was successful under very high pressure and temperature in the presence of a catalyst (nickel or its salt). In 1955, General Electric Company in the United States succeeded in synthesizing large crystal by the use of nickel catalyst. It was light yellowish green, which is due to the presence of nickel(II). In 1963, the Tokyo Shibaura Electric Company (presently Toshiba Corporation, Japan) applied for a patent for the preparation of diamond films, which is based on a gas-phase reaction. In 1967 C. Angus et al. published a paper on the vapor deposition of diamond seed in the Journal of Applied Physics. Later on, a number of research papers have been published based on the CVD method to synthesize a diamond polycrystalline films or coatings. At present this method provides a wide variety of practical uses such as diamond saws, slicing disks, laser beam windows, lens arrays, knives for surgery operation, saw blade gear foils, and high-fidelity speakers. Doping of boron into diamond makes diamond electron–conductive. It is used as an electrode for electrochemical synthesis and analysis.

Chemistry and Electrochemistry | Carbon

613

Fig. 23 Mutual relationship of carbon tetrahedrons in diamond. (a) Stereoscopic relationships, (b) hexagonal diamond, and (c) cubic diamond. Reproduced with permission from Inagaki, M. (ed.) Carbon Family, Tokyo, Japan: Agune Shofu Sha Publishing Co. 2001; p 63.

Fig. 24 Photograph of natural diamond ore host rock picked in Russia. Courtesy of Kiseki Museum, in Fujinomiya, Shizuoka, Japan.

1.3

Various types of carbon families

1.3.1 Amorphous carbon 1.3.1.1 Introduction Carbon materials showing no sharp peaks on the XRD pattern as in Fig. 25 (coke fired at 1000
C) is called amorphous carbon cokes, carbon blacks (CB), glass-like carbon, charcoal, and so on are included in this category. In the following sections, these materials are described.

1.3.2 Carbon black 1.3.2.1 Introduction Carbon black (CB), as soot, has a long history, that is, it was utilized in ancient Egypt and China for writing letters and for paints on papyrus or bamboo strip. It was prepared by the incomplete combustion of oils or gases. In the second century it was produced by a lamp method, and later on, produced by a “Channel Method” in 1892, and the “oil furnace method” was developed in 1947. TEM images of conventional CB are shown in Figs. 26 and 27; the average diameter of elementary CB particles is as small as about 30 nm but the interparticle attractive force is high, causing the formation of chain-like structure aggregation, which is called as “structure formation.” Pristine CB powder does not show any sharp peak in the XRD pattern implying it to be amorphous. Owing to the small size of each particle the specific surface area is high (50–150 m2 g−1); the surface is liable to react with the ambient air, forming oxygen-containing functional groups. Therefore, the surface can be modified by surface-treating reagents. The as-obtained powder is very bulky and the tap density is low. In general, the surface of carbon is hydrophobic and repels water, but it can be modified to be hydrophilic and applied to battery use. The intrinsic hydrophobic tendency is very useful for the strengthening of rubber. Vulcanization with CB makes rubber not only highly durable against weather, but also mechanically strong with high extensional

614

Chemistry and Electrochemistry | Carbon

002

Count (au)

10 004 11 20

30

40

50

60

70

80

90

2 (°)

Fig. 25 X-ray diffraction (XRD) pattern of coke. Reproduced with permission from Inagaki, M. (ed.) Carbon Family, Tokyo, Japan: Agune Shofu Sha Publishing Co. 2001; p 63.

Fig. 26 Transmission electron microscope (TEM) images of ordinary furnace carbon black. Courtesy of Tokai Carbon Co., Ltd.

Fig. 27 Transmission electron microscope (TEM) images of graphitized carbon black. Courtesy of Tokai Carbon Co., Ltd.

and compressive strength. This is due to the strong bonding between the rubber molecule and the CB surface. Without CB, the wheel tire could not have been realized, and accordingly, automobiles as well as airplanes would not have existed for practical use. Acetylene black (AB) and Ketjen black also belong to the CB group, and heating at high temperature can make the pristine powder electrically conductive, when partial graphitization proceeds. Nanosized particles are aggregated to form “structure,” where the particles are bound together to form a chain having branches, which is caused by the strong affinity among particles due to van der Waals forces. The structure-forming tendency is quite effective for a nonconductive matrix to change to an electrically conductive one.

615

Chemistry and Electrochemistry | Carbon

1.3.2.2 Physicochemical properties Examples of the physicochemical properties of CB are shown in Table 2. The special characteristics of CB are attributed to the very small particle size of individual carbon particles whose diameter ranges from 15 to 50 nm. In general, the surface of CBs is chemisorbed with oxygen containing functional groups such as carboxylic, quinonic, lactonic, phenolic groups, and so on. Most of these surface oxygen groups can be removed by heat treatment and therefore, referred to as volatile content.

Table 2

Carbon blacks, products and properties.

Tokai carbon black

Sample A

Sample B

Sample C

Sample D

Sample E

Sample F

Mean diameter  1 (nm) Specific surface area by N2 adsorption  2 (m2 g−1) I2 adsorption amount (mg g−1) Amount of dibutylphlthalate (DBP) absorption  3 (method A) (cm3 100 g−1) Tinning strength ⁎4 (%) Volatile content  5 (%) Ash  6 (%) Sieving residue (over 150 mm) (%) Tap density (Kg m−3)

18 142 139 130 129 0.4 0.3 0.002 310

19 142 139 115 129 0.4 0.3 0.002 370

19 126 120 125 124 0.4 0.3 0.002 330

22 119 121 114 115 0.4 0.3 0.001 330

23 106 111 75 117 0.4 0.3 0.002 450

22 99 104 129 106 0.4 0.3 0.001 320

Denka black

Mean diameter (nm) Surface area (m2 g−1) Amount of I2 adsorbed (mg g−1) Tap density (g mL−1) HCl absorbed (mL 5 g−1) Electric resistivity (Ocm) Ash (%) Water content (%) Coarse particle (ppm) pH  7

As obtained

Pressed

35 68 92 0.04 16.8 0.21 0.01 0.01 9  10

Granular

50%

75%

100%

FX35

HS100

36 65 88 0.08 15.8 0.20 0.01 0.04 6.0 mass%.25a In the presence of catalytically active metal residues, spillover-surface diffusion-chemisorption can occur, i.e., a phenomenon of migration of atomic hydrogen from a catalytically active metal to the carbon material. Thus, in addition to the molecular hydrogen adsorption, its chemisorption in the form of atoms can occur, although H2 dissociation to atoms cannot proceed on carbon surfaces. Doping with transitional metals, such as platinum (Pt), palladium (Pd), and nickel (Ni), on hydrogen storage materials is observed to be increased the hydrogen storage capability and stability due to spillover phenomenon.25b The electrochemical response of highly dispersed carbon materials is largely due to the nonfaradaic process of charging-discharge of the capacitor formed by the electric double layer (its capacitance is 10–100 F g−1) and redox transitions involving the electroactive groups of the material. The carbon-based material with light weight and high electronic conductivity can be used to store atomic hydrogen in its interlayers and pores when an aqueous solution is electrolyzed on the cathodic polarized carbon. The stable chemisorption of atomic hydrogen on single layer graphene looks as a good example showing a theoretical maximum hydrogen storage capacity of 7.7 mass%.23a The phase diagram and the structural studies of the water-hydrogen system evidence the formation of clathrate structures at low temperatures (below 250 K) and high pressures (100–360 MPa). The labile frame of this structure is built of host (water) molecules containing embedded guest molecules (hydrogen). The maximum theoretical stoichiometry of a clathrate corresponds to the formula H2 2.125H2O (5.2 mass% hydrogen). Hydrogen clathrate hydrate is now become a promising technology for H2 storage

646

Chemistry and Electrochemistry | Hydrogen

with specific features such as low energy utilization for charging and discharging, low fabrication costs, safety, and most preferably no any negative environmental effect.26 The mechanism of hydrogen storage via hydrate formation is similar but not identical to physisorption mechanism. Double clathrates of different compositions can be formed with organic additives (like methyl tert-butyl ether (MTBE), methylcyclohexane (MCH), 2,2,3-trimethylbutane (2,2,3- TMB), and 1,1-dimethylcyclohexane (1,1-DMCH)), water-soluble hydrating agents (ethylene oxide (EO), tetrahydrofuran (THF)), or gaseous fuels (methane, propane), which are stable under less severe conditions. An organic clathrate compound made of a Fullerene (C60) guest and a hydroquinone host framework shows an enhanced hydrogen-storage capacity and a good structural stability under 10 GPa pressure and 438 K temperature.26 Chemical methods of hydrogen storage use materials of different types. The first group includes substances (ammonia, unsaturated hydrocarbons, fullerenes, etc.) that contain hydrogen in their composition and are capable of its liberation under certain conditions (temperature, catalysts). The second group includes substances not obligatorily containing chemically bound hydrogen but capable of its liberation on exposure to water (iron, silicon, aluminium, and its alloys, e.g., aluminium-gallium alloy).27 Hydrogen can also be obtained in the reaction of certain hydrides formed by metals and alloys with water. Storage of hydrogen in the form of metal hydrides attracted attention since the discovery of hydrogen dissolution in palladium (T. Graham) in 1866. The boom in the active research was observed in the 1960s when intermetallic compounds capable of reversible absorption of hydrogen at moderate temperatures and pressures were discovered.28 Today, several metal hydrides and intermetallic compounds are known.29 The optimal material for hydrogen storage should fulfil the following requirements: the highest possible hydrogen capacity per mass and volume units; low temperature of hydride dissociation; moderate dissociation pressure; low heat of formation to minimize the energy consumed in hydrogen liberation; low heat of dissipation during the exothermal formation of a hydride; reversibility of sorption-desorption processes; fast kinetics; high stability (also toward oxygen and moisture); high cyclability; the lowest possible cost (including recycling and development of the service infrastructure); and high safety. In electrochemical applications, corrosion resistance and fast response to current and potential variations are also important. Reversible formation of metal hydrides can proceed via the direct interaction of a material with gaseous hydrogen or as a result of electrochemical discharge of water molecules or hydrogen ions in electrolytes.30a The second method is used in metal hydride power sources. Hydrides are divided into three classes with respect to the bond type and the structure. An ionic hydride is a salt-like compound that contains hydrogen in the form of a hydride ion H− and a metal cation (hydrides of alkali and alkali earth metals, rare earth elements, gallium, indium, thallium).29 In the solid state, ionic hydrides are dielectrics; however, they dissociate in nonaqueous polar solutions and upon melting. As a result, their solutions and melts conduct current and produce hydrogen on the anode during electrolysis. As a rule, ionic hydrides are Lewis bases. They react with the other acidic hydrides to produce complex compounds (e.g., lithium aluminium hydride (Li[AlH4]) and sodium borohydride (Na[BH4])). Complex metal hydrides form a special class of ionic hydrides, of which sodium alanate Na[AlH4] is the best known.30b Alanates doped with titanium (Ti[AlH4]) and doped with zirconium (Zr[AlH4]) can reversibly absorb and evolve hydrogen. Doping with suitable catalysts such as TiF3, TiH2, TiCl4 and TiCl3 are most demonstrated to exhibit an appropriate alter effect in alanates.30c The maximum hydrogen theoretical gravimetric capacity of sodium alanate is 5.5 mass% and 5.0 wt% hydrogen release achieved so far. However, in practice, the resulting data cannot be realized because of the stepwise nature of hydride dissociation (NaAlH4 ! 1/3Na3AlH6 + 2/3Al + H2; Na3AlH6 ! 3NaH + Al + 3/2H2). They are amongst the advanced and intensely studied materials for the hydrogen storage applications. Still, if it looks clear that its performance will be not suitable for on-board applications, it is considered as promising solution for stationary applications. Metal nitrides (M3Nx), imides (M2(NH)x), and amides (M(NH2)x), where x is the valence of metal, are examples showing continuous substitution of metal by hydrogen in the M-N-H system. Lithium nitride (Li3N) can store hydrogen according to reactions Li3N + H2 ! Li2NH + LiH; Li2NH + H2 ! LiNH2 + LiH. Most of the amide-elemental metal hydride composites desorb hydrogen endothermically and show full or partial reversibility in hydrogen absorption and desorption. Interestingly, such examples of mixed-metal amides, LiNH2-LiH, CaNH-CaH2, Mg(NH2)2-LiH, and Ca(NH2)2-LiH are showing significantly a good reversibility in desorbing and absorbing hydrogen, which is likely due to the important structural similarities between amides and their corresponding dehydrogenated products. However, all hydrogen can be extracted only at high temperatures. Because of the high theoretical capacity, the BdNdH compounds with the general formula NHxBHx, where x ¼ 4, 3, 2, or 1, which are called solid inorganic analogs of hydrocarbons, attract interest. Thermal dissociation of these compounds is a complex process, and some of these compounds are capable of spontaneous polymerization. First-principles calculations applied to study of the structural and electronic properties of bilayer hexagonal boron nitride (h-BN) by computational techniques for storage of H2 molecules. However, implementation of the reversible hydrogen storage in BdNdH compounds poses other problems. These include a physical adsorption process which driven through a combination of weak van der Waals interaction and minimal charge transfer, and the adsorption energy and the desorption temperature parameters may not match to each other. In the view of solution, there are different ways to enhance the hydrogen storage capacity like by doping or decoration and geometrical alternation of certain spatial characteristics of the BdNdH compounds. Aluminium and titanium-decorated boron nitride nanotubes (BNNTs) system demonstrate remarkable affinity toward H2 molecules with thermodynamically favorable binding energies. In covalent hydrides, a metal (semiconductor) and hydrogen atoms are bound by covalent bonds to form tin(IV) hydride (SnH4), lead(IV) hydride (PbH4), selenium(II) hydride (SeH2), polonium(II) hydride (PoH2), aluminium(III) hydride (AlH3), beryllium(II) hydride (BeH2), etc.; hydrogen compounds with nonmetals methane, ammonia, hydrazine (N2H4), water can also be

Chemistry and Electrochemistry | Hydrogen

647

classified with covalent hydrides for formal reasons.31 Many covalent hydrides are unstable and some are prone to polymerization. The light elements and more covalent hydrides (nitrogen and boron compounds with hydrogen) like BH3, AlH3, NaBH4, NH2-NH2 and NH3 have high energy densities, but are very difficult to handle safely and they decompose to very stable elements which are very demanding to re-fuel with hydrogen on board a vehicle. All these compounds readily react with ionic hydrides forming LiBH4, NaAlH4 and LiNH2.31 The most abundant class is formed by interstitial hydrides in which hydrogen donates an electron to the conduction zone of a metal and finds itself localized in an octahedral or tetrahedral void of the lattice. Interstitial hydrides can also form intermetallic compounds and solid solutions. They exhibit high thermal and electric conductivities. Their composition is as a rule nonstoichiometric. Their stability varies over a wide range depending on material type. Many interstitial hydrides reversibly sorb and desorb hydrogen at room temperature and pressures below the atmospheric pressure. The best known are materials of AB5 type, namely LaNi5−xMx (where M ¼ Cr, Fe, Co, Cu, Mn, Al, Si, Sn, Ge, In, Tl, Pd, Ag) and La1−xRxNi5 (where R]Ce, Mm (Misch metal), Ca, Y, Gd, Zr, Pr, Nd, Sm), and AB2 type (alloys and Laves structures), namely, ZrMn2, ZrCr2, TiMn2, TiCr2, ZrCo2, ZrMo2, ZrNi2, MgNi2, HfCr2, and their substituted derivatives, and also AB-type alloys based on vanadium, titanium, iron like TiFe, etc.32a These materials are used in commercial nickel-metal hydride batteries. Their hydrogen capacity reaches 2 mass% hydrogen. Body centered cubic (BCC) metals and alloys intrinsically based on vanadium, chromium, titanium etc. have large hydrogen capacity because they have more interstitial sites in the lattice than face-centered cubic (FCC) and hexagonal closest packed (HCP) structures. Certain hydrides, e.g., magnesium hydride, and some rare earth elements have bonds of the ionic-metallic type. Magnesium hydride (MgH2) with a hydrogen content of 7.66 mass% and a volume density 1.5 times higher than the density of liquid H2 attracts attention.(7780–7808) However, the slow kinetics of hydrogen sorption-desorption, low stability, and high desorption temperatures prevent its practical use. In order to improve the thermodynamics and kinetics for hydrogen release and uptake in MgH2, some approaches are implemented as nanoconfinement, nanostructuring by ball milling, utilization of catalytic additives or alloying with different metals like Sc, Ti, V, Cr, etc. These compounds can hold promise as the candidates for practical application. In contrast, magnesium in combination with a heat storage material (phase-change material) gives a safe and efficient mode for stationary large-scale hydrogen storage (up to 700 kg), long lifetime (>5000 absorption/desorption cycles), without degradation of the hydrogen uptake capacity (>6.6 mass%).32b A new unusual type of hydride is formed by certain inert gases such as neon, argon, krypton, xenon and very recently reported hydro-helium cation (HeH+).33 The reversible change in the bond type with variation in the hydrogen content in solid hydrides leads to reversible modification of electric, optical, and other properties. The formation of hydrides brings about structural changes, an increase in the unit cell volume (within 10–30%), embrittlement and fracturing of the material, and, finally, the formation of fine-grain powders. This property is used in the production of dispersed materials and, at the same time, brings about the necessity of developing special technologies for the preparation of dimensionally stable electrodes from interstitial hydrides for the electrochemical devices. In the first approximation, electrodes of interstitial hydride powders can be considered as systems with twice distributed parameters, viz., with respect to the particle radius and the electrode thickness.30a Hydrogen solubility, diffusion mobility, as well as physical and structural properties of hydrides depends on crystal size (and/or film thickness, diameter of nanoparticles or nanowires). Size effects are usually observed below 100 nm. There are also some similar structural effects resulting from matrix defectiveness. A promising method of utilizing hydrogen as the environmentally clean energy carrier consists of the combined use of a hydrogen-air fuel cell and an electric motor.29,34 The implementation of such a system and its competitiveness requires the development of hydrogen accumulators with high hydrogen capacity (not less than 6.5 mass % hydrogen per accumulator mass) for a hydrogen density of no less than 62 kg m−3. Moreover, the temperature of hydrogen generation should be in the range of 60–120
C and the kinetics of hydrogen adsorption/desorption should be sufficiently fast. The high purity hydrogen storage through compact metal hydrides accumulators is the best way of feeding low-temperature hydrogen-air fuel cells.47 The system is more attractive from the viewpoint of safety, especially at failures with the destruction of hydrogen tanks. To feed 0.5–10 kW hydrogen-air fuel cells system should liberate hydrogen at a pressure of 2–5 atm with a rate of 10–100 min−1, and for 2–100 W portable fuel cells, at a pressure of 1.1–1.5 atm with a rate of 0.2–5 min−1. The European (EU)-funded project demonstrates the key element of any hydrogen fuel supply chain, acting as the interface between various possible hydrogen supply concepts and end users’ vehicles.35 It is based on the use of hydro-pneumatic accumulators, which provide a more reliable and economical solution. Hydrogen/metal systems (in form of powder or film) can be used in different technologies for energy conversion and the control over heat flows (thermal desorption compressors, heat engines and pneumatic actuators, refrigerating units, heat transformers, systems for long-distance heat transfer, systems for hydrogen extraction from gas mixtures and its purification, systems for controlled high-precision hydrogen leak-in, etc.). A model of a film hydrogen accumulator with a tape foil carrier with coatings of titanium hydride and magnesium hydride is developed and manufactured as an energy carrier.36

6

Chemical and electrochemical processes involving hydrogen

Hydrogen enters into innumerable chemical reactions, sometimes with exceptional activity. Its reaction with fluorine to form hydrogen fluoride is accompanied by explosion even at low temperatures. The stoichiometric hydrogen-oxygen mixture explodes at its contact with a catalyst, flame, or under the action of an electric spark.

648

Chemistry and Electrochemistry | Hydrogen

Hydrogen-air mixtures with volumetric hydrogen content of 4–75% are inflammable. This makes it possible to organize combustion of hydrogen in place of gasoline in internal combustion engines. The process turns out to be sufficiently efficient and is accompanied by evolution of water and small amounts of nitrogen oxides. The combination ‘liquid hydrogen-liquid oxygen’ is a very good solution for space programs. Hydrogen combustion in pure oxygen affords pure water vapor. Such hydrogen steam generators find practical applications. Steam microgenerators are used in medical technology and biotechnology. By choosing the appropriate catalyst, it is possible to design hydrogen catalytic burners and heaters to be used at home and in offices. For cooking purposes, direct conversion from H2 to heat has the potential for a higher efficiency and low cost than by conversion from H2 to electricity for electrical cooking. With similar intension, a highly scalable, powerful, clean and save cooking stove based on catalytic H2 combustion without H2 and air premix technology is developed.37a The reaction of hydrogen with nitrogen occurs on heating under pressure in the presence of a catalyst. Hydrogen displaces certain heavy metals from solutions of their salts, and reduces oxides and other compounds; in the presence of catalysts, hydrogen adds to unsaturated, aromatic, and many other organic compounds. This is why hydrogen is widely used as a chemical, first of all, in the synthesis of ammonia (nearly 50% of manufactured hydrogen), petroleum refining (35% of hydrogen), hydrogenation of fats, production of methanol, and also in powder metallurgy, metal-working, production of glass and synthetic rubies, and microelectronics.37b Of great importance is the reaction of hydrogen with carbon monoxide (CO + H2 mixture is called synthesis gas) to form alcohols, hydrocarbons, and other products. In processes involving hydrogen, hydrogen chemisorption and dissolution in solid matrices play an important role. These phenomena are fundamental in the chemistry and electrochemistry of hydrogen.38 Once molecular hydrogen comes in contact with certain solid surfaces, its spontaneous adsorption occurs, namely, H2 + 2 M $ 2MH(ads).39 This reaction is possible owing to the strong interaction of hydrogen atoms with the surface atoms. Kinetics of the establishment of adsorption equilibrium depends on the nature of the metal, the crystallographic structure of the surface, its defectiveness, and other factors. The same factors determine the adsorption isotherm type; in the liquid phase, the additional factors are the nature of solvent and the solution composition. Hydrogen adsorption leads to surface relaxation and/or reconstruction and changes its properties. On hydrogen-saturated surfaces, the number of hydrogen atoms is usually equal to the number of atoms on the metal surface (or higher, even on ideal surfaces). Hydrogen atoms can penetrate into the second and even third metal layers to form subsurface hydrogen. The heat of hydrogen adsorption decreases with the increase in the surface coverage with hydrogen, often by a linear law for intermediate coverages (Temkin’s surfaces).39 Dissolution of hydrogen in metals and intermetallic compounds involves the following steps: diffusion, physical adsorption, dissociation to atoms, transition into the subsurface state, diffusion of atoms into the substrate bulk, formation of interstitial solid solutions (a-phase), and then a concentrated hydride (b-phase). The equilibrium concentration of hydrogen in the a-phase is described by the Henry-Sieverts law. The ratio of hydrogen atoms to metal atoms, at constant temperature, is H/M ¼ const(T ) PH21/2. According to the Gibbs phase rule, the b-phase is formed at a constant hydrogen pressure Pa$b, i.e., the sorption isotherm has a plateau corresponding to the a$b transition. In the ideal case, this plateau should be horizontal. In fact, this is not strictly fulfilled and, moreover, a hysteresis is observed, i.e., the b-phase formation and destruction occur at somewhat different pressures. This process occurs via the formation of a-phase nuclei in the b-phase volume. Based on the temperature dependence of Pa$b and using the van’t Hoff equation, it is possible to determine the standard enthalpy and entropy of hydride formation. For different metal-hydrogen systems, the latter quantity roughly corresponds to the entropy change with the transition from gaseous H2 to the solid state (120 Jmol−1 K−1). After the completion of the b-phase formation, the concentration of dissolved hydrogen persists to increase (due to the formation of a solid solution of H in the b-phase) and asymptotically approaches a limit determined by the number of sites in the metal matrix accessible for H. As the temperature increases, the equilibrium Pa$b increases and the a$b transition plateau shortens. Above a certain critical temperature Tc, the region of coexistence of two phases disappears (Fig. 2). The a$b transition pertains to phase transitions of the first kind. In aqueous solutions, on hydrogen-sorbing materials, a potential difference is established owing to the equilibria H2(gas)$ 2H(ads) 2H3O++2e−. Thus, the reversible hydrogen electrode (RHE) is realized. The RHE potential is independent of the electrode material. The choice of materials for RHE is determined by the kinetics of establishment of equilibrium and the corrosion stability. The fastest rates of establishment of equilibrium are typical of platinum group metals with well-developed surfaces. The most common choice is platinum covered with a layer of platinum black (Pt/Pt). Its high specific surface ensures electrode stability toward contaminations present in both solution and hydrogen. The RHE potential depends on the hydrogen ion activity a+H and the pressure PH2 of hydrogen gas over the solution according to the Nernst equation EH+/H2 ¼ EoH+/H2 + (RT/F)log(aH+/PH2–1/2). Here, EoH+/H2 is a constant conditionally taken to be equal to zero at all temperatures and a hydrogen pressure equal to 1 atm. The redox potential scale is conventionally related to the standard hydrogen potential. The potential range E > 0 V (from 0.05 V to 0.4 V) with respect to RHE is called the region of hydrogen underpotential deposition (UPD). The range E  0, where the evolution of H2 occurs, is the region of hydrogen overpotential deposition (OPD). Using the Nernst equation and experimental equilibrium charging curves (or voltammograms), it is possible to plot the electrochemical isotherms of hydrogen adsorption and dissolution in the UPD region, and even in some certain interval in the OPD region (Fig. 3). A comparison carried out for certain systems has shown that isotherms of hydrogen dissolution from the gas phase and from solutions coincide. The similarity of hydrogen adsorption isotherms measured in the gas phase and in solutions containing no specifically adsorbed ions was also ascertained, despite the involvement of water molecules and other solution components in the establishment of equilibrium in aqueous solutions.39

Chemistry and Electrochemistry | Hydrogen

649

Fig. 2 Schematic phase diagram of M-H system (a) and the dependence of phase transition pressure (Pa$b) on the inverse temperature (b) illustrating the hydride formation enthalpy DH
and entropy DS
.

Fig. 3 Charging curves for electrochemical desorption (1) and electrochemical absorption (2) of hydrogen by palladium black at 20
C; 0.5 mol L−1 H2SO4 solution. Er, potential in RHE scale; Q, electric charge spent for absorption/desorption. Reproduced from Newman, J.; Thomas-Agea, K. E. Electrochemical Systems, 3rd Edn. Wiley: New York (2004).

The equilibrium behavior of platinum group metal and hydrogen systems in the UPD region made it possible to elaborate the thermodynamic theory of surface phenomena on electrodes. The latter, in contrast to ideally polarized electrodes, are called perfectly polarized electrodes. The behavior of a perfectly polarized electrode is unambiguously determined by the quantity of electricity (charge) passed through the electrode/solution interface.40 Voltammograms in the hydrogen UPD region (in effect, differential adsorption isotherms) represent ‘fingerprints’ of the surface, insofar as the hydrogen adsorption is structure-sensitive (Fig. 4). The hydrogen adsorption value is used in the determination of the true surface area of electrocatalysts. Hydrogen sorption plays an important role in the hydrogenation of organic compounds and the electrochemical evolution and ionization of molecular hydrogen (hydrogen evolution reaction (HER) and hydrogen oxidation reaction (HOR), respectively).41a Hydrogenation is carried out in both gas and liquid phases. In the former case, gaseous hydrogen is used. In the liquid phase, a chemical mechanism similar to the gas phase is possible and in addition the following mechanisms are considered: electrochemical reduction of an organic molecule, which is preceded or followed by protonation (sometimes, protonation can be the limiting stage; the simultaneous transfer of a proton and an electron is also occasionally admitted); and electrocatalytic hydrogenation in which

650

Chemistry and Electrochemistry | Hydrogen

i(PA cm–2)

i(PA cm–2)

80 40

50

0

0.5

1.0

1.5

–40

0

0.2

0.6

1.0 Er(V)

Er(V) –50

–80 –120 (a)

(b)

i(PA cm–2)

i(PA cm–2) 200

150 100

100

50 0

0.2

0.5

0

0.8 Er(V)

–100

0.2 –50

0.5

0.8

Er(V)

–100 –200

–150

(c)

(d)

Fig. 4 The voltammograms of smooth polycrystalline Pt (a), Pt(111) (b), Pt(100) (c), and Pt(110) (d) electrodes in 0.5 mol L−1 H2SO4 solution. Scan rate 50 mV s−1.

chemisorbed hydrogen is generated from the reduction of water molecules or hydroxonium ions on the cathode to enter into the subsequent reaction with the organic substance. The advantage of these methods lies in the fact that they need not use gaseous hydrogen but generate hydrogen by varying the cathode potential. For E > 0 V (RHE), UPD hydrogen takes part in the reaction. Electroreduction and electrohydrogenation can often be carried out under milder conditions as compared with gas-phase hydrogenation. Sometimes, it is possible to hydrogenate substances that fail to react with gas-phase hydrogen. However, the necessity of isolation of reaction products from the medium is essential. This may be avoided, for example, by using systems with membranes made of palladium or its alloys on one side of which hydrogen is electrochemically generated to hydrogenate a given substance on the other side of the membrane. In addition to the elucidation of the electrohydrogenation mechanism, it is important to find the optimum electrode materials and conditions aimed at the enhancement of the efficiency and the selectivity (stereoselectivity) of this process. The HER and HOR lie in the basis of the electrochemical generation of gaseous H2 and the operation of fuel cell anodes. These processes depend on the electrode material and are electrocatalytic and structure-sensitive. The rates of HER and HOR are different for different hydrogen isotopes.41b The HER pertains to the most thoroughly studied electrochemical processes. Presumably, it involves the following stages: (1) Electrochemical sorption of hydrogen (the Butler-Volmer reaction) H+ + M + e− ! MHðadsÞ ðacidic solutionsÞ or H2 O + M + e− ! MHðadsÞ + OH− ðalkaline solutionsÞ (2) Electrochemical desorption (Heyrovsky reaction) MHðadsÞ + H+ + e− ! M + H2 ðacidic solutionsÞ or MHðadsÞ + H2 O + e− ! M + H2 + OH− ðalkaline solutionsÞ (3) Chemical desorption (Tafel reaction) 2MHðadsÞ ! 2 M + H2

Chemistry and Electrochemistry | Hydrogen ?

Pt 1

1

Pt

–Log io (A cm–2)

3 Rh Au 5

Cu

7

9

651

Ti

Zn

In 30

Ir

Ni Co Fe

5 W 7

Sn Bi Pb Cd

3

Re

Ag Ga

Mo

Ti Nb

Ta 9

In 50 70 EM-H (kcal mol L–1)

90

Fig. 5 Correlation of the exchange current densities for hydrogen reaction on various metals, i0, and the adsorption energies of hydrogen on corresponding metals, EM−H. Arrows demonstrate some tendencies of ‘volcano’ deformation when using more correct experimental values of i0 or EM−H. Reproduced from Niessen, R. A. H.; Vermeulen, P.; Notten P. H. L. The Electrochemistry of Pd-Coated MgySc(1−y) Thin Film Electrodes: A Thermodynamic and Kinetic study. Electrochim. Acta 2006, 51(12), 2427–2436.

Hence, the process rate (or the exchange current) should depend on the energy of hydrogen adsorption on a given metal. Under certain assumptions, this dependence is described by a volcano-shaped curve with platinum and the other members of the platinum group located in the maximum (Fig. 5).41b Similar dependence was proposed for the rate of gas-phase hydrogenation on different materials. Particular kinetic equations for HER depend on the adsorption isotherms of intermediates, the rate ratio of different stages, the presence of certain substances in solution and on the surface, etc. The problem of the identity of HER intermediates is unambiguously solved for the hydrogen UPD region. For catalysts in the OPD region, the nature of intermediates is unknown. The slope of a voltammetric curve depends on the porosity of the electrode material. Hydrogen dissolution in the electrode material is reflected in the specific features of HER. On highly active materials, the possibility of oversaturation of the near-electrode layer with hydrogen should be taken into account. The comparison of HER rates on different metals and alloys and the observed synergistic effects require that the true surface areas of electrode be taken into account. The HOR was studied less comprehensively because of the interfering effect of diffusion limitations for the delivery of H2 to the surface. The number of materials that allow one to observe HOR is limited to those on which the RHE potential is established. In the latter case, the HOR rate is as a rule substantially higher than the oxidation rates of other fuels. The directions of the search for catalysts for HOR are determined by the necessity of solving the problem of their tolerance toward carbon monoxide present as an impurity in the reforming hydrogen. In fuel cells with carbonate melts as the electrolyte, where the conductivity is determined by CO32− ions, the overall reaction on the anode can be written as H2 + CO32− ! H2O + CO2 + 2e−. In solid oxide fuel cell with electrolytes conductive with respect to oxygen ions, the following process occurs on the anode: H2 + O2− ! H2O + 2e−.42

7

Hydrogen sensors and detectors

Recently, from the medical research, the H2 gas expelled (>20 ppm) from human breath is being analyzed and quantified using gas chromatography/mass spectrometry and proton-transfer reaction mass spectrometry for the noninvasive diagnosis of gastrointestinal diseases (GIDs).43 The problem of hydrogen detection can be solved by different analytical procedures or by using hydrogen sensors of different kinds. These analytical techniques need the pretreatment of the exhaled volatile organic compounds and the instruments. Thermocatalytic sensors are based on measuring the heat efficiency of hydrogen combustion in the presence of a catalyst. Such sensors operate in the range of 0.1–10 vol% hydrogen in air. Scientists are now focusing on developing highly sensitive and accurate sensors based on solid-state transducers which are working based on chemiresistive or amperometric mechanisms. It has several advantages, such as excellent miniaturization potential, simple operation, low power consumption, fast measurement capacity, portability and low cost. In this finding, metal-oxide semiconductor-based catalysts such as TiO2, ZnO, SnO2 and WO3 are used as sensitive layer for recording the changes in the electrophysical properties of materials at the chemisorption of the analyzed gas on their surface or in the presence of a surface redox process involving hydrogen.44a The threshold of response of such detectors can reach 1 ppm. Thus, main efforts should be

652

Chemistry and Electrochemistry | Hydrogen

focused on developing sensors with uniform long-range ordered crystals (LROCs) or photonic crystals (PCs) in order to achieve consistent response from across the surface.44b,c A unique combination of materials and novel layered structure enables the detection of H2 gas down to 50 ppm with highly promising limit of detection (LoD) capabilities. Electrochemical sensors use two-electrode electrochemical cells with a solid electrolyte. The sensor based on SnO2 nanospheres shows an ultra-sensitive H2 detection ability lower to 1 ppm. Amperometric sensors measure either the current at a fixed potential difference between the electrodes, which is proportional to the concentration of the analyzed component substance, or the quantity of electricity consumed in a hydrogen-involving process in a fixed time interval. Only one electrode is in contact with the gas to be analyzed. Potentiometric sensors measure the working electrode potential with respect to the reference electrode. Several materials are studied for sensing electrodes (SE) in mixed-potential sensors. These include metals such as Au, Pt, and Mo or binary systems with Ag, Au, Ni, and Cu based on Pt material. Based on NiO, WO3, ZnO, and ZnO/CuO materials are recently investigated.44d,e The changes in the optical properties of films of rare earth element alloys with the hydrogen sorption can be used in the development of fiber-optic hydrogen detectors. WO3-Pd2Pt-Pt nanocomposite films are deposited on a single mode fiber as the hydrogen sensing material, which changes its reflectivity under different hydrogen concentration. At a room temperature of 25
C, the hydrogen sensor has a significant response toward 10 ppm hydrogen in nitrogen atmosphere, and may detect tens of ppb hydrogen changes when the hydrogen concentration is between 10 and 60 ppm. Moreover, a fiber optic hydrogen sensor with fast response fabricated from a graphene-Au-Pd sandwich nanofilm and an ultrashort fiber Bragg grating. The sensor has a response time of 4.3 s at a hydrogen concentration of 3.5%.

8

Conclusions

Hydrogen is considered as the most likely and promising energy carrier in the future, because it can be converted into different forms of energy by environmentally benign methods with efficiency higher than that of other fuels. This stimulates the development of concepts of hydrogen-based power systems. (The idea of wide use of hydrogen synthesized in water electrolysis was first formulated by Jules Verne in his science fiction novel ‘The Mysterious Island’ in 1874.) In such a concept, the involvement of chemistry and electrochemistry becomes clear from Fig. 6.9c,45

Solar (electricity) Hydro Wind Geothermal

Solar energy (heat)

Production

Renewable

Biomass

2.01 %

2.54 %

Ele ctro lys is

9.29 %

4.95 %

4%

Natural gas

Nuclear (heat)

22.88 %

Coal 26.88 %

Nuclear (heat)

H2

ICE 28.9 %

or

du

sp

29 %

Offices

Chemical processes

8.1 %

Mu

nicip a

l e cono 21.2 %

Living quarters ICE - Internal Combustion Engine Fig. 6 Hydrogen production & consumption.

In

t

F u e l c e ll s

FC

str

y

n Tra

63 %

Power units on turbines and ICE

my

Power supply Service quarters

Consumption

Chemistry and Electrochemistry | Hydrogen

653

It is also evident that the energy carriers such as hydrogen and electricity are capable of satisfying most of our energy needs. The electrolytic production of hydrogen from water using electric power generated by renewable energy sources and/or nuclear stations (power-to-gas) can be considered as the priority direction. Thus, the enhancement of the efficiency of electrolytic hydrogen production is currently the central problem. At power consumption sites, hydrogen can be either burnt in gas turbines or electrochemically oxidized in fuel cells. The latter approach seems preferable because of its higher efficiency, compactness, and convenience. The service life and the performance of fuel cells should be enhanced and their cost should be reduced. So far, the synthetic functionalization for novel materials and their use through technological development can be promising tool to provide a way of efficient hydrogen production at convenient cost. In the case of reconversion to electricity, hydrogen serves as storage medium for electricity. Hydrogen stored in underground salt caverns will be used in the future for large scale electricity storage. Furthermore, the hydrogen can be used for the chemical and steel industry.

References 1. (a) Brady, J. E.; Holum, J. R. Fundamentals of Chemistry; Wiley: New York, 1988; (b) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements, (2nd Ed.); Butterworth-Heineman Press: Oxford, 1997; (c) Parker, S. P. Encyclopaedia of Chemistry; McGraw-Hill: New York, 1993. 2. Lee, J. W. Hydrogen. In Encyclopaedia Britannica; 2020. https://www.britannica.com/science/hydrogen. 3. (a) Williams, W. D. Döbereiner’s Hydrogen Lighter. Bull. Hist. Chem. 1999, 24; (b) Wisniak, J. Electrochemistry and Fuel Cells: The Contribution of William Robert Grove. Ind. J. Hist. Sci. 2015, 50 (3), 476–490; (c) Williams, K. R. Francis Thomas Bacon. Biogr. Mem. Fellows R. Soc. 1994, 39, 3–18. 4. (a) Hanley, H. J. M.; McCarty, R. D.; Intemann, H. The Viscosity and Thermal Conductivity of Dilute Gaseous Hydrogen from 15 to 5000 K. J. Res. Natl. Bur. Stand A Phys. Chem. 1970, 74A (3), 331–353; (b) Assael, M. J.; Mixafendi, S.; Wakeham, W. A. The Viscosity and Thermal Conductivity of Normal Hydrogen in the Limit of Zero Density. J. Phys. Chem. Ref. Data Monogr. 1986, 15 (4), 1315–1322. 5. Dagdougui, H. Hydrogen Logistics: Safety and Risks Issues. In Hydrogen Infrastructure for Energy Applications: Production, Storage, Distribution and Safety; Dagdougui, H., Sacile, R., Bersani, C., Ouammi, A., Eds.; Academic Press, 2018; pp. 127–148. 6. Hoge, H. J.; Lassiter, J. W. Critical Temperatures, Pressures, and Volumes of Hydrogen, Deuterium, and Hydrogen Deuteride. J. Res. Natl. Bur. Stand. 1951, 47 (2). 7. (a) Energy (n.d.) Hydrogen and Fuel Cell Technologies Office. In: Hydrogen Storage. https://www.energy.gov/eere/fuelcells/hydrogen-storage; (b) Wiebe, D.; Donev, J. Energy Education-Types of Hydrogen Fuel; 2021. https://energyeducation.ca/encyclopedia/Types_of_hydrogen_fuel. 8. Purwanto Deshpande, R. M.; Chaudhari, R. V.; Delmas, H. Solubility of Hydrogen, Carbon Monoxide, and 1-Octene in Various Solvents and Solvent Mixtures. J. Chem. Eng. Data 1996, 41 (6), 1414–1417. 9. (a) Parangi, T. A Review on Electrochemical and Photochemical Processes for Hydrogen Production. Comments Inorg. Chem. 2022, 42; (b) Abdin, Z.; Ali, Z.; Rafiee, A.; Merida, W.; Lipinski, W.; Khalilpour, K. R. Hydrogen as an Energy Vector. Renew. Sustain. Energy Rev. 2020, 120, 109620; (c) Osman, A. I.; Mehta, N.; Elgarahy, A. M.; Hefny, M.; Al-Hinai, A.; Al-Muhtaseb Ala’a, H.; Rooney, D. W. Hydrogen Production, Storage, Utilization and Environmental Impacts: A Review. Environ. Chem. Lett. 2022, 20, 153–188. 10. TNO (n.d.) From Grey and Blue to Green Hydrogen, TNO. https://www.tno.nl/en/focus-areas/energy-transition/roadmaps/towards-CO2-neutral-industry/hydrogen-for-asustainable-energy-supply/. 11. (a) Green Hydrogen Cost Reduction: Scaling up Electrolysers to Meet the 1.5
C Climate Goal; International Renewable Energy Agency (IRENA): Abu Dhabi, 2020; (b) The Future of Hydrogen; International Energy Agency (IEA): Paris, 2019; (c) DOE. Hydrogen and Fuel Cells Program; Renewable Energy Laboratory (NREL), 2019. 12. (a) Hydrogen from Renewable Power: Technology Outlook for the Energy Transition; International Renewable Energy Agency (IRENA): Abu Dhabi, 2018; (b) Sonnichsen, N. Statista-2021. Global Hydrogen Production 2018 & 2030; 2021. https://www.statista.com/statistics/1121207/global-hydrogen-production. 13. Hirscher, M.; Yartys, V. A.; Baricco, M.; von Colbe, J. B.; Blanchard, D.; Bowman, R. C., Jr.; Broom, D. P.; Buckley, C. E.; Chang, F.; Chen, P.; Cho, Y. W.; Crivello, J.-C.; Cuevas, F.; David, W. I. F.; de Jongh, P. E.; Denys, R. V.; Dornheim, M.; Felderhoff, M.; Filinchuk, Y.; Froudakis, G. E.; Grant, D. M.; Gray, E. M.; Hauback Bjø, C.; He, T.; Humphries, T. D.; Jensen, T. R.; Kim, S.; Kojima, Y.; Latroche, M.; Li, H.-W.; Lototskyy, M. V.; Makepeace, J. W.; Møller, K. T.; Naheed, L.; Ngene, P.; Noreus, D.; Nygard, M. M.; Orimo, S.-I.; Paskevicius, M.; Pasquini, L.; Ravnsbæk, D. B.; Sofianos, M. V.; Udovic, T. J.; Vegge, T.; Walker, G. S.; Webb, C. J.; Weidenthaler, C.; Zlotea, C. Materials for Hydrogen-Based Energy Storage - Past, Recent Progress and Future Outlook. J. Alloys Compd. 2020, 827, 153548. 14. Scott, K. Chapter 1: Introduction to Electrolysis, Electrolysers and Hydrogen Production. In Electrochemical Methods for Hydrogen Production; Royal Society of Chemistry: Cambridge, 2019; pp. 1–27. 15. (a) Rivard, E.; Trudeau, M.; Zaghib, K. Hydrogen Storage for Mobility: A review. Mater 2019; (b) Moradi, R.; Groth, K. M. Hydrogen Storage and Delivery: Review of the State of the Art Technologies and Risk and Reliability Analysis. Int. J. Hydrogen Energy 2019, 44 (23), 12254–12269; (c) Barthelemy, H.; Weber, M.; Barbier, F. Hydrogen Storage: Recent Improvements and Industrial Perspectives. Int. J. Hydrogen Energy 2017, 42 (11), 7254–7262. 16. Karlsdottir, S. N. Chapter 7: Corrosion, Scaling and Material Selection in Geothermal Power Production. In Comprehensive Renewable Energy; Sayigh, A., Ed.; Elsevier, 2012; pp. 241–259. 17. (a) Chen, Z.; Li, P.; Anderson, R.; Wang, X.; Zhang, X.; Redfern, L. R.; Moribe, S.; Islamoglu, T.; Gomez-Gualdron, D. A.; Robison, L.; Stoddart, J. F.; Farha, O. K. Balancing Volumetric and Gravimetric uptake in highly Porous Materials for Clean Energy. Science 2020, 368, 297–303; (b) Ren, J.; Musyoka, N. M.; Langmi, H. W.; Mathe, M.; Liao, S. Current Research Trends and Perspectives on Materials-Based Hydrogen Storage Solutions: A Critical Review. Int. J. Hydrogen Energy 2017, 42 (1), 289–311; (c) Andersson, J.; Gronkvist, S. Large-Scale Storage of Hydrogen. Int. J. Hydrogen Energy 2019, 44, 11901–11919. 18. (a) Schlapbach, L.; Zuttel, A. Hydrogen-Storage Materials for Mobile Applications. Nature 2001, 414 (15), 353–358; (b) Lee, S.-Y.; Lee, J.-H.; Kim, Y.-H.; Kim, J.-W.; Lee, K.-J.; Park, S.-J. Recent Progress Using Solid-State Materials for Hydrogen Storage: A Short Review. Processes 2022, 10, 304; (c) Al-Ahmed, A.; Khan, M. M. A. Hydrogen Storage by Physisorption: An Overview. In: Advanced Materials Research: Al-Ahmed, A.; Hossain, M. K.; Afzaal, M.; Bahaidarah, H. M., Trans Tech Publications Ltd.; Switzerland: 1116, 157-172, 2015. 19. (a) Samantaray, S. S.; Putnam, S. T.; Stadie, N. P. Volumetrics of Hydrogen Storage by Physical Adsorption. Inorganics 2021, 9, 45; (b) Zacharia, R.; Rather, S. U. Review of Solid State Hydrogen Storage Methods Adopting Different Kinds of Novel Materials. J. Nanomater. 2015, 2015, 914845. 1-18; (c) Suh, M. P.; Park, H. J.; Prasad, T. K.; Lim, D.-W. Hydrogen Storage in Metal-Organic Frameworks. Chem. Rev. 2012, 112, 782–835; (d) Langmi, H. W.; Ren, J.; North, B.; Mathe, M.; Bessarabov, D. Hydrogen Storage in Metal-Organic Frameworks: A Review. Electrochim. Acta 2014, 128, 368–392; (e) Baumann, A. E.; Burns, D. A.; Liu, B.; Thoi, V. S. Metal-Organic Framework Functionalization and Design Strategies for Advanced Electrochemical Energy Storage Devices. Commu. Chem. 2019, 2 (86), 1–13; (f ) Lin, R.-B.; Xiang, S.; Zhou, W.; Chen, B. Microporous Metal-Organic Framework Materials for Gas Separation. Chem 2020, 6, 337–363; (g) Shet, S. P.; Priya, S. S.; Sudhakar, K.; Tahir, M. A Review on Current Trends in Potential use of Metal-Organic Framework for Hydrogen Storage. Int. J. Hydrogen Energy 2021, 46, 11782–11803. 20. (a) Dong, J.; Wang, X.; Xu, H.; Zhao, Q.; Li, J. Hydrogen Storage in Several Microporous Zeolites. Int. J. Hydrogen Energy 2007, 32 (18), 4998–5004; (b) Das, N.; Das, J. K. Chapter 6: Zeolites: An Emerging Material for Gas Storage and Separation Applications. In Zeolites - New Challenges; Margeta, K., Farkas, A., Eds.; IntechOpen: London, 2020; pp. 914–971.

654

Chemistry and Electrochemistry | Hydrogen

21. Tianyu, L. Energy Focus: Empirical Equations Identify Metal-Organic Frameworks with Unprecedented Hydrogen-Storage Capacity. MRS Bull. 2020, 45 (6), 411. 22. (a) Grunker, R.; Bon, V.; Muller, P.; Stoeck, U.; Krause, S.; Mueller, U.; Senkovska, I.; Kaskel, S. A New Metal-Organic Framework with Ultra-High Surface Area. Chem. Commun. 2014, 50, 3450–3452; (b) Farha, O. K.; Yazaydin, A. O.; Eryazici, I.; Malliakas, C. D.; Hauser, B. G.; De Kanatzidis, M. G. Novo Synthesis of a Metal-Organic Framework Material Featuring Ultrahigh Surface Area and Gas Storage Capacities. Nat. Chem. 2010, 2, 944–948; (c) Kaye, S. S.; Dailly, A.; Yaghi, O. M.; Long, J. R. Impact of Preparation and Handling on the Hydrogen Storage Properties of Zn4O(1,4-benzenedicarboxylate)3 (MOF-5). J. Am. Chem. Soc. 2007, 129, 14176–14177. 23. (a) He, Q.; Zeng, L.; Han, L.; Sartin, M. M.; Peng, J.; Li, J.-F.; Oleinick, A.; Svir, I.; Amatore, C.; Tian, Z.-Q.; Zhan, D. Electrochemical Storage of Atomic Hydrogen on Single Layer Graphene. J. Am. Chem. Soc. 2021, 143 (44), 18419–18425; (b) Rimza, T.; Saha, S.; Dhand, C.; Dwivedi, N.; Patel, S. S.; Singh, S.; Kumar, P. Carbon-Based Sorbents for Hydrogen Storage: Challenges and Sustainability at Operating Conditions for Renewable Energy. ChemSusChem 2022, 15 (11); (c) Mohan, M.; Sharma, V. K.; Kumar, E. A.; Gayathri, V. Hydrogen Storage in Carbon Materials - A Review. Energy Storage 2019, 1 (2), 35; (d) Daulbayev, C.; Lesbayev, B.; Bakbolat, B.; Kaidar, B.; Sultanov, F.; Yeleuov, M.; Ustayeva, G.; Rakhymzhan, N. A Mini-Review on Recent Trends in Prospective Use of Porous 1D Nanomaterials for Hydrogen Storage. J. Chem. Eng. 2022, 39, 52–61; (e) Rostami, S.; Pour, A. N.; Izadyar, M. A Review on Modified Carbon Materials as Promising Agents for Hydrogen Storage. Sci. Prog. 2018, 101 (2), 171–191. 24. (a) Vidyaev, D. G.; Savostikov, D. V.; Boretsky, E. A.; Verkhorubov, D. L. Hydrogen Sorption by Carbon Nanostructured Materials. J. Ind. Pollut. Control 2016, 32 (2), 411–413; (b) Fomkin, A. A.; Pribylov, A. A.; Murdmaa, K. O.; Pulin, A. L.; Shkolin, A. V.; Men’shchikov, I. E.; Zhedulov, S. A. Adsorption of Hydrogen in Microporous Carbon Adsorbents of Different Origin. Prot. Met. Phys. Chem. Surf. 2019, 55 (3), 413–419. 25. (a) Improvement of Hydrogen Storage with Nanomaterials. https://nanografi.com/blog/improvement-of-hydrogen-storage-with-nanomaterials-ed328c/; (b) Karim, W.; Spreafico, C.; Kleibert, A.; Gobrecht, J.; VandeVondele, J.; Ekinci, Y.; van Bokhoven, J. A. Catalyst Support Effects on Hydrogen Spillover. Nature 2017, 541, 68–71. 26. (a) Gupta, A.; Baron, G. V.; Perreault, P.; Lenaerts, S.; Ciocarlan, R.-G.; Cool, P.; Mileo, P. G. M.; Rogge, S.; Speybroeck, V. V.; Watson, G.; Voort, P. V. D.; Houlleberghs, M.; Breynaert, E.; Martens, J.; Denaye, J. F. M. Hydrogen Clathrates: Next Generation Hydrogen Storage Materials. Energy Storage Mater. 2021, 41, 69–107; (b) Ghaani, M. R.; Schicks, J. M.; English, N. J. A Review of Reactor Designs for Hydrogen Storage in Clathrate Hydrates. Appl. Sci. 2021, 11, 469; (c) Chan, K. S. Hybrid Carbon-Based Clathrates for Energy Storage. J. Carbon Res. 2018, 4, 7; (d) Davoodabad, A.; Mahmoudi, A.; Ghasemi, H. The Potential of Hydrogen Hydrate as a Future Hydrogen Storage Medium. iScience 2020, 24, 101907; (e) Yu, C.; Fan, S.; Lang, X.; Wang, Y.; Li, G.; Wang, S. Hydrogen and Chemical Energy Storage in Gas Hydrate at Mild Conditions. Int. J. Hydrogen Energy 2020, 45 (29), 14915–14919. 27. (a) Alefeld, G.; Volkl, J. Hydrogen in Metals I. Basic Properties. In Topics in Applied Physics; Springer-Verlag: Berlin-Heidelberg, 1978; p. 28. Hydrogen in Metals II. Application-Oriented Properties and Applications. In: Topics in Applied Physics: Springer-Verlag: Berlin-Heidelberg: 29, 1978; (b) Wipf, H. Hydrogen in Metals. III. Properties and Applications. In Topics in Applied Physics; Springer-Verlag: Berlin-Heidelberg, 1991; p. 73. 28. (a) Schlapbach, L. Hydrogen in Intermetallic Compounds. I. Electronic, Thermodynamic, Crystallographic Properties, Preparation. In Topics in Applied Physics; Springer-Verlag: Berlin-Heidelberg, 1988; p. 63; (b) Schlapbach, L. Hydrogen in Intermetallic Compounds. II. Surface and Dynamic Properties, Applications. In Topics in Applied Physics; Springer-Verlag: Berlin-Heidelberg, 1992; p. 67. 29. von Colbe, J. B.; Ares, J.-R.; Barale, J.; Baricco, M.; Buckley, C.; Capurso, G.; Gallandat, N.; Grant, D. M.; Guzik, M. N.; Jacob, I.; Jensen, E. H.; Jensen, T.; Jepsen, J.; Klassen, T.; Lototskyy, M. V.; Manickam, K.; Montone, A.; Puszkiel, J.; Sartori, S.; Sheppard, D. A.; Stuart, A.; Walker, G.; Webb, C. J.; Yang, H.; Yartys, V.; Zuttel, A.; Dornheim, M. Application of Hydrides in Hydrogen Storage and Compression: Achievements, Outlook and Perspectives. Int. J. Hydrogen Energy 2019, 44, 7780–7808. 30. (a) Stolten, D.; Emonts, B. Hydrogen Science and Engineering: Materials, Processes, Systems and Technology; Wiley-VCH Verlag GmbH & Co. KGaA, 2016; (b) Zhao, L.; Xu, F.; Zhang, C.; Wang, Z.; Ju, H.; Gao, X.; Sun, L.; Liu, Z. Enhanced Hydrogen Storage of Alanates: Recent Progress and Future Perspectives. Prog. Nat. Sci.: Mater. Int. 2021, 31, 165179; (c) Milanese, C.; Garroni, S.; Gennari, F.; Marini, A.; Klassen, T.; Dornheim, M.; Pistidda, C. Solid State Hydrogen Storage in Alanates and Alanate-Based Compounds: A Review. Metals 2018, 8, 567. 31. (a) Lale, A.; Bernard, S.; Demirci, U. B. Boron Nitride for Hydrogen Storage. ChemPlusChem 2018, 83 (10), 893–903; (b) Mananghaya, M. R. Titanium-Decorated Boron Nitride Nanotube for Hydrogen Storage: A Multiscale Theoretical Investigation. Nanoscale 2019, 11, 16052–16062. 32. (a) Akiba, E. Chapter 14: Solid Hydrogen Storage Materials: Interstitial Hydrides. In Hydrogen Energy, Engineering, Green Energy and Technology; Sasaki, K., Li, H.-W., Hayashi, A., Yamabe, J., Ogura, T., Lythet, S. M., Eds.; Springer: Tokyo, 2016; pp. 191–206; (b) Ley, M. B.; Jepsen, L. H.; Lee, Y.-S.; Cho, Y. W.; von Colbe, J. S. M.; Dornheim, M.; Rokni, M.; Jensen, J. O.; Sloth, M.; Filinchuk, Y.; Jorgensen, J. E.; Besenbacher, F.; Jensen, T. R. Complex Hydrides for Hydrogen Storage - New Perspectives. Mater. Today 2014, 17, 122–128. 33. Grandinetti, F. Cationic Noble-Gas Hydrides: From Ion Sources to Outer Space. Front. Chem. 2020, 8, 462. 34. (a) Tarasov, B. P. Metal-Hydride Accumulators and Generators of Hydrogen for Feeding Fuel Cells. Int. J. Hydrogen Energy 2011, 36, 1196–1199; (b) Srinivasan, S.; Rivera, L.; Escobar, D.; Stefanakos, E. Light Weight Complex Metal Hydrides for Reversible Hydrogen Storage. In Advanced Applications of Hydrogen and Engineering Systems in the Automotive Industry; Cocco, L., Aziz, M., Eds.; IntechOpen: London, 2022. 35. European-funded Project (n.d.) Advanced New Compression and Buffering Solution for Hydrogen Refuelling Stations (HRS). https://cordis.europa.eu/article/id/418239-safe-costeffective-technology-dispenses-hydrogen-to-fuel-cell-vehicles. 36. Yurchenkov, M. I.; Ivanov, A. G.; Karpov, D. A.; Chebukov, E. S. Experimental Research of Film Hydrogen Accumulator Model. J. Phys.: Conference Series. 2021, 1954, 012055. 37. (a) Fumey, B.; Stoller, S.; Fricker, R.; Weber, R.; Dorer, V.; Vogt, V. F. Development of a Novel Cooking Stove Based on Catalytic Hydrogen Combustion. Int. J. Hydrogen Energy 2016, 41 (18), 7494–7499; (b) Brown, A. Uses of Hydrogen in Industry. https://www.thechemicalengineer.com/tags/clean-energy-sig-the-hydrogen-economy. 38. (a) Newman, J.; Thomas-Agea, K. E. Electrochemical Systems, 3rd Ed.; Wiley: New York, 2004; (b) Appleby, A. J.; Kita, H.; Chemla, M.; Bronoel, G. Electrochemistry of Hydrogen. In Encyclopedia of Electrochemistry of the Elements; Bard, A. J., Stratman, M., Wilson, G. S., Eds.; vol. 9; Marcel Dekker: New York, 1982; p. 383A. 39. (a) Frumkin, A. N. Chapter 2: Hydrogen Overvoltage and Adsorption Phenomena: Part I. In Advances in Electrochemistry and Electrochemical Engineering; Delahay, P., Tobias, C. W., Eds.; InterScience Publishers: New York, 1961; p. 265; (b) Frumkin, A. N. Chapter 5: Hydrogen Overvoltage and Adsorption Phenomena: Part II. In Advances in Electrochemistry and Electrochemical Engineering; Delahay, P., Tobias, C. W., Eds.; Interscience Publishers: New York, 1963; pp. 287–391; (c) Krishtalik, L. I. Hydrogen Overvoltage and Adsorption Phenomena: Part III. Effect of the Adsorption energy of Hydrogen on Overvoltage and the Mechanism of the Cathodic Process. In Advances in Electrochemistry and Electrochemical Engineering; Delahay, P., Ed.; Wiley: New York, 1970; pp. 283–339. 40. (a) Frumkin, A. N.; Petrii, O. A. Potentials of zero total and zero free charge of platinum group metals. Electrochim. Acta 1975, 20, 347; (b) Niessen, R. A. H.; Vermeulen, P.; Notten, P. H. L. The Electrochemistry of Pd-Coated MgySc(1−y) Thin Film Electrodes: A Thermodynamic and Kinetic study. Electrochim. Acta 2006, 51 (12), 2427–2436; (c) Bao, J.; Macdonald, D. D. Oxidation of Hydrogen on Oxidized Platinum. Part I: The Tunneling Current. J. Electroanal. Chem. 2007, 600 (1), 205–216. 41. (a) Parsons, R. The rate of electrolytic hydrogen evolution and the heat of adsorption of hydrogen. Trans. Faraday Soc. 1958, 54, 1053; (b) Petrii, O. A.; Tsirlina, G. A. Electrocatalytic Activity Prediction for Hydrogen Electrode Reaction: Intuition, Art, Science. Electrochim. Acta 1994, 39, 1739. 42. (a) Justi, E. W.; Winsel, A. W. Kalte Verbrennung - Fuel Cells; Franz Steiner Verlag GMBH: Wiesbaden, 1962; (b) Breiter, M. W. Electrochemical Processes in Fuel Cells; Springer-Verlag: Berlin, 1969; (c) Markovic, N. M.; Ross, P. N. Surface Science Studies of Model Fuel Cell Electrocatalysts. Surf. Sci. Rep. 2002, 45, 117–229; (d) Trasatti, S.; Petrii, O. A. Real Surface Area Measurements in Electrochemistry. J. Electroanal. Chem. 1992, 327, 353–376. 43. Alenezy, E. K.; Sabri, Y. M.; Kandjani, A. E.; Korcoban, D.; Rashid, S. S. A. A. H.; Ippolito, S. J.; Bhargava, S. K. Low-Temperature Hydrogen Sensor: Enhanced Performance Enabled through Photoactive Pd-Decorated TiO2 Colloidal Crystals. ACS Sens. 2020, 5 (12), 3902–3914. 44. (a) Manjavacas, G.; Nieto, B. Chapter 10: Hydrogen Sensors and Detectors. In Compendium of Hydrogen Energy: Hydrogen Use, Safety and the Hydrogen Economy; Ball, M., Basile, A., Veziroglu, T. N., Eds.; Woodhead Publishing, 2016; (b) Goepel, W.; Hesse, J.; Zemel, J. N. Sensors: A Comprehensive Survey; VCH: Weinheim, 1996; (c) Yang, M.; Dai, J. Fiber Optic Hydrogen Sensors: A Review. Photo. Sens. 2014, 4 (4), 300–324; (d) Zhang, Y.; Peng, H.; Qian, X.; Zhang, Y.; Guowen, A.; Zhao, Y. Recent Advancements in Optical Fiber Hydrogen Sensors. Sens. Actuators B 2017, 244, 393–416; (e) Chachuli, S. A. M.; Hamidon, M. N.; Md Shuhazlly, M.; Ertugrul, M.; Abdullah, N. H. Hydrogen Gas Sensor Based on TiO2 Nanoparticles on Alumina Substrate. Sens. 2018, 18, 2483. 45. (a) Mengdi Jia, M.; Jianlong, W. Review and Comparison of various Hydrogen Production Methods based on Costs and Life Cycle Impact Assessment Indicators. Int. J. Hydrogen Energy 2021, 46 (78), 38612–38635; (b) Hydrogen. U.S. Department of Energy Hydrogen Program Plan-2020; 2020. https://www.hydrogen.energy.gov/pdfs/hydrogenprogram-plan-2020.pdf; (c) TECH (2022) Hydrogen Energy: A Paradigm Shifter, Shifting Towards a Hydrogen Society https://tech.hyundaimotorgroup.com/fuel-cell/hydrogenenergy/ (Accessed on 14th July, 2022).

Chemistry and Electrochemistry | Iron James A Behan and Frédéric Barrière, Université de Rennes, CNRS, Institut des Sciences Chimiques de Rennes, Rennes, France © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of A.K. Shukla, B. Hariprakash, CHEMISTRY, ELECTROCHEMISTRY, AND ELECTROCHEMICAL APPLICATIONS | Iron, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 744–750, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00063-0.

1 Introduction 2 Chemistry of iron 2.1 Magnetic properties of iron 3 Thermodynamic considerations 4 Applied electrochemistry of iron 5 Nickel–iron batteries 5.1 Iron–air batteries 5.2 Silver–iron batteries 5.3 Super-iron batteries 5.4 Lithium–iron sulfide batteries 6 Lithium-ion batteries 7 Iron materials in fuel cells 8 Other applications of iron 9 Conclusion Acknowledgments References Further reading

656 657 658 658 659 659 661 662 662 662 663 663 664 664 665 665 665

Abstract Iron is the second most abundant metal in the earth’s crust and among the most commonly used metals in industry. Iron is of fundamental importance in electrochemistry due to its rich redox chemistry, which has seen applications in bioelectrochemistry as well as materials in fuel cell electrocatalysts, batteries, and capacitors. This chapter briefly outlines the basic physical and chemical properties of iron including its fundamental thermodynamics and its redox behavior in solution. The electrochemistry of iron is briefly discussed with special emphasis on power sources, including nickel–iron, and iron–air batteries, FeS2 cathodes, iron-nitrogen electrocatalysts for fuel cells and microbial fuel cells based on iron redox proteins.

Glossary Bioelectrochemistry Electrochemical processes originating from some form of biological activity, such as catalysis by a redox-active enzyme or the coupling of respiratory processes of bacteria to electrodes. Ferromagnetism Ferromagnetism is defined as the phenomenon by which a material, such as iron, in an external magnetic field becomes magnetized and subsequently tends to remain magnetized to some extent afterwards due to hysteresis. Ferrimagnetism is a related phenomenon whereby populations of alternating magnetic moments in a material are unequal in magnitude, resulting in a net magnetization. Ligand A ligand is either an atom, ion, or molecule that binds to a central metal, generally involving donation of one or more of its electrons. Metalloprotein A protein which contains a metal atom or metal-containing complex as a cofactor. Microbial Fuel Cell A fuel cell system which utilizes the metabolic activity of micro-organisms to produce electricity. An example is the oxidation of acetate to CO2 by bacteria coupled to oxygen reduction at an abiotic cathode. Paramagnetic A material that is attracted by a magnetic field and has a relative magnetic permeability greater than 1. Pourbaix diagram Potential/pH diagram that maps out possible stable (equilibrium) phases for an aqueous electrochemical system. Pyrophoric A substance that ignites spontaneously with its autoignition temperature below room temperature. Solvothermal synthesis A high temperature and normally high pressure chemical synthesis protocol typically used for the preparation of nanomaterials or crystalline inorganic structures. Hydrothermal synthesis is a special case of the solvothermal method employing water as the solvent. Superparamagnetism A form of magnetism observed in nanoscale ferro or ferrimagnetic materials whereby the magnetization state randomly flips its direction under the influence of temperature. Superparamagnetic materials have the property of possessing zero net magnetic moment in the absence of external magnetic fields.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00214-0

655

656

Chemistry and Electrochemistry | Iron

Key points

• • • • • •

Iron is an abundant metal with wide-ranging applications due to its rich redox chemistry Iron and Iron-oxide materials are frequently applied as anodes in a variety of battery applications NidFe, Fe-air and LidFeS2 battery devices are industrially relevant power sources Iron supported on carbon forms the basis of an important class of non-precious metal electrocatalysts Iron-based materials also have broad uses in the related energy domains of photoelectrochemistry, value-added product synthesis and hydrogen production. Iron is among the most important metals in energy transformations in bioelectrochemical systems such as microbial fuel cells due to its presence in a wide range of conductive metalloproteins.

Nomenclature E E0 Ferric Ferrous

Electrode potential Standard electrode potential Fe(III) Fe(II)

Abbreviations HER MFC ORR RHE SHE

1

Hydrogen evolution reaction Microbial fuel cell Oxygen reduction reaction Reversible hydrogen electrode Standard hydrogen electrode

Introduction

Iron is the second most abundant metal in the earth’s crust after aluminum and the fourth most abundant overall. The overwhelming majority of iron on the surface of the earth exists as ores, with only small amounts of metallic iron found as nickel alloys of primarily meteoric origin. Along with a small amount of iron silicates, the major iron ores are oxides including magnetite (Fe3O4), hematite (a-Fe2O3) and limonite (2Fe2O3
3H2O), the iron carbonate ore siderite (FeCO3), and sulfur-rich iron pyrite (FeS2). Magnetite is the richest iron-containing ore with an iron content of ca. 72% by weight. It is typically found as lustrous black or brownish-black deposits and is the most magnetic of the naturally-occurring iron oxides species. Hematite normally varies in color from steel gray to black, but takes its name from the red or reddish-brown color of certain deposits which have often been used in the past as pigments. Unlike magnetite, which is ferrimagnetic, hematite is antiferromagnetic and shows only a weak response to external magnetic fields. Maghemite (g-Fe2O3), a related hematite polymorph, is ferrimagnetic and has a similar spinel structure to magnetite. Limonite is a hydrated iron oxide-hydroxide of varying composition and is typically brown or yellow in color. Siderite, also known as spathic iron ore, varies in color from pale yellow to shades of brown and is less useful as an ore for smelting due to the lower iron content and the presence of carbonate. Pyrite, also known as ‘fool’s gold’ due to it yellow luster, is not commonly smelted because of its sulfur content. The application of iron materials in electrochemistry varies from direct use as electrodes to its application as catalysts. Iron can be directly applied in metallic form in some cases, though this is limited typically due to facile oxidation and passivation under conditions relevant to battery operation. Iron oxides (both Fe2O3 and Fe3O4) are attractive materials particularly in nanostructures such as nanoparticles but usually require formulation as composites (e.g. with porous carbon materials) due to conductivity issues. Supported Fe on carbon scaffolds doped with heteroatoms such as nitrogen have seen applications as fuel cell electrocatalysts, most notably for the oxygen reduction reaction (ORR) as applied in hydrogen/oxygen fuel cells. Other Fe-based catalysts for varied processes including hydrogen evolution (HER) and CO2 reduction have been explored as other methods of contributing to the energy transition. Semiconducting iron oxides can be applied in photoinduced oxidation processes in photoelectrochemical cells. Finally, in the domain of biology, the rich and facile redox chemistry of Fe has led to its ubiquity in coordination complexes such as heme groups associated with metalloproteins like cytochromes. Such proteins serve roles in fundamental energy transformation processes in nature and are largely responsible for electron transport even over large distances in bioelectrochemical systems.

Chemistry and Electrochemistry | Iron Table 1

657

Some properties of iron.

S. No.

Property

Value

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Atomic number Electronic configuration Atomic weight (g at−1) Atomic volume (cm3/g iron) at 293 K Naturally occurring isotopes Total number of isotopes reported Electronic configuration Electronegativity (Pauling’s scale) Metal radius (12-coordinate; A˚ ) Effective ionic radius (A˚ ) Fe(II) Fe(III) Melting point (K) Boiling point (K) Heat capacity at (cal/g atom/ C) at 293 K Density (g cm−3) at 293 K Electrical resistivity (O cm) at 293 K Molar volume (cm3) Coefficient of linear thermal expansion, (K−1) 293–373 K Magnetic permeability

26 1s2 2s2 2p6 3s2 3p6 3d6 4s2 55.847 7.094 54 56 57 58 26Fe , 26Fe , 26Fe , 26Fe 10 [Ar]4s23d6 1.83 1.26 0.61 (low spin), 0.78 (high spin) 0.55 (low spin), 0.645 (high spin) 1811 3134 5.98 7.874 9.71  10−8 7.09 12.3  10−6 88,400

Sources: In: Silver, J., Ed., Chemistry of Iron. Springer Netherlands: Dordrecht, 1993. https://doi.org/10.1007/97894-011-2140-8 and Nicholls, D. Iron. In The Chemistry of Iron, Cobalt and Nickel. Elsevier, 1973; pp. 979–1051. https://doi.org/10.1016/B978-0-08-018874-4.50005-4.

2

Chemistry of iron

Pure iron is a soft, silvery white metal that rusts rapidly in moist air and is pyrophoric when divided into a fine powder. Chemically pure iron is obtained by the reduction of pure iron oxide using hydrogen, via electrolysis of aqueous solutions of iron salts or the thermal decomposition of iron compounds such as Fe(CO)5. Known oxidation states for iron compounds range from −II (d10) to +VIII (d0), but by far the most common oxidation states are II (d6) (ferrous) and III (d5) (ferric). Some important properties of pure iron are summarized in Table 1.1,2 Successful extraction of iron from ores by reducing the iron oxide with carbon marked a significant development in modern civilization. Reducing the iron ore in a blast furnace produces iron. The basic process involves burning coke to produce temperatures as high as 2000  C near the hearth of the furnace, facilitating the reduction of iron oxides to iron. The detailed chemistry of the blast furnace is complex and is probably still not fully understood. However, it can be assumed that the oxygen in the blast furnace combines with carbon at elevated temperatures (2000  C) in the hearth to produce carbon monoxide that rises through the furnace. Subsequent oxidation of CO to CO2 accompanies the production of pure iron from the oxide. The main reactions occurring in the blast furnace are: Fe2O3 + 3C ! 2Fe + 3CO

[I]

Fe2O3 + 3CO ! 2Fe + 3CO2

[II]

Iron produced by this method contains carbon, silicon, manganese, phosphorus, and some sulfur as impurities. The purity and physical nature of the iron active material has a direct effect on the material properties. Accordingly, efforts have been expended to work out suitable and easily controllable manufacturing processes for iron. Iron active material is prepared from pure iron by a process in which iron slugs are dissolved, recrystallized, roasted, reduced, and then partially re-oxidized. These steps further purify the material, producing a fine powder with a large surface area and good conductivity. This material has been found to be a mixture of a-Fe and Fe3O4. The conversion of iron into steel is among its most important commercial applications, as the pure metal itself is of limited commercial value. Carbon steels are strengthened with the addition of carbon to between 0.1 wt% and 1.5 wt%. Microalloyed steels employ different metals in order to impart different properties depending on the desired application.3 For example, vanadium acts as a strengthener, chromium and copper improve atmospheric corrosion resistance, nickel increases fracture toughness, and molybdenum and tungsten increase heat resistance. Steel microalloys are also strengthened by the precipitation of carbides and nitrides of niobium, titanium and vanadium. Stainless steels are widely used in cutlery and medical equipment due to their superior corrosion resistance; common stainless steel contains 18 wt% chromium and 8 wt% nickel.

658

Chemistry and Electrochemistry | Iron

Iron combines with fluorine and chlorine (on mild heating) to give iron(III) halides. It also reacts readily with many other non-metals including the rest of the halogens, boron, carbon, silicon, phosphorus, and sulfur. Hypochlorite oxidation of iron(III) nitrate in strong alkaline solutions has been found to yield alkali metal salts of iron(VI). As discussed above for steels, the formation of carbide, nitride and silicide phases are central to the metallurgy of iron. Outside of such manufactured materials, iron is overwhelmingly encountered as its oxides. Rusting, or open-air oxidation of iron in the presence of moisture, involves the formation of hydrated iron (III) oxides (Fe2O3
nH2O) and oxyhydroxides (Fe(OH)3 and FeOOH). Rusting is an electrochemical phenomenon and requires the presence of water, oxygen, and an electrolyte. The hydrous oxides formed typically flake off rather than passivate the surface, exposing fresh metal surface to further corrosion. The rate of corrosion is typically accelerated at low pH and in the presence of electrolytes (notably chlorides). Synthetic iron oxides are normally prepared by simple co-precipitation methods of ferric and ferrous salts or via thermal decomposition of ferrous complexes such as ferrous oxalate, ferrous carbonate, or ferrous nitrate. Values. Solubility product constants Ksp are on the order of 10−38 for ferric [Fe(OH)3] but rises to around 10−15 for the ferrous species [Fe(OH)2] at 25  C and around physiological pH.4 Hence, under typical conditions in soil solutions ferrous iron is the predominant iron species with an activity which is highly sensitive to both pH and to the soil redox potential. Dilute mineral acids attack iron, forming ferrous solutions if non-oxidizing acids are used under anaerobic conditions. In the presence of oxygen or oxidizing acids such as nitric acid (HNO3), oxidation of Fe(II) to Fe(III) can occur, allowing for the additional formation of ferric compounds. In the absence of oxygen (or at relatively low oxygen partial pressures) iron is generally stable at higher pH values, but will be attacked by concentrated sodium hydroxide.

2.1

Magnetic properties of iron

In addition to its strength and chemical versatility, iron has distinctive magnetic properties. As given in Table 1, the electronic configuration of elemental iron is [Ar]3d64s2. Depending on the oxidation state and the nature of surrounding ligands, unpaired electrons in iron compounds normally vary in number from 0 to 5. It is these unpaired electrons that give rise to the magnetic properties of metallic iron and its related compounds. Ferromagnetism describes the presence of a distinctive magnetic permeability and coercivity of a material in response to external magnetic fields. It results from a combination of the material electronic structure (i.e. the presence of unpaired electrons) and interactions between the magnetic moments of adjacent atoms in so-called magnetic domains. This relatively uncommon phenomenon derives its name from iron, which is undoubtedly the most famous ferromagnetic material. Iron (specifically a-iron, a body centered cubic (bcc) phase at room temperature is ferromagnetic up to its Curie temperature of 1043 K, where it becomes paramagnetic. The earliest exploited magnetic materials by humans were lodestones, minerals comprised mostly of magnetite. Magnetite is among the most magnetic naturally-occurring minerals on earth. It is an iron oxide with the chemical formula Fe3O4, and consists of a mixed Fe(II)/Fe(III) oxide with an inverse Spinel structure: Fe2+Fe3+ 2 O4. It is a ferrimagnetic complex, possessing a net magnetic moment up to its Curie temperature of 853 K. Synthetic magnetite can be produced by various methods including co-precipitation and thermal decomposition as outlined above; such methods can be used to prepare superparamagnetic magnetite nanoparticles with magnetic properties useful in nanomedical or diagnostic applications.5 The low-temperature oxidation of magnetite can yield maghemite (g-Fe2O3) which has a similar Spinel structure to magnetite and also has ferrimagnetic properties. The magnetic properties of iron are useful for characterization of unknown iron-rich minerals or compounds. The number of unpaired electrons in an iron compound or complex gives information on the chemical state of the iron species as well as the nature of bonding around iron centers. The oxidation state and chemical environment of iron in minerals (including meteorite samples) as well as in iron-catalysts and iron-containing redox proteins is commonly identified using Mössbauer spectroscopy, due to the high abundance of the 57Fe isotope necessary for analysis using the technique.

3

Thermodynamic considerations

Iron is thermodynamically unstable in water and, hence, has a natural tendency to corrode. It is necessary to consider the stability of water with respect to hydrogen and oxygen evolution: H2 ! 2H+ + 2e−

[III]

4OH− ! O2 + 2H2O + 4e−

[IV]

Various possible equilibria for iron in aqueous media are possible depending on the pH and redox potential. For the pure metal: Fe ! Fe2+ + 2e−

[V]

Fe ! Fe3+ + 3e−

[VI]

Chemistry and Electrochemistry | Iron

659

Fe + H2O ! FeO + 2H+ + 2e−

[VII]

Fe + 2H2O ! Fe(OH)2 + 2H+ + 2e−

[VIII]

Fe + 2H2O ! HFeO−2 + 3H+ + 2e−

[IX]

While other relevant processes involving hydroxides or Fe2+/Fe3+ include: Fe(OH)2 + H2O ! Fe(OH)3 + H+ + e−

[X]

Fe2+ + 2H2O ! Fe(OH)2 + 2H+

[XI]

Fe(OH)2 ! HFeO−2 + H+

[XII]

2+

Fe

+ 3H2O ! Fe(OH)3 + 3H + e +

[XIII]

HFeO−2 + H2O ! Fe(OH)3 + e−

[XIV]

Fe(OH)3 ! FeO−2 + H+ + H2O

[XV]

HFeO−2 ! FeO−2 + H+ + e−

[XVI]

These reactions can be employed to construct the Pourbaix (potential/pH) diagram for the iron/water system shown in Fig. 1.6 The region lying between 0.4 and −1.4 V versus SHE and corresponding to pH values >14 is of interest for battery applications. The region of thermodynamic stability for iron at standard temperature and pressure lies under line a (below −0.6 V vs standard hydrogen electrode (SHE)), with no part in common with water. Hydrogen evolution occurring as the conjugate reaction during the corrosion of iron can occur only below line (a) There are two nearly triangular regions for the corrosion of iron: one corresponding to Fe2+ and the other to HFeO2−. The iron electrode operation, however, only includes the corrosion domain of the latter. Deep discharge can shift the potential to a passive region where an oxide film can form. Thermodynamic characteristics of iron and its compounds have been studied extensively in the literature. The stability region for Fe(OH)2 is well within the stability region of magnetite; therefore, Fe(OH)2 is thermodynamically unstable with regard to Fe3O4 and transforms to the latter through the following reaction: 3Fe(OH)2 ! Fe3O4 + 2H2O + H2

[XVII]

Phase composition, morphology, and crystallinity greatly influence the operational behavior of iron electrodes. Changes in the structure and volume of the active mass can influence the porosity, electrical conductivity, and mechanical strength of the electrodes. Similar considerations exist when applying Fe and related materials as electrocatalysts for energy-relevant processes, particularly concerning the particle size and modulation of the surface by ligands or by a support material. Crystal chemistry data for various oxides and hydroxides of iron are presented in Table 2.7

4

Applied electrochemistry of iron

Iron is among the most applied metals in both industry and day-to-day activities. The tendency to form coordination compounds readily makes iron an important material in biological and medical applications. In the field of applied electrochemistry of iron, corrosion, electrochemical machining, electrodeposition, and power sources are of utmost economic importance. The electrochemistry of iron with special emphasis on commercially important power sources is briefly discussed below. The basic electrochemistry of iron can be found elsewhere in this encyclopedia (see Secondary batteries: Fe-air, NidFe). Iron electrodes constitute the negative plates of nickel–iron (NidFe), iron–air (Fe–air), and silver–iron (AgdFe) batteries. Iron sulfide has been applied as the cathode material in lithium–iron sulfide batteries. Iron compounds in their higher oxidation states have also been shown to be appropriate as cathode materials in super-iron batteries with aqueous KOH electrolyte. Iron phosphate has also gained attention as a cathode material for lithium-ion batteries. Other applications of iron materials in photoelectrochemistry, catalysis and bioelectrochemistry are also briefly discussed in order to give a broad overview of the significance of iron in energy applications.

5

Nickel–iron batteries

The first NidFe battery was commercialized by Edison in 1901. The NidFe battery is based on the use of nickel oxyhydroxide (NiOOH) at the cathode and iron or iron oxide materials at the anode.

660

Chemistry and Electrochemistry | Iron

Fig. 1 Potential/pH diagram for Fe/H2O system; inset shows regions of passivity, corrosion, and immunity. SHE, standard hydrogen electrode. Source: Pourbaix, M. Atlas of Electrochemical Equilibria in Aqueous Solutions, Pergamon Press Ltd: London, 1966 [Online]. Available: https://cir.nii.ac.jp/crid/ 1572543024422924800. Reproduced with permission from Pergamon. Table 2

Crystal chemistry data for various oxides and hydroxides of iron of electrochemical interest.

S. No.

Species

Descriptions

1 2 3 4 5 6 7

a-Fe Fe3O4 (magnetite) a-FeOOH (goethite) b-FeOOH g-FeOOH (lepidocrocite) d-FeOOH Fe(OH)2

Body-centered cubic structure Inverse spinel structure: (Fe83+)t(Fe82+ Fe83+)oO32 (where o and t refer to octahedral and tetrahedral sites, respectively) Hexagonal close-packing with OH− hydrogen-bonded to O2− ions; Fe3+ in o-sites Body-centered cubic structure akin to a-MnO2; Fe3+ in o-sites; stable only in the presence of certain interstitial impurities Cubic close-packed structure with Fe3+ in o-sites; hydrogen bonding between OH− ions Hexagonal close-packed structure with random distribution of Fe3+ on o-sites as in Fe(OH)2; 20% of Fe3+ in t-sites Hexagonal close-packed structure with Fe2+ in o-sites as in Mg(OH)2 lattice

Source: Vijayamohanan, K.; Balasubramanian, T. S.; Shukla, A. K. Rechargeable Alkaline iron Electrodes. J. Power Sources, 1991, 34(3), 269–285. doi: 10.1016/0378-7753(91) 80093-D.

Chemistry and Electrochemistry | Iron

661

Operational conditions are typically alkaline and may be summarized according to reaction XVIII: 2NiOOH + Fe + 2H2O ! 2Ni(OH)2 + Fe(OH)2 (Ecell ¼ 1.37 V)

[XVIII]

Under deep discharge, a nickel–iron cell with a negative-limited configuration will undergo a further discharge reaction at a potential that is lower than the first step represented by reaction [XVIII], i.e. [XIX], NiOOH + Fe(OH)2 ! Ni(OH)2 + FeOOH (Ecell ¼ 1.05 V)

[XIX]

The cell reactions are highly reversible in the alkaline electrolyte, particularly if the discharge is limited to the first step, making these batteries robust with a long cycle life. In addition, the use of earth abundant metals and aqueous electrolytes is attractive from the point of view of cost, toxicity and safety. The key limitation of NidFe batteries is their relatively low energy and power densities, much of which is attributed to phenomena such as self-discharging and passivation of metallic iron electrodes, along with generally poor electrical conductivity in the case of iron oxides. Recent advances in NidFe batteries have improved both conductivity and capacity through composite materials, such as iron oxide nanoparticles supported on nanocarbons such as graphene or carbon nanotubes.8

5.1

Iron–air batteries

Iron electrodes also constitute the negative plates of the Fe–air battery. The Fe–air battery is an attractive concept due to its simplicity, low cost and relatively high theoretical specific energy (764 Wh kg−1) which is nearly three times that of NidFe and four times that of lead–acid batteries. Fe-air batteries are also thought to be less prone towards dendrite formation than other devices (such as zinc-air batteries) across numerous charge-discharge cycles. Overall discharge of the battery involves the oxidation of metallic Fe to Fe(II) and Fe(III) surface films by a mechanism which is still not fully understood. The electrolyte employed is typically alkaline, in order to limit dendrite formation.9 One overall cell reaction can be summarized as: 2Fe +2 H2O + O2 ! 2Fe(OH)2 (Ecell ¼ 1.28 V)

[XX]

However during discharge, Fe(OH)2 may be converted to magnetite: Fe + 2OH− > Fe(OH)2 + 2e− E0’ ¼ −0.88V vs SHE at pH 14 −

−

3Fe(OH)2 + 2OH > Fe3O4 + 4H2O + 2e E ¼ −0.76V vs SHE at pH 14 0’

[XXI] [XXII]

When coupled with the 4-electron reduction of O2 (E0’ ¼ 0.401 V vs SHE at pH 14) the overall net reaction may therefore also be: 3Fe + 2O2 ! Fe3O4

[XXIII]

Therefore, considering Eqs. [XX] and [XXII], the theoretical Ecell actually varies in a range from 1.16 to 1.28 V. In practice, the operating voltage is only about 0.7 to 0.8 V, mostly due to polarizational losses at the air electrode. Indeed the difficulty in catalyzing oxygen reduction on discharge and oxygen evolution on charge is a severe limitation for Fe-air batteries both in terms of performance and cost, as state-of-the art catalysts with low overpotentials (ca 200–300 mV) for these processes utilize precious metals such as Pt, Ru and Ir. The poor performance of more base metals is typically compensated using two air-breathing electrodes for each iron electrode in a single cell. However, the 4-electron reduction of O2 also suffers from the possibility of an alternative 2 + 2 electron pathway: O2 + H2O + 2e− > HO−2 + OH− E0 ¼ −0.076 V vs SHE

[XXIV]

HO−2 + H2O + 2e− > 3OH− E0 ¼ 0.878 V vs SHE

[XXV]

The direct 4-electron pathway is preferable in terms of overall cell potential and for limiting the concentrations of peroxyl-anion intermediates, which may damage the cell over time. On the anode side, the Fe electrode suffers from hydrogen evolution as a side reaction during charging and a poor discharge rate due to passivation by insulating Fe(II) hydroxide. Recently, approaches involving nanostructured iron electrodes, such as supported Fe3O4 nanoparticles have been employed to increase Fe electrode utilization and specific discharge capacity. The chemistry, operating principle, and performance characteristics of Fe–air batteries can be found elsewhere in this encyclopedia (see Secondary batteries: Fe-air).

662 5.2

Chemistry and Electrochemistry | Iron Silver–iron batteries

Iron has also been used as anode material in AgdFe batteries. Silver–iron batteries have been limited in use due to their high cost. The theoretical energy density for AgdFe battery is essentially equal to that of silver–zinc (AgdZn) battery. Silver–iron batteries have longer cycle lives than AgdZn batteries and provide high reliability, and better durability where high specific energy content is required. The charge–discharge reactions for the AgdFe battery are as follows: Fe + 2AgO + H2O > Fe(OH)2 + Ag2O (first plateau)

[XXVI]

Fe + Ag2O + H2O > Fe(OH)2 + 2Ag (second plateau)

[XXVII]

2Fe + 2AgO + 2H2O ! 2Fe(OH)2 + 2Ag (Ecell ¼ 1.34 V)

[XXVIII]

The net cell reaction is.

The detailed chemistry and performance characteristics of silver–iron batteries can be found elsewhere in this encyclopedia.

5.3

Super-iron batteries

Li2FeO4, Na2FeO4, K2FeO4, and BaFeO4 electrodes, where iron is in Fe(VI) state, have also been tested as cathodes in aqueous KOH electrolyte with zinc or metal hydride anodes to form primary or secondary batteries, respectively. These cathode materials exhibit high specific capacity values because of a three-electron change in their reduced state10 as shown below: − − 0 FeO2− 4 + 3H2O + 3e > FeOOH + 5OH (E ¼ 0.66 V vs SHE)

[XXIX]

Theoretical specific capacities for Li2FeO4, Na2FeO4, K2FeO4, and BaFeO4 are 601, 485, 406, and 313 mAh g−1, respectively. The characteristics of aforementioned Fe(VI) compounds have not been studied extensively because these materials are highly unstable. Although Li2FeO4 and Na2FeO4 are soluble in aqueous KOH, BaFeO4 and K2FeO4 have shown evidence of low solubility.11 Electrochemically, FeO42− species have a reduction potential of about 0.66 V vs SHE. Against the zinc anode, the open-circuit potential is found to be 1.75 and 1.85 V for the K2FeO4 and BaFeO4 cells, respectively. The proposed discharge reaction mechanism is as follows: MFeVIO4 + 32Zn ! 12FeIII 2 O3 + 12ZnO + MZnO2

[XXX]

where M ¼ K2 or Ba. The discharge characteristics and specific energy of experimental primary alkaline button cells, with zinc anodes and Fe(VI) compound cathodes, were measured and compared with MnO2 cathodes. These data illustrate higher energy output for the cells fabricated with the Fe(VI) cathodes. Similar results were obtained in cells with the conventional cylindrical construction. Iron(VI) compounds have also been shown to be rechargeable. A button cell using a metal hydride anode and a capacity-limited K2FeO4 cathode has been discharged for several cycles to 75% depth of discharge (DoD) and for >400 cycles to 30% DoD. The open-circuit voltage of the cell was 1.3 V and the midpoint voltage was 1.1 V, similar to the voltage characteristics of a nickel–metal hydride (Ni–MH) cell. Iron(VI) compounds are promising cathode materials for both primary and rechargeable alkaline batteries. The reported results demonstrate their higher specific energy when compared to other cathode materials presently being used in alkaline batteries. The question of long-term stability, shelf life, and other critical performance characteristics, large-scale manufacture, and cost of materials for these systems needs to be resolved (see Super-Iron Batteries; SECONDARY BATTERIES: Super-Iron Batteries).

5.4

Lithium–iron sulfide batteries

Iron sulfide, in both the monosulfide (FeS) and disulfide (FeS2) forms, has been considered as the cathode material for lithium batteries.12 The primary Li–FeSx battery uses a lithium anode, iron sulfide cathode, and lithium iodide in an organic solvent blend as the electrolyte. The cell reactions are. At the anode: 4Li ! 4Li+ + 4e− −

[XXXI]

At the cathode: FeS2 + 4e ! Fe + 2S

2−

Net cell reaction: 4Li + FeS2 ! Fe + 2Li2S

[XXXII] [XXXIII]

The overall cell reaction with the monosulfide electrode is. 2Li + FeS ! Fe + Li2S

[XXXIV]

Chemistry and Electrochemistry | Iron

663

Only the disulfide battery has been commercialized because of its performance advantage due to its higher sulfur content and higher voltage. These batteries have a nominal voltage of 1.5 V and can therefore be used as replacements for aqueous batteries having a similar voltage. Button-type LidFeS2 batteries were developed as a replacement for AgdZn batteries but are no longer marketed. These batteries are now manufactured in a cylindrical configuration and have better high-drain and low-temperature performance than the Zn–alkaline MnO2 batteries. Secondary Li–FeSx cells developed for electric vehicle applications incorporate cold-pressed FeS or dense FeS2 positive electrode pellets, two-component Li–Al/Li5Al5Fe2 negative electrode pellets, and LiCl-rich LiCl–LiBr–KBr/MgO electrolyte/separator pellets. Electrolyte of the same composition is incorporated in both the positive and negative pellets. The overall electrochemical reactions for the LidFeS and LidFeS2 cells are 2Li–Al + FeS ! Li2S + Fe + 2Al (Ecell ¼ 1.33 V)

[XXXV]

2Li–Al + FeS2 ! Li2FeS2 + 2Al (Ecell ¼ 1.73 V)

[XXXVI]

Recently 2-dimensional flakes of crystalline FeS2 were produced by liquid phase exfoliation of naturally-occurring pyrite minerals. The resulting nanoflakes have a lateral dimension on the order of 120 nm and showed high capacities >1200 mAh g−1 when applied as lithium ion battery anodes in co-fabrication with carbon nanotubes.13 The production of 2D pyrite nanosheets was associated with a higher capacity and cycling stability compared to bulk FeS2.

6

Lithium-ion batteries

With a high theoretical specific capacity of 170 mAh g−1 and excellent thermal stability, LiFePO4 is a promising material for application as the cathode in lithium-ion batteries, in particular, large-format cells for electric vehicles. Among the several candidates of choice for lithium-ion battery cathode materials, LiFePO4 is reported to be inexpensive, nontoxic, and environmentally benign with a flat potential of 3.5 V versus lithium. During the extraction of lithium, the FePO4 framework is found to retain its structure, which is an added advantage of the material. The key disadvantage of the resulting batteries is undoubtedly the limited electrical conductivity and the slow diffusion of Li in the solid-state, which may be mitigated somewhat by compositing these materials with different types of carbon nanomaterials.14 The extraction of lithium from LiFePO4 while charging the cathode may be written as. LiFePO4 − xLi+ − xe− ! xFePO4 + (1 − x)LiFePO4

[XXXVII]

The reaction for the introduction of lithium into FePO4 on discharge is written as. FePO4 + xLi+ + xe− ! xLiFePO4 + (1 − x)FePO4

[XXXVIII]

As LiFeSiO4 exhibits very low volume changes during the charge–discharge processes, it has also been found to be attractive cathode material for lithium-ion batteries. Structure and performance characteristics of LiFePO4 as cathodes in rechargeable lithium-ion batteries can be found elsewhere in this encyclopedia (see SECONDARY BATTERIES – LITHIUM RECHARGEABLE SYSTEMS – LITHIUM-ION: Overview).

7

Iron materials in fuel cells

Iron materials can also be employed as electrocatalysts for the oxygen reduction reaction in proton-exchange membrane fuel cells. Carbon-supported PtdFe alloys formed by various techniques can confer methanol tolerance, facilitating the use of these materials in direct methanol fuel cells. Fe materials are also attractive for Pt-free, non-precious metal electrocatalysts, normally in the form of Fe-supported on nanocarbon scaffolds. Synthetic protocols for the preparations of such materials typically involve hydro- or solvothermal synthesis or pyrolysis of precursors in the presence of metal salts such as Fe(II) acetate or Fe(III) chloride.15 Pyrolyzed iron macrocycles, such as iron(II) phthalocyanine, iron(II) tetra-methoxy phenyl porphyrin, and iron(III) octa-ethyl porphyrin chloride, also exhibit specific oxygen reduction with complete methanol tolerance. The resulting materials are typically nanostructured carbons incorporating either Fe nanoparticles or coordinated Fe centers, where the Fe atom is supported at incorporated heteroatoms such as nitrogen. Such FedNdC materials have complex structures and hence varied oxygen reduction activity depending on the nature of the coordination site.15 Analysis of these materials is complicated by the variety of different nitrogen possible sites coordinated to atomically-dispersed Fe centers, but it has recently been reported that long-term cathode stability in accelerated stress tests is more attainable using pyridinic N-coordinated FeN4 sites compared to pyrrolic N-coordinated FeN4.16 Fe-phosphides have been reported as electrocatalysts for hydrogen evolution. A variety of nanostructured FeP and FeP on carbon materials have been reported with low overpotentials through relatively simple synthetic routes such as pyrolysis or chemical vapor

664

Chemistry and Electrochemistry | Iron

deposition.17 Undoped Fe nanoparticles have also been employed in hydrogen evolution reactions with Pt-like overpotentials under acidic conditions, where Fe dissolution is avoided by encapsulation in a layer of carbon.18 Recently, polyoxometalates, such as K7[(P2W17O61)FeIII(H2O)], have been demonstrated as promising hydrogen oxidation materials. Iron-based perovskite materials are stable in highly reducing atmosphere and possess catalytic activity for oxidation of hydrogen and carbon monoxide in the temperature range between 500 and 700  C, and have been used in solid oxide fuel cells. In the domain of bioelectrochemical systems such as microbial fuel cells, metalloproteins employing mainly heme-type iron cofactors are commonly employed by electroactive bacteria to accomplish extracellular electron transfer to and from electrodes.19 This allows the metabolic processes of microorganisms, such as the oxidation of organic matter, to produce electricity. These iron-rich proteins may be found in the outer membrane of these bacteria and hence rely on direct electrode contact or redox mediators in order to facilitate electron transfer, but the recent discovery of conductive, biological nanowires composed of self-assembled multi-heme iron metalloproteins has demonstrated that electroactive microbes are even capable of long-range electrical connectivity across many microns.20 These capabilities most likely evolved under anaerobic soil environments where they serve the role of coupling respiratory activities of bacteria to solid extracellular metal oxides as terminal electron acceptors instead of oxygen. As iron oxides are among the most commonly occurring minerals in natural environments, this illustrates another aspect of iron redox chemistry and its application in the energy domain. Details on the operating principles of fuel cells can be found elsewhere in this encyclopedia.

8

Other applications of iron

Electrocatalysts coordinating Fe in nitrogenous carbon nanomaterials have been employed for the selective reduction of CO2 to CO or commodity hydrocarbons. Such catalysts have hydrogen evolution as the main competing process under aqueous conditions, but CO efficiency at moderate overpotentials is high and the modification of the local chemical microenvironment of the catalyst through ligand modification or tuning the chemical properties of the support material can disfavor proton reduction and hence increase the Faradaic efficiency of CO2 reduction.21 Photoelectrochemical cells employing semiconductor electrodes have stimulated the investigations of iron oxides as potential candidates for such devices. a-Fe2O3 is an n-type semiconductor on which oxidations can be photoinduced with light of wavelength < 550 nm. This has been applied along with catalyst layers in photoelectrochemical anodes for water oxidation.22 New approaches to increase capacity in lithium-ion batteries include the use of conversion electrodes, which involve chemical bond breaking and formation during the lithiation/delithiation process. Iron difluoride, FeF2 and trifluoride FeF3 are among the most promising materials for conversion cathodes, as FeF3 has a theoretical capacity of 712 mAh g−1.23 In recent years, hydrogen has come to be seen as undisputable fuel for the future. Ironically, natural H2 generation is rather rare in nature, but it can be achieved by the interaction of water and ferrous rocks under anoxic conditions.24 H2 generation is accomplished with concomitant oxidation of Fe2+ to Fe3+ in banded iron formations rich in Fe2+ silicates, carbonates and magnetite.25 This opens the possibility of natural H2 sourcing using naturally-occurring iron rich deposits, or of exploiting the reactivity of the water/magnetite system under anoxic conditions as a form of in situ H2 production under near ambient conditions. In industrial settings hydrogen produced through redox cycling between Fe(II) and Fe(III) oxidation states is referred to as chemical-looping hydrogen, while the process itself is known as the steam-iron process.26 Initially reactors used in this process are fed a stream of reducing gases (chiefly CO and H2) which may arise from thermal decomposition of organic compounds with a high H/C molar ratio, such as ethanol. This reduces Fe(III) oxides such as Fe2O3 to Fe3O4, which can be subsequently reduced to FeO and metallic Fe. Afterwards during the oxidation step steam is fed into the reactor, producing H2 according to Eq. [XXXIX]: 3Fe + 4H2O ! Fe3O4 + 4H2

[XXXIX]

While some H2 is consumed in the reducing step, the overall process results in a net production of H2. Restoring the metal oxide in the reactor to its most oxidized Fe2O3 state requires a final step of air or dioxygen recirculation. The capability of Fe to undergo facile redox cycling between its oxidation states makes the steam-iron process among the most attractive methods for the production of chemical-looping hydrogen.

9

Conclusion

Iron is among the most important elements in the energy domain owing to its abundance, rich and varied redox chemistry and the versatility of techniques for its manufacturing and fabrication into electrode materials. Batteries employing iron or its oxides as both the anode and the cathode have been commercialized for decades and innovations in nanostructured iron-based materials have driven optimization in battery capacitance and cycle stability. The challenges of conductivity and competing reactions such as H2 evolution during charging need to be addressed. Iron in fuel cell technology has mainly served the role of a modulator of the activity of other materials, either as alloys or as a dopant in the scaffold of non-precious metal cathode materials. These materials are versatile, facile to prepare and show activity and durability sufficient to merit testing in automotive applications as an alternative to the state-of-the-art Pt-based cathode.

Chemistry and Electrochemistry | Iron

665

Many of the drawbacks or limitations of iron-based materials in batteries, catalysts and related applications are being addressed through the application of nanostructures and nanocomposites of iron, iron oxides and iron compounds to improve conductivity and long-term stability.

Acknowledgments J.A.B. acknowledges the support of the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 899546.

See also: Batteries – Battery Types – Iron Batteries: Iron-Air; Lithium Batteries – Lithium Secondary Batteries – Li-ion battery: Overview

References 1. Silver, J., Ed. Chemistry of Iron; Springer Netherlands: Dordrecht, 1993. https://doi.org/10.1007/978-94-011-2140-8. 2. Nicholls, D. Iron. In The Chemistry of Iron, Cobalt and Nickel; Elsevier, 1973; pp. 979–1051. https://doi.org/10.1016/B978-0-08-018874-4.50005-4. 3. Villalobos, J.; Del-Pozo, A.; Campillo, B.; Mayen, J.; Serna, S. Microalloyed Steels through History until 2018: Review of Chemical Composition, Processing and Hydrogen Service. Metals 2018, 8 (5), 351. https://doi.org/10.3390/met8050351. 4. Spiro, T. G.; Saltman, P. Polynuclear complexes of iron and their biological implications. In Structure and Bonding; Jørgensen, C. K., Neilands, J. B., Nyholm, R. S., Reinen, D., Williams, R. J. P., Eds.; vol. 6; Springer Berlin Heidelberg: Berlin, Heidelberg, 1969; pp. 116–156. https://doi.org/10.1007/BFb0118856. 5. Sheridan, E.; Vercellino, S.; Cursi, L.; Adumeau, L.; Behan, J. A.; Dawson, K. A. Understanding Intracellular Nanoparticle Trafficking Fates through Spatiotemporally Resolved Magnetic Nanoparticle Recovery. Nanoscale Adv. 2021, 3 (9), 2397–2410. https://doi.org/10.1039/d0na01035a. 6. Pourbaix, M. Atlas of Electrochemical Equilibria in Aqueous Solutions; Pergamon Press Ltd, 1966. [Online]. Available https://cir.nii.ac.jp/crid/1572543024422924800. 7. Vijayamohanan, K.; Balasubramanian, T. S.; Shukla, A. K. Rechargeable Alkaline iron Electrodes. J. Power Sources 1991, 34 (3), 269–285. https://doi.org/10.1016/0378-7753 (91)80093-D. l8. Yang, J.; et al. High-Capacity Iron-Based Anodes for Aqueous Secondary Nickel−Iron Batteries: Recent Progress and Prospects. ChemElectroChem 2021, 8 (2), 274–290. https://doi.org/10.1002/celc.202001251. 9. McKerracher, R. D.; Ponce de Leon, C.; Wills, R. G. A.; Shah, A. A.; Walsh, F. C. A Review of the Iron-Air Secondary Battery for Energy Storage. ChemPlusChem 2015, 80 (2), 323–335. https://doi.org/10.1002/cplu.201402238. 10. Licht, S.; et al. Analysis of Ferrate(VI) Compounds and Super-Iron Fe(VI) Battery Cathodes: FTIR, ICP, Titrimetric, XRD, UV/VIS, and Electrochemical Characterization. J. Power Sources 2001, 101 (2), 167–176. https://doi.org/10.1016/S0378-7753(01)00786-8. 11. Licht, S.; Wang, B.; Ghosh, S. Energetic Iron(VI) Chemistry: The Super-Iron Battery. Science 1999, 285 (5430), 1039–1042. https://doi.org/10.1126/science.285.5430.1039. 12. Linden, D.; Reddy, T. B. Handbook of Batteries, 3 edn.; McGraw-Hill: New York, NY, 2002. 13. Kaur, H.; et al. Production of Quasi-2D Platelets of Nonlayered Iron Pyrite (FeS2) by Liquid-Phase Exfoliation for High Performance Battery Electrodes. ACS Nano 2020, 14 (10), 13418–13432. https://doi.org/10.1021/acsnano.0c05292. 14. Eftekhari, A. LiFePO4/C Nanocomposites for Lithium-Ion Batteries. J. Power Sources 2017, 343, 395–411. https://doi.org/10.1016/j.jpowsour.2017.01.080. 15. Domínguez, C.; Behan, J. A.; Colavita, P. E. Electrocatalysis at nanocarbons: model systems and applications in energy conversion. Nanocarbon Electrochem. 2020, 201–249. https://doi.org/10.1002/9781119468288.ch7. 16. Liu, S.; et al. Atomically Dispersed Iron Sites with a Nitrogen–Carbon Coating as Highly Active and Durable Oxygen Reduction Catalysts for Fuel Cells. Nat. Energy 2022, 7 (7). https://doi.org/10.1038/s41560-022-01062-1. Art. no. 7. 17. Xu, S.; et al. Iron-Based Phosphides as Electrocatalysts for the Hydrogen Evolution Reaction: Recent Advances and Future Prospects. J. Mater. Chem. A 2020, 8 (38), 19729–19745. https://doi.org/10.1039/D0TA05628F. 18. Tavakkoli, M.; et al. Single-Shell Carbon-Encapsulated Iron Nanoparticles: Synthesis and High Electrocatalytic Activity for Hydrogen Evolution Reaction. Angew. Chem. Int. Ed. 2015, 54 (15), 4535–4538. https://doi.org/10.1002/anie.201411450. 19. Lovley, D. R.; Walker, D. J. Geobacter Protein Nanowires. Front. Microbiol. 2019, 10, 2078. https://doi.org/10.3389/fmicb.2019.02078. 20. Behan, J. A.; Louro, R. O.; Barrière, F. In Respiration in Electroactive Bacteria: Bioinorganic Aspects. In Encyclopedia of Inorganic and Bioinorganic Chemistry; Scott, R. A., Ed.; 2024. https://doi.org/10.1002/9781119951438.eibc2792. 21. Fan, Q.; Bao, G.; Chen, X.; Meng, Y.; Zhang, S.; Ma, X. Iron Nanoparticles Tuned to Catalyze CO2 Electroreduction in Acidic Solutions through Chemical Microenvironment Engineering. ACS Catal. 2022, 12 (13), 7517–7523. https://doi.org/10.1021/acscatal.2c01890. 22. Katsuki, T.; Zahran, Z. N.; Tsubonouchi, Y.; Chandra, D.; Hoshino, N.; Yagi, M. P–N Junction Formation between CoPi and a-Fe2O3 Layers Enhanced Photo-Charge Separation and Catalytic Efficiencies for Efficient Visible-Light-Driven Water Oxidation. Sustain. Energy Fuels 2023, 7 (12), 2910–2922. https://doi.org/10.1039/D3SE00346A. 23. Kotal, M.; Jakhar, S.; Roy, S.; Sharma, H. K. Cathode Materials for Rechargeable lithium Batteries: Recent Progress and Future Prospects. J. Energy Storage 2022, 47, 103534. https://doi.org/10.1016/j.est.2021.103534. 24. Otsuka, K.; Kaburagi, T.; Yamada, C.; Takenaka, S. Chemical Storage of Hydrogen by Modified Iron Oxides. J. Power Sources 2003, 122 (2), 111–121. https://doi.org/10.1016/ S0378-7753(03)00398-7. 25. Geymond, U.; et al. Reassessing the Role of Magnetite during Natural Hydrogen Generation. Front. Earth Sci. 2023, 11, 1169356. https://doi.org/10.3389/feart.2023.1169356. 26. De Filippis, P.; D’Alvia, L.; Damizia, M.; Caprariis, B.; Del Prete, Z. Pure Hydrogen Production by Steam-Iron Process: The Synergic Effect of MnO2 and Fe2O3. Int. J. Energy Res. 2021, 45 (3), 4479–4494. https://doi.org/10.1002/er.6117.

Further reading 1. Ait-Itto, F.-Z.; Behan, J. A.; Martinez, M.; Barrière, F. Development of Bioanodes Rich in Exoelectrogenic bacteria Using Iron-Rich Paleomarine Sediment Inoculum. Bioelectrochemistry 2024, 156, 108618. 2. Benoit, C.; Bourbon, C.; Berthet, P.; Franger, S. Chemistry and Electrochemistry of Nanostructured iron Oxyhydroxides as Lithium Intercalation Compounds for Energy Storage. J. Phys. Chem. Solids 2006, 67, 1265–1269.

666 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

Chemistry and Electrochemistry | Iron

Brousse, T.; Belanger, D. A Hybrid Fe3O4–MnO2 Capacitor in Mild Aqueous Electrolyte. Electrochem. Solid-State Lett. 2003, 6, A244–A248. Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry, 5th edn; John Wiley and Sons: New York, 1988. Evans, I. O. Rocks, Minerals and Gemstones; The Hamlyn Publishing Group Ltd: London, 1972; pp. 68–72. Evans, U. R. An Introduction to Metallic Corrosion; Arnold: London, 1979; p. 153. Falk, S. U.; Salkind, A. J. Alkaline Storage Batteries; John Wiley and Sons: New York, 1969. Itoh, T.; Katoh, K.; Hashimoto, T. Platinum Skeleton Alloy-Supported Electrocatalyst, Electrode Using the Electrocatalyst, and Process for Producing the Electrocatalyst; 1999. US Patent 5,876,867. Keusler, K. E. Iron. In Encyclopedia of Electrochemistry of the Elements; Bard, A. J., Ed.; vol. IX; Marcel Dekker: New York, 1982; pp. 229–381. Part A. Kuo, M.-C.; Stanis, R. J.; Ferell, J. R., III; Turner, J. A.; Herring, A. M. Electrocatalyst Materials for Fuel Cells Based on the Polyoxometalates—K7 or H7[(P2W17O61)FeIII (H2O)] and Na12 or H12[(P2W15O56)2FeIII4(H2O)2]. Electrochim. Acta 2007, 52, 2051–2061. Latimer, W. M. Oxidation Potentials; Prentice-Hall: New Jersey, 1961. Padhi, A. K.; Nanjundaswamy, K. S.; Goodenough, J. B. Phospho-Olivines as Positive-Electrode Materials for Rechargeable Lithium Batteries. J. Electrochem. Soc. 1997, 144, 1188–1194. Schwochau, K. Extraction of Metals from Sea Water. Top. Curr. Chem. 1984, 124, 91–133. Shukla, A. K.; Raman, R. K.; Scott, K. Advances in Mixed-Reactant Fuel Cells. Fuel Cells 2005, 5, 436–447. Toda, T.; Igarashi, H.; Uchida, H.; Watanabe, M. Enhancement of the Electroreduction of Oxygen on Pt Alloys with Fe, Ni and Co. J. Electrochem. Soc. 1999, 146, 3750–3756. Ujimine, K.; Tsutsumi, A. Electrochemical Characteristics of iron Carbide as an Active Material in Alkaline Batteries. J. Power Sources 2006, 160, 1431–1435. Wang, S.-Y.; Ho, K.-C.; Kuo, S.-L.; Wu, N.-L. Investigation on Capacitance Mechanisms of Fe3O4 Electrochemical Capacitors. J. Electrochem. Soc. 2006, 153, A75–A80. Licht, S. Secondary Batteries | Super-Iron Batteries. In Encyclopedia of Electrochemical Power Sources; Garche, J., Ed.; Elsevier, 2009, pp 262–284, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00932-1.

Chemistry and Electrochemistry | Lead Krzysztof Maksymiuk and Jadwiga Stroka, University of Warsaw, Warsaw, Poland © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of K. Maksymiuk, J. Stroka, Z. Galus, CHEMISTRY, ELECTROCHEMISTRY, AND ELECTROCHEMICAL APPLICATIONS | Lead, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 762-771, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5. 00898-4.

1 Introduction 2 Chemistry of lead 3 Thermodynamic considerations 4 Electrode behavior of lead 4.1 Electrode behavior of lead and its compounds dissolved in different media 4.2 Voltametric behavior of lead and its non-soluble compounds in different solutions 4.3 Voltametric behavior of lead in sulfuric acid solutions 4.4 Voltametric behavior of lead oxides 4.5 Electrochemical properties of lead sulfide 4.6 Lead-acid batteries 5 Other lead batteries (sulfuric acid free) 6 Other electrochemical applications of Pb and Pb oxides 7 Conclusions Acknowledgment References Further reading

667 667 669 670 670 672 673 673 674 674 677 677 678 678 679 679

Abstract A short description of chemical properties and thermodynamics of lead and its electrochemically important compounds is presented. Electrochemical behavior of lead and its technologically significant compounds, mainly oxides, in different media is reviewed, with special emphasis on voltametric characteristics. More attention was paid to lead-acid batteries: construction, electrochemical properties, performance limitations and new trends in lead batteries development.

Key points

• • • • •

1

Chemical properties of lead Thermodynamic properties of lead Electrochemical properties of lead and its selected compounds Basic properties of lead-acid batteries Properties of other lead batteries

Introduction

Lead is a rare element present in the earth’s crust in a 1.4 10−3 w/w %. It is a toxic heavy metal which exists only in a cubic close-packed metallic form. It shows the preference for the divalent state with electronic configuration Xe (4f145d 106s2) and low stability of the Pb-Pb covalent bond. Lead (IV) in solution is unstable (electronic configuration Xe (4f145d10). The most important Pb ore is galena (PbS). Other lead ores are anglesite (PbSO4), cerussite (PbCO3), mimetesite (Pb5(AsO4)3Cl) and pyromorphite (Pb5(PO4)3Cl).

2

Chemistry of lead

Chemical properties of lead are described in many textbooks, e.g..1–3 Lead is usually obtained from galena. After concentration by flotation, PbS is converted in air to PbO:

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00076-1

667

668

Chemistry and Electrochemistry | Lead PbS + 1:5O2 ! PbO + SO2

(1)

which is mixed with coke and limestone and reduced in the blast furnace, according to the reactions: PbO + C ! PbðliqÞ + CO

(2)

PbO + CO ! PbðliqÞ + CO2

(3)

Pb may be produced also by reduction of the roasted ore with fresh galena: PbS + 2PbO ! 3PbðliqÞ + SO2ðgÞ

(4)

Such obtained Pb contains several metal impurities like: Cu, Ag, Au, Zn, Sn, As and Sb, which have to be removed. The final purification is carried out by electrolytic purification using massive cast Pb anodes. This procedure yields a cathodic deposit with 99.99% Pb. It can be further purified electrochemically to obtain Pb with impurities level below 1 ppm. Pure lead is silver-grey very soft and low-melting metal with a high density which can be easily hammered. Pb is used for production of batteries, cable sheathing, sheets, pipes, foils, tubes, solders, pigments and chemicals. The physicochemical and atomic properties of metallic lead are presented in Table 1. Finely divided Pb powder is pyrophoric. The reactivity of lead at ambient temperature is usually inhibited by formation of protective layers of insoluble products such as oxide, carbonate or sulfate on the surface. Fluorine reacts with Pb at room temperature to give PbF2, while reaction with Cl2 gives PbCl2 on heating. Molten Pb reacts with chalcogens to give PbS, PbSe and PbTe. PbS is an important semiconductor material, with low band gap energy depending on the size of PbS particles. The band gap energy is equal to 0.41 eV (for bulk PbS), resulting in almost metallic conductivity at room temperature. PbS finds applications in photoelectric switching devices in the IR range (1–2.5 mm), it also exhibits photocurrents. Natural galena is usually n-type semiconductor. PbS is unstable in air, it can be oxidized by oxygen with elemental sulfur formation and possible production of Pb1-xS forms. Methods of PbS synthesis are described in literature. Aqueous solutions of HCl react slowly with Pb to give sparingly soluble PbCl2, while nitric acid reacts quite rapidly and forms soluble Pb(NO3)2. Acetic acid also dissolves Pb in the presence of air and forms Pb(CH3COO)2. Lead(II) halides, PbX2, are stable crystalline compounds which can be prepared by mixing water-soluble Pb(II) salt with HX or halide ions to precipitate ill-soluble PbCl2 or PbBr2 which are photo-sensitive and deposit metallic Pb on irradiation with ultraviolet or visible light. PbI2 is a photoconductor and decomposes on exposure to green light (lmax ¼ 494.9 nm). PbX2 as many other Pb(II) compounds is more stable thermally and chemically than PbX4. The only stable tetrahalide is yellow PbF4 (m.p. 600  C). PbCl4 is a yellow oil (m.p. -15  C) which is stable below 0  C but decomposes to PbCl2 and Cl2 above 50  C. PbBr4 is even less stable. Stability of Pb(IV) can be increased by coordination with Cl− ions to formation of M2PbCl6 (M ¼ Na, K, Rb, Cs, NH4). Depending on the method of preparation PbO exists in two forms: red tetragonal PbO form (litharge, a-PbO) with SnO structure stable at room temperature and yellow orthorhombic PbO form (b-PbO, massicot) stable above 488  C. PbO is amphoteric and dissolves in acidic and alkaline solutions. PbO and PbO2 are photoactive semiconductors with band gaps of 1.92 and 2.7 eV, respectively. These compounds are used as passivation layers in electrochemistry, oriented lead oxide on silica and lead oxide-matrix composites. PbO2 has two crystallographic forms, black orthorhombic a-PbO2 and maroon-colored b-PbO2 which has the tetragonal rutile structure. Table 1

Selected atomic and physical properties of lead.

Property

Pb

Atomic number Atomic weight Electronic structure Number of naturally occurring isotopes Ionization energy/kJ mol−1 I II III IV rII (ionic radius, 6-coordinate)/pm rIV (covalent radius)/pm rIV (ionic radius, 6-coordinate)/pm Pauling electronegativity Melting point/oC Boiling point/oC Density (20  C)/g cm−3 Electrical resistivity (20  C)/ohm cm

82 207.2 [Xe] 4f145d106s26p2 4 715.4 1450.0 3080.7 4082.3 119 146 78 1.9 327 1757 11.342 2  10−5

Chemistry and Electrochemistry | Lead

669

PbO2 is commercially produced by the oxidation of Pb3O4 in alkaline slurry with Cl2 and can be also obtained electrochemically. Nanostructured PbO2 obtained by hydrolysis of Pb(CH3COO)4 has been found to exhibit high electrochemical activity. PbO2 decomposes when is heated in air: 293o

3510

374o

605o

PbO2 ! PbO1219 ! Pb12 O17 ! Pb3 O4 ! PbO

(5)

PbO2 is a strong oxidizing agent used in production of chemicals, dyes, matches and pyrotechnics. It can slowly react with acids according to the relations: warm

PbO2 + H2 SO4 ! PbSO4 + H2 O + 0:5O2

(6)

PbO2 + 2HNO3 ! PbðNO3 Þ2 + H2 O + 0:5O2

(7)

Warm HCl reacts similarly in the redox process: warm

PbO2 + 4HCl ! PbCl2 + 2H2 O + Cl2

(8)

In the reaction of PbO2 with glacial acetic acid Pb(CH3COO)4 is obtained, which is a strong oxidizing agent used in organic chemistry. PbO2 reacts with concentrated NaOH according to the reaction:   (9) PbO2 + 2NaOH + 2H2 O ! Na2 PbðOHÞ6 Mixed oxides of Pb(IV) with other metal oxides can be obtained by heating PbO2 or PbO in air with an oxide, hydroxide or oxoacid salts. The obtained products are dependent on the stoichiometry of used substrates: MIIPbIVO3 or MeII2 PbIVO4, where M ¼ Ca, Sr, Ba. Mixed Pb(II) oxides are also important ferromagnetics: Pb•nFe2O3 (n ¼ 0.5, 1, 2.5, 5, 6), yellow PbTiO3 (ferroelectric below 490  C), PbZrO3 and PbHfO3 (antiferroelectric). Red Pb3O4 which is obtained by heating PbO and PbO2 in air behaves chemically as a mixture of PbO and PbO2. Pb3O4 is used as a surface coating to prevent corrosion of iron and steel, for production of leaded glasses, as an activator and a pigment. Pb(II) has a well-defined cationic chemistry, in aqueous perchlorate solution Pb(II) ions are partially hydrolyzed: Pb2+ + H2 O⇆PbOH+ + H+ log K ¼ −7:9

(10)

2+

At higher concentrations of Pb , polymerized species with 3, 4 or 6 Pb atoms are formed. Lead and its compound are very toxic. Lead is a cumulative toxicant that affects multiple body systems and introduces irreversible damages in a living organism. It is particularly harmful to young children, because children can absorb 4–5 times more lead than adults from a given source. Lead in the body is distributed mainly to the brain, but also to liver, kidney and bones, it can cause higher blood pressure. It is stored in the teeth and bones, where it can accumulate over time. Symptoms of lead poisoning include headaches, stomach cramps, constipation, muscle/joint pain, trouble sleeping, fatigue and irritability. For lead concentration within the range 40  80 mg dL−1, serious health damage may occur, even if there are no symptoms. There is no safe blood level of lead. A child whose levels become too high — generally 45 mg dL−1 or higher — should be treated. Centers for Disease Control and Prevention (CDC) uses a blood lead reference value of 3.5 mg dL−1 to identify children with blood lead levels that are higher than most children’s levels. Due to health risk the admissible lead level in water is 10 mg L−1.

3

Thermodynamic considerations

Metallic lead is thermodynamically stable in the presence of neutral or alkaline aqueous solutions free from oxidizing agents. In acidic solutions lead should decompose water with evolution of hydrogen, but this process is very slow on account of high hydrogen overpotential and passivation effects. As a rule the electrode reactions equilibria in the Pb(II)/Pb system are attained quickly, therefore the potential values for the lead systems, measured under well selected conditions should bring accurate results. The potentials obtained for different lead systems in aqueous solutions reported in literature are listed in Tables 2 and 3. The reactions given in Tables 2 and 3 and several others were used to construct the Pourbaix pH-potential diagrams (see Fig. 1).4 Such diagrams for lead in solutions containing sulfate ions of unit activity are presented in Fig. 2.5 In the latter case at 2 < pH < 8 the passivation layer of PbSO4 was formed on Pb surface, at higher pH in this layer the compounds PbO•PbSO4, 3PbO•PbSO4 and PbO were also present. The final product of oxidation of all these phases is PbO2 (a-PbO2 and b-PbO2 forms). The line showing Pb2+/PbO2 potential was constructed for b-PbO2 form. One should add that since the potential of the PbO2/PbSO4 couple, which is present in lead storage batteries, is about 0.47 V more positive than the potential of the oxygen reaction, PbO2 should be unstable in water. However, decomposition of water is very slow because of high oxygen overpotential on PbO2 and protection of the PbO2 by adsorbed sulfate ions or films. In fact, stability of PbO2 in H2SO4 is very high. In the absence of impurities the self-discharge reaction represented by Eq. (6) is very slow, though the reaction of b-PbO2 is faster than that of a-PbO2.

670

Chemistry and Electrochemistry | Lead Table 2

Standard potentials of the Pb(II)/Pb system given versus the standard hydrogen electrode.

Reaction

E0/V

Pb ¼ Pb2+ + 2e Pb + 2H2O ¼ Pb(OH)2 + 2H+ +2e Pb + H2O ¼ PbOred + 2H+ +2e Pb + 2OH− ¼ PbOred + H2O + 2e Pb + 2Cl− ¼ PbCl2 + 2e Pb + 2Br− ¼ PbBr2 + 2e Pb + 2I− ¼ PbI2 + 2e Pb + 2F− ¼ PbF2 + 2e Pb + HPO42− ¼ PbHPO4 + 2e Pb + S2− ¼ PbS +2e Pb + CO32− ¼ PbCO3 + 2e

−0.125 0.277 0.248 −0.580 −0.268 −0.280 −0.365 −0.344 −0.465 −0.954 −0.509

Table 3

Standard potentials of the Pb(IV)/Pb(II) system given versus the standard hydrogen electrode.

Reaction

E0/V

PbSO4 + 2H2O ¼ a-PbO2 + SO42− + 4H+ +2e for b-PbO2 Pb2+ + 2H2O ¼ a-PbO2 + 4H+ +2e for b-PbO2 Pb2+ ¼ Pb4+ + 2e PbOred + 2OH− ¼ PbO2 + H2O + 2e HPbO2− + OH− ¼ b-PbO2 + H2O + 2e HPbO2− ¼ PbO1.33 + 0.33H2O + 0.33OH− + 0.67e HPbO2− + H2O ¼ PbO32− + 3H+ +2e 3Pb2+ + 4H2O ¼ Pb3O4 + 8H+ +2e

1.698 1.690 1.468 1.460 1.69 0.249 0.208 0.024 1.547 2.094

Lead(II) forms various complexes in aqueous solutions, with many different inorganic and especially organic ligands. Electroanalytical methods played quite important role in the study of such complexation equilibria, supplying reliable results. The electrode potential of the Pbsol2+/Pb(Hg) electrode in non-aqueous solvents measured versus so-called solvent independent electrode (bis(biphenyl) chromium(0/+1)) appeared to be linearly dependent on the donor number of the used solvents. Such experimental dependence is given by Eq. (11). E0 ¼ 1:016 − 0:0254DN

ðcorrelation coefficient, r ¼ 0:945 for 16 solventsÞ 0

(11)

2+

With the increase of the donor number of solvents the formal potential, E , of the system Pbsol /Pb(Hg) couple moves to more negative values in agreement with expectation. In the measurements of these potentials the lead amalgam was used. Also the potentials of Pb2+/Pb electrode in molten salts were measured, especially in chlorides of alkaline and alkaline earth metals. In addition to the single molten salts also molten mixed halide electrolytes were used in such studies. Early experiments in the molten salts were summarized by Sharpe.6 Sometimes in such studies the chlorine electrode was used as a reference one. The potential of lead in molten PbCl2 given versus chlorine electrode is equal to −1.27 V at 500  C and −1.17 V at 700  C. These values are similar to the standard potential of the Pb(II)/Pb electrode in aqueous solutions expressed versus the chlorine electrode (−1.49 V), if these potentials are compared at similar temperatures.

4 4.1

Electrode behavior of lead Electrode behavior of lead and its compounds dissolved in different media

Electrochemical properties of lead were described in detail e.g. in Refs. 6,7 and in our chapter.8 The kinetics and mechanism of the oxidation/reduction process of the Pb(II)/Pb system were studied in various environments. In non-complexing media this electrode reaction is very fast. It was studied in detail using mercury electrodes (lead amalgam) and for the Pb(II)/Pb(Hg) system in aqueous solutions the standard rate constants of the order of 2  3 cm s−1 are reported. At mercury electrodes electroreduction of numerous lead(II) complexes was studied. Since, as a rule, also in these cases the electrode process is very fast, the complex equilibria in such systems could be elucidated and stability constants of various complexes determined. In addition to the complexes formed with common ligands (early works were reported by Kuhn7) also lead complexes with cyclodextrines and their derivatives, amino acids, cryptands and other macrocycles as well as many other ligands were studied. The behavior of Pb(II)/Pb(Hg) system was studied also

Chemistry and Electrochemistry | Lead

671

Fig. 1 Potential - pH equilibrium diagram for the system lead-water at 25  C.4

in non-aqueous and mixed solvents. There is no space and necessity to discuss in the present chapter these works in a greater detail. We would like only to point that such studies were less frequent than those carried out in aqueous solutions. The kinetics of the Pb(II)/Pb(Hg) systems in non-aqueous media was found to depend on the donor properties of the solvent used in the study according to the empirical equation. log ks ¼ 3:4 − 0:17DN

ðr ¼ 0:997 for 3 solventsÞ

(12)

Logarithm of the standard rate constant, ks, of the lead electrode reaction decreases linearly with the donor number of the solvent. This shows that the solvation energy of lead(II) influences the activation energy of the lead electrode process. The electrode behavior of lead in mixed solvents is more complex. Mostly mixtures of water with organic solvents were used. It was found that at least in some mixtures the rate constant changed from values characteristic for one solvent to that corresponding to the other pure solvent but not according to the bulk mixed solvent composition but to the composition of the surface phase. In liquid ammonia lead(II) can be reduced to Zintl ions such as Pb94−. Alkali metal was incorporated in the intermetallic phase formed. The redox processes of lead(II) to liquid lead have been also studied in molten halides. In recent decades a lot of published works was devoted to the lead underpotential deposition (Pb UPD), very frequently on well defined substrate metal faces. Since UPD process precedes the massive deposition of metal (lead), we will present very briefly these results. Pb UPD on Pt(111) was studied in solutions of perchloric and sulfuric acids. A large influence of sulfuric acid on this process was found, since the UPD of lead occurred on the surface covered by HSO−4 and SO42− ions. The influence of electrode smoothness and rate of potential scanning in early stages of lead deposition were also studied. The structure of the deposit with Pb-Pb and Pb-Pt domains has been considered. Pb UPD has been intensively studied on polycrystalline gold and defined Au(h,k,l) planes and comparison of voltametric curves recorded on different planes has been discussed.

672

Chemistry and Electrochemistry | Lead

2.0 PbO32–

Pb4+ 1.8 1.6 1.4

O

2

PbO2

+4

H+

1.2

+4

e

2H

2O

0.8

Pb2+

0.6 Pb3O4

0.4

PbO.PbSO4

Potential vs. N.H.E (V)

1.0

PbSO4 0.2

2H + 0

+2

e

3PbO.PbSO4

H

2

–0.2

PbO HPbO–2

–0.4 –0.6 Pb –0.8 –1.0 –2

0

2

4

6

8 pH

10

12

14

16

18

Fig. 2 Potential - pH diagram of lead in the presence of sulfate ions at unit activity at 25  C.5

On polycrystalline gold in HClO4 acid solutions a combined quartz crystal microbalance and probe beam deflection methods have pointed to three stages of lead deposition. Lead layer structures, changing with surface coverage of lead, with consideration of the formation of AuPb2, have been detected and studied using LEED and Auger spectroscopy. The Pb UPD on different Au(h,k,l) planes has been reviewed by Herrero et al.9 More details on the Pb underpotential deposition on different substrates one may find in the literature. In the literature there is a lot of papers devoted to electrode reactions of different lead species. There is no space to discuss these papers to a larger extent. We will only very briefly summarize the works related to voltametric behavior of lead and its solid compounds (such as PbO2 or PbSO4) which play (or may play) a role in a storage of electrical energy and in other technological applications.

4.2

Voltametric behavior of lead and its non-soluble compounds in different solutions

Voltametric behavior of lead in various electrolyte solutions is usually complicated by precipitation /dissolution of ill-soluble lead compounds.8 In phosphoric acid a pair of peaks was recorded due to oxidation/reduction of the couple PbHPO4/Pb. In the oxidation step, the PbHPO4 deposit covers the electrode what results in its passivation. The phosphate deposit inhibits also lead surface oxidation to PbO2. In Na2HPO4 solution oxidation of lead results in deposition of Pb(OH)2 coupled with partial dehydration process to PbO, while at higher potentials growth of PbO and PbO2 occurs. This was represented by a few peaks on the voltammograms. In cathodic scan reduction of these solid forms occurs. In nitric acid solutions two anodic peaks were observed, representing PbO and lead nitrate formation (first peak) and oxidation to PbO2 (more anodic peak). In concentrated HNO3 solutions Pb dissolution process can be inhibited owing to the presence of PbO2 layer. In course of cathodic scan two peaks or one peak followed by a plateau can be observed. In sodium borate solutions deposition/dissolution of lead oxides is coupled with deposition of lead borate. Deposition of lead was also studied from room temperature ionic liquids. In the case of nickel electrodes, the nucleation mechanisms of Pb was determined to be three-dimensional instantaneous nucleation with diffusion-controlled growth.10

Chemistry and Electrochemistry | Lead

673

Below we intend to describe also briefly electrochemical behavior of several solid lead compounds which are important in the energy storage devices.

4.3

Voltametric behavior of lead in sulfuric acid solutions

Owing to obvious applications in lead-acid batteries, voltametric characteristics of lead in sulfuric acid solutions is very important. Typically, voltametric curves of lead, limited by hydrogen and oxygen evolution processes, exhibit several peaks, an example is presented in Fig. 311 with following peaks: (a) - oxidation of Pb to PbSO4, (b) - oxygen evolution, (c) - oxidation of PbSO4 to PbO2 (usually b form), (d) - reduction of PbO2 to PbSO4, (e) - oxidation of Pb to PbO, (f) - reduction of PbO to Pb, (g) - reduction of PbSO4 to Pb and (h) - hydrogen evolution. The inner PbO layer is formed due to impermeability of PbSO4 layer for SO42− ions, only Pb2+, OH− and H+ ions can move across this film. Thus, for anodic scan, H+ ions can flow from the reaction site into the solution, resulting in pH increase close to the electrode surface. With increasing H2SO4 concentration, the lead sulfate layer is more compact and electrolyte ions access to the internal layer is hindered. A characteristic feature is the presence of a small anodic peak (anodic excursion peak), located close to the peak (d). The most probable explanation is based on large increase of molar volume accompanying the reduction peak (d): from 25 cm3.mol−1 for b-PbO2 to 48 cm3.mol−1 for PbSO4, resulting in cracks of the film on electrode surface. Thus, metallic lead surface is exposed, which can be oxidized by sulfuric acid. A more complicated behavior is observed in Na2SO4 solutions,8 where effects typical for both sulfuric acid and alkaline solutions occur. Anodic peaks corresponding to PbSO4, PbO and PbO2 formation can be recognized; moreover, intermediate oxides and soluble HPbO−2 can be also produced. In cathodic scan peaks corresponding to reduction of PbO2 to PbSO4, PbO2 to PbO as well as Pb(II) compounds (PbO and PbSO4) to metallic Pb are recorded.

4.4

Voltametric behavior of lead oxides

PbO can occur as a transition form in oxidation/reduction process of the couple PbO2/Pb(II).8 It is particularly stable in alkaline solutions. Anodic oxidation of Pb results in growth of b-PbO which can be then transformed to a-PbO. The structure of PbO deposit is dependent on hydroxide (NaOH) concentration. In more concentrated solutions amorphous or submicrocrystalline forms are obtained. In 0.05  0.5 M solutions highly oriented (110) a-PbO is obtained, while in dilute solutions (0.01 M NaOH) b-PbO is also present. The stability of PbO2 structure is dependent on solution acidity and deposition conditions. a-PbO2 is more stable mechanically and can be obtained electrochemically in alkaline or neutral solutions of lead(II) salts. b-PbO2 is thermodynamically more stable, it can be obtained in acidic solutions. The latter form predominates if low current density is applied or cyclic polarization is prolonged. PbO2 was found to be easily reduced using electrochemical methods, the obtained products are dependent on solution pH and applied overpotential. In acidic media reduction results in Pb2+ ions (in the absence of anions giving ill-soluble compounds), while in alkaline solutions various hydroxo- and oxide forms are postulated as adsorbed PbO and Pb(OH)−3 species.

(a) PboPbSO4

300

First cycle Current density, i / Am–2

200

(b) H2OoO2

Second cycle Third cycle (c) PbSO4oPbO2

100

(e) PboPbO 0 (f) PbOoPb –100

(h) H+oH2 (g) PbSO4oPb

–200

(d) PbO2oPbSO4 –1000

–500

0

500

1000

Potential, E/ mV vs. Ag/AgCI Fig. 3 Cyclic voltammetric curves of Pb electrode in 4.5 M H2SO4 solution.11

1500

2000

674

Chemistry and Electrochemistry | Lead

Lead dioxide coatings on inert substrates such as titanium and carbon offer new opportunities for this material. It is recognized that electrodeposition allows the preparation of stable coatings with different phase structures and a wide range of surface morphologies. In addition, substantial modification of the physical properties and catalytic activities of the coatings are possible through doping and the fabrication of nanostructured deposits or composites. In addition to applications as a cheap anode material in electrochemical technology, lead dioxide coatings provide unique possibilities for probing the dependence of catalytic activity on layer composition and structure.12 The main application of PbO2 is a positive electrode in lead-acid cells as well as electrode material for electrochemical processes of practical significance (see Sections 4–6).

4.5

Electrochemical properties of lead sulfide

PbS can be obtained electrochemically by deposition at constant potential or under conditions of cyclic voltammetry from buffered aqueous solutions of Pb(II) (low concentration of the Pb2+ ions) and Na2S (or other sulfur containing compounds such as Na2S2O3). Organized PbS structures can be obtained electrochemically using electrochemical atomic layer epitaxy method, where separate underpotential deposition of elements (Pb, S) is carried out layer-by-layer. The rate of anodic oxidation of PbS in various media differs significantly, the common main oxidation product is elemental sulfur. In recent years considerable and growing interest concerns PbS nanoparticles/quantum dots. Numerous chemical methods of PbS nanoparticles production have been proposed using micelles, polymers, sol-gel or hydrothermal technique as well as microwave or ultrasonic irradiation. In a simple electrochemical method fast scan cyclic voltammetry can be applied, with Pb anode being oxidized to Pb2+ ions reacting with S2− present in the solution, in the presence of poly(N-vinyl alcohol) protecting from agglomeration.

4.6

Lead-acid batteries

The lead acid cell was constructed by French physicist Gaston Planté in 1859 and now these cells belong to the most widely used electrochemical power sources. Details of construction and performance of such cells can be found, for instance, in the chapter written by Bullock and Vincent13 and earlier books or chapters (see8). Although there are rapidly developing other rechargeable electrochemical power sources, mainly lithium-ion batteries, lead-acid batteries are still very popular. The main advantages of lead-acid batteries concern low costs, simple and known technology, no necessity of using rare elements, very effective recycling as well as low self-discharge rate. Lead-acid cell is a reversible cell consisting of negative electrode, usually of porous lead (lead sponge) and a positive electrode of lead dioxide, PbO2, immersed in solution of sulfuric acid: PbðsÞ | PbSO4 ðsÞ | H2 SO4 ðaqÞ | PbSO4 ðsÞ | PbO2 ðsÞ | PbðsÞ +

Because sulfuric acid dissociates into H and represented by equation:

HSO−4

(13)

ions under applied conditions, the overall discharging/charging process can be

PbðsÞ + PbO2 ðsÞ + 2H+ + 2HSO4 − ⇆2PbSO4 ðsÞ + 2H2 O

(14)

In the course of a discharge process sulfuric acid is consumed to deposit lead sulfate and water is formed as by-product. Therefore, the acid concentration decreases from about 40 w/w % (full charge) to about 16 w/w % (full discharge). This is accompanied by decrease of both electrolyte solution density and open circuit voltage from 2.15 V (1.30 g.mL−1) to 1.98 V (1.10 g. mL−1), respectively. Lead-acid batteries can be characterized by different configuration of electrodes. In a conventional battery, electrons generated in the oxidation process move from the active material to a current collector and then to the next cell through the outer circuit. However, in a different setup, in bipolar batteries,14 electrons move directly to the next cell through the substrate (partition wall) as the positive and the negative active material are placed on two opposite faces of the bipolar substrate. The bipolar configuration simplifies the battery design by elimination external circuit components and the ohmic power loss is reduced due to shortened conductive paths. In a typical configuration the battery contains a stack of bipolar electrodes separated by separators containing the electrolyte. Harned and Hamer15 considered the emf of the lead-acid storage battery as a function of the H2SO4 concentration and the temperature from 0 to 60  C. Their equation is of the form: E ¼ E0 + at + bt 2 0

(15)



where E is the emf at 0 C, t is the temperature in the Celsius scale, while a and b are constants different for each H2SO4 concentration. These constants are given in Table 4. The discharge process at the positive electrode is: PbO2 + 3H+ + HSO4− + 2e −

discharge

!

PbSO4 + 2H2 O

(16)

and the charging reaction proceeds in the opposite direction. The discharge process follows according to dissolution – precipitation mechanism, with the formation of Pb2+ in charge transfer reaction followed by precipitation of PbSO4. In charging process, Pb2+

Chemistry and Electrochemistry | Lead Table 4

675

Constants for the equation of E ¼ E0 + at + bt2.

m (H2SO4)/mol kg−1

E0/V

106 a/V K−1

108b/V K−2

0.05 0.10 0.20 0.50 1.0 2.0 3.0 4.0 5.0 6.0 7.0

1.7687 1.8021 1.8349 1.8791 1.9174 1.9664 2.0087 2.0479 2.0850 2.1191 2.1507

−310 −265 −181 −45 56.1 159 178 177 167 162 153

134 129 128 126 108 103 97 91 87 85 80

ions, from PbSO4, are oxidized to Pb(IV) on b-PbO2 of relatively high electronic conductivity and react with water to deposit as b-PbO2. Oxidation of PbSO4 occurs inside the salt layer and b-PbO2 is formed at the interface between the support electrode and the PbSO4 layer. The oxide layer grows and then PbSO4 deposit close to the interface between the electrolyte solution and the electrode is oxidized. In order to obtain high current densities it is necessary to use highly porous structure of PbO2 in the b-form, diminishing the role of passivation of the positive electrode by poorly conducting PbSO4. The porosity is additionally important because it enables the expansion during the solid phase volume increase, which accompanies the transformation of PbO2 to PbSO4. In the most popular construction SLI (SLI - starting, lighting, ignition) the electrode paste material (mixture of metallic lead with lead oxides) is hold in a framework composed of lead/lead alloys. The discharge process at the negative electrode is: Pb + HSO4−

discharge

!

PbSO4 + H+ + 2e −

(17)

The discharge reaction also occurs according to dissolution – precipitation mechanism with formation of Pb ions in the charge transfer step, followed by PbSO4 precipitation. For charging process it is assumed that Pb2+ ions are transported from the PbSO4 crystal surface to the Pb surface and then reduced. The reaction rate is usually controlled by mass transfer – diffusion in a narrow gap between the PbSO4 and Pb surface. Negative electrodes are pasted plates using grids covered with perforated lead foil and the same paste as that used in positive electrode plates. Under specified conditions the paste material is reduced to lead sponge of high porosity with a high electrode surface area. Additions of so-called expanders as surface active substances (e.g. lignosulfonic acid) is useful to lower the surface energy of lead and thus to reduce large crystals formation. Improvement of properties of electrode materials can be obtained by using lead alloys as a grid material, with additions of tin, antimony, selenium and calcium. Antimony, the most popular additive, improves the mechanical stability; however, it increases the resistance and facilitates self-discharge of the battery. Better results are obtained using low antimony content and/or lead-calcium alloys. Calcium addition can be advantageous, however, higher content of this element can result in higher corrosion rate. Tin or silver addition to lead-calcium alloys is advantageous, owing to significant decrease of corrosion rate (formation of calcium-tin intermetallic compounds); moreover, tin improves also mechanical properties of the alloys. More significant improvement of mechanical properties can be also obtained by barium addition to lead-calcium-tin alloys. Concentration of H2SO4 affects the processes occurring in lead-acid cells. Three acid concentration ranges, important from the point of view of cell operation, can be distinguished.16 The first is active H2SO4 concentration range (from 0.5 to 5 M), where the concentration of HSO−4 is the highest and b-PbO2 phase is formed. This phase is reach in hydrated gel areas (PbO(OH)2) which are ionically (H+) and electronically conductive, necessary for electrochemical process to occur, e.g. for hydrated PbO2 reduction: 2+

PbOðOHÞ2 + 2H+ + 2e − ! PbðOHÞ2 + H2 O

(18)

+ PbðOHÞ2 + HSO 4 + H ! PbSO4 + 2H2 O

(19)

The highest electrochemical activity of PbO2 is observed for sulfuric acid concentration range from 1.7 to 4.8 M. On the other hand, concentrations higher than 5 M and lower than 0.5 M correspond to a passive region, where the amount of gel zones is low, resulting in low electroactivity. Separators used in lead-acid batteries, between the half-cells of opposite polarity, are usually produced from synthetic polymers such as sintered poly(vinyl chloride) and extruded polyethylene, glass or microglass fiber. The partially hydrophobic properties of synthetic separators can be overcome by the presence of organic surfactants. The recharging process of the cell can be accompanied by a side process of water electrolysis resulting in water loss. Additional source of gases (hydrogen and oxygen) can be spontaneous processes occurring at electrodes: lead dissolution in sulfuric acid solution (hydrogen release at the negative electrode Eq. 20) and PbO2 decomposition associated with oxygen evolution (at positive electrode, see Eq. 6):

676

Chemistry and Electrochemistry | Lead Pb + H+ + HSO4− ! PbSO4 + H2

(20)

The rate of these processes is dependent on temperature, electrolyte composition and presence of impurities which can catalyze hydrogen evolution reaction, e.g. antimony removed from the positive electrode grid, deposited on the negative electrode or impurities present in the electrolyte solution (e.g. Fe2+ ions). These processes should be minimized, especially in maintenance free (MF) batteries, because they lead to undesirable water losses. MF batteries have grid with reduced antimony amount, other metals such as calcium, strontium, tin or aluminum are used. Valve regulated lead-acid (VRLA) batteries have been constructed to enhance the recombination of the oxygen at the negative electrode. In VRLA batteries oxygen diffuses from the positive to negative electrode, where it oxidizes lead, preventing it to obtain hydrogen evolution potential. On the other hand, the lead electrode has relatively low potential, enabling reduction of the oxidized lead back to the metallic form. In such batteries the electrolyte amount is reduced, it can be immobilized by gel formation, using silica or calcium sulfate or by incorporation of microporous glass separators. Under some conditions as e.g. a prolonged period in an uncharged state, operation at high temperatures or at too high acid concentration, a so-called sulfation process can occur. PbSO4 can gradually recrystallize into a dense, coarse-grained form, resulting in passivation especially of the negative electrode. This effect can be reduced by addition of phosphoric acid. Three main types of lead-acid batteries can be distinguished: SLI batteries, industrial batteries and small portable batteries. SLI batteries are the most popular, they represent about 80% of total production. These batteries should supply for short time currents of a large intensity (charge over 5C). They usually contain thin pasted plates, thin separator layers to reduce the internal resistance. Typical SLI batteries for car applications have voltage of 12 V, capacity of 30  100 Ah and specific energy around 30 Wh kg−1. In recent decades, a significant improvement of properties of the most popular automotive starting batteries has been obtained due to following technical modifications of the classical construction: (i) application of thin-walled polypropylene cases, allowing more space for the plates, (ii) optimization of electrode geometry and connections, lowering the internal resistance and also allowing more space for active elements, (iii) application of thinner polymeric separators and (iv) use of lightweight plastic grids. Industrial batteries should fulfill more rigorous requirements (compared to SLI batteries) concerning high specific energy and capacity (20  30 Wh.kg−1 and 100  1500 Ah, respectively), voltage 12  240 V and resistance to vibrations. As positive electrodes tubular plates can be applied, pasted grids are also used; however, in this case special separators are required to absorb shocks and vibrations and to immobilize the electrode active material. Portable batteries of capacity 2  30 Ah should be vibration resistant and capable of working at any position without electrolyte leakage. Positive and negative electrodes are usually produced with “honeycomb” grid support filled with electroactive material. They contain also separators as thin films of porous insulating material, retaining the electrolyte, as well as heat and oxidation resistant glass microfiber mat. A significant challenge is development of electrode materials characterized by good performance and low cost. From this point of view carbon-based materials are very promising.17–22 Therefore carbon is often used in lead-acid batteries, mainly in three components: the active mass, the current collector and also the negative plate. Application of carbon as additive of the active mass concerns mainly the negative electrode, where it is used as “expander”. This addition is particularly useful for batteries working at high rate/discharge currents, lead-acid batteries with carbon additives show a significant improvement in the high-rate partial stage of charge (HRPSoC) conditions, applicable e.g. in hybrid vehicles. Among possible factors concerning advantageous role of carbon role as additive to the negative electrode, three of them seem the most important.19 (i) Carbon can change the structure of the active mass and form a conductive skeleton. In the course of discharge non-conductive lead sulfate is formed, however, carbon presence helps to retain conductivity and it provides larger surface area, even in a discharge level. (ii) Carbon provides mechanical restriction for the growth of lead sulfate crystals, formation of big crystals would slow down their reduction to lead, This “sulfation” process would lead to exclusion of some parts of the active mass from the process and thus to loss of capacity. Addition of carbon inhibits growth of large lead sulfate crystals due to steric effect and therefore it improves the cyclability of the battery. (iii) Carbon additive can act as a capacitor storing charge in the electric double layer. This results also in higher power when using high currents. As carbon materials various forms may be used as activated carbon, carbon black, graphite powder or nanomaterials as carbon nanotubes, graphene carbon nanofibers, the amount of additives is usually a few %. Typically used carbon additives have specific surface area and capacitance greater by order of magnitude than those of a typical negative electrode (mass). However, disadvantageously, application of carbon materials results in acceleration of the hydrogen evolution reaction, leading to losses of electrolyte in the course of battery charging. This effect can be mitigated by doping with non-metals as boron, nitrogen, phosphorus, also some lead-carbon composites may be used. Carbon materials can be also used as an additive to the positive electrode mass, however, this is not so widespread, due to sometimes less stable working conditions and not so significant improvement as in the case of negative electrodes. Carbon materials can improve also properties of current collectors. In this case a typical grid cast from lead alloy can be replaced by much lighter carbon material, e.g. reticulated vitreous glassy carbon (RVC), as proposed by Czerwi nski et al. (e.g.23). This material is resistant to corrosion when working in the negative plate, however, it can undergo corrosion when used with the positive plate. For this purpose also other carbon materials may be used as e.g. graphite foams or foils.

Chemistry and Electrochemistry | Lead

677

The carbon material may be also used to replace the whole standard negative plate with a capacitor plate. This solution results in higher power but lower capacity compared to a typical battery. The carbon capacitor plate may be also used parallel with a standard negative lead plate, this setup was applied in the Ultrabattery developed by Australian Commonwealth Scientific and Industrial Research Organisation. The properties of batteries with capacitor plate are promising for applications in electric and hybrid vehicles.

5

Other lead batteries (sulfuric acid free)

A serious problem related to lead-acid batteries operation is passivation of electrodes by deposited PbSO4. Therefore, other constructions have been proposed, with PbSO4 replaced by a soluble lead(II) salt. As electrolytes HClO4, HBF4 or H2SiF6 have been used to obtain primary cells; however, with limited practical significance. Also safety hazards associated with above mentioned acids seriously limit their practical applications. Some new concepts of secondary cells have been also proposed, using e.g. HBF4, with charging/discharging processes occurring as follows: Pb + PbO2 + 4HBF4 ⇆2PbðBF4 Þ2 + 2H2 O

(21)

The main problems encountered in such a cell relate to the positive electrode as anodic oxidation of the electrode material, oxygen evolution at the end of charging process, passivation at higher charging current density and release of PbO2 particles. At the beginning of 21st century, a novel kind of flow lead-acid battery with single electrolyte and no separator, lead(II) in methanesulfonic acid (MSA) has been proposed by Hazza et al.24 Carbon was found to be a suitable material for both electrodes, ensuring good adhesion of both Pb and PbO2. The appropriate electrolyte composition for uncharged cell is 1.5 M lead(II) methanesulfonate with addition of 0.9 M methanesulfonic acid. CðsÞ | PbðsÞ | PbðIIÞ in 0:9M CH3 SO3 HðaqÞ | PbO2 ðsÞ | CðsÞ

(22)

The overall discharging/charging process is: Pb + PbO2 + 4H+ ⇆2Pb2+ + 2H2 O

(23)

The electrode reactions occurring during the discharge are: Pb ! Pb2+ + 2e ðat negative electrodeÞ

(24)

PbO2 + 4H+ + 2e ! Pb2+ + 2H2 O ðat positive electrodeÞ

(25)

The battery voltage is close to 1.5 V. The charge transfer reactions are rapid and charging/discharging process can be repeated with significant efficiency. The major source of energy losses are overpotentials associated with deposition and dissolution of PbO2. Additional drawback can be uneven deposition of lead at the negative electrode and growing across the interelectrode gap, possibly resulting in shortening of the cell. Some additives can be used for levelling of the Pb electrodeposition, e.g. sodium ligninsulfonate or polyethyleneglycol. On the other hand, there are still attempts to catalyze effectively the process of the PbO2/Pb2+ couple at the positive electrode, using e.g. inorganic ions such as Ni(II). These batteries are designed to operate in a flow mode (soluble-lead flow battery, SLFB). In the simplest SLFB design, Pb2+ ions are dissolved in an aqueous MSA electrolyte and this is then pumped through an undivided electrochemical cell. Typical SLFB electrolytes offer up to 40 Wh kg−1 of storage, with performance on the 100 cm2 electrode scale reaching 90% charge and 80% voltage efficiencies across 100 cycles. However, the SLFB has also been tested on the 1000 cm2 electrode, four-cell stack scale.25 Other lead battery concepts represent substitution of the negative or positive electrode by non- lead materials, e.g. the systems with the following charging/discharging reactions have been proposed:

6

Cd + PbO2 + 2H2 SO4 ⇆CdSO4 + PbSO4 + 2H2 O ðvoltage : 2:16VÞ

(26)

Cu + PbO2 + 2H2 SO4 ⇆CuSO4 + PbSO4 + 2H2 O ðvoltage : 1:325V Þ

(27)

Pb + Ag2 O⇆PbO + 2Ag ðvoltage : 0:92V Þ

(28)

Other electrochemical applications of Pb and Pb oxides

Lead is used also as a cathode material for electroreduction of different organic compounds, because the hydrogen evolution overpotential on lead is quite large. Such lead cathodes have been used, for instance, in the production of aniline from nitrobenzene. Lead anodes have been used also for cathodic protection of ships and different constructions which are in contact with water, especially seawater. Since the long use of such Pb (PbO2 at positive potentials) electrodes depends on the stability of the PbO2, coating additions of different metals (Ag, Sb, Bi, Sn and others) were used to improve the long-lasting electrochemical work of such electrodes.

678

Chemistry and Electrochemistry | Lead

Due to the high oxygen evolution overpotential, stability and relatively high conductivity, PbO2 was also used as an electrode material working at highly positive potentials to oxidize different organic compounds (also harmful to environment) as numerous organic acids, phenol, 4-chlorophenols, benzene, some nitrocompounds, cyanides. Such electrodes were also applied for electrolytic production of different inorganic compounds such as perchlorates, bromates and chlorine. PbO2 can be also used for electrochemical ozone generation as an alternative to a classical cold corona discharge (CCD) technique. This method has some advantages related to higher in situ ozone concentration without by-products and secondary pollution.12 The electrolytic generation of ozone requires a high positive potential (+1.5 V) and is accompanied by significant oxygen evolution. Therefore, high current efficiency and inhibition of O2 evolution are advantageous. The anodic generation of ozone is carried out at lead dioxide anodes in various aqueous media, including e.g. neutral sodium sulfate, perchloric acid, phosphoric acid, sulfuric acid etc. Because of very positive potential necessary for ozone generation, the only alternatives to lead dioxide is vitreous carbon or diamond, however, lead dioxide is more flexible in cell design and is cheaper.12 A significant advance in electrolytic ozone production was a generator with proton conducting polymer electrolyte (Nafion®) and PbO2 anode, known under the trademark Membrel®. It is used for disinfection and purification of water, to supply ultra-pure water, e.g. for pharmaceuticals, fine chemicals, electronics and food supply industry. Ozone electrogenerated at PbO2 has been shown to be effective for destruction of many organic compounds as e.g. phenols, reactive dyes etc. Lead compounds can find also photoelectrochemical applications, mainly in solar cells. A promising way of solar energy conversion is application of nanoscale materials, e.g. nanocrystal excitonic solar cells of Pb compounds as PbS, PbSe and PbSSe. They have gained interest due to their large exciton Bohr radii and low cost fabrication. The open circuit voltage of such cells depends on the energy levels of nanocrystals, which vary with their size. Similarly, non-stoichiometric oxides, PbOx, can be used in fabrication of solid state Schottky junction solar cells. In this case higher work function metals as Pt, Au or Pd are suitable candidates to realize a Schottky junction with n-PbOx.26 Lead oxide (PbO) is a semiconductor of particular interest, due to progress in its preparation and applications, e.g. in batteries, solar cells, gas sensors, photocatalytic applications etc. PbO as semiconductor is characterized by band gap of 1.9–2.2 eV and 2.6 eV, for tetragonal and orthorhombic phases, respectively. The tetragonal PbO form was found to be more photoactive than the orthorhombic phase. Therefore, methods of annealing oxidation of electrodeposited lead were used and optimized to obtain lead oxide mainly in the advantageous tetragonal form.27 Application of lead oxide in nanoscale is advantageous also form the point of view of other photoelectrochemical applications. It was found that orthorhombic, b-PbO quantum dots can be fabricated as a working electrode in a photoelectrochemical typed detector that exhibits significantly high photocurrent density and excellent stability under ambient conditions.28 Lead is also applied in oxygen sensors. The conventional oxygen sensor is a galvanic cell, which uses an anode of lead wool and an inert cathode. In this simple and low cost setup the reactions proceed spontaneously (without external voltage): O2+2H2O+4e−!4OH− (cathode)

(29)

2Pb+4OH !2PbO+2H2O+4e (anode)

(30)

-

-

with net oxidation of Pb to PbO and thus consumption of the anode material. After lead is oxidized or the solid PbO/Pb(OH)2 completely covers the Pb surface, the sensor’s work stops. The life time of the sensor depends on the rate of oxygen diffusion which is controlled by the inlet capillary diameter. A narrow capillary enables long application of the sensor, up to 3 years.

7

Conclusions

Lead is an important element from electrochemical point of view. Its significance related mainly to its applications in lead-acid batteries. Although the concept of this electrochemical power source has been known for over 160 years, lead-acid batteries are still very popular. The main advantages of these batteries concern low costs, simple and known technology, available raw materials, very effective recycling, low self-discharge rate. Application of carbon materials has been a significant trigger for lead-acid batteries development due to lower weight, higher capacitance and better active mass utilization. Therefore these batteries can effectively compete with newer battery types as e.g. lithium-ions batteries and there are promising perspectives of their further development. Additional advantages related to electrochemistry of lead is development of soluble lead flow batteries as well as applications of lead dioxide as advantageous anode material in electrochemical technology and lead oxides in photoelectrochemical applications.

Acknowledgment The authors are grateful to Professor Zbigniew Galus for his help, support and valuable comments concerning this chapter.

Chemistry and Electrochemistry | Lead

679

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; part devoted to chemistry of lead; Wiley: New York, 1999. all editions. Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements; Pergamon Press: Oxford, 1985; pp. 243–289. Overton, T. L.; Rourke, J. P.; Weller, M. T.; Armstrong, F. A. Inorganic Chemistry, 7th ed.; Oxford University Press: Oxford, 2018. Pourbaix, M.; De Zoubov, N.; Vanleugenhaghe, C.; Van Rysselberghe, P. Lead. In Atlas of Electrochemical Equilibria in Aqueous Solutions; Pourbaix, M., Ed.; Pergamon Press: New York, 1966; pp. 485–492. Mathieson, R. T. In Batteries 2; Barnes, S. C., Collins, D. H., Eds.; Pergamon: New York, 1967; p. 163. Sharpe, T. F. Lead. In Encyclopedia of Electrochemistry of the Elements; Bard, A. J., Ed.; 1; M. Dekker: New York, 1973; pp. 235–347. Kuhn, A. T. The Electrochemistry of Lead; Academic Press: London, 1979. Stroka, J.; Maksymiuk, K.; Galus, Z. Electrochemistry of lead. In Encyclopedia of Electrochemistry; Bard, A. J., Stratmann, M., Eds.; 7; Wiley-VCH: Weinheim, 2006; pp. 804–837. Herrero, E.; Buller, L. J.; Abruña, H. H. Underpotential Deposition at Single Crystal Surface of Au, Pt, Ag and Other Materials. Chem. Rev. 2001, 101, 1897–1930. Tsai, R. W.; Hsieh, Y. T.; Chen, P. Y.; Sun, I. W. Voltametric Study and Electrodeposition of Tellurium, Lead, and Lead Telluride in Room-Temperature Ionic Liquid 1-Ethyl-3-Methylimidazolium Tetrafluoroborate. J. Electroanal. Chem. 2014, 137, 49–56. Taguchi, M.; Sugita, H. Analysis for Electrode Oxidation and Reduction of PbSO4/Pb Electrode by Electrochemical QCM Technique. J. Power Sources 2002, 109, 294–300. Li, X.; Pletcher, D.; Walsh, F. C. Electrodeposited Lead Dioxide Coatings. Chem. Soc. Rev. 2011, 40, 3879–3894. Bullock, K. R.; Vincent, C. A. Secondary Lead-Acid Cells. In Modern Batteries. An Introduction to Electrochemical Power Sources; Vincent, C. A., Scrosati, B., Eds.; Arnold: London, 1997. Pradhan, S. K.; Chakraborty, B. Substrate Materials and Novel Designs for Bipolar Lead-Acid Batteries: A Review. J. Energy Storage 2020, 32, 101764. Harned, H. S.; Hamer, W. J. The Molal Electrode Potentials and the Reversible Electromotive Forces of the Lead Accumulators from 0 to 600 Centigrade. J. Am. Chem. Soc. 1935, 57, 33–35. Pavlov, D.; Kirchev, A.; Stoycheva, M.; Monahov, B. Influence of H2SO4 Concentration on the Mechanism of the Processes and on the Electrochemical Activity of the Pb/PbO2/ PbSO4 Electrode. J. Power Sources 2004, 137, 288–308. Hao, Z. D.; Xu, X. L.; Wang, H.; Liu, J. B.; Yan, H. Review on the Roles of Carbon Materials in Lead-Carbon Batteries. Ionics 2018, 24, 951–965. Lach, J.; Wróbel, K.; Wróbel, J.; Podsadni, P.; Czerwinski, A. Applications of Carbon-Lead in Lead-Acid Batteries: A Review. J. Solid State Electrochem. 2019, 23, 693–705. Lach, J.; Wróbel, K.; Wróbel, J.; Czerwinski, A. Applications of Carbon in Rechargeable Electrochemical Power Sources: A Review. Energies 2021, 14, 2649. Moseley, P. T.; Rand, D. A. J.; Davidson, A.; Monahov, B. Understanding the Functions of Carbon in the Negative Active-Mass of the Lead-Acid Battery: A Review of Progress. J. Energy Storage 2018, 19, 272–290. Wang, F.; Hu, C.; Zhou, M.; Wang, K.; Lian, J.; Yan, J.; Cheng, S.; Jiang, K. Research Progresses of Cathodic Hydrogen Evolution in Advanced Lead-Acid Batteries. Sci. Bull. 2016, 61, 451–458. Mahajan, V.; Bharj, R. S.; Bharj, J. Role of Nano-Carbon Additives in Lead-Acid Batteries: A Review. Bull. Mater. Sci. 2019, 42, 21. Czerwinski, A.; Z˙ elazowska, M. Electrochemical Behavior of Lead Deposited on Reticulated Vitreous Carbon. J. Electroanal. Chem. 1996, 410, 55–60. Hazza, A.; Pletcher, D.; Wills, R. A Novel Flow Battery: A Lead Acid Battery Based on an Electrolyte with Soluble Lead(II). Phys. Chem. Chem. Phys. 2004, 6, 1773–1778. Krishna, M.; Fraser, E. J.; Wills, R. G. A.; Walsh, F. C. Developments in Soluble Lead Flow Batteries and Remaining Challenges: An Illustrated Review. J. Energy Storage 2018, 15, 69–90. Patel, D. B.; Mukhopadhyay, I. Schottky Junction Solar Cells Based on Non-Stoichiometric PbOx Films. J. Physics D: Appl. Phys. 2015, 48, 025102. Zerguine, W.; Abdi, D.; Habelhames, F.; Lakhdari, M.; Derbal-Habak, H.; Bonnassieux, Y.; Tondelier, D.; Choi, J.; Nunzi, J. M. Annealing Effect on the Optical and Photoelectrochemical Properties of Lead Oxide. Eur. Phys. J. Appl. Phys. 2018, 84, 30301. Huang, W.; Jiang, X.; Wang, Y.; Zhang, F.; Ge, Y.; Zhang, Y.; Wu, L.; Ma, D.; Li, Z.; Wang, R.; Huang, Z. N.; Dai, X.; Xiang, Y.; Li, J.; Zhang, H. Two-Dimensional Beta-Lead Oxide Quantum Dots. Nanoscale 2018, 10, 20540.

Further reading 1. Blair, T. L. Lead Oxide Technology—Past, Present and Future. J. Power Sources 1998, 73, 47–55. 2. Dhanabalan, K.; Raziq, F.; Wang, Y.; Zhao, Y.; Mavlonov, A.; Ali, S.; Qiao, L. Perspective and Advanced Development of Lead-Carbon Battery for Inhibition of Hydrogen Evolution. Emerg. Mater. 2020, 3, 791–805. 3. Galus, Z. Carbon, Silicon, Germanium, Tin and Lead. In Standard Potentials in Aqueous Solution; Bard, A. J., Parsons, R., Jordan, J., Eds.; M. Dekker: New York and Basel, 1985. Chapter 8. 4. Galus, Z. Electrochemical Reactions in Nonaqueous and Mixed Solvents. In Advances in Electrochemical Science and Engineering; Gerischer, H., Tobias, C. W., Eds.; 4; VCH: Weinheim, 1995; pp. 217–295. 5. Prengaman, R. D. Lead-Acid Technology: A Look to Possible Future Achievements. J. Power Sources 1999, 78, 123–129. 6. Perry, D. I.; Wilkinson, T. J. Synthesis of High Purity of a- and b-PbO and Possible Applications to Synthesis and Processing of Other Lead Oxide Materials. Appl. Phys. 2007, 89, 77–80. 7. Takehara, Z. On the Reaction in the Lead-Acid Battery (as the Special Review-Article by the 2005’ Gaston Planté Medal Recipient). J. Power Sources 2006, 158, 825–830. 8. Tolstoi, V. P.; Tostobrov, E. V. Synthesis of Highly Oriented Alpha-PbO2 on the Surface of Single-Crystal Silicon and Quartz by Successive Ionic Layer Deposition. Russ. J. Appl. Chem. 2002, 75, 1529–1531. 9. Torimoto, T.; Takabayashi, S.; Mori, H.; Kubata, S. Photoelectrochemical Activities of Ultrathin Lead Sulfide Films Prepared by Electrochemical Atomic Layer Epitaxy. J. Electroanal. Chem. 2002, 522, 33–39.

Chemistry and Electrochemistry | Lithium Zawar Alam Qureshia,b, Tasneem Elmakkia,c, Jeffin James Abrahama, Hanan Abdurehman Tariqa, Buzaina Moossaa,b, Leena Al-Sulaitid, Dong Suk Hana,b,c,⁎, and Rana Abdul Shakoora,e,⁎, aCenter for Advanced Materials, Qatar University, Doha, Qatar; b Department of Chemical Engineering, College of Engineering, Qatar University, Doha, Qatar; cMaterials Science & Technology Program, College of Arts & Sciences, Qatar University, Doha, Qatar; dDepartment of Mathematics, Statistics and Physics, College of Arts and Sciences, Qatar University, Doha, Qatar; eDepartment of Mechanical and Industrial Engineering, College of Engineering, Qatar University, Doha, Qatar © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of R.J. Brodd, Chemistry, Electrochemistry, and Electrochemical Applications | Lithium, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 772–783, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00065-4.

1 Introduction 1.1 Properties 2 Electrochemistry 2.1 Lithium electrode potential 2.2 Electric double layer 2.3 Kinetic properties 3 Major applications 4 Primary batteries 4.1 Types of lithium primary batteries PBs 4.1.1 Liquid electrolyte lithium cells 4.1.2 Solid-state electrolyte lithium cells 4.1.3 Liquid cathode lithium batteries 4.1.4 Solid cathode lithium batteries 4.1.5 Other types 5 Secondary lithium batteries 5.1 Lithium cathodes 5.1.1 Layered oxides 5.1.2 Spinel oxides 5.1.3 Polyanionic cathode types 5.2 Lithium metal as anode 5.3 Other types 5.3.1 Anode-free lithium metal batteries (AFLMB) 5.3.2 Lithium-sulfur batteries 5.3.3 Lithium-air batteries 5.4 Electrolytes 5.4.1 Solid electrolyte interphase 6 Circular economy of lithium 7 Outlook Acknowledgment References

681 682 684 684 684 684 685 686 686 686 687 687 687 688 688 690 690 691 691 692 693 693 693 694 694 695 695 697 698 698

Abstract Lithium is a chemical element that heads the alkali group of metals with the chemical symbol Li and an atomic number of 3. It is a soft, lustrous, white-silverish metal that has the lowest density of all metals or solids. Elemental Lithium, like other alkali metals, exhibits extreme reactivity with water, nitrogen, and oxygen. Consequently, storing and transporting Lithium is challenging. Thus, the highly reactive and flammable metal must be stored in an inert or vacuum environment. Lithium, due to its low density, cannot be submerged in kerosene or mineral oil as it simply floats and reacts with air. Hence, it is commonly stored under a layer of petroleum jelly or paraffin wax. On exposure to air, the metal corrodes rapidly, losing its metallic sheen and forming a grayish-black tarnish, a consequence of reactions with nitrogen, oxygen, moisture, and carbon dioxide. Due to its extreme reactivity, lithium element does not occur in nature. Lithium is a rare element typically found in molten rock, salt marshes, and brines. It is believed to be non-essential to human biological processes, despite its widespread usage in medication therapies for its beneficial effects on the human brain. Lithium has the highest oxidation potential due to its tendency to lose electrons, and the electrochemical applications of Lithium and its compounds have been predominantly in primary and secondary lithium-ion batteries for energy storage. Twenty-five years have passed since lithium-ion batteries (LIBs) were commercialized in 1991. With the rapid growth of portable electronic devices, LIBs are indispensable for our comfortable living today. However, the increasing demands for high energy density impose us on developing advanced types of LIBs and so-called beyond LIBs.

⁎

Corresponding authors.

680

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00008-6

Chemistry and Electrochemistry | Lithium

681

Key points

• • • • •

Minerals/Rocks were the original sources of lithium, whereas plants yielded the majority of the other main alkali metals. Hence, the term lithium is derived from the Greek word for stone, “lithos.” A soft, silvery metal. It has the lowest density of all metals. It reacts vigorously with water. Lithium metal has the lowest electrochemical potential (3.04 V vs the hydrogen electrode), making it suitable to be utilized as a negative electrode in Lithium based batteries. The most common application for lithium is in rechargeable (secondary) batteries for portable electronics and electric vehicles. Lithium is also utilized in non-rechargeable (primary) batteries used in devices such as cardiac pacemakers, toys, and clocks. Batteries account for 71% of worldwide lithium usage. Australia, Chile, and China have the most reserves and the greatest production rates globally.

Abbreviations AEM CEM CNT DEC DMF DMSO EMC FEC LCO LFP LIB LLO LMB LMO LMS LNMO LPB MCDI NASICON NCA NMA NMC PDMS PVP SED SEI SVO THU TLL

1

Anion exchange membrane Cation exchange membrane Carbon nanotubes Diethyl carbonate Dimethylformamide Dimethyl sulfoxide Ethyl methyl carbonate Fluoroethylene carbonate Lithium cobalt oxide Lithium iron phosphate Lithium-ion batteries Lithium-rich layered oxides Lithium metal batteries Lithium manganese oxide Lithium manganese orthosilicate Lithium nickel manganese oxide Lithium primary batteries Membrane capacitive deionization Natirum Super Ionic CONductor Lithium nickel cobalt aluminum oxide Lithium nickel aluminum oxide Lithium nickel manganese cobalt oxide Poly(dimethylsiloxane) Polyvinyl pyridine Selective electrodialysis Solid electrolyte interphase Silver vanadium oxide Thiourea Transplantable LiF-rich Layer

Introduction

Lithium is an alkali metal found in tiny amounts in rocks. It does not exist in its elemental form, but it is present as a component of minerals and salts in rocks and seawater brine. The element Lithium derives its name from the Greek word lithos, which means stone. In a Swedish iron mine, Johan August Arfwedson discovered Lithium in 1817. Lithium was also found in petalite ore and minerals, including spodumene, amblygonite ores, and lepidolite. Overall, it constitutes 0.002% of the earth’s crust, while seawater contains 0.1 ppm of Lithium (ppm). Arfwedson found Lithium, but he could not separate it from mineral salts. In 1818, William Thomas Brande and Sir Humphrey Davy used electrolysis of lithium oxide to separate Lithium. Typically, Lithium is found in conjunction with aluminum, silicon, and oxygen in Lithium bearing pegmatite ore minerals such as spodumene (LiAlSi2O6) and petalite (LiAlSi4O10), and lepidolite. Before 1990, most lithium production came from mineral

682

Chemistry and Electrochemistry | Lithium

resources in the United States, but at the turn of the 21st century, Australia, Chile, and Portugal developed into the world’s top producers. Despite its vast lithium reserves, Bolivia is not a big metal producer. Lithium carbonate (Li2CO3) and Lithium and Lithium hydroxide (LiOH) are is the most common commercial Lithium product and is made in various ways from ores and brines. To break down the Lithium ores, they are heated to temperatures between 1200 and 1300 K. After this step, Lithium may be extracted using one of three techniques.1

• • •

Iron and aluminum are removed from the ore using sulfuric acid and sodium carbonate, and Lithium is extracted as lithium carbonate precipitate. For further processing, hydrochloric acid is used to create lithium chloride. Lithium hydroxide is produced by exposing the ore to limestone and then leaching it with water. Lithium chloride is created by reacting lithium hydroxide with hydrochloric acid. Lithium sulfate monohydrate is produced by adding sulfuric acid to the refined ore and then leaching it with water. First, lithium carbonate is created by reacting sodium carbonate with Lithium, and then lithium chloride is created by reacting hydrochloric acid with lithium carbonate.

The electrolysis of a fusion of lithium and potassium chlorides yields lithium metal. Graphite anodes and steel cathodes are employed in the electrolytic manufacture of Lithium. Coalescing pure Lithium at the cathode forms a molten pool shielded from air reaction by a thin coating of the electrolyte on the surface. The solidified electrolyte is poured into a mold at a temperature just over the melting point, and the Lithium is ladled out of the cell and cast. Afterward, the solidified Lithium is remelted, and the insoluble elements either float to the surface or sink to the bottom of the pot. Lithium metal’s body-centered cubic crystal structure, which can be pulled into wire and rolled into sheets, is softer than lead but harder than the other alkali metals.1 Lithium carbonate may also be obtained as lithium chloride from brines and bodies of saltwater. The greatest levels of Lithium are found in briny lakes, often called salars, represented in Fig. 1c. Bolivia, Argentina, and Chile (also known as the Lithium Triangle) are home to the world’s densest salars, where the mineral may be found in extremely high concentrations. A year or more is spent letting saltwater drain in shallow ponds. Lithium and other salts are left behind when the water evaporates. The magnesium salt is extracted using lime, and lithium carbonate is precipitated by treating the solution with sodium carbonate.1 Table 1 lists the global Lithium availability per the U.S Geological Survey 2022.

1.1

Properties

Lithium, with the chemical symbol Li, of Group 1 (Ia) within the periodic table is essentially a very soft silvery-white metal that showcases similar characteristics to other alkali metals like Sodium (Na) and Potassium (K). It has a density lower than water and is one of only three metals that can float on water, along with sodium and potassium. Lithium is a powerful heat and electrical

Fig. 1 (a) Lithium metal stored in paraffin oil. (b) Extremely finely ground lithium reacts aggressively with water. When combined with lithium, the hydrogen produced explodes into a ball of fire and sparkles. Lithium hydroxide, a highly alkaline compound, is generated when lithium metal dissolves in water. Even though the carmine red flame is typical for lithium and its salts, lithium metal’s intense burning produces a white flame. (c) In recent years, salt flats (salars), particularly those in Bolivia, have become extremely desirable due to their abundance of lithium metal. Lithium metal is key in developing the most widely used portable electronics. Courtesy: CyberChemist, 2007, Flickr. Accessed at: https://flic.kr/p/4CGcHe, Courtesy: CyberChemist, 2007, Flickr. Accessed at: https://flic.kr/p/ 3gHtW2, Courtesy: Barabara Dalmazzo, 2007, Flickr. Accessed at: https://flic.kr/p/uFJ6xs.

Chemistry and Electrochemistry | Lithium Table 1

683

Reserves for Argentina, Australia, and “other countries” based on the U.S. Mineral Commodity Summaries 2022. Productiona

United States Argentina Australia Brazil Chile China Portugal Zimbabwe Other countriesd World total (rounded)

Reservesb

2020

2021

Withheld 5900 39,700 1420 21,500 13,300 348 417 – 82,500e

Withheld 6200 55,000 1500 26,000 14,000 900 1200 – 100,000f

750,000 2,200,000 5,700,000c 95,000 9,200,000 1500,000 60,000 220,000 2,700,000 22,000,000

a

All values in Metric Tons. All values in Metric Tons. c For Australia, Joint Ore Reserves Committee or equivalent reserves were 3.8 million tons. d Other countries with reported reserves include Austria, Canada, Congo (Kinshasa), Czechia, Finland, Germany, Mali, Mexico, and Serbia. e Excludes U.S. production. f Excludes U.S. production. From U.S. Geological Survey. Mineral commodity summaries 2022: U.S. Geological Survey; 2022; p. 202; https://doi.org/10.3133/mcs2022. b

Table 2

Lithium metal properties.

Lithium Group Period Block Atomic number (Z) Electron configuration Relative atomic mass Physical properties Phase at STP Melting point (K) Boiling point (K) Density (g cm−3) Specific heat capacity (J kg−1 K−1) Atomic properties Atomic radius (A˚ ) Covalent radius (A˚ ) Electron affinity (kJ mol−1) Ionization energies (kJ mol−1) Electronegativity (Pauling scale) Other properties Crystal structure Magnetic ordering

1 2 s 3 [He] 2s1 6.94 Solid 453.65 1615 0.534 3582 1.82 1.30 59.633 1st: 520.222 2nd: 7298.150 3rd: 11815.044 0.98 Body-centered Cube (BCC) Paramagnetic

Data obtained from the CRC Handbook of Chemistry and Physics; Haynes, W. M., Ed.; CRC Press, 2014. https://doi. org/10.1201/b17118.

conductor similar to other alkali metals owing to their single valence electron. However, it’s the least reactive alkali metal due to the closeness of Lithium’s valence electron to its nucleus. Table 2 lists the properties of Lithium. Lithium readily reacts with water, albeit with less intensity than other alkali metals. The reaction results in the formation of hydrogen gas and lithium hydroxide (LiOH). Lithium compounds emit a beautiful crimson flame when ignited (Fig. 1b), but when the metal burns vigorously, the flame turns silver. Upon exposure to humidity, Lithium quickly oxidizes into a black film consisting of lithium hydroxide, lithium nitride (Li3N), and lithium carbonate.

684

Chemistry and Electrochemistry | Lithium

Due to its reactivity with oxygen, water, and even nitrogen, lithium metal is often kept in a hydrocarbon sealant; See Fig. 1a, typically petroleum jelly. Although larger alkali metals may be kept submerged in mineral oil, Lithium is too thick to immerse in these liquids completely. Its melting temperature of 453.65 K and boiling point of 1615 K is the highest of all alkali metals.1 Lithium has a higher thermal expansion coefficient than aluminum and iron. Lithium is superconductive below 400 K at normal pressure and above 9 K at >20 GPa. Lithium, like sodium, displays diffusion-less phase transition below 70 K. At 4.2 K, it possesses a rhombohedral (nine-layer) crystal structure; at higher temperatures, it becomes face- and body-centered cubic. High-pressure Lithium is allotropic. Lithium has the largest mass-specific heat capacity of all solids, 3.58 kJ kg−1 K−1, and is widely utilized in heat transfer coolants. In many ways, Lithium resembles elements of the alkaline-earth group, particularly magnesium, which has comparable atomic and ionic radii. This resemblance may be observed in the oxidation characteristics since the monoxide is generally produced in both instances. Organolithium compound reactions resemble organomagnesium compounds.1

2 2.1

Electrochemistry Lithium electrode potential

Lithium interacts spontaneously with water, hence electrode potential can’t be measured in aqueous solutions. Indirect methods must be used to determine the potential at a lithium electrode. The lithium amalgam electrode’s potential is evaluated against a known reference electrode, such as saturated calomel in an aqueous electrolyte. Voltage and reaction at a lithium electrode are described by Eq. (I). These results are consistent with those obtained by calculating the thermodynamic free energy of a lithiumlithium ionlithium-ion pair in water. Li $ Li+ + e− E0 ¼ −3:028 V

(I)

It is possible to apply the same method for electrolytes other than water measurements. The lithium potential is affected by several factors, including the solvent-electrolyte combination and the lithium ion’s degree of solvation. Table 3 provides the voltage of Lithium in common electrolyte solvents. In the same solution, the voltage of a lithium metal electrode may be measured relative to that of lithium amalgam. The potential of the amalgam electrode is then measured in an aqueous electrolyte against a standard. Polarography using a mercury electrode can detect and quantify Lithium in neutral or alkaline aqueous electrolytes, despite Lithium’s highly negative reduction potential. This is feasible because amalgam production causes a positive shift in the half-wave potential, and mercury has a very high hydrogen overpotential. Non-aqueous electrolytes, including dimethoxyethane, acetonitrile (AN), and N, N-dimethylformamide, have produced well-formed polarographic waves. The standard voltage of the lithium electrode in propylene carbonate (P.C.), dimethylformamide (DMF), and dimethylsulfoxide (DMSO) using ferrocene/ferricene as the reference electrode is shown in Table 3.

2.2

Electric double layer

The immersion of a conductor in an electrolyte causes the electrolyte’s surface ions to induce an opposing charge inside the conductor. As a result, the interface behaves like a capacitor. Compared to aqueous electrolytes, the capacitance of the double layer at a mercury amalgam electrode in dry organic electrolytes is roughly 20 mF cm−2. Similarly, in organic electrolytes, the capacitance of the electrical double layer at the surface of a lithium metal electrode is around 20 mF cm−2.

2.3

Kinetic properties

The kinetic parameters for lithium amalgams in aqueous chloride electrolytes have been calculated using faradaic rectification methods. Reported values for the transfer coefficient (a) and exchange current (i0) are 0.65 and 72 mA cm−2, respectively. When Table 3

Standard lithium potential in various solvents.

Solvent

Lithium potential (V)

Methanol Ethanol Acetonitrile Formic acid Ammonia Propylene carbonate Dimethylformamide Dimethylsulfoxide n-Methylformamide

−3.095 −3.042 −3.230 −3.480 −2.240 −2.906 −3.163 −3.237 −3.124

Chemistry and Electrochemistry | Lithium

685

working with lithium metal, kinetic measurements need to be taken swiftly because the lithium surface interacts with the electrolyte to build a protective coating on the surface. In contrast to aqueous electrolytes, lithium metal is stable enough to be employed in real-world battery applications. Because of its high capacity and voltage is employed as the negative electrode in various high-energy battery systems. A measurement of the lithium exchange current has been made in organic electrolytes. Lithium metal interacts rapidly with organic non-aqueous electrolytes to generate a thin layer that slows down the process. This barrier film is extremely reliable and allows Li+ ions to pass through it easily; therefore, this film is called solid-electrolyte-interphase (SEI). Topochemical reactions allow Lithium to intercalate into the crystal structure of certain compounds, including nonmetallic solids like graphite and transition metal oxides like nickel and cobalt oxides. Electrochemical processes allow for reversible insertion into the crystal structure. The Li-ion battery technology, which will be detailed later, is predicated on this quality. Similarly, the composition of the electrolyte does not affect the reduction of TiS2 in organic lithium electrolytes. The crystal lattice also grew by 10% due to the reduction. Van der Walls forces confine Lithium to the spaces between the S-Ti-S layers.

3

Major applications

Principal industrial uses of lithium metal are in metallurgy, where the active element is utilized as a scavenger (impurity remover) in refining iron, nickel, copper, zinc, and their alloys. Lithium absorbs many nonmetallic elements, including oxygen, hydrogen, nitrogen, carbon, Sulfur, and halogens. Lithium is exploited extensively in organic synthesis in laboratory and commercial operations. N-butyllithium (C4H9Li) is a binding reagent manufactured commercially on a huge scale. Its primary commercial application is as a polymerization initiator, for instance, in creating synthetic rubber. Lithium-magnesium and lithium-aluminum alloys, which are lighter and stronger than aluminum alone, have structural uses in aircraft and other sectors. Lithium metal is utilized to produce chemicals such as lithium hydride. Additionally, it is utilized widely in synthesizing various organic compounds, particularly medicines. Fig. 2 illustrates the variety of lithium applications in the real world. Lithium hydride (LiH), a gray crystalline solid generated by the direct combination of its constituent elements at extreme temperatures, is a readily available hydrogen source that, following water treatment, releases the gas instantaneously. It is also used to produce lithium aluminum hydride (LiAlH4), which rapidly converts aldehydes, ketones, and carboxylic esters into alcohol. Lithium is a strong reducing agent, with an electrode potential of −3.040 V vs hydrogen. It generates just one positively charged cation, Li+, and its ionization energy is rather low. Low equivalent weight and high voltage make it an attractive negative electrode for high-energy primary and rechargeable batteries and other processes requiring reduction reactions. Since the early 1990s, significant research has been conducted on high-power lithium storage batteries for electric cars and energy storage. The most effective of these involves the separation of an anode and a cathode, such as LiCoO2 (LCO), by a solvent-free conducting polymer that allows the lithium cation, Li+, to migrate. Numerous cell phones, cameras, and other electronic gadgets employ rechargeable lithium batteries of a smaller size.

Fig. 2 Global Lithium end-user markets illustrate the rising demand for portable electronics, which has resulted in a surge in the usage of rechargeable lithium-ion batteries in a wide variety of other applications, such as electric tools and electric cars, and grid storage. Ore concentrations of lithium minerals were utilized in the production of ceramics and glass. Reworked from: U.S. Geological Survey. Mineral commodity summaries 2022: U.S. Geological Survey; 2022; p. 202; https://doi. org/10.3133/mcs2022.

686

4

Chemistry and Electrochemistry | Lithium

Primary batteries

Lithium primary batteries (LPBs) are a source of power that was first introduced in the late 1960s. When transistors and integrated circuits were being developed, portability, miniaturization, and thinning were trends in electronic appliances. Long lifetimes, high volumetric energy, and reliable anti-leakage performance batteries were urgently needed to meet portability demands. The lithium fluoride carbon primary batteries were launched in 1973, and lithium manganese oxide primary batteries were established in 1976. The primary batteries are non-rechargeable, discarded after use, and known as dry cells. In contrast to the secondary lithium-ion batteries (LIB), the primary lithium-ion batteries possess 4–8 times higher energy per unit volume and mass, a wide operating temperature range, and long storage life of up to 5–10 years. Their role is inevitable in medical applications, microprocessor data storage, micropower supply equipment, and military systems. Lithium primary batteries (LPB) can provide higher operating voltage, higher energy density, and a wide temperature range than zinc-manganese and zinccarbon dry batteries. The LPBs utilize the active materials that are readily available, do not release anything harmful to the environment like heavy metals, cadmium, mercury, and lead, and hence do not harm the environment.2,3

4.1

Types of lithium primary batteries PBs

Primary lithium batteries have been developed based on various chemical compositions and construction methods. They are divided into two types based on the electrolyte used and are then classified according to the cathode type, as exhibited in Fig. 3.

4.1.1 Liquid electrolyte lithium cells 4.1.1.1 Lithium iron disulfide batteries In this kind of battery, the Li-Aluminum (Al) alloy is used as the anode, and the iron di sulfide (FeS2) is used as the cathode. LiCl-KCl binary eutectic and the ternary LiF-LiCl-LiI lithium halides are used as the electrolytes. They work at high operating temperatures of 673–773 K. The prismatic flat plate construction, damage resistance to many freeze-thaw cycles, cell protection in short circuit conditions, and ability to withstand any overcharge are advantages of these batteries. They also have the advantage of being low-cost materials with available construction techniques. A drawback associated with them is lower performance. Energizer introduced the commercial model, which provides 1.5 V in AA and AAA configurations.4 4.1.1.2 Lithium chloride batteries In these batteries, the anode is the liquid Li, a porous carbon with a gas of chlorine at high pressure that works as the cathode, and the electrolyte is molten lithium chloride. The cell has an open-circuit voltage of 3.46 V and is also capable of working at high

Primary lithium batteries

Liquid electrolyte type

Solid cathode

Solid electrolyte type

Liquid cathode

Solid cathode

Iron disulfide

Sulfur dioxide

Iodine

Carbon fluoride

Thionyl Chloride

Lead Iodide

Manganese dioxide (MnO2)

Sulfur chloride

Silver chromate

Me4Ni5-C cell Bromine Trifluoride

Copper oxide

Fig. 3 Hierarchal representation of lithium primary battery technologies. Reproduced from Julien, C.; Mauger, A.; Vijh, A.; Zaghib, K. Lithium Batteries. In Lithium Batteries; Springer International Publishing: Cham, 2016; pp. 29–68. https://doi.org/10.1007/978-3-319-19108-9_2.

Chemistry and Electrochemistry | Lithium

687

temperatures of 923.15 K. The theoretical capacity of this type of cell is 2.18 kWhkg−1. Corrosion of the various parts of the cell is a major disadvantage of this type of battery, and there is challenging to develop high-performance seals to combat this problem.4

4.1.2 Solid-state electrolyte lithium cells 4.1.2.1 Lithium iodine cells Solid-state electrolyte cells have a long shelf life of 5–10 years and can withstand broader environmental conditions of temperature, pressure, and acceleration. The Li iodine cell is famous for its well-behaved terminal voltage decay characteristics, allowing the anticipation of the end of life of these batteries, making them notable for use in pacemakers. They use Li as the anode, a thin film of LiI as the electrolyte, and the cathode is a polyphase polyvinyl pyridine (PVP). Using Br instead of the I, there are LIBr batteries developed. However, the low conductivity of the LIBr films formed limits the practical applications.4 4.1.2.2 Lead iodide cells These batteries meet long life and low open circuit voltage applications and have cathodes in various materials, including PbI2 + Pb or PbI2 + PbS + Pb. High energy density new models use TiS2 + S combination. They provide a low open circuit voltage of only around 1.9 V but have a high energy density (75–150 Whkg−1).4 4.1.2.3 Carbon tetramethyl-ammonium penta iodide batteries These batteries use the cathode used as a mixture of Me4NI5 and carbon; the electrolyte is the LiI (SiO2, H2O). With a voltage of 2.5 V, they are evaluated for use in cardiac pulse makers.4 4.1.2.4 Lithium bromine trifluoride batterytrifluoride battery A super high energy density battery has been developed using Li as an anode, the cathode is a catalyst with liquid bromine trifluoride, and the electrolyte is antimony pentafluoride. A high voltage of 5 V is possible with this battery, and a theoretical energy density value is 2680 Whkg−1. It faces the challenge of forming a byproduct layer at the anode electrolyte interface, which causes increased impedance.4

4.1.3 Liquid cathode lithium batteries 4.1.3.1 Lithium bromine trifluoride batterytrifluoride thionyl chloride battery Liquid thionyl chloride is used as the cathode electrode, Li is used as the anode, and the redox voltage available is 3.6 V with a volumetric energy density of 970 Wh/dm3. They have a low self-discharge rate, and the operating temperature range is 218–358 K.4 4.1.3.2 Lithium-sulfur dioxide battery These batteries have a cathode as the SO2 gas under pressure with a salt electrolyte. They find applications in military and aircraft use in cold weather conditions. The national aeronautics and space administration (NASA) uses them for their balloon and flight equipment. They have a high energy density of 280 Whkg−1 and an open-circuit voltage of 2.95 V and are characterized by having long shelf life and an operating temperature range of 219–344 K.4

4.1.4 Solid cathode lithium batteries 4.1.4.1 Lithium polycarbonate fluoride batteries With a general formula (CFx)n, they have a theoretical specific energy density of 2600 Whkg−1 and a voltage range of 2.8–3.3 V. They are commercially available in various forms like BR 435 cylindrical and spiral wound cylindrical cells. The energy density increases with the fluorine content x in the lithium battery at low and high discharge rates. Furthermore, the sub fluorinated compounds performed better than commercial (CFx)n.4 4.1.4.2 Lithium manganese oxide batteries This battery covers around 80% of the market share and is the most popular consumer-grade lithium primary battery. Heat-treated MnO2 is used as the cathode and LiClO4 in propylene carbonate/dimethoxy ethane as the electrolyte. LMO batteries are characterized by an energy density of 150–250 Whkg−1 and an operating temperature of 233–333 K. They are commercially available in various sizes, such as coin cells, button cells, and spiral wound cylindrical cells.4 4.1.4.3 Low-temperature lithium-ion batteries The lithium iron disulfide batteries Li//FeS2 use Li as the anode, FeS2 as the cathode, and the electrolyte is a solute including lithium iodide dissolved in an ether-containing solvent. Compared to the Zn/MnO2 alkaline A.A. cells, the commercial Li//FeS2 offers extended run time for applications requiring high power and low ambient temperatures. However, they have a challenge of limited rate capability.4 4.1.4.4 Silver vanadium oxide (SVO) cells The Ag2V4O11 as the cathode is used in Li primary batteries, and they find applications in implantable cardiac defibrillators. The Li// SVO batteries have two discharge plateaus at 3.24 V and 2.6 V. A combination of SVO and CFx as a hybrid cathode has also been developed.4

688

Chemistry and Electrochemistry | Lithium

4.1.5 Other types Various lithium oxide cells with CuO, Mn2O3, Bi2O3, or Pb3O4 as the oxides are used with Li and have a theoretical capacity of 670, 310, 350, and 310 mAhg−1 with open-circuit voltage in the range of 3.0–3.5 V and practical energy density of 500 Whdm−3. Of the various oxides, the highest value of 750 Whdm−3 is available with the Li//CuO system and is the highest value among all solid cathode-based lithium cells. The oxide-based cathodes, when compared with the sulfide-based cathodes, have the disadvantage of being less conductive and often require the addition of carbon.4

5

Secondary lithium batteries

Using insertion processes at positive electrodes lead to rechargeable energy storage technologies. Lithium-metal batteries (LMB) and Lithium-ion batteries are two typical lithium rechargeable batteries (LIB). LMB’s positive electrode is an insertion compound, while the negative electrode is pure lithium metal. Lithium metal supplies Li+ ions that intercalate into insertion compounds, which expel an electron to the external circuit to preserve electro-neutrality.4 Here, cell reversibility coincides with insertion compound stability. Fig. 4 depicts the LIB’s behavior. Instead of lithium metal, it uses a Li+ ion electrode. When electrons impact the external circuit, the electrolyte’s strong ionic conductivity permits Li + ions to travel from the cathode to the anode. The anode’s Li+ ions intercalate to the cathode through the electrolyte, creating electrons for the external circuit and powering the system. Cathodes are generally lithium-containing materials, while anodes can host Li+ ions.4 A few more variations employ Lithium batteries besides these two. Lithium-polymer batteries contain solid polymer as its electrolyte, Lithium-sulfur batteries use Sulfur as cathode electrode material with Lithium utilizing conversion reaction pathways, and Lithium-air batteries oxidize lithium metal and reduce air-oxygen to produce electrical current. All these forms of Lithium secondary batteries have benefits and limitations. LIB, the most prevalent lithium battery technology, contains three primary cathode varieties based on their crystal structure: layered oxide, spinel, and polyanionic materials. Crystal structure affects each material’s (de)intercalation behavior. More energy-dense batteries are needed to power Electric Vehicles (E.V.s), eventually replacing fossil fuel-powered vehicles. Recently, LIB technology has improved; however, next-generation applications such as grid storage and freight transportation demand highenergy-density batteries, which require novel electrode materials. Table 4 lists common cathode materials based on their electrochemical potential.

Fig. 4 Intercalation is the fundamental insertion method in Li-ion batteries, and the majority of electrode materials, both positive and negative, follow the same mechanism. Lithium-ion batteries have an electrolyte in the center and two insertion electrodes, an anode and a cathode. Because the electrolyte is an ionically conductive material, lithium ions may travel between the anode and cathode. Because a nonconductive membrane electrically divides the anode and cathode, lithium ions can travel between them. Electrons, on the other hand, are unable to flow through, preventing a short circuit. From: WP-052EN by Metrohm, p. 3, 2019. Accessed at: https://www.metrohm.com/en_us/applications/whitepaper/wp-052.html.

Table 4

Summary of the most prevalent available electrode materials and their Li + intercalation potentials, organized by crystal type.

Chemistry and Electrochemistry | Lithium

Reprinted with permission from Gao, J. (高健); Li, H. (李泓), S.-Q. S. Brief Overview of Electrochemical Potential in Lithium Ion Batteries. Chin. Phys. B. 18210.

689

690 5.1

Chemistry and Electrochemistry | Lithium Lithium cathodes

5.1.1 Layered oxides Layered oxides have the formula AxMO2, where A is Li or Na and M is a transition metal. Alkali metal ions are sandwiched between layers of MO2 slabs and edge-sharing MO6 octahedra. Mn3+ or Ni3+ as the transition metal can result in a monoclinic or orthorhombic phase in layered oxides. Current commercial LIBs use layered oxides as cathodes, and LiCoO2(LCO) is the most popular. Mobile phone and laptop batteries employ this electrode material. Li+ and Co3+ ions in alternating layers provide layered LiCoO2 exceptional structural stability.5 Due to Co-Co interactions, LCO has strong electrical conductivity and rapid Li-ion diffusivity. LCO’s 274 mAhg−1 theoretical capacity revolutionized energy storage technology. However, LCO isn’t used at higher potentials (>4.35 V vs Li/Li+) due to the instability and failure of the layered structure when most Li+ ions diffuse out from the crystal lattice.5 Hence, this lowers the practical capacity of LCO to 140 mAh g−1. In addition, the fact that Cobalt is a more expensive and scarce material raised the cost of LIBs utilizing LCO. With an increased demand for effective storage solutions driving up Co costs and LCO’s limited practical capacity, alternative transition metal layered oxides were researched as LCO substitutes. LiNiO2 was explored along with LCO. The well-ordered LiNiO2 structure is challenging to manufacture because Ni is reduced to Ni2+ during heat treatment, causing Li loss from the reaction product and Ni in the Li alkali layer, leading to irreversible capacity loss during cycling. Many considerations suggested replacing Co with Ni and Mn. Mn in LCO increases the material’s chemical stability. Ni introduces stable Mn4+ and Ni2+ oxidation states into the multilayer structure, stabilizing it during charge-discharge.5 The layered oxide with the formula LiNi1–y–zMnyCozO2 (NMC) has gained prominence recently. All three transition metals affect cathode materials differently. Mn is readily accessible and chemically stabilizes the substance when oxygen is released. Co0 s lack of cation mobility increases structural stability. Ni is structurally and chemically more stable than Mn and Co. Moreover, Ni0 s cheaper cost than Co makes replacing them economically feasible. Table 5 depicts the features of all three transition metals. Materials featuring all three transition metals, notably Ni, are explored for energy storage. High-Nickel NMC cathode materials have also been commercialized for electric vehicles. LiNi0.8Mn0.1Co0.1O2 (NMC811) layered oxide material was chosen for this because of its greater redox potential vs Li+/Li than LiFePO4 (LFP) and lithium manganese oxide (LMO) batteries. As the active redox transition element, Ni can boost cathode energy density. Commercialized viable compositions include LiNi0.5Mn0.3Co0.2O2 (NMC532) and LiNi0.6Mn0.2Co0.2O2 (NMC622). Increasing the Ni content in the cathode material reduces the battery’s cyclability due to weaker NidO bonds.6 Internal microcracks form in the cathode particle of Ni-rich layered oxides due to the H2-H3 phase transition at 4.2 V vs Li+/Li. Electrolyte penetrates particles through microcracks, damaging them when it combines with Ni4+ and producing impurity phases, raising cell impedance.6 Aluminum replacement for Mn in layered NMC provides outstanding reversible capacity for cost-efficient E.V. batteries. Al0 s advantages over Mn in certain setups make it a possible alternative. Al restricts Li/Ni mixing during (de)intercalation, whereas Mn increases Ni-alkali mixing. Al lowers battery transition metal dissolution, which Mn doesn’t.7 Adding 50 mol% Mn to Ni-rich layered oxide does not trigger phase change. A little greater Al content (>6 mol%) might influence structural stability and generate secondary phase forms. Presently, NMC and NCA Ni-rich layered oxide materials are refined and used in E.V.s like Tesla Model 3 and G.M. Bolts. However, these cathode materials lack long-range capacity and cycling performance. On the other hand, Lithium-rich layered oxides—(LLOs (also so called layered layered (LL) NMC, high energy (HE) NMC, or over lithiated NMC)) have 250–300 mAhg−1 discharge capabilities, compared to Ni-rich layered oxides (200–220 mAhg−1). LLOs have a greater energy density than other cathode materials and are considered LIB electrodes. Li1+xM1-x O2, 0 < x < 1/3 is the general formula for LLOs, they have high energy density and discharge capacity due to a reversible anionic redox pair. Oxide ions actively engage in the (de)intercalation process, where the oxidation and removal of oxides ions from the lattice induce the reduction of Mn when discharging.8 LLOs, like other layered oxides, consume less Co, reducing manufacturing costs. LLOs offer good qualities for a high-capacity electrode material, but their negative characteristics make commercialization problematic. LLOs suffer from voltage degradation. LLO material transitions to a spinel phase during long-term cycling, lowering the operating voltage and cell energy density. It’s also difficult to correctly gauge the battery’s charge. LLOs is slower than other layered oxides. Large particle size during synthesis, insulating Mn4+ ions, Solid electrolyte interphase (SEI) layer development during the cycle, and enhanced charge-transfer resistance where particles and electrolyte interact explain LLOs’ low ratelow-rate capability. LLOs have a considerable irreversible capacity loss during the first charge cycle. When oxygen loss plateaus remove Li2O, the alkali and transition metal layer in LLOs will contain Table 5

Comparison of Mn, Co, and Ni characteristics in NMC cathodes.

Parameter

Trend

Chemical stability Structural stability Electrical conductivity Abundance Environmental benignity

Mn > Ni > Co Co > Ni > Mn Co > Ni > Mn Mn > Ni > Co Mn > Ni > Co

From Manthiram, A. A Reflection on Lithium-Ion Battery Cathode Chemistry. Nat. Commun. 2020, 11(1), 1550. https://doi.org/10.1038/s41467-020-15355-0.

Chemistry and Electrochemistry | Lithium

691

Li and oxygen vacancies.8 Here, transition metal migration offsets these vacancies, preventing part of the Li-ions removed during the first charge from reinsertion, resulting in a high irreversible capacity after the first cycle. All these difficulties must be solved and refined to commercialize LLOs as next-generation LIB materials.

5.1.2 Spinel oxides As a cathode for lithium-ion batteries, spinel oxides are essentially confined to LiMn2O4 (LMO). Li and Mn exist in the tetrahedral and octahedral positions, respectively, in the anionic lattice of this spinel-like molecule. Li+ shuttles between tetrahedral sites in LMO by hopping through an intermediary octahedral site. The lithium diffusion route is three-dimensional, owing to the spinel structure and diffusion behavior. This form of robust three-dimensional shape improves Li-ion diffusivity and provides high-rate capabilities. LMO has a discharge potential of 4 V and a theoretical capacity of 148 mAhg−1. However, due to the Jahn-Teller distortion, the LiMn2O4 cathode has a substantial Mn dissolving difficulty. Adding Ni to the spinel structure of LiMn2O4 to generate LiNi0.5Mn1.5O (LNMO) increases the working voltage to 4.7 V vs Li+/Li. LNMO has garnered significantly greater interest than LMO due to its high operational potential, low synthesis cost, and Co-free nature. It has a specific reversible capacity of 148 mAhg−1.9 The molecule crystallizes in an ordered and disordered structure depending on the synthesis conditions. Due to its superior ionic and electronic conductivities, the disordered structure has a more stable cycling performance. However, the practical use of LNMO is hindered by the instability of the traditional carbonate-based electrolytes at their high working potential. Consequently, global research efforts are being undertaken to improve the cyclability of LNMO by using surface coatings, elemental doping, high-voltage electrolytes, or dimension reduction strategies.

5.1.3 Polyanionic cathode types The polyanionic cathode materials consist of polyanionic tetrahedral structures and their derivatives with strong covalent bonding with transition metal oxide. They are suited to LIBs for their high thermal stability compared to the layered transition metal oxides, safety, and long cycle life. They are of various types according to the polyanionic group involved, including the olivine, orthosilicate, tavorites, and borate. The poly anionic olivine LIFePO4 was the 1st reported in this category of cathode materials. 5.1.3.1 Olivine polyanionic Li cathode materials The olivine materials are of the form LiMPO4, where M stands for Mn, Ni, Co, and Fe. The olivines are known for high voltages; LiMnPO4, LiNiPO4, and LiCoPO4 have 4.1–4.8 V, and LiFePO4 has a voltage of 3.5 V. However, they are characterized by having lower capacities. The LiFePO4 introduced by Goodenough in 1997 is well known for its high cyclability and high cell voltage. For battery applications, the privilege of being cheap and non-toxic makes the olivine structures suitable for electric vehicles, hybrid electric vehicles, power tools, and electric bicycles. A theoretical capacity of 170 mAhg−1 is possible from this material, with the reversible electrochemical lithium insertion and extraction taking place at 3.5 V. The stability of the olivine LiFePO4 stems from the strong covalent bonding between (PO4)3− that stabilizes the oxygen in the fully charged state and prevents O2 discharge at a high state rendering them stable and safe. However, the material shows poor electronic conductivity because of the arrangement of the ions in the crystal lattice. Various methods have been introduced to improve the lower conductivity problem, including decreasing the particle size, doping, surface modification and coating, and selective doping with cations super valent to Li. The carbon-based materials used for this coating with LiFePO4 include carbon nanotube (CNT), graphene, reduced graphene oxide (rGO), nitrogen-doped carbon, and carbon black. Polypyrrole polymer was also used as a coating with the LiFePO4 to improve the rate capability and cycling stability. The polypyrrole being conductive gives this attribute to the LiFePO4.10 5.1.3.2 NASICON structured compounds The NASICON stands for NAtrium SuperIonic CONductor and is a 3D structured polyanionic compound well known for its structural stability and fast ionic conductivity. Though described in Na systems, Li-based NASICON structures are also in Fe, V, and Ti-based materials.11 They are famous for their high Li+ mobility, discharge capacities, and temporal stability. The M2(XO4)3 structure with Li has (XO4)n where (X ¼ Si4+, P5+, S6+, Mo6+) tetrahedral corner linked to octahedral site Mm+ where M is the transition metal. The open 3D nature of the structure facilitates the easy migration of Li+ ions. The octahedral site is occupied by Fe3+ ions and allows 2 Li-ion insertions per formula unit by undergoing Fe3+ to Fe2+ conversion. The Li3Fe2(PO4)3 at room temperature crystallizes in three crystallographic structures: monoclinic, rhombohedral, and orthorhombic. The rhombohedral phase crystallizes with the NASICON structure. In the V family, the Li3V2(PO4)3 is the most popular in the Ti family; it is the LiTi2(PO4)3. There are also mixed transition metal NASICON compounds that include Li3FeV(PO4)3, Li2NaFeV(PO4)3, LiNa2FeV(PO4)3, Li2FeTi(PO4)3, TiNb(PO4)3, and LiFeNb(PO4).11 5.1.3.3 Orthosilicates The orthosilicates with a general formula Li2MSiO4 (M ¼ Fe, Mn, Co, Ni) garnered research interest in 2000. The presence of the SiO−4 group theoretically allows the T.M. to change its valence from +2 to +4, thus causing two Li intercalation/deintercalation for one formula unit. They belong to the tetrahedral structures, and the Fe containing Li2FeSiO4 is the most popular orthosilicate. Compared to the LiFePO4, the covalent bonding between the SidO bonds is more stable and safer and has a high theoretical capacity of around 300 mAhg−1. At the same time, a suitable synthesized material gave a capacity of 140 mAh/g. Like the LiFePO4, various coatings have been tried on the Li2FeSiO4 using rGO and C, which showed higher discharge capacities. Another class of

692

Chemistry and Electrochemistry | Lithium

orthosilicate is the Li2MnSiO4 (LMS), which, however, displays lower cycling stability, capacity fading, and low ionic conductivity. Carbon-coated LMS shows much-improved performance with discharge capacity up to 209 mAhg−1.10 5.1.3.4 Tavorite The main disadvantages of polyanionic materials are poor electronic conductivity and low ionic diffusion. The substitution of oxygen with fluorine and other fluorine-coated materials has resulted in improved electrochemical performances and safer materials. Adding fluorine has the advantages of high voltage redox potential, protection from H.F. attack on the electrode material, stabilization of the host lattice, and easier Li+ migration. The material’s stability is from the F’s electronegativity.10 5.1.3.5 Fluorophosphates The LiFe(PO4)F was the first developed fluorophosphate. The fluorophosphate has a general formula LiM(PO4)F, where M is a transition metal with a +3 oxidation state. Phosphate tetrahedra interconnect the chain of metal octahedra for electron transport. The presence of Li tunnels in many directions allows for the easy movement of Li+ ions. The fluorophosphate LiFePO4F has ionic conductivity higher than the spinel LiFePO4 and a theoretical capacity of 152 mAhg−1 with a redox voltage of 2.75 V. Because it cannot be oxidized in LIBs, the Fe has been substituted with V resulting in the LiVPO4F. It shows a two-step redox process with voltage plateaus at 4.26 V and 4.30 V. The theoretical capacity obtained is 155 mAhg−1. Apart from Vanadium, other transition metals like Cobalt (Co) and Nickel(Ni) were also used with Li-based fluorophosphate. The LiCoPO4F gives a theoretical capacity of 120 mAhg−1 at 5.0 V. The Li2NiPO4F and the Li2CoPO4F have a theoretical capacity of around 310 mAhg−1 with a redox value of 5.3 V for the Li2NiPO4F. The LiVPO4F was analyzed with potassium doping and co-doping of Potassium (K) and Zirconium (Zr). Both have resulted in higher discharge capacities than the undoped sample, and the co-doping of K and Zr has resulted in high discharge rates at higher c rates enabling fast charging and discharging.10 5.1.3.6 Fluorosulphates They have a general formula LiMSO4F, and M stands for Fe, Mn, Co, Ni. The LiMSO4F with M ¼ Co, Ni, and Mn was electrochemically inactive till 5.0 V. This shows that the fluorophosphate redox couples of M2+/M3+lie above 5 V. The theoretical capacity of LiMSO4F was 151 mAhg−1. The (SO4)2− has a low negative charge corrected by the presence of the F− anion. The LiFeSO4 material has been analyzed with G.O. coating and shows improved discharge capacity and lower capacity fading performance.10 5.1.3.7 Borates Borates are also among the class of polyanionic materials for LIBs. The borates with a general formula LiMBO3 where M stands for Fe, Co, Mn. The Fe and Co-containing borate show the same structure, and the LiMnBO3 system has Li occupying the tetrahedral sites. The Fe-based borate has a limited theoretical capacity of 179 mAhg−1. The hexagonal LiMnBO3 has a higher charge-discharge capacity of 75.5 mAhg−1 at C/40 in the window 1.0–4.8 V. The carbon-coated h-LiMnBO3 provides improved discharge capacities. The carbon coatings are also applied to LiFeBO3.10

5.2

Lithium metal as anode

The phrases anode and cathode are typically used for the negative and positive electrodes. Sony’s first commercial LIB in 1991 employed LiCoO2 and graphite in the positive and negative electrodes. LIBs are used in portable computers, electric and hybrid cars, and smart grids to integrate wind and solar electricity. These applications demand high energy, power, and safety. Standard graphite anode (299 Whkg−1) cannot meet electric car driving range requirements (500 km). Hence, alternate electrodes that match these parameters have been explored. Li metal has the highest theoretical capacity (3860 mAhg−1 or 2061 mAhcm−3) and the lowest electrochemical potential (3.04 V vs the hydrogen electrode); it’s a great anode for Li-Sulfur and Li-Air rechargeable batteries. Unfortunately, dendrites occur when Li is used as an anode material during charging. Dendrites develop and puncture the separator, resulting in a short circuit. Anode lithium plating/stripping occurs when charging and discharging. Dendrites, lithium corrosion, dead Lithium, and volume expansion are well-known charging/discharging problems. Cell short-circuits, negative reactivity amplification, increased polarization, and volume change are frequent lithium anode concerns. Discharging the Li metal anode generates isolated Li, limiting cycle performance. These flaws cause lithium metal batteries to lose capacity and constitute a safety risk (ignition and explosion).12 Most preliminary investigations demonstrate that lithium anodes can be stabilized by establishing an external barrier and regulating the anode process, avoiding dendrite, dead Lithium, corrosion, and expansion. Creating a strong solid-electrolyteinterphase (SEI) on the anode’s surface, using a stiff solid electrolyte, or isolating Lithium in the substrate might generate an external barrier. Controlling the anode process involves altering Li+ in the electrolyte and the Li+ solvation layer. Changing electrolyte composition, SEI, and anode structure can modify a lithium anode’s stability. During battery operation, SEI forms on Lithium’s surface. Additives, solvents, and salts in the electrolyte alter in-situ SEI properties. The proper electrolyte combination can generate an SEI with a favorable molecular structure. The conventional LiPF6/EC-EMC technology is inappropriate for lithium metal anodes. Fluorine-substituted cyclic carbonates were created to improve lithium stability, potentially increasing the LiF concentration in SEI. LiF has excellent chemical stability against Lithium, protecting Lithium from corroding. Suppression of Li dendrite formation by applying the curvature enhanced accelerator coverage (CEAC) technology with sulfide-conjugated molecules may efficiently suppress dendrite formation by serving as an electrolyte additive. FEC additives are used in high-energy-density LMB to safeguard a

Chemistry and Electrochemistry | Lithium

693

Li metal anode. Because of the shorter and thicker metaphase on the Li anode, voltage polarization was dramatically decreased, and Coulombic efficiency was significantly enhanced. Dendrite development is efficiently prevented by Li-rich composite alloy sheets produced in situ on Lithium using a simple and low-cost approach. This is due to rapid lithium-ion migration through Li-rich ion conductive alloys and an electrically insulating coating ingredient. A Li metal anode benefits from SEI coatings’ chemical and electrochemical stability. Transplantable LiF-rich layer (TLL) capable of suppressing electrolyte-lithium metal side reactions. This peelable coating is formed by electrochemically decreasing NiF2 electrodes and might be utilized to safeguard Li metal anodes. The PDMS film’s nanopores enable rapid lithium-ion transport. The film is mechanically and chemically robust during electrochemical cycling to prevent Li-dendrite production, making it compatible with various electrolytes.13

5.3

Other types

5.3.1 Anode-free lithium metal batteries (AFLMB) AFLMBs are a class of Lithium metal batteries (LMBs), and the term “anode-free” originates from the cell assembly perspective. In AFLMBs, the lithium anode is absent, and only the anode current collector (copper) is utilized. Anode-free cells make the traditional lithium metal used in LMBs redundant. Also, the configuration is cost-effective due to the omittance of the processing and handling of lithium foil. Moreover, the removal of the anode reduces the volume and weight of the battery, increasing energy density from 30% to 60%.14 The concept of AFLMBs was introduced by Neubdecker’s group in 2000; this was a thin-film battery with a few micro amp hours capacity.15 However, significant improvements in anode-free cell performance were realized in 2016 by utilizing a low voltage LFP cathode with excess electrolyte in a coin-cell configuration, providing a 2 mAh capacity.16 Recently, Dahn’s group fabricated an anode-free battery utilizing a single crystalline LiNi0.5Mn0.3Co0.2O2 (NMC 532) cathode in pouch cell format and a lean amount of dual-salt electrolyte (2 g/Ah), providing a capacity of 250 mAh.17 The performance of this AMFLB configuration provided 125 times more capacity while using 37% times less electrolyte compared to the 2016 report, marking a significant performance increase and highlighting fresh new perspectives on AFMBL development. On cycling an AFLMB, the formation of the lithium anode and the respective SEI layer occurs during the charging process (insitu). In comparison, anode dissolution and re-intercalation of lithium ions within the cathode structure occurs on discharge. Consequently, the battery behaves like a lithium metal battery after the first charge. Hence, similar drawbacks exist for both LMBs and AFLMBs. The limitations of AFLMB stem from the electrochemical plating process of lithium metal on the current collector. The deposition of lithium over the current collector requires high activation energy; the high overpotential needed for nucleation causes inhomogeneous deposition of lithium over the copper current collector. On cycling, this uneven deposition leads to dendrite formation. Further side reactions with the electrolyte can accelerate the growth of dendrites, reducing the availability of active lithium within the cell. Furthermore, the reaction of the metal with the electrolyte also forms the SEI layer. The mechanical strength, structure, and composition of this film, if not optimum, can lead to rupture, causing the repeated formation of the SEI interface. This reformation of the defective SEI layer, on cycling, can cause rapid consumption of the electrolyte and lithium, inducing inhomogeneous electric flux and exacerbating dendrite formation. The dendrites can convert into electrochemically inactive lithium or “dead lithium” upon the formation of a new SEI interface. Moreover, the dendrite growth can render the deposited anode porous, causing massive volume changes during cycling. These phenomena can severely degrade the cell cycle life and electrochemical performance. The lithium present in the anode-free battery is solely donated through the cathode. Hence, utilizing a Lithium rich cathode that can easily facilitate the intercalation/dissolution of lithium ions can be beneficial for such a battery configuration.

5.3.2 Lithium-sulfur batteries The Lithium-sulfur (Li-S) battery technology was first investigated in 1960 as a secondary battery with a very high energy density. This was further solidified mainly due to low sulfur cost decreasing the battery’s overall production cost. With both Li metal and Sulfur showing very high theoretical specific capacity values, combining them as anode and cathode respectively displays a conversion reaction behavior providing an average cell voltage of 2.15 V will result in a Li-S battery with a theoretical energy density of 2500 WhKg−1 overtaking many of the existing Li-ion battery technology. The Li-S battery technology utilizes a conversion reaction behavior where the lithium metal gets oxidized at the anode during discharge. This causes the movement of Li ions through the Lithium bis(trifluoromethanesulfonyl)imide (LTFSI) electrolyte to the sulfur cathode. Here, the electrons move through the external circuit as well. When the Li-ions and the electrons arrive and are accepted at the cathode side, the Sulfur is reduced to form lithium sulfide (Li2S). The exact reaction process involves multiple steps during discharge. Initially, the sulfur cyclo S8 gets lithiated to Li2S8, then Li2S6, and finally Li2S4. These products are soluble in the organic electrolyte used in this battery. At this stage, theoretically, 25% of capacity should be contributed from Sulfur which translates to 418 mAhg−1 capacity and an average potential of 2.3 V. With additional lithiation, short chain sulfides Li2S2 and Li2S form from the previously soluble Li2S4 and then reprecipitates back to the electrode. At this stage, theoretically, the remaining 75% of capacity should be contributed from Sulfur which translates to 1254 mAhg−1 and an average voltage of about 2.1 V. During the charging process, the reaction reverses. Lithium ions are released from Li2S to the electrolyte, and the intermediate lithium polysulfides described previously are reformed. Further release of lithium ions results in the reformation of the original sulfur cyclo S8, thus completing the reversible cycle.18 Li-S battery commercialization is uncertain owing to issues that limit viability. The sulfur cathode and discharged lithium sulfide have poor conductivity and insulating behavior, causing low reaction kinetics and specific capacity. Intermediate lithium polysulfides are dissolved in electrolytes for shuttling between the cathode and anode. This can reduce capacity and coulombic efficiency. Additionally, during sulfur lithiation to lithium sulfide, these batteries expand significantly. The electrode can destabilize

694

Chemistry and Electrochemistry | Lithium

with repeated cycling, resulting in low cyclability. Challenges on the lithium anode side include intermediate lithium polysulfides reacting with the lithium metal, causing side reactions that lower cyclic performance. Moreover, dendrite formation on the lithium metal damages both the electrode and electrolyte and causes internal short circuits, potential safety concerns, and volume change in the lithium metal anode, similar to the sulfur cathode causing loss of Lithium. In recent years, many strategies have been explored to improve cyclic performance, such as sulfur encapsulation, lithium anode protection, electrolyte, and separator modification. Li-S batteries might be used for future energy storage if these obstacles are overcome. Electrolyte amount affects lithium-ion battery energy density and capacity. Too little electrolyte diminishes capacity and longevity, whereas too much reduces energy density. A minimum amount of electrolyte related to pore volume must be established for optimum cell component wetting. The conventional lithium-sulfur cell functions with a liquid electrolyte in which sulfur-reduction products are soluble. Due to the complicated behavior of polysulfides, this technology has a high initial capacity but poor discharge retention and charging efficiency. Adding nitrate to the electrolyte composition can boost performance by decoupling the polysulfide shuttle from the lithium anode. Over numerous charge-discharge cycles, both electrodes consume nitrate, removing its impact. The lithium SEI’s dynamic exchange with solution makes shielding it from redox-active species difficult. The best path forward may be to switch from lithium to silicon or another high-energy material. The Li4.4Si|Si electrode’s somewhat less negative potential may reduce anode-TFSI interaction. This allows for the development of an SEI better suited to solution-based polysulfide chemistry. The modification of lithium-sulfur battery electrolytes focuses on lowering polysulfide solubility. Suo et al. recently showed arguably the most impressive improvement based only on electrolyte change. High concentrations of lithium salt restrict polysulfide solubility and prevent discharge capacity decline (compared with standard liquid electrolyte cells). Limited power rate, low sulfur usage, and unstable Li metal anode hinder high-energy-density Li-S batteries. Due to its improved wettability toward the electrode, 0.5 M LiTFSI showed better cycle stability than 1.0 M LiTFSI with high sulfur loading. Low-concentration electrolytes could increase the Li–electrolyte interface stability due to a greater organic component in the solid electrolyte interface (SEI) due to more solvent in the SEI’s development. Flexible organic components could better accommodate Li metal anode volume variations. Low-concentration electrolytes may be better for high-energy Li-S batteries.19

5.3.3 Lithium-air batteries Lithium-air (Li-air) batteries are a derivative of lithium batteries known for their very high theoretical specific energy. Compared to the existing Li-ion battery technology, the Li-air battery has 5–10 times greater theoretical energy density. The current Li-air battery has an estimated energy density value of 3500 WhKg−1 when considering the formation of Li2O2. This value is inflated mainly due to ignoring the cathode, electrolyte, and other cell components and focusing mainly on the lithium anode electrode and the inexhaustible O2 supply that supports the battery.20 The Li-Air battery usually consists of a porous cathode that can facilitate O2 gas, an electrolyte, and a Lithium metal anode. Depending on the electrolyte, there are aqueous, non-aqueous, hybrid, and solid-state Li-air batteries. The non-aqueous organic electrolyte Li-air batteries are the most common type encountered. The working principle of the Li-air battery follows a simple redox process. When the battery is discharged, the Li ions from the anode diffuse to the cathode and combine with the O2 molecules resulting in Li2O2. The Li2O2 is deposited on the cathode surface because of this process. On charging the battery, the reaction reverses, resulting in the decomposition of Li2O2 into the Li ions and O2 molecule, where the Li-ion will diffuse back to the anode, and the O2 molecule gets released.20 Li-air batteries have a high theoretical energy density but poor actual capacity and energy density. Li-air batteries have cathode restrictions. The porous cathode, which should allow oxygen diffusion and store discharged Li2O2, is blocked by the product, restricting oxygen flow and reducing discharge capabilities.21 Li-air batteries have a large charge-discharge voltage gap (>1 V). Catalysts improve the Li-air electrode battery kinetics and efficiency to limit overpotential; unreliable catalytic activity causes battery overpotential and a significant charge-discharge voltage gap. In Li-air batteries, atmospheric moisture causes unstable cathodes. The Li-air battery electrolyte is also crucial. During charge-discharge cycles, most organic carbonate electrodes decompose. Lithium oxide precipitation and O2 dissolution block the cathode, reducing the battery’s capacity. Aqueous electrolytes react strongly with Li metal anodes. Solid electrolytes were also advised; however, they had less ionic conductivity than liquids.22 Li-air batteries aren’t commercialized yet. It’s only possible if these diverse combinations are optimized.

5.4

Electrolytes

The electrolyte is essential for transferring positive lithium ions between the cathode and anode. Current collectors and polymer separators have developed substantially smaller over the years to enhance the volume fraction occupied by active electrode materials while lowering costs. Higher loadings may also be obtained by increasing active layer thicknesses, reducing the binder percentage, and decreasing porosity. These characteristics demand augmented electrolyte (ionic) transport to sustain rate capability. The literature on lithium battery electrolytes is pervasive, including thousands of publications and reviews. Like any other electrochemical system, a lithium-ion battery has two electrodes (anode and cathode) and an electrolyte: it is evident that the electrode materials constrain the electrolyte composition. In other words, the chemistry of the two electrodes in the battery dictates the appropriate composition of the electrolyte. However, in theory, one could describe an “ideal” electrolyte as having the following characteristics23: (i) incombustible, (ii) environment-friendly, (iii) wider voltage stability window, (iv) non-toxic, (v) sufficient availability, (vi) inert toward the battery components (vii) resistant to different forms of abuse, such as electrical, mechanical, and thermal and (viii) good wetting properties. Much research on lithium batteries indicates that a

Chemistry and Electrochemistry | Lithium

695

“perfect” electrolyte does not exist. Electrolytes are made from aprotic organic liquids like ethylene carbonate or dimethyl carbonate, which have a high dielectric constant and thus make appropriate salt solvents; they also have a wide electrochemical stability window. However, because of their high vapor pressure, compounds can cause fires and explosions in the event of accidental battery faults, primarily when low-stability cathodes like oxides are utilized. Such safety concerns are amplified in larger lithium-ion batteries used in electric automobiles, especially when high charge-discharge rates. The most widely used electrolyte is an organic solution of lithium salts such as lithium hexafluorophosphate (LiPF6) and lithium tetrafluoroborate (LiBF4). Today’s most common commercial electrolyte is a carbonate solvent solution combined with LiPF6 salt.24 LiPF6 is a conductive salt often employed in organic carbonate-based electrolytes in lithium-ion batteries and is regarded as the foundation chemistry for electromobility. While the LiPF6-based electrolyte system has good electrical conductivity and can produce a stable solid electrolyte interface (SEI), it is moisture sensitive and thermodynamically unpredictable. Despite its relatively strong thermal stability in a dry inert atmosphere up to 380 K, this salt degrades when exposed to traces of water, moisture, or alcohol, triggering parasitic reactions and electrolyte decomposition. The solvent molecules ethylene carbonate (E.C.) and diethyl carbonate (DEC) further change the equilibrium state, increasing H.F. formation. The electrolyte system based on LiBF4 has significant promise but is not frequently used due to its hydrolysis susceptibility, relatively poor conductivity, and impeded SEI formation. However, compared to the present LiPF6 system, LiBF4 offers many benefits. Although LiBF4 has low conductivity at low temperatures than LiPF6, it performs better because it has a lower electric charge transfer (Rct) than LiPF6. LIBF4 is less moisture sensitive at room temperature and has equivalent cycling characteristics to LiPF6. In contrast, the use of LiBF4 in high-temperature conditions is problematic. Despite having superior thermal stability to LiPF6, allowing it to be used in high-temperature situations, LiBF4 has poor film-forming capability. The rise in temperature of the LiBF4-based electrolyte system produces a quick fall in LIB capacity, resulting in a significant loss in the battery’s coulombic efficiency. Thus, a few fundamental alterations must be performed for LiBF4 to be used for wide-temperature electrolytes. First, the film-forming ability of LiBF4 must be improved, and second, the conductivity of the electrolyte must be raised to improve performance at both high and low temperatures. Transport properties and molecular-scale structures of novel solution chemistries (e.g., new solvent systems, highly concentrated salts) are known. Basic testing and calculations of innovative electrolyte compositions will give new materials and insights into their properties. The structure and stability of the SEI in different solutions and conditions (temperature, voltage) must also be characterized. Such insights will help create optimum additives/coatings for electrolytes and cell longevity. These systems still need intensive benchmarking and lifetime studies. Before large-scale usage, their cost and handling safety must be proved, the latter being an important yet understudied topic.

5.4.1 Solid electrolyte interphase No solvent will remain unaffected by the reduction process when exposed to lithium metal. The solid electrolyte interphase (SEI) layer is created when Lithium interacts directly with the solvents. To begin with, E. Peled postulated that the SEI layer is formed due to a direct interaction between the Lithium and the solvent. To stop the Lithium from reacting further with the solvent, the SEI on the negative electrode produces a durable protective layer. Li+ ions are the only ones that can go through the film since the film has a Li+ transference number of 1. Commercial lithium primary and rechargeable cells and batteries cannot function without the SEI layer forming on their surfaces. The SEI is made from the byproducts of the electrolyte’s reduction with lithium metal. Solvent effects make it difficult to pin down the film’s precise chemical makeup. Reductive polymerization or solvent reduction of electrolyte components are hypothesized to be its main constituents. Solvents sometimes include trace quantities of water, which might influence the final products. Once the SEI has formed, it stops the electrolyte from reacting directly with the lithium metal. It is responsible for primary batteries’ extended storage life, security, and high discharge efficiency. Primary lithium batteries typically have a 10-year lifespan. First cycle loss is a phrase used to characterize the creation, cycle life, shape of the deposit, and faradaic efficiency of the SEI layer in rechargeable batteries. This layer forms during the first charge. There is no electrical conductivity in the film; only Li+ is transferred through the SEI layer, and the film has high adhesion (wetting) to the negative lithium electrode. As for the negative electrodes, both lithium metal and lithium-intercalated carbon create stable SEI films that do not increase after formation.

6

Circular economy of lithium

With the increasing global demand for Lithium and lithium-based energy products, the use of the conventional linear economy would not be applicable, as it places the global supply chain of this high-value material at potential risk. This is why it is necessary to develop an effective, long-lasting circular economy of Lithium that can (1) effectively preserve the amount of Lithium recovered and accumulated, (2) decrease the risk of natural lithium reserves depletion, (3) extend the lifecycle of lithium-based products and offer disposition alternatives, and (4) eliminate the wastes and possible pollution associated with the end-of-life of lithium-based products.25 Fig. 5 presents a schematic diagram showcasing an all-inclusive circular economy of Lithium, emphasizing the LIB industry. The cycle starts with the raw liquid and solid elemental lithium resources, primarily including different types of brines, seawater, wastewater, and mineral ores. Most lithium resources are in the liquid phase, covering more than 85% of extractable Lithium (around 1.6 M tons).26

696

Chemistry and Electrochemistry | Lithium

Fig. 5 Schematic diagram illustrating the different stages of lithium recovery and use, with an emphasis on lithium-ion battery use and recycle, highlighting its role in forming a long-lasting circular economy of lithium.

Among them, high salinity brine comprises the highest lithium concentration, with geothermal brines presenting deep underground containing Li concentrations of 10–20 mg L−1, whereas salt lake brines, which are surface-bound, possess much higher Li concentrations of 100–1000 mg L−1. As for industrial brines, they mainly originate from seawater desalination processes. For instance, The reverse osmosis (R.O.) process releases up to 70 million cubic meters of brine daily. The released brine has vast amounts of recoverable Lithium, which, if utilized effectively, can increase the capital value of brine and simultaneously decrease the unadvisable disposal of such a highly saline stream (brines contain some dangerous pre-treatment chemicals, organics, and heavy metals) to the environment. Seawater contains 230 billion tons of Li, but its Li concentration (0.17 mg L−1) is considered too low for direct Li recovery. A similar low concentration with a high amount of Lithium persists in wastewater, where large amounts of lithium-containing wastewater are generated during the production of various lithium compounds, LIB cathode materials, and in the recovery process of spent LIBs. The solid lithium reserves contain around 0.11 M tons of Lithium, with an estimated extraction cost of US$ 6–8/kg.26 The comparably higher extraction cost of Li from solid ores compared to brine (US$ 2–3/kg) encourages its removal from the aforementioned liquid resources. However, extracting the hard-rock lithium content is still necessary to meet the escalating market demand for Lithium. Mineral ores like spodumene, tainiolite, petalite, and hectorite undergo a series of hard rock crushing, sequential heating/cooling cycles, powdering, acid treatments, and filtration systems followed by soda ash addition to effectively yield high purity lithium in the form of lithium carbonate. On the other hand, lithium extraction from liquid-based resources is quite vast and involves multiple

Chemistry and Electrochemistry | Lithium

697

techniques that can effectively and selectively recover Lithium. Solar evaporation, a solar-induced concentration process, is leading the commercial Li production worldwide, with around 75–85% of Li from Salt Lake brine recovered. A series of sequential processes that take around 12–18 months must be performed to recover Lithium with high selectivity and purity via solar evaporation. It includes a prolonged and climate-dependent solar evaporation rate, followed by lime-soda addition, precipitation, and high-volume post-treatment processes. Therefore, fast-smart techniques to recover Lithium need to meet the massive demand for Li extraction. As can be seen in Fig. 5, Li can also be recovered through chemical precipitation or solvent extraction. These processes involve the addition of particular salts or chelating reagents that can either precipitate Lithium in the form of lithium salts or selectively absorb Li+ ions. It is important to note, however, that chemical precipitation and solvent extraction produce large volumes of sludge to dispose of and are thus not environmentally friendly. Membrane-based lithium recovery techniques involve specialized membranes that allow the selective passage of Lithium through its porous channels, like in the case of Nanofiltration (N.F.) or the selective adsorption of Li through an intercalation mechanism, like in the case of Ion-imprinted-sieve-membranes. In terms of sustainability, membrane-based processes are promising due to their low footprint and operating costs; however, they often suffer from membrane fouling. Moving to electrochemical processes, electrochemical Lithium capturing systems use applied electrical current as a deriving force for lithium capture. These processes utilize Li intercalating electrodes that operate battery-like to charge and discharge Li effectively. Although not technically commercialized yet, these electrochemical systems demonstrate high performance in Li recovery selectivity and tunability.26 Moreover, hybrid technologies such as membrane-capacitive deionization (MCDI) and selective electrodialysis (SED) utilize electrical current as a driving force for ions movement and use different types of membranes to separate Lithium from the other competing ions. Here, the key difference between those two hybrid technologies is that: MCDI involves the use of Lithium intercalating electrodes to capture Lithium selectively, and the anion and cation exchange membranes (AEM and CEM) are used only; to facilitate the movement of the ions and trap the captured ions. On the other hand, in SED, the series of alternating membranes utilized work on concentrating Lithium in the brine, while the electrodes are only responsible for ions movement. After Lithium is captured and produced in the form of lithium metal or other lithium salts, it can be used in various manufacturing processes, including polymers, ceramics, and lubricants. That being said, around 65% of the extracted Lithium is extensively used in the energy storage industry. Approximately 88% of Lithium allocated for this industry is used to fabricate, design, and produce LIBs. Since these LIBs possess: (1) both easy and fast charging abilities making them great candidates as renewable energy sources, (2) high electrochemical performance and voltages to be abundantly used in the aerospace industry and power transmission, and (3) high energy densities, and are light in weight making the perfect candidates for consumer electronics and electric vehicles production. Consequently, the escalating demand for LIBs puts the lithium supply chain at risk of potential resource depletion. The critical LIB value chain focuses on closing the loop on LIB waste through the sustainable and economic recycling of spent LIBs. Where recycling spent LIBs is one of the most effective ways to ensure a sustainable value material circularity and address the issue of resource efficiency. Currently, no technologies can fully recover all the elements from used LIBs. However, over the years, some effective procedures were implemented to recover most lithium content. To prepare the spent LIBs for a recycling process that is more efficient and less energy-intensive, the spent LIBs first undergo a pre-treatment process. During the pre-treatment process, various components and elements of spent LIBs are segregated, classified, and separated. Then, the recovered spent LIB active materials undergo either a pyrometallurgical or hydrometallurgical treatment or a combination of both. Pyrometallurgy mainly involves melting the separated cathode material and reacting them with carbon to reduce them to metal, slag, and carbon dioxide. In contrast, hydrometallurgy is essentially a chemical leaching process followed by extraction.27 Although pyrometallurgy is a simple, well-tested process, it involves multiple energy-intensive and toxins-emitting procedures that predominantly have low elemental recovery rates. On the contrary, the hydrometallurgical methods require less capital and energy and can recover Lithium, but they also require large quantities of potentially harmful chemicals. Another evolving recycling method that can shorten the recycling loop and potentially eliminate the necessity to remanufacture the cathode structure is direct recycling. In this process, electrodes are lithiated rather than dissolved or smelted to recover the active elements. In short, it is crucial to develop an effective, durable, and long-lasting circular economy of Lithium that offers reliable deposition alternatives and ensures a lifelong lithium supply.

7

Outlook

Lithium supply security has been a significant worry for U.S. and Asian technology corporations. Continued strategic partnerships and joint ventures between technological businesses and exploration companies will ensure a dependable, diverse supply of Lithium for battery providers and vehicle makers, the main consumers of lithium resources. In Argentina, Bolivia, Chile, China, and the United States, brine-based lithium sources are in varying stages of development. In contrast, mineral-based lithium sources are being exploited in Australia, Austria, Brazil, Canada, China, Congo, Czechia, Finland, Germany, Mali, Namibia, Peru, Portugal, Serbia, Spain, and Zimbabwe. Despite the restricted energy density defined by the number of crystallographic sites accessible as well as structural and chemical instabilities in a deep charge state, existing lithium-ion technology based on insertion-reaction cathodes and anodes will continue for the foreseeable future. Although conversion-reaction anodes and cathodes provide orders of magnitude larger capacities than insertion-reaction electrodes, their practicality is fraught with difficulties. Recently, there has been renewed interest in using lithium metal as an anode and substituting liquid electrolytes with solid electrolytes because they can provide safer cells with greater

698

Chemistry and Electrochemistry | Lithium

working voltages and charge-storage capacity. However, only time will tell if they are practicable. Due to its high cost, lithium-ion batteries won’t be cost-competitive until at least 2027. The highest energy-to-mass densities required for the E.V. industry are displayed most obviously by lithium-ion NMC and NCA batteries. However, LFP, less susceptible to thermal runaway, is beginning to gain ground in electric vehicles, and solid-state lithium-ion will ultimately replace it. The energy storage sector is about to take off globally. By 2050, the world will have installed 12,147 GWh of storage energy, giving 3715 GW of power capacity worth of battery energy storage systems for storing energy for periods of less than 4 h for use by the grid and by individual households. By 2030, about 30% of this capacity would be expected to originate from non-traditional energy storage technologies, with lithium-ion batteries accounting for the vast majority. Given the difficulties encountered with the alternatives (conversion-reaction electrodes, lithium metal, and solid electrolytes), a viable near-term strategy would be to focus on high-nickel layered oxide cathodes, liquid electrolytes compatible with and capable of forming stable SEI on both graphite anode and high-Ni cathodes. Attention should also be given to cell engineering innovations to fabricate thicker electrodes, reduce inactive components, and realize safer, long-life, and affordable novel system integration. Also, other alkali metal battery systems such as Na-ion batteries could be an alternative.

Acknowledgment This publication was made possible by NPRP grants # [NPRP11S-1225-170128 & NPRP12S-0227-190166] and GSRA grant # [GSRA8-L-2-0414-21012] from the Qatar National Research Fund (a member of Qatar Foundation). The work was also supported via Qatar University High Potential Projects # [QP-H3P-CAM-2021-449]. The findings achieved are solely the responsibility of the authors.

References 1. Haynes, W. M., Ed. CRC Handbook of Chemistry and Physics; CRC Press, 2014. https://doi.org/10.1201/b17118. 2. Liu, Y.; Su, M.; Gu, Z.; Zhang, K.; Wang, X.; Du, M.; Guo, J.; Wu, X. Advanced Lithium Primary Batteries: Key Materials, Research Progresses and Challenges. Chem. Rec. 2022. https://doi.org/10.1002/tcr.202200081. 3. Sun, L.; Peng, C.; Kong, L.; Li, Y.; Feng, W. Interface-Structure-Modulated CuF2/CFx Composites for High-Performance Lithium Primary Batteries. Energy Environ. Mater. 2022. https://doi.org/10.1002/eem2.12323. 4. Julien, C.; Mauger, A.; Vijh, A.; Zaghib, K. Lithium Batteries. In Lithium Batteries; Springer International Publishing: Cham, 2016; pp. 29–68. https://doi.org/10.1007/978-3319-19108-9_2. 5. Manthiram, A.; Goodenough, J. B. Layered Lithium Cobalt Oxide Cathodes. Nat. Energy 2021, 6 (3), 323. https://doi.org/10.1038/s41560-020-00764-8. 6. Julien, C. M.; Mauger, A. NCA, NCM811, and the Route to Ni-Richer Lithium-Ion Batteries. Energies (Basel) 2020, 13 (23), 6363. https://doi.org/10.3390/en13236363. 7. Li, W.; Lee, S.; Manthiram, A. High-Nickel NMA: A Cobalt-Free Alternative to NMC and NCA Cathodes for Lithium-Ion Batteries. Adv. Mater. 2020. https://doi.org/10.1002/ adma.202002718. 8. Gent, W. E.; Lim, K.; Liang, Y.; Li, Q.; Barnes, T.; Ahn, S. J.; Stone, K. H.; McIntire, M.; Hong, J.; Song, J. H.; Li, Y.; Mehta, A.; Ermon, S.; Tyliszczak, T.; Kilcoyne, D.; Vine, D.; Park, J. H.; Doo, S. K.; Toney, M. F.; Yang, W.; Prendergast, D.; Chueh, W. C. Coupling between Oxygen Redox and Cation Migration Explains Unusual Electrochemistry in Lithium-Rich Layered Oxides. Nat. Commun. 2017, 8 (1). https://doi.org/10.1038/s41467-017-02041-x. 9. Qureshi, Z. A.; Tariq, H. A.; Shakoor, R. A.; Kahraman, R.; AlQaradawi, S. Impact of Coatings on the Electrochemical Performance of LiNi0.5Mn1.5O4 Cathode Materials: A Focused Review. Ceram. Int. 2021. https://doi.org/10.1016/J.CERAMINT.2021.12.118. 10. Jyoti, J.; Singh, B. P.; Tripathi, S. K. Recent Advancements in Development of Different Cathode Materials for Rechargeable Lithium Ion Batteries. J. Energy Storage 2021, 43, 103112. https://doi.org/10.1016/j.est.2021.103112. 11. Jian, Z.; Hu, Y.; Ji, X.; Chen, W. NASICON-Structured Materials for Energy Storage. Adv. Mater. 2017, 29 (20), 1601925. https://doi.org/10.1002/adma.201601925. 12. Heubner, C.; Maletti, S.; Auer, H.; Hüttl, J.; Voigt, K.; Lohrberg, O.; Nikolowski, K.; Partsch, M.; Michaelis, A. From Lithium-Metal toward Anode-Free Solid-State Batteries: Current Developments, Issues, and Challenges. Adv. Funct. Mater. 2021. https://doi.org/10.1002/adfm.202106608. 13. Zhu, B.; Jin, Y.; Hu, X.; Zheng, Q.; Zhang, S.; Wang, Q.; Zhu, J. Poly(Dimethylsiloxane) Thin Film as a Stable Interfacial Layer for High-Performance Lithium-Metal Battery Anodes. Adv. Mater. 2017, 29 (2). https://doi.org/10.1002/adma.201603755. 14. Schmuch, R.; Wagner, R.; Hörpel, G.; Placke, T.; Winter, M. Performance and Cost of Materials for Lithium-Based Rechargeable Automotive Batteries. Nature Energy 2018, 1, 267–278. https://doi.org/10.1038/s41560-018-0107-2. Nature Publishing Group April. 15. Neudecker, B. J.; Dudney, N. J.; Bates, J. B. Lithium-Free Thin-Film Battery With In Situ Plated Li Anode. J. Electrochem. Soc. 2000, 147 (2), 517. https://doi.org/ 10.1149/1.1393226. 16. Qian, J.; Adams, B. D.; Zheng, J.; Xu, W.; Henderson, W. A.; Wang, J.; Bowden, M. E.; Xu, S.; Hu, J.; Zhang, J.-G. Anode-Free Rechargeable Lithium Metal Batteries. Adv. Funct. Mater. 2016, 26 (39), 7094–7102. https://doi.org/10.1002/adfm.201602353. 17. Genovese, M.; Louli, A. J.; Weber, R.; Martin, C.; Taskovic, T.; Dahn, J. R. Hot Formation for Improved Low Temperature Cycling of Anode-Free Lithium Metal Batteries. J. Electrochem. Soc. 2019, 166 (14). https://doi.org/10.1149/2.0661914jes. 18. Chen, Y.; Wang, T.; Tian, H.; Su, D.; Zhang, Q.; Wang, G. Advances in Lithium–Sulfur Batteries: From Academic Research to Commercial Viability. Adv. Mater. 2021, 33 (29), 1–67. https://doi.org/10.1002/adma.202003666. 19. Barghamadi, M.; Best, A. S.; Bhatt, A. I.; Hollenkamp, A. F.; Musameh, M.; Rees, R. J.; Uther, T. R. Lithium-Sulfur Batteries-the Solution Is in the Electrolyte, but Is the Electrolyte a Solution?; 2014. https://doi.org/10.1039/c4ee02192d. 20. Chen, K.; Yang, D. Y.; Huang, G.; Zhang, X. B. Lithium-Air Batteries: Air-Electrochemistry and Anode Stabilization. Acc. Chem. Res. 2021, 54 (3), 632–641. https://doi.org/ 10.1021/acs.accounts.0c00772. 21. Geng, D.; Ding, N.; Hor, T. S. A.; Chien, S. W.; Liu, Z.; Wuu, D.; Sun, X.; Zong, Y. From Lithium-Oxygen to Lithium-Air Batteries: Challenges and Opportunities. Adv. Energy Mater. 2016, 6 (9), 1–14. https://doi.org/10.1002/aenm.201502164. 22. Rahman, M. A.; Wang, X.; Wen, C. A Review of High Energy Density Lithium-Air Battery Technology. J. Appl. Electrochem. 2014, 44 (1), 5–22. https://doi.org/10.1007/s10800013-0620-8. 23. Xu, G.; Shangguan, X.; Dong, S.; Zhou, X.; Cui, G. Key Scientific Issues in Formulating Blended Lithium Salts Electrolyte for Lithium Batteries. Angew. Chem. Int. Ed. 2020, 59 (9).

Chemistry and Electrochemistry | Lithium

699

24. Simon, P.; Gogotsi, Y. Materials for Electrochemical Capacitors. Nat. Mater. 2008. https://doi.org/10.1038/nmat2297. 25. Islam, M. T.; Iyer-Raniga, U. Lithium-Ion Battery Recycling in the Circular Economy: A Review. Recycling 2022, 7, 33. https://doi.org/10.3390/RECYCLING7030033. 26. Zavahir, S.; Elmakki, T.; Gulied, M.; Ahmad, Z.; Al-Sulaiti, L.; Shon, H. K.; Chen, Y.; Park, H.; Batchelor, B.; Han, D. S. A Review on Lithium Recovery Using Electrochemical Capturing Systems. Desalination 2021, 500, 114883. https://doi.org/10.1016/J.DESAL.2020.114883. 27. Zhou, L. F.; Yang, D.; Du, T.; Gong, H.; Luo, W. B. The Current Process for the Recycling of Spent Lithium Ion Batteries. Front. Chem. 2020, 8, 1027. https://doi.org/10.3389/ FCHEM.2020.578044/BIBTEX.

Chemistry and Electrochemistry | Magnesium Zhenyou Lia,c,∗, Liping Wanga, Sebastián Pinto Bautistaa,b, and Marcel Weila,b, aHelmholtz Institute Ulm for Electrochemical Energy Storage (HIU), Ulm, Germany; bInstitute for Technology Assessment and Systems Analysis (ITAS), Karlsruhe Institute of Technology, Karlsruhe, Germany; cQingdao Institute of Bioenergy and Bioprocess Technology, Chinese Academy of Sciences, Qingdao, Shandong, China © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 Introduction 2 Applied chemistry of magnesium 2.1 Magnesium compounds as refractory material 2.2 Desulfurization of steel 2.3 Organomagnesium compounds 3 Magnesium production and the market 3.1 Production methods 3.2 Market and cost development 3.3 Resource and environmental impacts 3.3.1 Resource criticality 3.3.2 Environmental considerations of magnesium metal 3.3.3 Recycling 4 Electrochemistry and electrochemical applications 4.1 Electrochemistry of magnesium 4.2 Primary magnesium batteries 4.2.1 Water-activated magnesium batteries 4.2.2 Magnesium-air batteries 4.3 Rechargeable magnesium batteries 4.3.1 Electrolytes for rechargeable magnesium batteries 4.3.2 Alternative anode materials 4.3.3 Cathode materials 4.4 Environmental considerations of rechargeable magnesium batteries 4.4.1 Battery pack 4.4.2 Electric vehicles 5 Summary and outlook Acknowledgments References

701 702 702 702 702 703 703 704 705 705 705 706 706 706 707 708 708 709 709 710 710 711 711 711 713 714 714

Abstract Magnesium (Mg) is one of the most abundant elements on Earth, with its compounds being applied in various industrial applications. Electrochemical energy storage (EES) systems based on Mg were mostly primary battery systems. Recent development of rechargeable Mg batteries (RMBs) has attracted widespread attention due to their potential to be more sustainable, cost-efficient, and safe, which are crucial factors for accessing renewable energy at a lower environmental impact. This chapter starts by outlining basic chemical and physical properties of Mg, followed by an assessment of Mg production and the market, with a focus on the environmental and economic considerations. As an entry point to Mg-based EES systems, a brief introduction to the electrochemical behaviors of Mg is presented, and its electrochemical applications, particularly RMBs, are highlighted. Based on a summary of recent progress in the field of RMBs and a prospective environmental life cycle assessment (LCA) of such systems, we provide our perspectives on further development of the emerging EES system.

Key points

• • • • • •

This chapter aims to provide an overview of electrochemical energy storage (EES) systems based on magnesium (Mg), with a focus on their economic and environmental impacts. As background information, basic properties and industrial applications of Mg and Mg-based compounds are discussed. The production of Mg, its market, and cost development in view of the sustainability of these activities are evaluated in terms of resource criticality, carbon footprint and energy consumption, as well as current recycling. Based on a brief introduction to the electrochemistry of Mg, its application in EES systems is highlighted. Recent progress of rechargeable Mg batteries (RMBs) and the development of key cell components are summarized. Life Cycle Analysis (LCA) of RMB technology is demonstrated in order to provide perspectives on its further development.

⁎

Corresponding author.

700

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00066-9

Chemistry and Electrochemistry | Magnesium

701

List of symbols and abbreviations APC BMS EES EoL EoL-RIR EV GE ICEV LCA RMB SHE THF TM TRL

1

all phenyl complex battery management system electrochemical energy storage. end of life end-of-life recycling input rate electric vehicle general electric internal combustion engine vehicle life cycle analysis rechargeable Mg battery standard hydrogen electrode tetrahydrofuran transition metal technology readiness level

Introduction

Magnesium (Mg) lies in the second row of the group alkaline earth metal in the periodic table. The metal has a silver-gray color in its pure form, which is highly reductive. Therefore, Mg normally appears with its oxidation state of 2+ in nature. Resources of Mg are abundant in the Earth’s crust as well as in seawater, amounting to 13% of the planet’s mass. In addition to magnesium chloride as the main source of Mg in the ocean, magnesium-bearing minerals of commercial importance include magnesite, dolomite, carnallite etc.1 The metal form of Mg was first discovered by Humphry Davy in 1808, through evaporating the mercury from a magnesium amalgam alloy, which was made by electrolyzing a mixture of magnesium oxide and mercury oxide. Two decades later, Antoine Bussy produced Mg metal with approximating purity via a chemical reduction approach, allowing for further study of the properties of Mg.2 Mg is a lightweight material with a density of 1.74 g cm−3, which is two-thirds that of aluminum. It has a melting point of 650
C and a boiling point of 1090
C, both of which are the lowest among the alkaline earth metals. In addition, Mg has a high specific strength (158 kN m kg−1), good ductility and castability, making it the third most used metal after iron and aluminum. However, as pure Mg is brittle and has the tendency of creep, especially at high temperature, it needs to be alloyed with other elements such as aluminum, zinc or rare earth metals in order to enhance its mechanical properties and processibility.3 Mg and Mg alloys are widely used in the field of aerospace and automotive industries. A complete substitution of aluminum alloys by magnesium alloys in an aircraft could save up to 28% of its weight.4 Chemically, Mg is sensitive to oxygen and water. The fresh Mg surface quickly gets tarnished once exposed to air. Fine Mg powder readily reacts with water to produce hydrogen gas. These reactions give rise to the formation of a thin oxide layer covering the Mg surface. The native oxide layer protects Mg from further oxidation, allowing handling and storage of Mg in an ambient environment, which is practical for application. Furthermore, both metallic and ionic forms of Mg are non-toxic and exhibit superior biocompatibility, which generates enormous application in the biomedical industry. On the downside, notable corrosion of Mg is found when in contact with aggressive electrolyte species. The process is accelerated if galvanic corrosion is triggered when Mg is alloyed with metals such as iron, nickel, copper etc. To improve the corrosive resistance, surface coating is an effective approach.5 The origin of Mg corrosion in aqueous electrolyte is its low redox potential, which is −2.37 V vs. SHE (standard hydrogen electrode). This property also makes Mg a promising metal anode for electrochemical cells. More importantly, a Mg metal anode provides high theoretical capacity both gravimetrically (2233 mA h g−1) and volumetrically (3833 mA h cm−3), based on the redox of Mg/Mg2+ couple. As a charge carrier, Mg2+ has a high charge density due to its bivalence, exhibiting the potential to build high-energy electrochemical energy storage (EES) systems. Therefore, primary Mg batteries were developed and commercialized back to the mid-20th century.6 As Mg is reactive, primary cells based on a Mg anode are mostly reserve battery types using (sea)water as the activating electrolyte. Among these, Mg-air batteries received considerable attention by providing a high theoretical energy density of 6800 Wh kg−1 with an operating voltage of 3.1 V.7 Shifting from an aqueous environment to non-aqueous solution enables reversible Mg plating/stripping at the anode side. By applying organic solvent with a large dielectric constant, high-voltage cathode materials can be coupled with a Mg anode to achieve high energy. Therefore, rechargeable Mg batteries (RMBs) are currently a research focus in the field of applied electrochemistry of Mg. This is particularly the case since 2000, when Aurbach et al. demonstrated the first RMB prototype.8 Compared to other rechargeable battery systems, RMBs allow the use of a metal anode with fewer safety concerns, because electrochemical deposition of Mg shows low tendency to form dendrite,9 which easily triggers short-circuiting and is fire hazardous. In addition, considering the resource abundance, high recycling rate and environmental friendliness of Mg, RMBs are regarded as a sustainable energy storage solution for large-scale applications.

702

Chemistry and Electrochemistry | Magnesium

This chapter starts with a summary of some important chemistries and compounds related to Mg. The production and market of Mg are then discussed, with a focus on the environmental and economic impacts. Subsequently, a brief introduction to the electrochemical behaviors of magnesium in both aqueous and non-aqueous media is given. Based on that, their electrochemical applications, particularly in battery fields, are presented. Finally, we bring our perspectives on further development of EES systems based on magnesium.

2

Applied chemistry of magnesium

Mg has an outer electron configuration of 3s2. It tends to lose or share two electrons, forming plenty of Mg(II) compounds. Due to the abundance of Mg, common inorganic Mg salts, MgO or Mg(OH)2 etc. are largely consumed for industrial and agricultural applications. Organomagnesium compounds are also frequently used in preparative chemistry for coupling reactions, organic transformations and enantioselective synthesis etc.10 In the following, some important Mg chemistries and their applications are discussed.

2.1

Magnesium compounds as refractory material

According to the annual report from the United States Geological Survey, about 40% of Mg compounds produced in the US were used for refractories in 2018.11 Mg compounds, primarily MgO and Mg(OH)2, are effective flame retardants, due to their thermal decomposition being highly endothermic. As illustrated in Eq. (1), the decomposition product of Mg(OH)2 is MgO, which has a high heat capacity, thereby further reducing the heating rate of the host material in a fire. Meanwhile, the reaction also produces water vapor, which is beneficial for diluting any flammable species in the fire zone.12 Mg ðOHÞ2 ! MgO + H2 O; DH ¼ 1224 J g −1

2.2

(1)

Desulfurization of steel

Another major application of Mg is as a desulfurizer in steelmaking processes to reduce FeS, which negatively affects the mechanical property and corrosive resistance of steel. The desulfurization process is normally carried out in hot metal. By dissolving Mg into liquid iron, it reacts with S forming MgS as per Eq. (2). The advantages of Mg over other desulfurization agents are its high efficiency and low cost. Desulfurization of steel by applying Mg is around 20 times and 3 times faster than with lime or calcium carbide, respectively. Moreover, 1 kg of Mg could in principle remove 1.32 kg of sulfur, making it cost-efficient compared to other counterparts. However, as Mg is sensitive to air, resulfurization via Eq. (3) may take place. To prevent this, other capture materials such as lime are added so that more stable compounds can be formed (see Eq. (4)).13

2.3

½Mg Fe + ½SFe ! MgS

(2)

2MgS + O2 + Fe ! 2MgO + ½SFe

(3)

MgS + CaO ! MgO + CaS

(4)

Organomagnesium compounds

In addition to the inorganic compounds, Mg also forms a variety of organic complexes, which are widely used in synthetic chemistry. The most common organomagnesium compounds are Grignard reagents with a general formula R-Mg-X, where R is an organic group and X stands for halogen atom. Grignard compounds are typically produced by reacting organic halides (R-X) with metallic Mg via a radical process, as illustrated in Eq. (5). As the reaction is highly exothermic, a large amount of heat release needs to be considered during the synthesis.14 R −X + Mg ! R −X  − + Mg + ! R + X − + Mg + ! RMg+ + X − ! RMgX

(5)

Grignard compounds are strong reducing reagents which are nucleophilic in nature, due to a polar C-Mg bond. They are frequently used to couple with electrophilic groups such as ketones or aldehydes for the addition reactions (see Eq. (6)). The importance of the reaction is featured by a Nobel prize awarded in 1912 to Victor Grignard for the discovery of the reaction. Notably, Grignard reagents are also used to prepare Mg electrolyte for RMBs. The all phenyl complex (APC) electrolyte8 based on phenyl magnesium chloride is still one of the standard electrolytes applied in the field.

Chemistry and Electrochemistry | Magnesium

703

ð6Þ

3 3.1

Magnesium production and the market Production methods

The electrolysis of molten magnesium chloride (MgCl2) and the silicothermic reduction of magnesium oxide (also called the Pidgeon process) are two techniques that dominate the Mg production industry. Although there are rich sources of Mg in seawater as well as in earth deposits, the abovementioned production processes are both energy intensive; the specific energy consumption and associated CO2 emissions are more than twice that of aluminum production. Compared to the electrolytic process, the Pidgeon process requires slightly more energy. However, the capital cost for the latter is only 1/3 that of the former as it allows easy operation and versatile production scale.15 As a result, the Pidgeon process is currently the most widely used method to produce Mg. Fig. 1 illustrates typical procedures applied to the electrolytic and Pidgeon techniques. The Pidgeon process was named after its inventor Lloyd M. Pidgeon in 1944. Dolomite (CaMg(CO3)2) and quartzite (mostly SiO2 with iron hydroxides) are typical minerals used for the thermal process. They are treated separately at high temperature to produce CaOMgO and ferrosilicon, respectively. The intermediate products are then mixed and briquetted before undergoing a silicothermic reduction via Eq. (7). As a thermodynamically unfavorable process, the reaction needs to proceed at high temperature under the condition that Mg vapor is continuously removed at a reduced pressure. Therefore, Mg with high purity can be obtained by controlling the pressure to limit the vapor of other solid impurities (e.g., Ca or Fe), which generally have a much higher boiling point than Mg.

Thermal process

Electrolytic process

Dolomite mining

Magnsite mining

Transport

Transport

Crushing

Leaching with HCI

Calcination

Purification Hydrocloridric acid plant

Briquetting

Dehydratation

Reduction

Electrolysis

Melting and fefining

Magnesium

Chlorine

Magnesium Ingots Fig. 1 Typical procedures of the electrolytic process and the Pidgeon process (thermal process). Reproduced with permission from Cherubini F; Raugei M; Ulgiati S, LCA of Magnesium Production: Technological Overview and Worldwide Estimation of Environmental Burdens. Resour. Conserv. Recycl. 2008, 52 (8), 1093–1100. Copyright 2008 Elsevier Inc.

704

Chemistry and Electrochemistry | Magnesium 2CaOMgO + ðxFeÞSi ! 2Mg + Ca2 SiO4

(7)

In general, the electrolytic process involves the preparation of feedstock (MgCl2), dehydration of the feedstock and electrolysis. The electrolysis of anhydrous MgCl2 follows Eq. (8). As a byproduct of the electrolysis process, chlorine is recycled by feeding back directly to the hydrochloric acid plant for further leaching process. The degree of MgCl2 dehydration determines the efficiency of hydrolysis by limiting the formation of unwanted oxide or oxychloride and the consumption of the carbon anode. Typically, anhydrous MgCl2 can be prepared either by direct dehydration of aqueous MgCl2 or by the chlorination of anhydrous MgO. MgCl2 ðsÞ ! Mg ðlÞ + Cl2 ðgÞ, E ¼ 3:74 V

3.2

(8)

Market and cost development

In 2020, the global production of primary Mg was estimated to be about 1060 kt, rounding similar values during the last years of the decade. As shown in Fig. 2, China is the lead supplier, with an estimated yearly production of about 886 kt and a total global market share of around 83%, followed by the United States (5.9%), Russia (4.5%), Israel (1.8%), Brazil (1.7%), Kazakhstan (1.5%), Turkey (1.1%) and Ukraine (95  C. After EMD electrolysis, the final product is removed from the anode, ground, washed, and neutralized with NaOH to remove residual electrolyte, followed by a drying step. A typical analysis of EMD sample is given in Table 3. In general, the purity of the electrolyte solution ensures the chemical purity of the resultant EMD, whereas current density, temperature, and solution conditions are responsible for the structure of EMD.

5.1

Cathodic reactions of manganese dioxide in neutral or acidic solutions

The manganese dioxide electrode was thoroughly investigated by N. C. Cahoon and W. C. Vosburgh in NH4Cl or NH4Cl and ZnCl2 solutions at various pH values in the 1960s. It was found that the initial product of the cathodic reaction at the manganese dioxide electrode was MnOOH and that during later stages of discharge, Mn2+ ions are formed. In ammonium chloride solution at pH 7, very little manganese (II) was present in the solution during the first third of the discharge, whereas in solutions of pH Li(s) Na+(aq) + e− > Na(s) Mg+(aq) + e− > Mg(s) Fe2+(aq) + 2e− > Fe(s) 2H+(aq) + 2e− > H2(g) Cu2+(aq) + 2e− > Cu(s) Ag+(aq) + e− > Ag(s) Au3+(aq) + 3e− > Au(s)

−3.03 −2.71 −2.37 −0.44 0 +0.34 +0.80 +1.50

From Vanýsek, P. Electrochemical Series. In Handbook of Chemistry and Physics, 93rd ed.; Haynes, W. M., Ed.; CRC Press: Boca Raton, FL, 2012; p. 5.

measured against the hydrogen redox couple. The terms standard half-cell potential, standard oxidation potential, standard reduction potential, standard redox potential and standard electrode potential are often used interchangeably in this context.

4 4.1

Primary reference electrodes Standard hydrogen electrode

The Standard Hydrogen Electrode (SHE) is the electrochemical half-cell that forms the basis of the thermodynamic scale of standard reduction potentials. At any given temperature, its potential is defined to be 0 V and the potentials of all other reference electrodes are compared with that of the SHE at the same temperature. The SHE consists of a platinized platinum electrode immersed in a solution of unit proton activity, through which is bubbled pure H2 gas at 1 bar pressure (Fig. 1). Platinization of the electrode produces a surface layer of platinum nanoparticles (known as platinum black) that improves the electrode kinetics by increasing the surface area and enhancing hydrogen absorption. The following electrochemical equilibrium is established at the electrode surface: 2H+ ðaqÞ + 2e− >H2 ðgÞ

(1)

According to the Nernst equation, the electrode half-cell potential, ESHE, for this reaction is: ESHE ¼ E0 H+ =H2 

p =p0 RT ln H2 2 2F ðaH + Þ

(2)

where E0 H+ =H2 is the standard reduction potential of the hydrogen redox couple (defined as 0 V), R is the gas constant, T is temperature, F is Faraday’s constant, pH2 is the partial pressure of hydrogen gas, p0 is standard pressure (1 bar)1 and aH+ is the proton activity in the electrolyte. It follows that the potential of the SHE depends only on the partial pressure of H2 gas and the H+ activity in the electrolyte, which are both fixed at unity, giving a potential of 0 V. Coupled with the fact that the hydrogen oxidation/proton reduction reactions are extremely fast and reversible on platinum, this means that the potential of the SHE is well defined and highly stable. The SHE is an idealized electrode/electrolyte interface and cannot be realized in practice due to the theoretical assumption that the proton activity has no interaction with any other ions, which is not strictly valid at this relatively high concentration. Furthermore, it is challenging to achieve both a hydrogen partial pressure of exactly 1 bar (due to the additional vapor pressure of water) and a hydrogen ion activity of exactly unity. Fortunately, when a solution of 1 M H+ is used these errors tend to cancel so this form of the SHE is used in practice. The main application of the SHE is as a primary standard against which secondary reference electrodes such as those described below can be calibrated.

5

Secondary reference electrodes

Several types of laboratory reference electrode are available commercially and are routinely used in electrochemical experiments across a range of applications. As mentioned above, they must be calibrated against a primary reference electrode (SHE) and are therefore known as secondary reference electrodes.

1 1 bar is the thermodynamic standard pressure according to the IUPAC definition. In practice 1 atm (1.013 bar) is often used erroneously but the associated correction is typically smaller than the uncertainty in the measurement.

4

Methods and Instruments | Reference Electrodes

H2(g) at 1 atm e–

Pt wire Pt electrode

Pt electrode H+ e–

H2(g) Outlet H2

e– H+

1 M H+(aq)

H+ e– H2

e– H+

half-reaction at Pt surface: 2H+ (aq) + 2e– H (g) 2

Fig. 1 Standard hydrogen electrode (SHE). Adapted from University of Oregon; https://stock.adobe.com/bg/images/a-standard-hydrogen-electrode-or-sheis-an-electrode-that-scientists-use-as-a-reference-electrode-with-potential-zero/579405189.

Secondary reference electrodes come in a range of different sizes, geometries and chemistries. All of them make use of a fast, reversible, redox couple whose potential can be reliably controlled by maintaining a constant electrolyte composition close to the electrode surface. In order to facilitate measurement of the potential of the working electrode, a reference electrode must have an ionic connection to the electrolyte in the test cell. This is typically achieved by inserting a porous ceramic frit between the two electrolyte solutions. The limited electrolyte pathways through the pores of the frit are sufficient to maintain ionic conductivity while minimizing cross-contamination of dissolved species from one solution to the other during the timescale of the measurement. Cross-contamination of the reference electrode and the test cell may be further reduced by adding a second electrolyte chamber and frit to the reference electrode; this is known as a double junction reference electrode. In some cases, e.g., for sensitive pH measurements, triple junction reference electrodes may be used.

Methods and Instruments | Reference Electrodes

5

An alternative method of separating the two solutions is via the use of a salt bridge. A salt bridge provides an ionically conducting path between two electrochemical half-cells by incorporating ions into a gel within a U-shaped glass tube. The gel, typically agar, is used to minimize mixing between the two solutions. The use of proton-conducting polymer electrolyte, e.g., Nafion, as a salt bridge has emerged in recent years for reference electrode measurements in low temperature fuel cells and electrolyzers. The choice of reference electrode for a particular measurement is often determined by chemical compatibility. Even with the above precautions, there will always be some non-zero rate of contamination of both the reference electrode and the test cell. For this reason, reference electrodes are often selected to minimize any mismatch in electrolyte chemistry, particularly where precipitation reactions can occur, or species are present that can poison the reference electrode surface or perturb the test cell. The most common types of secondary reference electrode are described below. Their potentials with respect to the SHE are listed in Table 2. Slight variations in these potentials are found depending on the literature source so the values in Table 2 are quoted to two decimal places for consistency. Conversion from one reference electrode scale to another is straightforward; it is essential when reporting results of electrode potential measurements that the type of reference electrode used and the electrolyte concentration (if applicable) are reported.

5.1

Saturated calomel electrode

The saturated calomel electrode (SCE) consists of a mixture of liquid Hg and HgCl2 (also known as calomel) paste supported on a rod immersed in KCl solution, as shown in Fig. 2, which leads to the following electrochemical equilibrium: HgCl2 ðsÞ + e− >2HgðlÞ + 2Cl− ðaqÞ

(3)

Since the activity of a pure solid or liquid phase is defined as unity, the electrode potential, ESCE, depends only on the activity of chloride ion in the electrolyte, aCl−, according to the Nernst equation: ESCE ¼ E0 HgCl2 =Hg 

RT ln aCl− 2F

(4)

where E0 HgCl2 =Hg is the standard reduction potential of the HgCl2/Hg redox couple. The most straightforward way to ensure a constant chloride concentration is to use a saturated KCl solution, which can be achieved by maintaining an excess of solid KCl crystals in the electrolyte chamber. These need to be replenished periodically to maintain saturation as electrolyte may be lost by diffusion through the reference electrode frit and dilution can also occur via diffusion of water from the test cell. The SCE is among the most robust and reliable of laboratory reference electrodes, although its use has declined in recent years due to the health risks associated with exposure to mercury.

5.2

Silver/silver chloride electrode

The Ag/AgCl electrode (Fig. 3) is based on a Ag wire with a AgCl coating, which is immersed in KCl solution, leading to the following electrochemical equilibrium: AgðsÞ + Cl− ðaqÞ >AgClðsÞ + e−

(5)

The electrode potential, EAgCl/Ag, varies according to: EAgCl=Ag ¼ E0 AgCl=Ag 

Table 2

RT ln aCl− F

(6)

Potentials of selected secondary reference electrodes vs SHE at 25  C.

Reference electrode

Potential (V vs SHE)

Mercury/mercurous sulfate (satd. K2SO4) CSE (satd. CuSO4) Silver/silver chloride (0.1 M KCl) Silver/silver chloride (seawater) Silver/silver chloride (3 M KCl) Silver/silver chloride (satd. KCl) SCE (satd. KCl) Mercury/mercuric oxide (1 M NaOH) Palladium hydride (1 M H+) NHE RHE

0.64 0.32 0.29 0.25 0.22 0.20 0.24 0.10 0.05 0.00 0.00–0.059 pH

Filling hole Saturated solution of KCI Platinum wire

Mercury Calomel Sintered glass KCI crystals Sintered glass Fig. 2 Saturated calomel electrode (SCE). Adapted from PNGITEM; https://www.pngitem.com/middle/hxwobbh_saturated-calomel-electrode-diagram-hd-pngdownload/.

Ag

Ag AgCl

KCl, 3.5 mol dm-3

Fig. 3 Silver/silver chloride electrode. Adapted from Chemistry Glossary; https://glossary.periodni.com/glossary.php?en=silver%2Fsilver-chloride+electrode.

Methods and Instruments | Reference Electrodes

7

Similar to the SCE, the potential of the reference electrode is constant when the chloride concentration is fixed, which can be achieved with a saturated solution of KCl. However, Ag/AgCl reference electrodes employing other chloride concentrations, e.g., 3 M, 0.1 M, are also commonly used. The chloride concentration is often chosen to match the concentration of the electrolyte in the test cell in order to minimize liquid junction potentials. N.B. It is essential to quote the KCl concentration used when reporting the results of measurements with this type of reference electrode. With the decrease in the availability and use of the SCE due to health concerns, the Ag/AgCl electrode is now the most commonly used laboratory reference electrode. It is also normally the reference electrode of choice for incorporation into pH electrodes. One of its advantages is that it is relatively insensitive to changes in temperature compared to other types of reference electrode. When used in dilute solutions care should be taken to minimize contamination of the test cell due to chloride leakage, particularly for corrosion measurements on stainless steel where chloride is a particularly aggressive species.

5.3

Mercury/mercurous sulfate electrode

In situations where chloride contamination of the test cell must be avoided at all costs, secondary reference electrodes based on other anions can be used. The Hg/HgSO4 electrode (Fig. 4) is suitable for use in neutral and acidic environments and consists of a platinum wire immersed in a mixture of liquid Hg and HgSO4. The following electrochemical equilibrium is established: Hg2 SO4 ðsÞ + 2e− >2HgðlÞ + SO4 2− ðaqÞ

(7)

The electrode potential, EMMS, varies as follows: EMMS ¼ E0 HgSO4 =Hg 

RT ln aSO4 2− 2F

(8)

where E0 HgSO4 =Hg is the standard reduction potential of the HgSO4/Hg redox couple and aSO4 2− is the activity of the sulfate ion, which can be maintained constant by using a saturated solution of K2SO4.

5.4

Mercury/mercuric oxide electrode

The Hg/HgO electrode (Fig. 5) is suitable for use in alkaline environments where contamination of the test cell with chloride would be undesirable. It consists of an inert tube containing liquid Hg and HgO immersed in a solution of NaOH or KOH. Polytetrafluoroethylene (PTFE) is generally used for the electrode body due to the risk of leaching of glass at high pH. The following electrochemical equilibrium is established: HgOðsÞ + H2 OðlÞ + 2e− >HgðlÞ + 2OH− ðaqÞ

(9)

Electrical connector

Platinum wire

Refill opening

Glass-to-metal seal Hg-drop Layer of Hg2SO4 Cotton plug Sulfate-containing solution Diaphragm

Fig. 4 Mercury/mercurous sulfate electrode. Adapted from Roscher et al. Reference Electrodes. Encyclopedia. 2023, 3, 478–489, Figure 7; https://www.mdpi. com/2673-8392/3/2/33#::text=A%20reference%20electrode%20is%20a,been%20developed%20for%20different%20applications.

8

Methods and Instruments | Reference Electrodes

Connector Hg/HgO electrode Hg HgO Cotton

Internal solution (x M MOH aq.)

Teflon outer body

Frit

Fig. 5 Mercury/mercuric oxide electrode. Adapted from Kawashima et al. ACS Catal. 2023, 13, 1893–1898, Figure 1; https://pubs.acs.org/doi/10.1021/acscatal. 2c05655.

The electrode potential, EMMO, varies according to: EMMO ¼ E0 HgO=Hg 

ða − Þ2 RT ln OH aH2 O 2F

(10)

where E0 HgO=Hg is the standard reduction potential of the HgO/Hg redox couple, aOH− is the activity of the hydroxyl ion and aH2O is the activity of water. A concentrated solution of NaOH or KOH is typically used to fix the hydroxyl ion concentration but the water activity must also be taken into account. This makes the use of the Hg/HgO electrode less straightforward than other secondary reference electrodes as computational approaches are often required.3 Mercury/mercury oxide reference electrodes are particularly suited to studies of alkaline water electrolyzers.

5.5

Copper/copper (II) sulfate electrode

The copper sulfate electrode (CSE) comprises a Cu rod immersed in a saturated solution of CuSO4 (Fig. 6). The solution must be acidified with H2SO4 (pH < 3) to maintain an active copper surface. The following electrochemical equilibrium is established: Cu2+ ðaqÞ + 2e− >CuðsÞ

(11)

The electrode potential, ECSE, varies according to: ECSE ¼ E0 Cu2+ =Cu +

RT ln aCu2+ 2F

(12)

where E0 Cu2+ =Cu is the standard reduction potential of the Cu2+/Cu redox couple and aCu2+ is the activity of the Cu2+ ion. The CSE is primarily used to measure the potential of cathodically protected steel structures in soil or seawater, such as pipelines, marine installations and reinforced concrete. It is a relatively rough and ready reference electrode that is required to be durable in harsh external environments and is not particularly responsive. It is therefore rarely used in analytical electrochemistry and has yet to be applied to the study of electrochemical energy conversion and storage devices.

5.6

Reversible hydrogen electrode

The Reversible Hydrogen Electrode (RHE) is a more practical and versatile form of the SHE in which the proton activity in the electrolyte is permitted to vary from unity. This allows the reference electrode to be directly immersed in any electrolyte. Unlike

9

Methods and Instruments | Reference Electrodes

O-Ring seal

S Saturated copper ssulfate solution Pure copper rod Electrode body C Copper sulfate ccrystals

Porous plug

O-Ring seal

Fig. 6 Copper/copper (II) sulfate electrode. Adapted from Wikipedia; https://en.wikipedia.org/wiki/Reference_electrode#/media/File:CopperSulphateElectrode.png.

other secondary reference electrodes, precise knowledge of the pH is required for it to be used reliably. The potential of the RHE varies with the pH of the solution in which it is immersed as follows: ERHE ¼ E0 H+ =H2 +

p =p0 RT ln H2 2 2F ðaH+ Þ

(13)

which can be reduced to a simple pH dependence in 1 bar H2: ERHE ¼ 0:000 V − 0:059 pH

(14)

The pH dependence tracks that of both the oxygen reduction and proton reduction reactions in the Pourbaix diagram, making the RHE a useful reference electrode in studies of catalyst activity. The development of commercial miniaturized hydrogen electrodes, supplied with hydrogen from a replaceable cartridge, has led to an increase in the use of RHEs, most notably for low temperature fuel cells and electrolyzers where there are no contamination issues in either direction. A special case of the RHE is the Normal Hydrogen Electrode (NHE), in which the H+ concentration is 1 N. The NHE was historically used as a primary reference electrode before the establishment of the SHE.

5.7

Palladium hydride electrode

The palladium hydride (PdH) electrode is similar in characteristics to the SHE but with the primary difference that Pd has the ability to absorb up to 1000 times its volume of dissolved hydrogen within its crystal lattice. This provides a convenient source of H2 that can generate a stable potential when in contact with H+ in solution over a wide range of absorbed hydrogen concentrations. This is advantageous when continuous bubbling of H2 through the solution is not feasible. The potential of the PdH electrode, EPdH, has an identical pH dependence to the RHE but with a positive offset of 50 mV, which is associated with the potential of the PdH a + b transition phase. EPdH ¼ 0:050 V − 0:059 pH

(15)

PdH electrodes can be fabricated by exposing a Pd wire to gaseous H2. The duration of exposure depends on the amount of time for which the reference electrode is required to be at a stable potential, as the hydrogen will subsequently desorb gradually from the Pd. Of course, this is not a concern if hydrogen is supplied continuously to the reference electrode during the measurement.

10

6

Methods and Instruments | Reference Electrodes

Pseudo reference electrodes

In some circumstances it is not possible to use a conventional secondary reference electrode to carry out a particular measurement. This is the case (i) in non-aqueous environments, e.g., organic solvents, ionic liquids, molten salts, due to their incompatibility with aqueous reference electrodes, (ii) in the presence of geometrical constraints, e.g., within miniaturized cells, or (iii) when the impact of contamination of the cell from the reference electrode electrolyte poses a significant risk. In such cases, a pseudo (or quasi) reference electrode is often used.2 This consists of a metal wire placed in direct contact with the test cell electrolyte. Since there is no control of the concentrations of electroactive species at the electrode/electrolyte interface, the electrode potential is determined by the local environment, which may vary with time in an unpredictable manner. On the other hand, contamination effects, ohmic resistance issues and the presence of liquid junction potentials are avoided. Reliable use of a pseudo reference electrode depends on the extent to which to the variation of its potential in the local environment can be confidently predicted or measured. For example, this can be achieved by calibrating the pseudo reference electrode with respect to a redox couple of known potential. The ferrocene/ferrocenium redox couple is often used to do this since both the oxidized and reduced forms are soluble and stable in a wide range of solvents and a Nernstian response is usually observed.4 Use of a pseudo reference electrode can also be justified when its potential can be shown to vary consistently with that of the working electrode. The most commonly employed pseudo reference electrode in electrochemical energy conversion and storage research is metallic lithium, which is routinely used in Li-ion battery three-electrode cell configurations due to the incompatibility of secondary reference electrodes with non-aqueous solvents. It is perhaps not always widely acknowledged that the Li+/Li reference electrode is not a true secondary reference electrode and it is important that its limitations are well understood by users.5

6.1

Dynamic hydrogen electrode

The dynamic hydrogen electrode (DHE) is a type of pseudo reference electrode that generates a proton/hydrogen gas equilibrium at the platinum electrode surface through the application of a cathodic current (Fig. 7). Its main advantage over the RHE is that it does

R V E

B A

WE

Fig. 7 Dynamic hydrogen electrode (DHE). WE: working electrode, A: reference electrode cathode, B: reference electrode anode. Adapted from Giner, J. J. Electrochem. Soc. 1964, 111, 376–377, Figure 1; https://iopscience.iop.org/article/10.1149/1.2426125/pdf.

There is some confusion in the research community regarding the terms ‘pseudo’ (meaning ‘false’ in Latin) and ‘quasi’ (meaning ‘almost’) reference electrode, which appear to be used interchangeably in the literature. Some authors have attempted to draw a distinction based on whether or not calibration with a known redox couple is performed, but there is no widely accepted standard definition and no evidence in the published literature or in textbooks that there is any real distinction between the two terms. The term ‘pseudo reference electrode’ appears to be in more common use at the present time and is adopted here.

2

Methods and Instruments | Reference Electrodes

11

not require its own hydrogen supply. However, it is relatively complex to set up as it requires a separate counter electrode and a source of power to pass a current to the DHE. The DHE is primarily used in fuel cell applications and typically consists of two platinized platinum electrodes in contact with the same electrolyte as the test cell. A small cathodic current (1 mA cm−2) is applied to one of the electrodes, which then adopts a potential 20–40 mV more negative than that of the RHE and can be used as a pseudo reference electrode.

7

Historical nomenclature

Historically, reference electrodes have been grouped into two categories known as electrodes of the first and second kind. Electrodes of the first kind are based on a reduced species (typically a metal but also including hydrogen gas) in equilibrium with a solution of its cations, while those of the second kind are based on a metal in contact with a solution of a sparingly soluble anion of the metal salt. Electrodes of the first kind are generally pseudo reference electrodes, e.g., Li+/Li, unless the cation concentration can be reliably controlled, as in the SHE and CSE. Most commercially available secondary reference electrodes are electrodes of the second kind, e.g., SCE, Ag/AgCl. The use of this particular nomenclature appears to be in decline and it is noted here merely for completeness.

8

Liquid junction potentials

A consequence of the ionic connection between the reference electrode and the test cell, which is required for the measurement of electrode potential, is that a liquid junction potential can develop at the interface between the two different electrolytes. This occurs due to diffusion of charged species across the interface, which establishes a steady-state potential difference between the two electrolytes. A liquid junction potential may also be referred to as a diffusion or concentration potential. It is important when using a reference electrode that the magnitude of any liquid junction potential is minimized, or at least maintained constant. Minimization of the liquid junction potential can be accomplished via the use of a salt bridge with a high concentration of ions of similar mobility, e.g., K+ and Cl−, inserted between the two electrolyte solutions. The liquid junction potentials at each salt bridge/electrolyte interface are dominated by the contribution from the KCl and are opposite in polarity, hence they tend to cancel each other. In this way the magnitude of the liquid junction potential between the reference electrode and the test cell can be reduced below 1 mV. Thermal liquid junction potentials can also be an issue when the reference electrode and the test cell are maintained at different temperatures. For example, for a room temperature reference electrode connected to a low temperature fuel cell operating at 80  C, the magnitude of this thermal liquid junction potential is expected to be of the order of tens of millivolts.6 Such errors can be minimized by maintaining the reference electrode at the same temperature as the test cell where possible.

9

Dilute solutions

When the electrolyte conductivity is low, the ohmic drop between the working electrode and the reference electrode can be significant. In such circumstances it is convenient to use a Luggin capillary as a salt bridge that terminates close to the working electrode surface. The diameter of the capillary should be as small as possible in order to minimize shielding effects on the current distribution at the working electrode. Since the inner diameter of the capillary is also small, it should be filled with solution of sufficient conductivity to facilitate the measurement.

10

Use of reference electrodes

10.1 Standards In general, there is a lack of standardized approaches to the use of reference electrodes. ASTM Standard G 2157 provides guidance on their use for corrosion-related measurements but, surprisingly, there is no international standard that sets out clear guidance for the use of reference electrodes in broader applications. This includes their use in energy conversion and storage devices, for which no widely accepted best practice or standardized guidance has been established. There is a clear need for the development and uptake of international standards to improve the quality and inter-comparability of data generated using reference electrodes for these applications.

10.2 Calibration It is good practice to calibrate a reference electrode before and after every experiment to check for any drift in its potential. Calibration is achieved by comparing the potential of the reference electrode with that of a laboratory standard reference electrode.

12

Methods and Instruments | Reference Electrodes

Every electrochemical laboratory should maintain at least two standard reference electrodes (calibrated against the SHE by the manufacturer), which should be permanently stored in clean electrolyte of the same chemistry and concentration, and must never be used for any other purpose. A typical criterion adopted by electrochemical research groups is that a given reference electrode is suitable for use when its potential is measured to be within 2 mV of that of a standard reference electrode. Regular comparison of the potential of the standard reference electrodes will indicate when either of them needs to be replaced.

10.3 Storage When not in use, it is important to store reference electrodes in a manner that minimizes contamination of the internal electrolyte and avoids dry-out of the porous frit due to evaporation. This is normally achieved by immersing the frit (or salt bridge) in a clean solution that is identical to the internal electrolyte, e.g., saturated KCl. The body of the reference electrode should be thoroughly rinsed with deionized water and dried with tissue paper before storage to prevent contamination of the storage solution. One disadvantage of storage of reference electrodes in saturated solutions is that salt crystals tend to form around the liquid/air interface over time as the solvent evaporates, which can be unsightly and can also lead to spillage of crystals on work surfaces and corrosion of metallic objects that come into contact with them. An alternative strategy is to seal the frit with an elastomeric cap, thereby slowing down the rate of evaporation. This is only effective for relatively short periods of time (up to several months) as it merely delays dry-out of the porous frit.

10.4 Maintenance Reference electrodes require regular inspection and maintenance to ensure they are operating reliably and to extend their useful life. This should involve checking the following: 1. Potential The potential of the reference electrode is the primary indicator of its state-of-health and should be checked regularly against that of a standard laboratory reference electrode (see Section 10.2). A deviation of >2 mV should prompt further checks and corrective action. If the correct potential cannot be restored the electrode should be discarded. 2. Filling solution The level of the filling solution should be checked to see if it has dropped. The filling solution can be topped up through the refill port, or drained and replaced completely if contamination or dilution is suspected. For saturated filling solutions, solid crystals of salt should always be visible in the electrolyte chamber. If the concentration of the solution has dropped below saturation, fresh salt crystals can be added through the refill port. 3. Frit The reference electrode frit should be kept wet at all times to prevent blockage due to salt precipitation when evaporation of electrolyte occurs. Measurement of the impedance of the reference electrode (using an impedance analyzer) can be used to check for this; the value should be 1 M) makes connection of the reference electrode to the cell relatively straightforward.

11.2 Batteries Despite many decades of research, the state-of-the-art in reference electrode development for high energy density battery applications is less well advanced than for aqueous-based electrochemical energy conversion and storage devices.5 This is primarily due to the fact that all of the common secondary reference electrodes described above are incompatible with the organic solvents used in Li-ion batteries and other emerging chemistries. Pseudo reference electrodes, such as lithium and sodium metal, are therefore almost exclusively used. Unfortunately, the potential of these electrodes is not stable over long periods of time and varies considerably at high current densities and in the presence of different electrolytes due to the SEI layer formed in the passivating reaction with the electrolyte. The issue is compounded by a general lack of awareness of these limitations among the research community. As a result, there is an urgent need for the development of more reliable and reproducible reference electrode materials for use in battery research. Many active materials have been investigated, with partially delithiated LiFePO4 and partially lithiated Li4Ti5O12 identified as the most promising candidates for Li-ion batteries due to the relatively wide and stable plateau in their charge/ discharge curves.12 This means that their potential does not change appreciably with the passage of current, i.e., they are relatively non-polarizable. On the downside, the process of partial (de)lithiation adds undesirable complexity and their use remains far from widespread. Reference electrode material development is also required for next generation battery chemistries, such as Li-S, solid-state, metal-ion and metal-air batteries.13

16

Methods and Instruments | Reference Electrodes

Materials discovery in Li-ion battery research tends to focus on cathode materials, which are routinely screened in half coin cell configuration. An example of a half coin cell is shown in Fig. 10a; it typically consists of a working electrode of the cathode material and a lithium metal counter electrode, which doubles as a pseudo reference electrode. In addition to the limitations of lithium metal as a reference electrode outlined above, a major disadvantage of half coin cells is that they are not fully representative of conditions in a real battery due to the use of lithium metal as the counter electrode, which provides an unlimited source of current.

Casing top (Negative terminal)

Cap (Negative)

Gasket

Spacer

Wave spring

Counter electrode

Disk spring

RE

Separator Reference electrode Separator

Stainless steel spacer

Gasket

Counter/ Reference Li electrode

Separator Working electrode

Separator

Case (Postive)

Casing bottom (Postive terminal) working electrode

A

B

Li metal reference (Ø 6 mm) Reference current collector

Anode (Ø 11 mm)

Glassfiber separator with electrolyte Cathode (Ø 11 mm)

Spring Anode current collector

Cathode current collector Insulation tube

C Fig. 10 (a) Half coin cell for electrochemical characterization of battery electrode materials. (b) Full coin cell (including lithium wire reference electrode) for laboratory scale testing of battery electrode materials. (c) Swagelok cell for three-electrode characterization of battery electrode materials. Panel a: Adapted from Sinha et al. J. Electrochem. Soc. 2011, 158, A1400–A1403, Figure 1, https://iopscience.iop.org/article/10.1149/2.080112jes/pdf; Panel b: Adapted from Juarez-Robles et al. J. Electrochem. Soc. 2017, 164, A837–A847, Figure 1(e), https://iopscience.iop.org/article/10.1149/2.1251704jes/pdf; Panel c: Adapted from Solchenbach et al. J. Electrochem. Soc. 2016, 163, A2265–A2272, Figure 1(a), https://iopscience.iop.org/article/10.1149/2.0581610jes/pdf.

Methods and Instruments | Reference Electrodes

17

Other non-representative conditions in half coin cells include the use of single layer electrodes and a relative excess of electrolyte, which can mask decomposition and dry-out issues. In battery R&D, the scale-up process consists of a progression from half coin cell to full coin cell to small format commercial cell (typically a pouch cell). In a full coin cell, both the positive and negative electrode contribute to the measured electrochemical response (Fig. 10b). They are sometimes fitted with a reference electrode, typically via the insertion of a lithium wire, but are more commonly tested in two-electrode configuration. A more common three-electrode cell used in battery material development is the Swagelok cell (Fig. 10c), which makes use of the T-shaped geometry of a commercial tubing connection. In all such cells, the geometry of the reference electrode with respect to the test cell is critical to the accuracy and reliability of the measurement of working electrode potential. Particularly for EIS measurements, any non-uniformity in the cell, including misalignment of the electrodes, imbalance in electrode kinetics or edge effects, can lead to errors in the measured potential.14 For this reason, the use of wire reference electrodes in full coin cells is discouraged. Coaxial (and reversed coaxial) geometries are often adopted15,16 but even then misalignment effects can be significant if care is not taken. In theory, the use of symmetrical coin cells, in which identical electrodes are employed on either side of the separator, provides artifact-free measurements; in practice, however, this is only practical for pristine cells as the study of cycled electrodes requires the electrodes to be removed from their original cell and assembled in the symmetrical cell for each measurement.17 In general, reference electrodes are not routinely used beyond coin cell or Swagelok cell level. Some attempts have been made to incorporate them into larger format cells18,19 but this is far from common practice.

11.3 Other devices Reference electrodes have been applied to a range of other electrochemical energy conversion and storage devices, including supercapacitors,20 redox flow cells,21 and high temperature fuel cells22 and electrolyzers.23 However, their use in these applications is extremely limited by comparison and guidance/best practice in the literature is sparse. The same general issues of chemical compatibility, reference electrode positioning and perturbation of cell performance apply to these devices.

11.4 Case study: Bipolar plate potentials in PEM fuel cells and electrolyzers The usefulness of reference electrodes in electrochemical energy conversion and storage devices is not restricted to characterization of electrode materials. They can be equally powerful in diagnosing performance and lifetime issues of other components, notably bipolar plates in PEM fuel cells and electrolyzers. For example, application of RHEs at single cell level in PEM fuel cells and electrolyzers has demonstrated conclusively that the local potential of the bipolar plate in such devices is decoupled from that of the nearest electrode,24,25 contrary to accepted wisdom in both industry and academia. Such insight not only informs the development of more representative ex situ test methods for bipolar plate materials, coatings and surface treatments, but has opened the possibility of replacing platinum-coated titanium in PEM water electrolyzers with much cheaper materials, such as carbon-coated stainless steel. This would significantly reduce the capital cost of the technology, accelerating the uptake of green hydrogen production.

11.5 Use in commercial devices Reference electrode development has not yet reached the stage where they can be routinely incorporated into commercial energy conversion and storage devices, such as fuel cell/electrolyzer stacks and battery cells, modules and packs. While this would undoubtedly prove useful, for example in condition monitoring and lifetime prediction, their use in this context is hampered by several factors, including (i) the additional cost and complexity of system design, (ii) challenges associated with calibration and long term stability of the reference electrodes, and (iii) the risk of perturbation of system performance due to their physical presence and additional mass/volume. This is an increasingly active area of research, which may lead to practical advances in the coming years.

12

Summary

Reference electrodes are an increasingly powerful research tool for the electrochemical characterization of materials, components and devices in fuel cell, electrolyzer, battery and supercapacitor research. However, data generated from reference electrodes should always be treated with caution as their use is prone to a range of experimental artifacts if best practice is not closely followed. There is an associated need for standardized approaches in the application of reference electrodes to such devices in order to provide greater confidence in the data generated. While reference electrodes are now relatively well established as a research tool at the material characterization and single cell level, their application in commercial cells and at stack/pack level remains very much a future goal.

18

Methods and Instruments | Reference Electrodes

Acknowledgments This work was supported by the National Measurement System of the UK Department of Science, Innovation and Technology.

References 1. Vanýsek, P. Electrochemical Series. In Handbook of Chemistry and Physics; Haynes, W. M., Ed, 93rd ed.; CRC Press: Boca Raton, FL, 2012; p. 5. 2. Kissinger, P. T.; Bott, A. W. Electrochemistry for the Non-Electrochemist. Curr. Sep. 2002, 20, 51–53. 3. Kawashima, K.; Márquez, R. A.; Son, Y. J.; Guo, C.; Vaidyula, R. R.; Smith, L. A.; Chukwuneke, C. E.; Mullins, C. B. Accurate Potentials of Hg/HgO Electrodes: Practical Parameters for Reporting Alkaline Water Electrolysis Overpotentials. ACS Catal. 2023, 13, 1893–1898. 4. Mozhzhukhina, N.; Calvo, E. J. Perspective—The Correct Assessment of Standard Potentials of Reference Electrodes in Non-Aqueous Solution. J. Electrochem. Soc. 2017, 164, A2295–A2297. 5. Raccichini, R.; Amores, M.; Hinds, G. Critical Review of the Use of Reference Electrodes in Li-Ion Batteries: A Diagnostic Perspective. Batteries 2019, 5, 12–25. 6 Lvov, S. N.; Macdonald, D. D. Estimation of the Thermal Liquid Junction Potential of an External Pressure Balanced Reference Electrode. J. Electroanal. Chem. 1996, 403, 25–30. 7. ASTM Standard G 215-17. Standard Guide for Electrode Potential Measurement; ASTM International: West Conshohocken, PA, 2017. 8. Zalitis, C. M.; Kramer, D.; Kucernak, A. R. Electrocatalytic Performance of Fuel Cell Reactions at Low Catalyst Loading and High Mass Transport. Phys. Chem. Chem. Phys. 2013, 15, 4329–4340. 9. Li, G.; Pickup, P. G. Measurement of Single Electrode Potentials and Impedances in Hydrogen and Direct Methanol PEM Fuel Cells. Electrochim. Acta 2004, 49, 4119–4126. 10. He, W.; Nguyen, T. Edge Effects on Reference Electrode Measurements in PEM Fuel Cells. J. Electrochem. Soc. 2004, 151, A185–A195. 11. Hinds, G.; Brightman, E. In Situ Mapping of Electrode Potential in a PEM Fuel Cell. Electrochem. Commun. 2012, 17, 26–29. 12. La Mantia, F.; Wessells, C. D.; Deshazer, H. D.; Cui, Y. Reliable Reference Electrodes for Lithium-Ion Batteries. Electrochem. Commun. 2013, 31, 141–144. 13 Cengiz, E. C.; Rizell, J.; Sadd, M.; Matic, A.; Mozhzhukhinaz, N. Review—Reference Electrodes in Li-Ion and Next Generation Batteries: Correct Potential Assessment, Applications and Practices. J. Electrochem. Soc. 2021, 168, 120539. 14. Levi, M. D.; Dargel, V.; Shilina, Y.; Aurbach, D.; Halalay, I. C. Impedance Spectra of Energy-Storage Electrodes Obtained with Commercial Three-Electrode Cells: Some Sources of Measurement Artefacts. Electrochim. Acta 2014, 149, 126–135. 15. Itagaki, M.; Honda, K.; Hoshi, Y.; Shitanda, I. In-Situ EIS to Determine Impedance Spectra of Lithium-Ion Rechargeable Batteries During Charge and Discharge Cycle. J. Electroanal. Chem. 2015, 737, 78–84. 16. Battistel, A.; Fan, M.; Stojadinovic, J.; La Mantia, F. Analysis and Mitigation of the Artefacts in Electrochemical Impedance Spectroscopy due to Three-Electrode Geometry. Electrochim. Acta 2014, 135, 133–138. 17. Waldmann, T.; Kasper, M.; Wohlfahrt-Mehrens, M. Optimization of Charging Strategy by Prevention of Lithium Deposition on Anodes in High-Energy Lithium-ion Batteries – Electrochemical Experiments. Electrochim. Acta 2015, 178, 525–532. 18 McTurk, E.; Birkl, C. R.; Roberts, M. R.; Howey, D. A.; Bruce, P. G. Minimally Invasive Insertion of Reference Electrodes into Commercial Lithium-Ion Pouch Cells. ECS Electrochem. Lett. 2015, 4, A145–A147. 19. Ahmed, Z.; Roberts, A. J.; Amietszajew, T. Ti-Based Reference Electrodes for Inline Implementation into Lithium-Ion Pouch Cells. Energy Technol. 2021, 9, 2100602. 20. Le Fevre, L. W.; Fields, R.; Redondo, E.; Todd, R.; Forsyth, A. J.; Dryfe, R. A. W. Cell Optimisation of Supercapacitors Using a Quasi-Reference Electrode and Potentiostatic Analysis. J. Power Sources 2019, 424, 52–60. 21. Huang, Q.; Li, B.; Song, C.; Jiang, Z.; Platt, A.; Fatih, K.; Bock, C.; Jang, D.; Reed, D. In Situ Reliability Investigation of All-Vanadium Redox Flow Batteries by a Stable Reference Electrode. J. Electrochem. Soc. 2020, 167, 160541. 22. Finklea, H.; Chen, X.; Gerdes, K.; Pakalapati, S.; Celik, I. Analysis of SOFCs Using Reference Electrodes. J. Electrochem. Soc. 2013, 160, F1055–F1066. 23. Kim, J.; Ji, H.-I.; Dasari, H. P.; Shin, D.; Song, H.; Lee, J. H.; Kim, B. K.; Je, H. J.; Lee, H. W.; Yoon, K. J. Degradation Mechanism of Electrolyte and Air Electrode in Solid Oxide Electrolysis Cells Operating at High Polarization. Int. J. Hydrog. Energy 2013, 38, 1225–1235. 24. Castanheira, L.; Bedouet, M.; Kucernak, A.; Hinds, G. Influence of Microporous Layer on Corrosion of Metallic Bipolar Plates in Fuel Cells. J. Power Sources 2019, 418, 147–151. 25. Becker, H.; Dickinson, E. J. F.; Lu, X.; Bexell, U.; Proch, S.; Moffatt, C.; Stenström, M.; Smith, G.; Hinds, G. Assessing Potential Profiles in Water Electrolysers to Minimise Titanium Use. Energy Environ. Sci. 2022, 15, 2508–2518.

Further reading 1. Compton and Sanders, 1996. Compton, R. G.; Sanders, G. H. W. Electrode Potentials. Oxford Chemistry Primers, 41; Oxford University Press: Oxford, 1996. 2. Inzelt et al., 2013. Inzelt, G.; Lewenstam, A.; Scholz, F. Handbook of Reference Electrodes; Springer Berlin: Heidelberg, 2013. 3. Ives and Janz, 1961. Ives, D. J. G.; Janz, G. J. Reference Electrodes, Theory and Practice; Academic Press: New York, 1961.

Methods and Instruments | Potential and Current Steps Rudolf Holze, State Key Laboratory of Materials-oriented Chemical Engineering, School of Energy Science and Engineering, Nanjing Tech University, Nanjing, Jiangsu Province, China; Chemnitz University of Technology, Chemnitz, Germany; Institute of Chemistry, Saint Petersburg State University, St. Petersburg, Russia © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of R. Holze, MEASUREMENT METHODS | Electrochemical: Potential and Current Steps, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 655–659, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00070-8.

1 2 3 4 References

Introduction Potential step experiments Current step experiments Conclusions

20 20 23 26 26

Abstract Current, electrode potential or cell voltage can be stepped from an initial value to a final value by using external current or voltage sources. Measurements of the associated parameters (current in case of potential or voltage step, potential in case of current step) provide transients. Evaluation of these transients yields data pertaining to the kinetics of the electrode process occurring as an effect of the applied step; in addition, double-layer capacity values and further electrochemical data can be derived.

Key points

• • •

Two classes of experimental methods in electrochemistry controlling either electrode potential or current are introduced Possibilities to determine structural and kinetic properties of electrochemical interfaces Features of step-methods in electrochemistry are illustrated

Symbols, units and acronyms A c Cd c 0O c 0R DO DR E E u0 F I Ic Id IF Iss j0 k n N O QC QF Qtot R

electrode surface area molar concentration double-layer capacity concentration of the reacting species in the bulk, oxidized form concentration of the reacting species in the bulk, reduced form diffusion coefficient of the reacting species, oxidized form diffusion coefficient of the reacting species, reduced form electrode potential electrode potential at equilibrium with no flow of current, formal potential Faraday constant current (total current), also flow of species capacitive current mass transport-limited current faradaic current steady-state current exchange current density rate constant of electrode reaction electrode reaction valency diffusion coefficients and concentrations of reactants oxidized species electrical charge charge associated with faradic process total charge reduced species

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00201-2

19

20

Methods and Instruments | Potential and Current Steps

R r0 Ru t T h hct u t

1

electrical resistance, gas constant radius of spherical electrode uncompensated electrolyte solution resistance time temperature overpotential charge transfer overpotential degree of coverage transition time

Introduction

In studies of electrode kinetics, electrochemical methods employing controlled potential or current (single or multiple) steps are used.1–7 The recorded response of the corresponding (dependent) variable of the electrochemical cell related to the behavior of an electrode (commonly called the working electrode) or the complete cell yields kinetic data; in addition, structural information can be derived from values of the double-layer capacity. Depending on the controlled experimental variable, methods are grouped into potentiostatic and galvanostatic methods. Besides, methods are also classified according to the recorded (dependent) variable, i.e., chronoamperometric in the former and chronopotentiometric in the latter case. Because in most cases the response of the system (i.e., the change in the dependent variable as a function of time after application of the change in the control variable) is recorded only for a short period of time, the methods in both groups are frequently called pulse experiments; more generally, they are also designated as step experiments. A further criterion particularly important in galvanostatic experiments is the state of the system before application of the stimulus (the current change): The system may be at open circuit, i.e., no current is flowing. In this case, a switch-on experiment is performed. Alternatively, the constant current applied for a time sufficient to establish steady-state conditions is interrupted: A switch-off or current interruption experiment is performed. Further, it is assumed that during the actual experiment the overall composition of the electrolyte solution is not changed to any significant extent. In the case of electrodes of only a few square centimeter active areas, cell volumes of several tens of milliliters will suffice this requirement. It is also assumed that only diffusion mass transport proceeds; convection (either that might be natural because of the density gradients after extended periods of operation and corresponding concentration and density changes in the solution, or might be artificial caused by, e.g., stirring) and migration (movement of charged species induced by electric fields, suppressed effectively by adding sufficient inert supporting electrolyte) are absent.

2

Potential step experiments

In an electrochemical system under open-circuit conditions, the relationship between the actual electrode potential and the concentration of electrochemically active species involved in establishing this electrode potential is given by the Nernst equation. Because the concentration of these species is constant in the system, no concentration gradient is established. In case the composition of the system does not support establishment of a well-defined electrode potential (e.g., only one component of a redox couple or only one reactant is present), the electrode potential may be established initially at a value at which no electrode reaction proceeds. In the example discussed following a species R which can be oxidized is present, the potential is kept at a value sufficiently negative where no oxidation takes place. At time zero, a sudden change in potential is applied; in this case, the new potential has to be in the range where the available reactant is consumed. The respective changes in electrode potential and current as a function of time, as well as the change in concentration, are depicted in Figs. 1–3. The current response to the changed electrode potential recorded as a function of time (thus the method is called chronoamperometry) can be rationalized by taking into consideration the effect of the changed electrode potential. When both redox species are present, the composition of the redox system close to the electrode surface must adapt to the new potential value; in the case discussed here, the amount of oxidized species O will simply increase by consumption of reduced species R proceeding through electrooxidation at the electrode. Thus, a positive current will flow. In addition, the electrochemical double layer has to be charged to a value in accordance with the new electrode potential – i.e., during a potential step in the positive direction, a positive charging current will flow. Because the charging current of the double layer is essentially capacitive (the more or less slight differences in behavior between an ideal condenser and the electrochemical double layer are of minor importance here), the initial charging current will be very large as given by Ic ¼ Cd A (dE/dt), where Cd is the double-layer capacity, A the electrode surface area, and dE/dt the rate of potential change. Because of the almost instantaneous potential change (requiring fast instrumentation), the current will be limited only by the uncompensated Ohmic resistance Ru of the electrolyte solution. The cell time constant given by CdRu is important in assessing the initial time span where capacitive currents are large enough to mask Faradaic currents. Usually, this is the case within 5CdRu. On the contrary, the Faradaic current will be limited by diffusion of both the consumed species (limited supply) and the generated species (slow removal by diffusion into the bulk of the solution). The concentrations of reactants

Methods and Instruments | Potential and Current Steps

Fig. 1 Change in electrode potential as a function of time.

Fig. 2 Change in electrode current as a function of time.

Fig. 3 Changes in concentration profiles as a function of time with t3 > t2 > t1.

21

22

Methods and Instruments | Potential and Current Steps

will change as a function of time as depicted. The evaluation of the potential response depends on the amplitude of the potential step, on the shape and size of the electrode (this is related to the type of diffusion, i.e., planar or spherical diffusion), and on the electrode kinetics (i.e., fast kinetics, also called reversible or Nernstian kinetics, or slow kinetics (sometimes further distinguished into (totally) irreversible and quasireversible kinetics)). Assuming in the first example that the electrode potential established after the step is in the range where all arriving species R will be consumed (diffusion-limited case), the concentration of R at the electrode surface will be zero. Species required for the electrode reaction will have to be supplied from the bulk of the solution. With time, the concentration gradient will extend further into solution, and it will become less steep (see Fig. 3). As this gradient constitutes the driving force for mass transport by diffusion, the rate of transport will decrease accordingly, and so will the current as depicted in Fig. 2. As a result, the local concentration and the concentration profiles of the consumed and the generated (not shown) species will change as depicted for the consumed species. Methods of evaluation of the obtained I–t curves depend on the experimental conditions: For a sufficiently large potential step, the diffusion-limited case applies; electron transfer kinetics are of no importance. For small steps, both redox forms (i.e., reactant and product) may be present, and the Butler–Volmer equation in its simplified linearized form may be applicable. In addition, the actual rate of electron transfer (as expressed by the rate constant k) has to be considered. For planar diffusion, the mass transport-limited current Id is given by the Cottrell equation I ðt Þ ¼ I d ðt Þ ¼

1=2

n  F  A  DR  c0R p1=2  t 1=2

where c0R is the concentration of the reacting species in the bulk of the solution, n is the electrode reaction valency, F is the Faraday constant, and DR is the diffusion coefficient of the reacting species. For a spherical electrode of radius r0, the equation changes to " # 1 1 0 + Id ðtÞ ¼ n  F  A  DR  cR r0 ðp  DR  t Þ1=2 For micro- and ultramicroelectrodes of different shapes, steady-state currents Iss can be obtained. For a disk electrode, it is Iss ¼ 4  n  F  DR  c0R  r 0 If potential steps with smaller amplitude toward electrode potential values at which the diffusion is not current-limiting are employed, the electrode kinetics should also be considered. Using the degree of coverage y h i nF y ¼ exp E − E0’ RT where E is the actual electrode potential and E00 the formal electrode potential, and assuming planar diffusion for a fast electrode reaction yields IðtÞ ¼

n  F  A  DR  c0R ð1 + yÞ  ðp  t  DR Þ1=2

and, for a slow reaction (with rate k of electrode reaction),   k2  t t 1=2  k IðtÞ ¼ n  F  k  c0R exp erfc 0 DR DR For short periods of time, the latter expression reduces to " IðtÞ ¼ n  F  A  k 

c0R

1-

2  k  t 1=2

#

1=2

p1=2  DR

For nonplanar diffusion, more complicated expressions have been derived. In analytical electrochemistry, double or multiple potential steps are employed. Thus, instead of recording the current as a function of time, the integrated current (i.e., the charge can be recorded using an electronic integrator; hence, the method is called chronocoulometry) is measured: QF ðtÞ ¼

1=2

2  n  F  A  DR  c0r  t 1=2 p1=2

Besides the charge QF associated with the Faradaic process, charging the electrochemical double consumes an additional amount of electricity QC. This contribution can easily be identified in a plot of Qtot ¼ QF + QC as a function of t1/2 (Fig. 4).

Methods and Instruments | Potential and Current Steps

23

Fig. 4 Plot of total charge Qtot vs. t1/2.

3

Current step experiments

A sudden change in the current driven through the electrochemical interface causes changes in the composition of the electrolyte solution depending on the polarity and magnitude of the current; accordingly, the electrode potential depending on this composition will change. The current step (perturbation) and a typical potential response are depicted in Figs. 5 and 6, respectively. Because of the continuous change in the electrode potential, the charging current Ic needed to keep the double-layer capacitance at its proper state will always consume part of the applied current I. Therefore, the faradic current. IF ¼ I − Ic Because the electrode potential is a function of time (dE/dt), both current components – the charging current and the Faradaic current – vary as a function of time. When the charge transfer reaction is fast and the charge transfer overpotential ct is low accordingly, a plot of E versus t1/2 yields an intercept equivalent to ct. Because consumption of the reacting species (in this example R) causes extension of the depletion region into the bulk of the electrolyte solution volume, a situation will be established where the surface concentration of R drops to zero; the corresponding time is called transition time t. The electric current still forced through the electrochemical interface is now maintained by another oxidation reaction, e.g., decomposition of the solvent. The transition time t and the current I are related by the Sand equation:

Fig. 5 Change in electrode current as a function of time.

24

Methods and Instruments | Potential and Current Steps

Fig. 6 Change in electrode potential as a function of time.

pffiffiffiffiffiffiffipffiffiffi pffiffiffi I t n  F  A  DR p ¼ 2 c0R The potential–time relationship for a fast charge transfer reaction is given by pffiffiffi pffiffi t- t RT E ¼ Et=4 + ln pffiffi nF t where Et=4 ¼ E00 -

RT D ln O DR 2nF

For slower reactions (quasireversible), the exchange current density influences the response    pffiffi  RT 2 t 1 1 1 pffiffiffi 0 pffiffiffiffiffiffiffi + 0 pffiffiffiffiffiffi + −¼ I F I0 F  A  p cO  DO cR  DR For small values of , a plot of  versus t1/2 yields a straight line with I0 as the intercept. To overcome the inherent limitations in determining ct especially for fast charge transfer reactions, the galvanostatic double-pulse method has been developed. In a sequence of two current pulses of the same sign but of significantly different current values and duration, the first short, high-current pulse serves mainly for charging the double layer to an electrode potential value roughly equivalent to the charge transfer overpotential, whereas the following longer pulse of much smaller current mostly maintains the Faradaic reaction. A typical current program and a schematic response for different ratios of current height of the first pulse are shown in Figs. 7 and 8, respectively.

Fig. 7 Change in electrode current as a function of time (double pulse program).

Methods and Instruments | Potential and Current Steps

25

Fig. 8 Change in electrode potential as a function of time.

The large charge in the first step (given by t1I1) results in a potential increase; the actual trace as depicted is frequently masked or distorted by electronic artifacts caused by the nonideal response of the employed instruments. At the end of the first step, the further change in the electrode potential depends on the amount of charge injected. In case of trace 1, the injected charge is too small, and the electrode potential increases immediately after the start of the second pulse. In case of trace 3, the charge is too large, and after the end of the first pulse the potential initially drops. Proper adjustment is achieved with the current I1 applied for trace 2. For sufficiently short times (to avoid excessive electrolysis besides the desired charging of the double layer), the overpotential  caused by the current I2 can be approximated by   RT 1 4  N pffiffiffiffi pffiffiffi t 1 +  I2 F j0 3 p where N represents diffusion coefficients and concentrations of reactants besides further details. Experiments are performed with various values of t1 and corresponding currents I1 are adjusted by trial and error; a plot of  at the beginning of t2 as a function of t1/2 1 and taking into account the electrode surface area yields as an intercept the exchange current density j0. Assuming that the charge in the first step is almost completely consumed for charging the double layer, the double-layer capacity can be determined according to   pffiffiffiffi−1 4  N  I0  t 1 F  t 1  I 0 I1 pffiffiffi Cd ¼ lim 1− t ð1Þ!0 R  T  A I2 3 p Because of the extrapolations, knowledge of diffusion coefficients for reactants and products is not required. Instead of current pulses, interruption of a galvanostatically applied current (current interruption or switch-off ) may be used to obtain information on electrode kinetics. The electrode potential decay is based on the discharge of the electrochemical double layer by the electrode process. The potential drop caused by the uncompensated electrolyte solution resistance disappears practically instantaneously, causing a sudden change in the observed electrode potential immediately after interruption, as depicted in simulated decay curves in Fig. 9.

Fig. 9 Change in electrode potential as a function of time after interruption of current I of different magnitudes.

26

Methods and Instruments | Potential and Current Steps

Depending on the current I forced across the interface before interruption, the decay is more or less steep; at higher currents I, the decay is more pronounced. From the linearized plots of the potential–time relation kinetic data, the exchange current density j0, in particular, can be derived.

4

Conclusions

Step methods employing instantaneous changes of electrode potential or current (pulses) are powerful method for the determination of structural and kinetic parameters of the electrochemical interface, i.e., of electrodes.

References 1. 2. 3. 4. 5. 6. 7.

Bard, A. J.; Faulkner, L. R. Electrochemical Methods, 2nd ed.; Wiley: New York, 2001. Delahay, P. New Instrumental Methods in Electrochemistry; Interscience Publisher: New York, 1962. The Southampton Electrochemistry Group. Instrumental Methods in Electrochemistry; Horwood Publishing: Chichester, 2001. Yeager, E.; Salkind, A. J. Techniques of Electrochemistry 1; Wiley: New York, 1972. Zoski, C. G., Ed. Handbook of Electrochemistry; Elsevier: Amsterdam, 2007. Bard, A. J., Stratmann, M., Unwin, P., Eds. Encyclopedia of Electrochemistry; WILEY-VCH: Weinheim, 2003. Kreysa, G., et al., Eds. Encyclopedia of Applied Electrochemistry; Springer: New York, 2014.

Methods and Instruments | Linear Sweep and Cyclic Voltammetry N Elgrishi, G Bontempelli, and R Toniolo, Louisiana State University, Baton Rouge, United States © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of G. Bontempelli, R. Toniolo, MEASUREMENT METHODS | Electrochemical: Linear Sweep and Cyclic Voltammetry, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 643–654, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00069-1.

1 Introduction 2 Overview of linear sweep and cyclic voltammetry 2.1 Basic description and cell components 2.2 Resulting data for a simple electron transfer process 2.3 How electrochemical reversibility impacts the data for a simple charge transfer process 2.4 Chemical reactions coupled to an electron transfer 2.5 Surface processes 3 Brief overview of applications 3.1 Analytical applications 3.2 Linear sweep voltammetry and cyclic voltammetry at microelectrodes 3.3 Linear sweep voltammetry and cyclic voltammetry application to battery studies 4 Conclusions Acknowledgments References

29 29 29 31 32 33 37 38 38 39 40 41 41 41

Abstract This chapter is an update/modernization of the chapter of the 1st edition by G. Bontempelli and R. Toniolo. It reports on the procedure adopted to perform linear sweep voltammetry (LSV) and cyclic voltammetry (CV) and summarizes the main information obtained using these electrochemical techniques. Several typical situations are described including of the presence of chemical steps coupled to the electron transfer steps affect the shape of voltammograms. Some major applications of LSV and CV to battery studies are also briefly summarized.

Glossary Capacitive current Component of the current flowing (at least transiently) at the electrode–solution interface that is not due to the occurrence of charge transfer reactions, but to the charge release or intake required by the electrode surface to compensate the charge variations caused by potential changes on the electrical double layer (i.e., the whole array of charged species and oriented dipoles existing in the solution layers close to the electrode surface). CE process Electrochemical process involving a chemical reaction preceding the charge transfer step that can be either reversible (CErev) or irreversible (CEirrev). Charge (or electron) transfer Step involving the transfer of electrons across the metal–solution interface. Charge transfer coefficient Parameter reflecting the symmetry of the activation energy barrier for the occurrence of the two half-reactions involved in an electron transfer. It depends on the experimental conditions, including the electrode potential and its value must lie between 0 and 1 (usually, it ranges from 0.3 and 0.7). Current function Current expressed as a function only of potential, i.e., in a dimensionless form making a single set of values suitable to provide voltammograms independent of the experimental conditions under which a specific redox process is tested. Diffusion coefficient It is a constant for each solute freely diffusing in solution, which comes out from the evaluation of the mean displacement of solute molecules by applying the usual probability distribution approach for a random walk process. Typical values of D for aqueous solutions are within 5
10−6cm2s−1. The displacement D of a dissolved species in the time t can be inferred from D by the equation D ¼ (2Dt)1/2. EC process Electrochemical process involving a charge transfer step followed by a chemical reaction that can be either reversible (ECrev) or irreversible (ECirrev). ECE process Electrochemical process involving a chemical reaction interposed between two subsequent charge transfers. Such a type of process is indicated as (ECrevE) or (ECirrevE) depending upon whether the chemical step is reversible or irreversible, respectively. Electrochemical process Overall process involving a charge transfer. It includes also possible chemical reactions and/or adsorptions accompanying the transfer of electrons across the electrode–solution interface. Faradaic current Component of the current that is due to an electron transfer across the electrode–solution interface.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00257-7

27

28

Methods and Instruments | Linear Sweep and Cyclic Voltammetry

Formal potential Potential measured for a redox couple when the concentration ratio of the partners is unity and other components of the medium are present in designated concentrations. Thus, it incorporates the standard potential E and some activity coefficients, as well as factors related to complexation (or other reactions subtracting at least one of the partners). Heterogeneous reaction This term is usually adopted for the transfer of electrons across the electrode–solution interface, with the meaning of charge (or electron) transfer. Homogeneous reaction Chemical reaction occurring in the solution phase. Pseudo-first-order conditions Experimental conditions allowing the rate of a homogeneous chemical reaction to depend on the concentration of only one of the reactants, because a much higher concentration of other reacting species is adopted that it remains unaltered during the reaction run. Reversibility degree The extent at which a redox process approaches the reversible behavior. Reversible (Nernstian) behavior Behavior of a redox system following the Nernst equation (or an equation derived from it), i.e., characterized by a very high value of the standard heterogeneous rate constant ksh that, at any potential different from the equilibrium potential, currents limited only by diffusion are recorded. Quasi-reversible behavior Intermediate behavior between reversible and totally irreversible behaviors. Totally irreversible behavior Behavior of a redox system characterized by a value of ksh low enough that currents can flow only when a significant overpotential is applied with respect to the equilibrium potential. In this case, currents limited by the electron transfer rate are recorded at overpotentials not very high.

Key points

• • • • • •

Understand the difference between LSV and CV. Overview of the electrochemical cell. Expected LVS and CV data for a Nernstian electron transfer. Impact of electrochemical reversibility on the data. Impact of the presence of chemical steps coupled to the electron transfer step. Brief overview of some relevant applications.

Symbols and units

a A Cdl C0 CZ DO DR E Ei Ep Epb Epf Ep/2 Et El E E1/2 F i ic if ilim ip ipb ipf

Normalized scan rate ¼ nFu/RT (s−1) Electrode area (cm2) Double-layer differential capacity (mA s V−1) Bulk concentration of the oxidized species (mol cm−3) Bulk concentration of a non-electroactive analyte (mol cm−3) Diffusion coefficient of the oxidized species (cm2 s−1) Diffusion coefficient of the reduced species (cm2 s−1) Electrode potential (V) Initial potential (V) Peak potential (V) Backward peak potential (V) Forward peak potential (V) Half-peak potential Instantaneous potential applied at time t (V) Switching potential (V) Standard potential (V) Half-wave potential (V) Faraday constant (charge on 1 mol of electrons) ¼ 96.487C mol−1 Current (mA) Capacitive current (mA) Faradic current Kinetically controlled plateau current (mA) Peak current (mA) Backward peak current (mA) Forward peak current (mA)

Methods and Instruments | Linear Sweep and Cyclic Voltammetry

kb kf k0 f ksh K n q R t tl T y a x(at) DEp g C

29

Homogeneous rate kinetic constant for backward reactions (s−1) Homogeneous rate kinetic constant for forward reactions (s−1) Pseudo-first-order rate kinetic constant for forward reactions ¼ kfCZ (mol cm−3 s−1) Standard heterogeneous rate constant (m s−1) Equilibrium constant for homogeneous reactions Overall number of electrons involved in a charge transfer Charge (C) Ideal gas constant ¼ 8.31 J K−1 mol−1 Time (s) Switching time (s) Absolute temperature (K) Scan rate (V s−1) Charge transfer coefficient Current function Peak potential separation ¼ Epb − Epf (V) (DO/DR)1/2 Dimensionless kinetic parameter for charge transfers ¼ gaksh/(paDO)1/2

Abbreviations CV DL HPLC LSV

1

Cyclic voltammetry Detection limit High-performance liquid chromatography Linear sweep voltammetry

Introduction

Linear sweep voltammetry (LSV) and cyclic voltammetry (CV) are the most widely used voltammetry techniques for studying redox reactions of both organic and inorganic compounds. This is because they are unmatched in their ability to provide information on the steps involved in electrochemical processes with only a modest expenditure of time and effort in the acquisition and interpretation of data. These electroanalytical methods require rather simple and inexpensive instrumentation and provide information not only on the electrochemical quantities typical of a redox process, but also on possible chemical reactions coupled with charge transfer steps. This is because the electrode can be used as a tool for producing reactive species in a small solution layer surrounding its surface and at the same time for monitoring chemical reactions in which they are involved. Moreover, as the relevant responses can be obtained within a few milliseconds after stimulation of the electrode, they may be used for studying even very fast reactions, thus allowing detection of short-lived transient intermediates. The LSV and CV techniques were proposed in the early 1950s, together with some theoretical approaches for rationalizing the simplest responses, and several reviews and overview papers were published around the 1980s.1–7 However, the use of these electroanalytical methods has received considerable renewed interest more recently, thanks to increased knowledge of more subtle criteria for interpreting the relevant responses and to greater availability of theoretical tools for processing experimental data.8–11 This article briefly summarizes the performance of these techniques, as well as the criteria followed to gain information on participants in electrochemical processes from the recorded responses. More information on theoretical and practical aspects can be found in other works.8,12–15

2 2.1

Overview of linear sweep and cyclic voltammetry Basic description and cell components

In both LSV and CV, the potential of a working electrode, E, is controlled and changed linearly over time while the resulting current response, i, is recorded. The absolute value of the rate at which the potential is changed over time is called the scan rate or sweep rate, u, with units in V s−1. In LSV, the potential is changed to be either more and more oxidizing (anodic sweep) or more reducing (cathodic sweep) over time. At any given time point t during the LSV experiment, the potential applied to the working electrode E applied at time t is given by Eq. (1): Et ¼ Ei  ut

(1)

Methods and Instruments | Linear Sweep and Cyclic Voltammetry

Applied Potential

30

Ef

Ei

Applied Potential

Time Eλ

Ei tλ Time

Fig. 1 Evolution of applied potential over time for LSV (top) and CV (bottom).

in which Ei is the initial potential. The scan rate u is constant during the sweep and corresponds to dE/dt. The voltage waveform applied to the working electrode in LSV and CV is shown in Fig. 1. In CV, the potential scan direction is varied linearly with time exactly as in LSV until a switching potential El is reached at time tl, after which the scan direction is reversed, such that an anodic sweep is performed immediately following the cathodic sweep (or vice versa). The switching potential El is both the final potential for the forward scan and the initial potential for the reverse sweep. As a result, the instantaneous potential over time takes the same form as LSV in the forward sweep, and in the return sweep it takes on the slightly more complex form in Eq. (2): Et ¼ Ei  ut l  uðt − t l Þ ¼ Ei  2ut l  ut

(2)

where upper signs refer to forward anodic sweeps and lower signs refer to forward cathodic sweeps. This equation applies to symmetrical triangular waveforms, which are the most commonly adopted in CV, even though other waveforms have also been proposed. The rest of the chapter will primarily focus on CV which is more widely used. The electrochemical cell for CV contains three electrodes: a working electrode where the reaction of interest occurs, a counter electrode to close the electrical circuit, and a reference electrode to measure the potential and convert it to an absolute scale. The electrodes are placed in a solution containing the solvent of choice, the dissolved analyte under study, and typically a dissolved salt called supporting electrolyte. The large excess of supporting electrolyte represses migration of charged reactants and products. The working electrode, with a typical surface area 7, quasi-reversible for 7 > C > 10−3, and totally irreversible for C < 10−3. Consequently, the same redox system displays a reversible behavior for high-enough C values (low values of u, when the system is allowed a long time to attain equilibrium conditions). As the scan rate is increased the system can become quasi-reversible for intermediate values of u, and it becomes totally irreversible at sufficiently high scan rates. Such a transition is shown in Fig. 5 reporting the effect of C on the response for a one-electron redox system characterized by a ¼ 0.5. In this figure, normalized current–potential plots are indeed reported, to make voltammograms independent of the experimental variables typical of the particular redox process considered. Thus, current is replaced by the dimensionless current function p1/2w(at), that is, a function of both a (related to the scan rate) and t (related to the potential), whose generalized form is given as Eq. (6): 1/2

p1=2 wðat Þ ¼ iðatÞ=nFAC0 ðDO Þ1=2 a1=2 a1=2

(6)

0

where i(at) is the current, A the electrode area, C the bulk concentration of the analyte, and other symbols have the meaning defined earlier. For reversible electrode reactions, a unity value must be inserted in this equation for the charge transfer coefficient a, even though such a parameter makes no sense in this case. This current function exhibits a maximum value (peak current function) of 0.4463 for reversible charge transfers and of 0.4958 for totally irreversible redox reactions, but the higher value for this last case must not be misleading, because from Eq. (6) the peak-current ratio in Eq. (7) is inferred for these two limiting cases.     (7) ip irrev = ip rev ¼ ð0:4958=0:4463Þa1=2 Thus, this peak-current ratio becomes 0.7855 when a ¼ 0.5, as frequently occurs, pointing out that under the same experimental conditions higher peaks are found for Nernstian charge transfers, in agreement with the progressive peak-lowering and peakbroadening accompanying the decrease of the reversibility degree (C) apparent in Fig. 5. Finally, the linear dependence of the peak current on both the analyte concentration and the square root of u must be emphasized. The dependence on the analyte concentration is the basis of the application of LSV to quantitative analysis, whereas the dependence on u1/2 suggests the possibility of modulating suitably the sensitivity of LSV and CV measurements. Table 1 summarizes the main characteristics of LSV and CV responses for the different types of uncomplicated charge transfers, as well as their trend with u, whose suitable use enables their characterization and evaluation of their typical quantities (ksh, a, and E1/2).

2.4

Chemical reactions coupled to an electron transfer

Reactants and products of electrode processes may be involved in chemical reactions following, preceding, or parallel to electron transfers. Their effect on voltammograms depends on both their nature and their rate, compared with the timescale of the experiment.16 Three common scenario are briefly discussed below: a CE process in which chemical step (C) is followed by and

34

Methods and Instruments | Linear Sweep and Cyclic Voltammetry

π1/2Χ(at) 0.3 0.2

0.1 –0.4

–0.2

E – E⁰ (V) 0.2

0.4

–0.1 –0.2 –0.3 Ψ = 10

–3

Ψ = 10–2 Ψ = 10–1

–0.4

Ψ = 108 Fig. 5 Effect of the dimensionless parameter C on CV responses for an uncomplicated one-electron system characterized by a ¼ 0.5. The starting species is the oxidized one.

Table 1

Trend of the parameters characterizing LSV and CV responses for uncomplicated electrode processes.

Parameter

Reversible process

Totally irreversible process

Quasi-reversible process

ip/u1/2

Independent of u; 0 ip/u1/2 ¼ 2.688
108n3/2 AD1/2 0 C  at 25 C Unity and independent of u Independent of u

Independent of u; 0  ip/u1/2 ¼ 2.987
108a1/2n3/2AD1/2 o C at 25 C

Slightly dependent on u; it changes from 1.00 to 1.11a1/2 for about a 108-fold increase in u

ipb/ipf Epa Ep − Ep/2a

Independent of u and equal to: −2.199RT/nF mV (−56.5/n mV at 25  C)

Epb−Epfa

Independent of v and equal to: 2.218RT/ nF mV (57.0/n mV at 25  C)

i at the foot of the peak

It depends linearly on u1/2

Lower than 1 Unity only for a ¼ 0.5 Dependent on u; it shifts cathodically by 1.15RT/a Dependent on u; for a ten-fold increase in u the nF mV for a ten-fold increase in u (29.5/an mV cathodic shift is smaller than 1.15RT/anF mV at 25  C) Independent of v and equal to −1.857RT/anF mV Slightly dependent on u; it changes from −2.199RT/nF mV to −1.857
RT/anF mV (−47.7/an mV at 25  C) (from −56.5/n to 47.7/an mV at 25  C) for about a 108-fold increase in u Dependent on v according to: Dependent on u DEp ¼ {RT/[a(1 −a)nF]}[0.780 − lnksh + ln(DOa)1/2] + (RT/nF) [ln(1 − a)[1/2(1−a)] + lna(1/2a)] Independent of u It depends on u½2 in a nonlinear way

a

All trends indicated refer to cathodic processes; anodic processes are characterized by symmetrical shifts and dependences.

electron transfer (E) (Scheme 1b), an EC process in which an electron transfer is followed by a chemical step (Scheme 1a), and finally a simple electrocatalytic process EC’ in which the chemical step regenerates the starting analyte (Scheme 1c). In the EC scheme, the existence of the follow up chemical step influences the concentration of R. As R is consumed in the chemical step at each potential the extend of reduction of O increases following the Nernst equation. In the case of an irreversible chemical step, this leads to a voltammogram exhibiting an irreversible reduction wave, shifted towards more positive potentials in the case of a reduction. The evolution of the chape of the voltammogram as a function of chemical step rate constant, scan rate and equilibrium constant can be given in “zone diagrams”. See reference 17 for further reading on the topic. The scan rate is a critical parameter to affect the time scale of the experiment. In particular, at scan rates so high as to prevent the occurrence of chemical reactions in the time scale of the experiment, CV responses remain unaltered by the chemical step. Conversely, chemical reactions occurring at rates higher than the potential scan result in potential shifts of CV profiles. Under intermediate conditions, when

Methods and Instruments | Linear Sweep and Cyclic Voltammetry

35

a) EC O + n e– kf

R

kb

R

E1/2

Z

K

b) CE kf

Z

kb –

O+ne

O

K

R

E1/2

R

E1/2

c) EC’ O + n e– R+Z

kf

O+P

Scheme 1 Three common schemes: (a) EC, (b) CE and (c) EC’.

π1/2Χ(at)

kf/a = 500

10 ≤0.1

0.4

0.0 0.1 0.01

180

120 60 0 n(E – E1/2) (mV)

–60

Fig. 6 Cyclic voltammograms for first-order irreversible chemical step (ECirrev) processes characterized by the indicated values of the dimensionless kinetic parameter kf/a.

chemical reaction rates are comparable with the sweep rate, voltammograms are affected by both kinetic and thermodynamic characteristics of coupled reactions, so that kinetic investigations by LSV and CV become particularly profitable. Example normalized voltammograms are shown in Fig. 6 for an ECirrev electrochemical process, i.e., the case in which the chemical step is irreversible following the steps: O + ne− >R kf

R!Z The normalized voltammograms are shown with different values of kf/a, in a range so wide as to achieve a complete transition from one to the other limiting situation. In the case of the irreversible CV response in which the peak potential is shifted significantly (past the initial E1/2), the peak potential position can be used to extract kinetic information from the data following Eq. (8):17–19   kf RT RT RT EP ¼ E1=2 − (8) ð0:78Þ + ln nFu nF 2nF For cases where the follow up chemical step is reversible, ECrev, some key parameters of the expected voltammograms are summarized in Table 2.

36

Methods and Instruments | Linear Sweep and Cyclic Voltammetry Characteristics of CV responses for ECrev processes.

Table 2 Experimental conditions

Reversible chemical step (ECrev) O + ne− > R kf R > Z (K ¼ kf/ kb > > 1) kb

kf/a ! 0 Comparable values of kf and a kf/a > > 1 kb/a < > 1

Uncomplicated reversible process. [p1/2w(at)]p ¼ 0.4463; n(Ep/2 − Ep) ¼ 56.5 mV at 25  C; dEp/dlogu ¼ 0; ipb/ipf ¼ 1 [p1/2w(at)]p: decreases from 0.4958 to 0.4463 at 25  C; n(Ep/2 − Ep): increases from 48.0 to 56.5 mV at 25  C; dEp/dlogu: ranges from −29.6/n to 0 mV at 25  C; ipb/ipf: increases from 0 to 1 Superreversible process [p1/2w(at)]p ¼ 0.4958; n(Ep/2 − Ep) ¼ 48.0 mV at 25  C; ipb/ipf ¼ 0; dEp/dlogu ¼ −29.6/n mV at 25  C [p1/2w(at)]p: increases from 0.446 to 0.4958; n(Ep/2 − Ep): decreases from 56.5 to 48 mV at 25  C; dEp/dlogu: ranges from 0 to −29.6/n mV at 25  C; ipb/ipf: decreases from 1 to 0 Kinetically uncomplicated reversible process. The response is anodically shifted of an extent: (RT/nF) ln(1 + K) with respect to E0O/R

Reproduced from Bontempelli, G., Toniolo, R. Linear sweep and cyclic voltammetry. In: Worsfold, P.J., Townshend, A., Poole, C.F. (eds.) Encyclopedia of Analytical Science, 2nd edn., vol. 9, 2005, pp. 188–197. London: Elsevier.. The potential scan rate u is decreased progressively from the top to the bottom of the table.

Similarly, in the CE scheme (Scheme 1b) the impact of the chemical step on the electron transfer varies depending on the scan rate and rate or equilibrium constant of the chemical step. In the extreme in which the chemical step is fast and irreversible, the analyte is fully transformed from Z to O before the start of the experiment and the system behaves as a simple reversible 1 electron transfer. In the other extreme of a small equilibrium constant, the amount of O available for reduction at the electrode depends on the equilibrium constant. Similar to Figs. 6 and 7 illustrates the different responses recorded for CErev electrochemical processes, when the normalized scan rate a is changed in a range so wide as to achieve a complete transition from one to the other limiting situation. The characteristics displayed by the responses for the types of usual processes more frequently encountered are reported in Table 3 summarizing the criteria adopted for characterizing the process involved and for evaluating the relevant kinetic quantities.20 Another common electrochemical system involves a simple electrocatalytic process (Scheme 1c). It involves the homogeneous reaction of the reduced product R of a reversible charge transfer with a non-electroactive substrate Z, allowing regeneration of the electroactive species O and the concomitant formation of a non-electroactive product P. The condition CbZ  CbO is usually adopted because it makes responses easier to be interpreted, thanks to the attainment of pseudo-first-order conditions. This allows a pseudofirst-order rate constant kf0 ¼ kfCbZ to be adopted. To account for the effect of the homogeneous regeneration reaction on the CV profiles, it is convenient to compare this rate constant with the normalized potential scan rate by adopting the dimensionless kinetic parameter kf’/a. Two limiting cases can be recognized. Similar to the EC process discussed above, in the case where scan rates are so high as to make kf’/a very low, the chemical reaction cannot occur significantly in the timescale of the experiment, thus leading to responses typical for uncomplicated charge

K = 10–2

K = 10–5 (ka + kb)/ = 109

K = 10–4 (ka + kb)/ = 10

8

0.3

0.03

π1/2Χ(at) (ka + kb)/ = 10–3

0.2

×10–2

0.2 ×10–4

1

0.1

×10–6 1

1

0.02

0.1

1

1

0.01

0.5 0.1

0

0 10

0 –10 –20 –30 (nF/RT)(E – E1/2)

0.1

0.07

0

0 10

0 –10 –20 –30 (nF/RT)(E – E1/2)

0

0 10

0 –10 –20 –30 (nF/RT)(E – E1/2)

Fig. 7 Voltammograms for reversible electrode step (CErev) processes calculated for the indicated values of the equilibrium constant K and different values of the dimensionless kinetic parameter (kf + kb)/a. In all cases, the scale at the right refers to the curve calculated for the highest (kf + kb)/a value, whereas the scale at the left-hand side is relative to other voltammograms.

Methods and Instruments | Linear Sweep and Cyclic Voltammetry Table 3

37

Characteristics of Voltametric responses for CE processes.

Experimental conditions

Reversible electrode step (CErev) kf

Irreversible electrode step (CEirrev) kf

Z>O

Z>O

(K ¼ kf/ kb)

kb O + ne− > R kf/a ! 0 (or kf/aa!0) Comparable values of kf and a (or aa) (Steady state for the concentration of species O) kf/a> > 1 (or kf/aa> > 1)

(K ¼ kf/ kb)

kb ksh

O + ne− ! R

Uncomplicated totally irreversible process O/R; ip depends Uncomplicated totally irreversible process O/R; linearly on CbO ip depends linearly on CbO Pure kinetic process Sigmoidal shaped response with a limiting current independent of u and equal to: i ¼ nFAD1/2 O Cb K (kf + kb)1/2 dEp/2/dlogu ¼ 29.6/n mV at 25  C; ipb/ipf > 1; Uncomplicated reversible process Z/R. (E1/2)Z/R ¼ (E1/2)O/R − Uncomplicated totally reversible process Z/R. (RT/nF)lnK/(1 + K); ip depends linearly on Cb (¼CbO + CbZ) (E1/2)Z/R ¼ (E1/2)O/R − (RT/anF)lnK/(1 + K); ip depends linearly on Cb (¼CbO + CbZ)

Reproduced from Bontempelli, G., Toniolo, R. Linear sweep and cyclic voltammetry. In: Worsfold PJ, Townshend A, and Poole CF (eds.) Encyclopedia of Analytical Science, 2nd edn., vol. 9, 2005, pp. 188–197. London: Elsevier. https://doi.org/10.1016/B0-12-369397-7/00647-6.

transfers. At the second limit, for u so slow as to make kf’/a high, the regeneration reaction is allowed to occur quantitatively during the potential sweep and the response is no longer peak-shaped but assumes a sigmoidal shape, as shown in Fig. 8. This is because a steady state for the electroactive species is attained by mutual compensation of its subtraction by the electrode reaction with its chemical regeneration. Under these kinetic conditions, the plateau current attains the limiting value given in Eq. (9) for high-enough overpotentials: qffiffiffiffiffiffiffiffiffiffiffi ilim ¼ nFACO DO k0f (9) Such a current, independent of u, enables the homogeneous kinetic constant to be evaluated by using its ratio with the corresponding diffusion-controlled current (Eq. 4), which can be recorded either in the absence of the substrate Z or at scan rates high enough to cancel the effect of the chemical step. In such a way, the knowledge of all the experimental quantities accompanying kf’ in equation (9) is not required and the analysis simplifies to Eq. (9): sffiffiffiffiffiffiffiffiffiffiffiffi 0 ilim RT kf ¼ (10) ip F u With respect to CV responses, they display the backward peak typical of uncomplicated reversible processes only in the first limiting case (kf’/a ! 0), whereas no backward peak is found for high values of kf’/a. Once again, intermediate situations are encountered for intermediate values of kf’/a.

2.5

Surface processes

In some electrochemical processes, reactants, intermediates, or products can be confined onto either the electrode surface (e.g., adsorbed species, oxide layers, covalently attached species, redox polymer films) or even the electrode material itself, as in electrodeposition and electro-dissolution. Surface-involving processes cause different effects on voltammograms, depending on their characteristics (e.g., adsorption isotherm, adsorption kinetics, electron transfer coupled to adsorption), so that a generalized treatment is rather problematic. Therefore, only a brief survey of surface processes is reported here, aimed at recognizing their presence and the reader is directed to further resources for more special cases.13 A rough distinction can be based on the degree of interaction with the electrode surface, in that weakly adsorbed species cause only enhancement of peak currents. Thus, LSV and CV forward peaks higher than those expected for uncomplicated charge transfers (Eq. 4) are found when the weakly adsorbed species is the reactant, whereas only enhancement of CV backward peaks is recorded for weak adsorption of the electrode product. On the contrary, a separate adsorption peak is displayed before or after the diffusion-controlled peak when redox products or reactants, respectively, are strongly adsorbed. These adsorption-controlled peaks can be identified because they are symmetrical about ip, unlike diffusion-controlled peaks, and a linear dependence of their height with CO is usually observed in a narrow range of low concentrations alone, whereas a constant value for ip is attained at higher CO. Moreover, ip for adsorption peaks increases linearly with u, instead of u1/2 (see Eq. 4), because the electrode–solution interphase displays in the presence of adsorbed species the capacitive current given in Eq. (11).

38

Methods and Instruments | Linear Sweep and Cyclic Voltammetry

π1/2Χ(at) 1.0

kf’/a = 1.0

0.8

0.6

0.6 0.2 0.4 0.01 0.2

0.0 120

60 0 –60 n(E – E1/2) (mV)

Fig. 8 Theoretical linear sweep voltammetry (LSV) curves for electrocatalytic processes involving reversible charge transfers under pseudo-first-order conditions. Potential scale is in millivolts.

ic ¼ dq=dt ¼ Cdl A ðdE=dtÞ ¼ Cdl Au

(11)

where Cdl is the double-layer differential capacity depending on both the electrode material and the solution composition, as well as on the electrode potential and the analyte concentration, although moderately. Comparison between Eqs. (11) and (4) points out that capacitive currents are usually negligible with respect to faradaic currents at high-enough analyte concentrations, but they prevail at very high u.

3 3.1

Brief overview of applications Analytical applications

Analytical applications of LSV and CV include characterization and quantitative determination of electroactive analytes. In general, the potential provides information on the identity of the analyte, while the current provides information on the concentration. Both LSV and CV are used to identify the presence of analytes in solution or on an electrode surface. Characterization is mainly based on the response position on the potential scale usually estimated by LSV, but CV provides some peculiar additional information about the reversibility of the charge transfer and the presence of adsorptions or coupled chemical reactions. Conversely, quantitative determinations, based on the response height, are usually performed by resorting to sole LSV, as no additional information is provided by CV. The detection limit (DL) in LSV is mainly governed by the ratio of the faradaic signal to charging current background. By restricting the considerations to uncomplicated charge transfers, faradaic currents can be written as if ¼ kCu1/2 (Eq. 6), while charging currents can be written as ic ¼ k0 u (Eq. 11), where k and k0 are suitable constants. Their ratio if/ic ¼ k/k0 Cu−1/2 points out that the less favorable DL is achieved at slow u and high concentrations. Typical scan rates used for analytical purposes are in the range of 10–1000 mV s−1 and offer acceptable compromise for achieving close to optimum performance. Under such conditions, approximately an order of magnitude improvement in DL is gained over conventional dc voltammetry, so that concentrations as low as 10−6 mol L−1 can be determined by LSV in aqueous media. More complicated considerations must be adopted for analytes undergoing adsorption processes. In most of these cases, capacitive currents lead, however, to profitable determinations characterized by low DL (up to 10−8 mol L−1). The analyte limit of detection and quantification is dependent on the signal to noise ratio which here takes into account the faradaic to capacitive current ratio. Methods such and stripping voltammetry have been developed to further increase the signal and improve DL.13

Methods and Instruments | Linear Sweep and Cyclic Voltammetry

39

i (Ep)2

(Ep)1

(ip)2 Extrapolation required

(ip)1

E Fig. 9 Voltammogram relative to the reduction (or oxidation) of the analyte 2 in the presence of another species 1 more easily reduced (or oxidized) when a fairly little difference exists between (Ep)2 and (Ep)1.

As far as the effect of the reversibility degree on sensitivity is concerned, despite the dependence of ip on charge transfer kinetics (Eqs. 6 and 8), it is very moderate because LSV peak heights are almost independent of the charge transfer rate. Indeed, only a decrease of 25% is found on passing from a Nernstian to a totally irreversible process with a ¼ 0.5. In routine analytical work, the main advantages of LSV are (1) increased sensitivity, (2) quickly recorded current–potential curves, (3) specificity attributable to both a large available timescale and improved resolution of peak-shaped curves. This last advantage is, however, rather questionable in some cases (see Fig. 9) because the presence of quite close preceding peaks makes the identification of the correct baseline difficult. A recent advanced analytical application of LSV and CV is their introduction in electrochemical detectors for flow analysis (HPLC, CE, etc.). Fast-scan LSV and CV (20–1000 Vs−1), starting at the rising portion of the typical peaks provided by these hydrodynamic methods and recorded at microelectrodes suitably positioned at the outlet of the flowing system, make the achievement not only of additional information for analyte characterization possible, but also of additional resolving power. Thus, for instance, peaks separated incompletely in time can be resolved voltammetrically.

3.2

Linear sweep voltammetry and cyclic voltammetry at microelectrodes

A recent development in LSV and CV has been achieved by the introduction of microelectrodes, which are metal or carbon electrodes with diameters usually in the range 0.1–100 mm, though smaller electrodes have been reported for more specialist applications. These electrodes were originally developed for in vivo biological and medical measurements.21 The major benefit is that the decrease in surface area results in improved performance due to the expansion of the inherently small diffusion layer on the timescale of the experiment, which is greater than the dimensions of the electrode. Consequently, relatively large diffusion layers develop some time after current starts, so that more electroactive particles reach the electrode per both unit time and surface area than in the case of conventional electrodes. Thus, although currents are indeed lower at microelectrodes, current densities are higher by several orders of magnitude compared to those at conventional electrodes, so that the faradaic-to-capacitive current ratio is drastically improved, allowing lower detection limits (10−8−10−9 mol L−1) to be attained. Moreover, microelectrodes enable high-quality voltammograms to be recorded even in poorly conducting media (e.g., gas phases or nonpolar solvents), thanks to the small currents (in the nano- and picoampere range) making negligible the relevant ohmic drop. They are virtually nondestructive of the species analyzed, thus becoming suitable for refined applications, such as in vivo measurements, lithography, and electrochemical scanning tunneling microscopy. Microelectrodes require very effective shielding and instrumentation designed suitably to avoid ripple and noise effects, which are very critical with small currents. When CV are recorded at stationary microelectrodes with slow u, sigmoidal current–voltage curves are found in both forward and backward scans, which are usually coincident with each other, except for processes involving coupled chemical reactions that display more or less marked hystereses. This sigmoidal shape (steady-state current) can be accounted for by considering the radial diffusion to the edges of ultramicroelectrode surfaces, which is very important at slow u, so as to make the diffusion rate of analyte molecules to the electrode surface comparable with the charge transfer rate. On the contrary, when fast scan rates are used to record cyclic voltammograms at microelectrodes, the radial diffusion to the edges of electrode surfaces becomes negligible and the majority of the diffusion is perpendicular to the electrode surface (usual planar diffusion), thus leading to peak-shaped responses similar to those at conventional electrodes. However, the little current flowing at these small electrodes result once more in a very small ohmic drop and this profitable advantage, accompanied by the very favorable ratio between faradaic and capacitive current mentioned earlier, permits the use of extremely higher scan rates. Thus, while at conventional electrodes u of hundreds of volts per second cannot be overcome, microelectrodes enable sweep rates up to 105 V s−1 to be used, allowing shorter timescales to be explored and faster reactions to be studied.

40 3.3

Methods and Instruments | Linear Sweep and Cyclic Voltammetry Linear sweep voltammetry and cyclic voltammetry application to battery studies

The advent of miniaturized electronic equipment, such as for cell phones and wearable devices, has increased the use of batteries and the demands for them. In addition, the search for new energy sources has created renewed interest in the research and development of energy storage devices such as batteries and fuel cells. These devices are electrochemical cells designed to provide a source of current and voltage, so that their testing and evaluation involves a variety of activities that can be profitably conducted by resorting to voltammetry techniques such as LSV and CV. These methods help investigate electrode materials, study compatibility among electrode couples, gather energy density information, and develop charge-discharge testing protocols. A systematic and complete description of the application of LSV and CV to battery studies is rather problematic because several and quite different subjects fall under the purview of the estimation of battery efficiency, even though most of them pertain to the electrochemistry of surface processes. The main aspects relevant to battery performance that can be profitably investigated by LSV and CV are briefly discussed below:

•

Emf evaluation for batteries assembled by using either unconventional electrode materials (e.g., polyconjugated polymers with tailor-made properties) or micro- and nanostructured electrodes prepared with the aim of taking advantage of both their definite morphology and the effect caused by the addition of small amounts of specific metals or other conductive materials such as carbon and graphite

Such an evaluation can be simply performed by recording repetitive LSV or CV profiles. Some important redox characteristics can be estimated through these measurements, using criteria similar to those reported earlier. These include for example electrical capacitance vs potential relationships, charge–discharge curves, reversibility of battery electrode materials, or transport properties. In these cases however, the interaction of reactants, intermediates, and products with the electrode surface must be taken into account. These LSV and CV investigations are frequently coupled to other experimental methods such as scanning probe microscopy measurements, galvanostatic charge–discharge experiments, quartz crystal oscillator determinations, or impedance spectroscopy detections. This helps provide integrative information on some important phenomena, e.g., hysteresis effects involved when systematic variations of positive and negative reversal potentials are applied, as well as kinetic and transport properties of thin films deposited onto the electrodes.

•

Evaluation of the effect of impurities and of the performance of battery electrolytes, as well as characterization of expander materials

Information on the effect of impurities can be once again achieved from base voltammetry studies, which make it possible to detect either additional peaks located in specific potential regions or the suppression of peaks that are usually present instead. Similarly, i–E curves provide great insights into electrode reactions and reaction rates involving electrode materials in the presence of different electrolytes (see previous sections). In particular, CV can be effectively adopted to characterize solid and gel-type polymer electrolytes, which are the subject of an ever-growing interest for the research and development of thin-film solid-state batteries. Expander materials also benefit from CV and LSV studies. Exander materials are mixtures of inorganic and organic substances added frequently to the negative plate of lead–acid batteries with the aim of increasing the apparent volume of the active material, so as to achieve increased performance at high discharge rates and cyclability. Characterization can be reliably performed by cycling repeatedly the electrode potential, in suitable ranges, in the presence and in the absence of the expander. The comparison of the two trends observed for voltammograms thus recorded makes estimation of the expander effectiveness possible, because the slope of the relevant voltammograms is affected by the porosity and thickness of the resistive lead sulfate layer formed during the cycling process. Most of the information gained on these topics is based on LSV and CV responses relative to surface processes discussed earlier.

•

Acquisition of information about hydrogen and oxygen adsorption on electrode materials

The LSV and CV profiles recorded for the adsorption–desorption of hydrogen at different electrode materials depend on the scan rate use in the voltammetry study, as well as on the electrolyte composition, surface preparation, and crystallographic plane exposed. The number of peaks observed helps diagnose that hydrogen presents different adsorption energies and interaction with neighboring adsorbates on the electrode surface. Thus, the evaluation of the electric charge of hydrogen adsorbed in each of the corresponding potential regions provides careful information about the attained hydrogen coverage and the state of the surface as a function of potential. This is critical information in connection with both fuel cells (control of the possibility of preparing active surface of metal electrocatalysts with reproducible procedures) and hydrogen storage in multicomponent alloys planned to assemble potential energy storage devices for portable applications that involve electrochemical processes taking place for adsorbing–desorbing hydrogen in the bulk of the alloy. Similar remarks can be made about the adsorption of other potential fuels such as methanol or carbon monoxide, as well as for the cathode coverage with oxygen.

•

Recognition of electrocatalytic processes involved in electrolyzers and fuel cells for both the oxidation and reduction of small molecules (e.g., hydrogen, carbon monoxide, methanol, formic acid, oxygen)

Several important fuel cell reactions occur through electrocatalytic processes of the type reported in Scheme 1, either with homogenous catalysts, or with a heterogeneous reaction of the product of the charge transfer with the non-electroactive substrate Z (fuel species), allowing regeneration of the electroactive species and the concomitant formation of the product P. The electroactive mediator can thus be in solution or a species immobilized onto the electrode surface (e.g., metal adlayers conveniently deposited onto the electrode, adsorption layers, conductive polymers). This growing field uses LSV and CV to determine kinetic and thermodynamic information about the electrocatalytic processes, and a full discussion is beyond the scope of this overview.9

Methods and Instruments | Linear Sweep and Cyclic Voltammetry

4

41

Conclusions

In conclusion cyclic voltammetry and linear sweep voltammetry are powerful techniques to study processes involving electron transfers and chemical steps. These are fundamental to energy storage technologies. Theoretical frameworks have been developed to analyze voltammograms and extract kinetic and thermodynamic information. More specific equations and analysis methods are provided in the references cited for this work and are beyond the scope of this short chapter.

Acknowledgments NE acknowledges support from the National Science Foundation (award numbers 2119435 and 2046445) as well as from Louisiana State University through the College of Science and the Office or Research and Economic Development.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

Evans, D. H.; O’Connell, K. M.; Petersen, R. A.; Kelly, M. J. Cyclic Voltammetry. J. Chem. Educ. 1983, 60 (4), 290. https://doi.org/10.1021/ed060p290. Faulkner, L. R. Understanding Electrochemistry: Some Distinctive Concepts. J. Chem. Educ. 1983, 60 (4), 262. https://doi.org/10.1021/ed060p262. Moran, P. J.; Gileadi, E. Alleviating the Common Confusion Caused by Polarity in Electrochemistry. J. Chem. Educ. 1989, 66 (11), 912. https://doi.org/10.1021/ed066p912. Baca, G.; Dennis, A. L. Electrochemistry in a Nutshell A General Chemistry Experiment. J. Chem. Educ. 1978, 55 (12), 804. https://doi.org/10.1021/ed055p804. Kissinger, P. T.; Heineman, W. R. Cyclic Voltammetry. J. Chem. Educ. 1983, 60 (9), 702. https://doi.org/10.1021/ed060p702. Mabbott, G. A. An Introduction to Cyclic Voltammetry. J. Chem. Educ. 1983, 60 (9), 697. https://doi.org/10.1021/ed060p697. Birss, V. I.; Truax, D. R. An Effective Approach to Teaching Electrochemistry. J. Chem. Educ. 1990, 67 (5), 403. https://doi.org/10.1021/ed067p403. Elgrishi, N.; Rountree, K. J.; McCarthy, B. D.; Rountree, E. S.; Eisenhart, T. T.; Dempsey, J. L. A Practical Beginner’s Guide to Cyclic Voltammetry. J. Chem. Educ. 2018, 95 (2), 197–206. https://doi.org/10.1021/acs.jchemed.7b00361. Costentin, C.; Savéant, J.-M. Multielectron, Multistep Molecular Catalysis of Electrochemical Reactions: Benchmarking of Homogeneous Catalysts. ChemElectroChem 2014, 1 (7), 1226–1236. https://doi.org/10.1002/celc.201300263. Rountree, E. S.; McCarthy, B. D.; Eisenhart, T. T.; Dempsey, J. L. Evaluation of Homogeneous Electrocatalysts by Cyclic Voltammetry. Inorg. Chem. 2014, 53 (19), 9983–10002. https://doi.org/10.1021/ic500658x. Costentin, C.; Drouet, S.; Robert, M.; Savéant, J.-M. Turnover Numbers, Turnover Frequencies, and Overpotential in Molecular Catalysis of Electrochemical Reactions. Cyclic Voltammetry and Preparative-Scale Electrolysis. J. Am. Chem. Soc. 2012, 134 (27), 11235–11242. https://doi.org/10.1021/ja303560c. Zoski, C. G. Handbook of Electrochemistry; Oxford: Elsevier, 2007. https://doi.org/10.1017/CBO9781107415324.004. Bard, A. J.; Faulkner, L. R.; White, H. S. Electrochemical Methods: Fundamentals and Applications, 3rd ed.; 2022. Compton, R. G.; Banks, C. E. Understanding Voltammetry; Imperial College Press, 2010. https://doi.org/10.1142/p726. Bontempelli, G.; Toniolo, R. Voltammetry | linear sweep and cyclic. In Encyclopedia of Analytical Science; Elsevier, 2005; pp. 188–197. https://doi.org/10.1016/B0-12-3693977/00647-6. Molina, A.; Laborda, E. Detailed Theoretical Treatment of Homogeneous Chemical Reactions Coupled to Interfacial Charge Transfers. Electrochim. Acta 2018, 286, 374–396. https://doi.org/10.1016/j.electacta.2018.07.142. Savéant, J.-M. Elements of Molecular and Biomolecular Electrochemistry; John Wiley & Sons, Inc.: Hoboken, NJ, USA, NJ, USA, 2006. https://doi.org/10.1002/0471758078. Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals a(1). In Electrochemical Methods: Fundamentals and Applications; Bard, A. J., Faulkner, L. R., Eds, 2nd Ed.; John Wiley & Sons, Inc.: Hoboken, NJ, 2001. Elgrishi, N.; Kurtz, D. A.; Dempsey, J. L. Reaction Parameters Influencing Cobalt Hydride Formation Kinetics: Implications for Benchmarking H 2 -Evolution Catalysts. J. Am. Chem. Soc. 2017, 139 (1), 239–244. https://doi.org/10.1021/jacs.6b10148. Savéant, J. M.; Xu, F. First- and Second-Order Chemical-Electrochemical Mechanisms. J. Electroanal. Chem. Interfacial Electrochem. 1986, 208 (2), 197–217. https://doi.org/ 10.1016/0022-0728(86)80535-6. Wightman, R. M.; Wipf, D. O. Voltammetry At Ultramicroelectrodes. In Electroanalytical Chemistry; Bard, A. J., Ed.; 1988.

Methods and Instruments | Electrochemical Impedance Spectroscopy J Odrobina, W Strunz, and C-A Schiller, Zahner-Elektrik GmbH & Co. KG, Kronach, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of Z.B. Stoynov, D.E. Vladikova, MEASUREMENT METHODS | Electrochemical: Impedance Spectroscopy, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 632–642, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00068-X.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 References

Introduction Impedance studies of electrochemical power sources Impedance techniques for the study of energy sources Measurement instrumentation Prevention of measurement artifacts Artifacts in context with the state of an ECS, changing in time The influence of drift on the EIS data during the acquisition of a single low frequency sample: Online drift compensation “ODC” The influence of drift on the EIS data during the acquisition of a complete EIS Understanding the results of impedance measurements Differential impedance analysis Distribution of relaxation times Validity considerations Measurement model ZHIT: Minimum phase system conformity test Summary/conclusions

44 48 49 49 49 50 51 51 53 54 55 56 56 57 58 59

Abstract Impedance spectroscopy is a powerful analytical technique used to investigate electrochemical properties of materials and systems across a wide range of frequencies. The impedance, composed of resistance (R) and reactance (X), represents the opposition to the flow of AC excitation and includes components such as capacitance and inductance. Through equivalent circuit modeling, experimental impedance data can be interpreted to extract information about underlying physical and chemical processes. Impedance spectroscopy finds wide-ranging application in fields such as materials science, electrochemistry, and biomedical research, enabling progress in electronic devices, energy storage systems, and diagnostics. Impedance spectroscopy helps researchers to deconvolute the individual processes, which determine the electrochemical properties of investigated systems. Impedance spectroscopy finds a variety of applications for studies of different electrochemical power sources – batteries, fuel cells, supercapacitors, electrolysis, and other systems – and has different targets – studies of basic kinetic processes, materials research, quality control, matching, and diagnostics. The principle of the method and a set of useful techniques providing for reliable experiments and data interpretation are described. The selection of the measurement setup configuration and details of the experimental program are discussed. Techniques for correction of the errors due to cable inductance and to possible time evolution of the object are described. Methods for automatic knowledge acquisition from the primary measurement data like distribution of relaxation times and differential impedance analysis, are also explained. Examples of experimental data – raw and processed – illustrate the recommended techniques.

Glossary Frequency response analyzer An instrument that generates a sine wave perturbation to measure the periodical signals in the object’s input and output and to calculate the transfer function. Typically, the Fourier transform is used for this purpose. Fuzzy object An object that typically has a kernel with well-exhibited properties and that has boundaries which are not well defined. Galvanostat An instrument that controls the current proportional to the set input and independent of the load. Least squares method An estimation method in which the distance between the object and the model is measured as the sum of the squares of the differences between them. Perturbing signal An external signal that causes deviation of the system from equilibrium. Potentiostat An instrument that controls the voltage proportional to the set input and independent of the load. Shunt Calibrated resistance used for conversion of the current signal into voltage signal convenient to be measured by the frequency response analyzer.

42

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00263-2

Methods and Instruments | Electrochemical Impedance Spectroscopy

Key points This chapter gives a short introduction about different aspects of impedance spectroscopy, focusing on.

• • • •

Principles for modeling of impedance spectra Detection of measurement artifacts Advanced methods for enhanced analysis of spectra Validation of data

Nomenclature

Symbols and units

C C^ f G g k L P^ r R R^ T T^ Z |Z| Z0 Z00 w v

capacitance, F ¼ s O−1 estimate of the capacitance, F frequency, Hz transfer function transfer coefficient reaction rate constant, mol s−1 inductance, H ¼ O s−1 matrix of the LOM parameters estimates as functions on the log frequency electrolyte additive resistance, O resistance, O estimated resistance, O time constant (s) estimated time constant (s) impedance (O) modulus of Z (O) real component of Z, O imaginary component of Z, O phase angle (rad) angular frequency (s−1)

Abbreviations and acronyms a.c. BW CNLS CPE d.c. DIA DRT ECS EIS ESA EQC FRA HOS IS LKK LOM MF MPS NASA NFRA

Alternating current Bounded Warburg element Complex nonlinear least-squares method Constant phase element, general constant phase element Direct current Differential impedance analysis Distribution of relaxation times Electrochemical system Electrochemical impedance spectrum European Space Agency Equivalent circuit Frequency response analyzer High oversampling technique Impedance spectroscopy Linear Kramers-Kronig transform Local operating model Model free analysis approach Minimum phase system National Aeronautics and Space Administration Nonlinear frequency response analysis

43

44

Methods and Instruments | Electrochemical Impedance Spectroscopy

ODC SoC SoH TD TF W WHA ZHIT

1

Online drift compensation State-of-charge State-of-health Time drift Transfer function Warburg element Weighted harmonics autocorrelation Two-pole logarithmic Hilbert transform

Introduction

Impedance spectroscopy (IS) has been emerging in recent years as one of the most powerful and informative tools for investigating the complicated processes in electrochemical power sources. The increasing interest arises from the unique advantage of the method’s capability to separate and characterize in a single measurement the different steps involved in a complex process. Although introduced in the end of the nineteenth century, the method evolved later and has found advanced applications owing to the achievements obtained in theoretical electrochemistry and in technical cybernetics and to the promotion of precise computerized measurement equipment supported by advanced software for data analysis. The best instruments can cover more than 11 frequency decades and are capable to measure resistances from microohms up to teraohms and capacitances from picofarads up to kilofarads. IS is based on the classical method of transfer function (TF) analysis. The TF G(s) is the quotient between the Laplace transform Y ¼ ℒ(y) of a systems response y, and the Laplace transform X ¼ ℒ(x) of an excitation signal x, applied in the time domain G(s) ¼ ℒ(y)/ℒ(x), with the complex transform frequency s. If certain conditions apply, in particular causality, linearity and periodicity, the Laplace transform can be replaced by the Fourier transform and the complex transform frequency simplifies to the pure imaginary frequency jo. If a linear system is perturbed by a sinusoidal wave input x(o), the output signal y(o) is also sinusoidal with the same frequency, but with different phase and amplitude. The ratio of the output to the input is called transfer coefficient for the corresponding frequency g(o) ¼ y(o)/x(o). The transfer coefficient is a complex number and depends on the frequency and on the properties of the investigated system, which is stationary, when the state without perturbation during and at the end of the experiment is equal to the initial state. An electrochemical system (ECS) is preferentially characterized by the voltage v across, and the current i flowing through the electrical connection by two electron conducting terminals – on a discussion of the role of further (reference) electrodes is renounced here due to space reasons. An ECS should be kept first at a certain bias operating point, when determining the TF of an ECS, in order to maintain stability. Second, the voltage v (“potentiostatic mode”) or the current i (“galvanostatic mode”) may now be modulated to a small extend around the bias point, while the corresponding current i respectively voltage v is recorded as the system’s response. Both voltage v and current i may generally depend on each other plus a set of characteristic system properties pi, like temperature, pH, concentration: i ¼ i ðv, p1 , p2 , . . . , pn Þ v ¼ v ði, p1 , p2 , . . . , pn Þ If one intends to calculate the TF theoretically, in these equations the dependencies from the properties pi underlying the modulation, usually addressed as “state variables,” must be eliminated, keeping besides the static properties, usually addressed as “kinetic parameters,” only the dependencies from i and v. As a simplification, for small deviations from the bias, the resulting differential equation for every pi may be substituted by a Taylors expansion, which can be terminated after the linear term. For sufficient small sinusoidal excitations ~i ¼ bi  ejot and v~ ¼ b v  ejðot+’Þ with the phase shift ’ and the amplitude values bi and b v one yields:     ~i ¼ ∂i  v~ + . . . v~ ¼ ∂v  ~i + . . . ∂v ∂i The over line magnitudes i and v indicate the steady state values of i and v at the operating bias point and their differential quotients represent the slope of the corresponding current-voltage characteristic. v¼b v  ejðot+’Þ is identical to the impedance The TF G(s) ¼ ℒ(y)/ℒ(x) for small periodical sinusoidal excitations ~i ¼ bi  ejot and ~ Z(o) and reads now ZðoÞ ¼

LðvÞ ~ v b v ¼ ¼ ∙ ej’ LðiÞ ~i bi

The choice of the amplitude of the small perturbing signal depends on the degree of the object’s nonlinearity at the selected working point. In the general case of electrochemical kinetic studies, amplitudes around 5 mV are acceptable. However, when sharp volt–ampere characteristics are investigated, a decrease in the amplitude down to 1 mV is necessary for obtaining informative

Methods and Instruments | Electrochemical Impedance Spectroscopy

45

performance measurements. In the opposite case the measurements in quasi-linear regions can be carried out at amplitudes of 50–100 mV or even higher. The decrease in the signal amplitude is limited by the relative increase in the noise measurements, resulting in deterioration of the measurement precision. On the contrary, an increase in the signal amplitude is associated with an increase in the nonlinearity errors. Thus, the choice of the signal amplitude is a compromise between the noises and errors and depends on the aim of the study, the degree of linearity at the selected working points, the quality of the instrumentation, and the applied methods for analysis of the experimental data. The selection of the operation point, which defines the region of the impedance measurement, is also of interest. It is well known that two main types of phenomena govern the operational behavior of practically all batteries and fuel cells: kinetic activities (charge transfer processes) and transport limitations. They differ significantly in their physical nature and behavior. For small load currents, the kinetic processes dominate the object’s behavior and the result is the well-exhibited nonlinear character of the voltage–current characteristics. For large loads – currents typical for the nominal operation of the energy source – the dominant limitations are of a transport type. As a result, the load characteristic is quasi-linear. Thus, the selection of the working point, defined as the DC bias, depends on the aim of the investigation: for studies of electrochemical kinetics, the convenient experimental working points are in the initial nonlinear region of small loads; whereas for screening evaluation of the overall cell behavior, the working point has to be selected in the range of the large loads. The first step in the classical analysis of experimental data is the graphical representation. A problem of data monitoring is that the measurement generates a three-dimensional (3D) data set (angular frequencies oi and real Zi0 and imaginary Zi00 components of the impedance for every frequency), which should be shown in a 2D picture. The impedance can be represented in linear Cartesian coordinates as ZðoÞ ¼ Z0 + j  Z00

(1)

or logarithmic in polar coordinates as ln½ZðoÞ ¼ lnjZj + j  ’ (2) pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi   02 00 2 −1 Z0 where jZj ¼ Z + Z is the modulus and ’ ¼ tan Z00 is the phase, corresponding to a given angular frequency o. The most common form of representation, based on Eq. (1), is the complex-plane impedance (Nyquist) diagram, Fig. 1(a). Every sample of the graph corresponds to a given frequency (o/s−1 or f/Hz, where o ¼ 2pf). The plot is more illustrative when the frequency of some sample points, for instance those corresponding to decades, are denoted explicitly, as shown in Fig. 1(a). An alternative representation is the so-called Bode plot, which uses Eq. (2) and consists of two graphs, illustrating the dependencies of log |Zi|/O and ’i on log fi/Hz, Fig. 1(b). Impedance spectra displayed in Nyquist diagrams may be more intuitive due to the typical occurrence of semicircle arrangements, but they emphasize high impedance values and may mask small ones. This appears as a problem, if the TF course over the frequency spans several orders of magnitude, what is normal for instance in the fields of corrosion research. However, this is usually not the case in the fields of electrochemical power sources, what leads here to the dominance of the Nyquist plot. Logarithmic Bode diagrams on the other hand display small and big magnitudes of the TF with equally weighted accuracy, but the TF of electrochemical power devices with their typically small impedance to frequency dynamics may exhibit only poor structure. A unique strength of the Bode diagram is its capability to assist a first glance judgement of the measurement quality: the slope of the logarithmic impedance modulus course over the logarithm of the frequency must resemble the course of the phase angle. This is an inherent property of all ECS impedance TF, as they must behave as so called two-poles. This relationship can be derived from the logarithmic Hilbert transform ZHIT, what will be discussed a little closer in the later Section 12. Besides Nyquist- and Bode-diagrams, further forms of graphical representations are found in literature, like Mott-Schottky-Plot, Cole-Cole-Plot, complex

Fig. 1 Graphical impedance presentation of a time-constant model (R, 200 O and C, 20 mF, in parallel connection) simulated in the frequency f range 104–10−3 Hz: (a) linear complex-plane impedance (Nyquist) diagram and (b) logarithmic Bode diagram.

46

Methods and Instruments | Electrochemical Impedance Spectroscopy

modulus, complex dielectric constant. Different from Nyquist- and Bode-diagrams they imply certain model concepts, how to interpret the components of the complex impedance and shall therefore not be discussed here. Measured using powerful instrumentation in a wide frequency range, the experimental data are rich in information about the topology of the electrode components and the processes differing in velocities and time constants such as very fast electronic conduction, fast electrochemical kinetics, diffusion of gases in porous electrodes, ion transport in solid-state ionic, and formation and growth of new phases. However, the experimental data represent the response of the system to the external perturbation with an a.c. signal and thus do not concern an immediately measured physico-chemical property. The challenge is how to decode and rationalize quantitatively the useful information from the experimental data. The classical approach for impedance data analysis is based on the comparison of the experimental results with a model accepted in advance according to a chosen working hypothesis that is supposed to describe the properties of the investigated system. The model structure is confirmed by a parametric identification procedure, followed by verification of its correspondence to the experimental data. Excellent programs based on the complex nonlinear least-squares (CNLS) method are developed and largely disseminated. In practice, the structure of the working model is based on previous experience or information from the literature. The shape of the impedance diagrams could also be decisive. After testing a number of possible models, the one that yields the best fit between the measured impedance and that of the model is chosen. The final stage of the parametric identification is the validation of the selected model, performed by frequency analysis of the so-called residuals – the final distances between the measured and the simulated model’s impedance. They should be completely statistical. The derived mathematical equations for the impedance of the hypothetical model can be shown in the frequency domain as a construction of elements, which are connected by applying different laws in accordance with the chosen working hypothesis. Thus, the complete model can be described as an electrochemical equivalent circuit (EQC). For instance, the total impedance of a series circuit containing the partial impedance Z1 and Z2 is the sum of the partial impedances, while the total impedance of a parallel circuit is the reciprocal of the sum of the two reciprocals 1/Z1 and 1/Z2 (with the abbreviations “−” for series and “//” for parallel arrangement): Z ¼ Z1 + Z2 ,

Z== ¼ 1=ð1=Z1 + 1=Z2 Þ

(3)

More complicated networks can be described as arrangements of meshes and nodes and analyzed by means of Kirchhoff’s laws (Fig. 2). The Nyquist plot of a parallel RC element is geometrically represented as an ideal semicircle with a diameter equal to the value of the resistance and the semicircle apex located at the angular frequency as the reciprocal of the time constant trc ¼ RC: fapex ¼ (2pRC)−1 [Hz] (Fig. 1(a)). This is one of the most frequently observed impedance shapes. An ECS leading to such an RC circuit is often a Faradaic electrochemical reaction represented in the simplest case by the charge transfer resistance Rct, in parallel with the omnipresent electrostatic capacity, formed by the electrochemical double layer Cdl. The latter appears as the activation energy barrier at the phase boundary between ion conduction in the electrolyte and electron conduction at the electrode terminal. One should not mix the electrochemical reaction rate time constant tk ¼ k−1, the reciprocal of the kinetic rate constant k, which is associated solely to Rct, with the RC time constant trc ¼ RC, noticeable in the impedance TF. trc depends besides on R, also directly on the capacitive component, but Cdl has no immediate kinetic meaning. It is therefore questionable to intuitively associate the time constants observable in the lower or higher frequency ranges of the impedance TF directly with slow or fast electrochemical processes. The variety of electrochemical phenomena cannot be represented only with combinations of the classical lumped electrical elements resistance, capacitance, and inductance. For studies of real systems, a new class of the so-called electrochemical elements is developed. They ensure modeling of the behavior of frequency-dependent parameters, which is frequently observed in the experimentally measured impedances. For instance, the “Special Warburg Impedance” element (W) represents the second Fick’s law in the frequency domain. If a charge transfer reaction takes place, a sinusoidal modulation of the electrode’s potential resp. current flow, causes sinusoidal concentration changes of the reactants in front of the electrode’s surface, which propagates as a

Fig. 2 Visualization of Kirchhoff’s laws: The sum of all voltage drops vi in a network mesh adds up to zero. In a similar way the sum of all currents ii flowing to a network node adds up to zero.

Methods and Instruments | Electrochemical Impedance Spectroscopy

47

Fig. 3 Impedance spectra of three diffusion models under different boundary conditions (a: Nyquist plot, b: Bode plots). The Special Warburg Impedance W (red) exhibits an imaginary/real quotient of −1 overall frequencies, what is similar to the Nernstian N and the Bounded Warburg Impedance BW course at higher frequencies. At lower frequencies BW turns into capacitive (battery) behavior and N becomes Ohmic (fuel cell behavior).

Fig. 4 Impedance diagrams of a modified time-constant model (R, 200 O and CPE, 20 mF−a, a ¼ 0.8, in parallel connection), simulated in the frequency f range 104–10−3 Hz (a: Nyquist plot and b: Bode plot).

damped wave into the electrolyte bulk. The model of W assumes a potentially infinite length of the damped concentration wave in the direction of one spatial dimension. For the modeling of the diffusion under spatially limitation or under damping by additionally processes several extensions to the W were evaluated, characteristic for the particular boundary situation. Modern simulation programs should provide many readymade different impedance elements for diffusion, for instance the “Gerischer’s Diffusion Impedance,” which takes into account a homogeneous chemical reaction in the solution coupled to the electrochemical charge transfer at the electrode. In Fig. 3 the impedance spectra of the three most basic diffusion boundary situations are compared. An often-applied electrochemical element for the representation of frequency-dependent behavior is the constant phase element (CPE). The replacement of the capacitance in Eq. (3) with CPE produces a depressed semicircle in the Nyquist plot (Fig. 4a) and limits the absolute phase angle to less than 90 in the Bode plot (Fig. 4b). The electrical equivalent of the CPE is a fine distribution of RC time constants (chain ladder) produced potentially by many different origins. Electrode inhomogeneities or porosity are well known as reasons, but even in a lateral perfect homogeneous electrode CPE behavior can occur, if a dielectric with a conductive gradient perpendicular to the electrode surface is involved. In a more generalized meaning, the term CPE may be applied not only to the upper mentioned distributed- or loss capacity, but to all impedance elements, which exhibit a constant phase angle over the frequency, what applies also for inductance, resistance and Special Warburg impedance W: d lnjZðoÞj 2 ZðjoÞ ¼ CðjoÞa , a ¼ ’, a ¼ p d ln o In the meaning of the general CPE, the exponent a may get values between 1, corresponding to phase angles of – 90  ’  + 90 and impedance modulus slopes of −1  d lnd jlnZðooÞj  1. This inherent relationship between impedance and phase is the starting point for the evaluation of the later discussed ZHIT transform.

48

Methods and Instruments | Electrochemical Impedance Spectroscopy

Fig. 5 One of the numerous equivalences between impedance models with identical impedance spectrum, but representing different mechanisms. Left: two stacked partially conductive surface layers, right: a passive layer (C4), partially penetrated by holes filled with electrolyte (R4) with active corrosion (R3) taking place at the pore ground. A further interpretation of the EQC at the right is a charge transfer reaction (R4) into an adsorbed species with the crystallization impedance (R3, C3), and the double layer capacity (C4).

It must be mentioned here, that the correct interpretation of the mechanisms in an ECS must always lead to an unequivocally assigned equivalent circuit EQC, while the inverse assignment from a well-fitting EQC to a certain mechanism is not unequivocal: in general, EQC representing different mechanisms may exhibit the same TF shape. Fig. 5 illustrates this on hand of a simple example: an EQC consisting of two parallel RC elements in series (representing for instance two stacked surface layers) can have exactly the same spectrum like one RC element with one R in series plus one C in parallel to the total (representing for instance a protective layer, containing holes, were corrosion takes place in parallel). The spectrum of the circuit (1 O//1F) − (2 O//2F) is identical to the one from ((1 O//3F) − 2 O)//0.667F.

2

Impedance studies of electrochemical power sources

From the viewpoint of a system’s analysis, energy sources, sinks and storage devices are nonlinear and quasi-irreversible large systems with micro- and macro distribution of the characterizing parameters. Processes of energy and mass transport take place at their electrodes, changing the activity, the chemical composition, and often the morphology of the electrodes. As a result, those objects have memory effect: the details of their external behavior depend on the previous history, that is, fast charges or deep discharges or other coarse interventions. IS finds a large variety of applications for studies of different energy sources – primary cells and batteries, fuel cells, and supercapacitors – as well as in the growing field of electrolysis. As far as the impedance measurements can supply a large amount of valuable information, there are also various main targets of these applications:

• • • • •

study of processes of the electrochemical kinetics and their evolution; materials research related to the development of active materials and their degradation and corrosion; quality control during the production and matching of cells for important applications; diagnostics of SoC and state-of-health (SoH) during operation of energy sources; and creation of a knowledge base, necessary for a better understanding of the impedance results.

The first successful measurement of battery impedance has been performed in the mid-1970s. Since then, IS has found many applications for studying the nature of the processes at the electrodes of lead–acid, lithium, and other batteries; the porosity of the active mass; and degradation the cells. IS has been successfully introduced in National Aeronautics and Space Administration (NASA), European Space Agency (ESA), and the Russian space programs for matching of cells before launching as well as for evaluation of the SoC and SoH during their operation in space. In recent years, testing standards using IS as a tool for assessment of SoH of different batteries have been discussed and approved, in particular the attractive field of testing electromobility batteries for “second life” (reuse) considerations has been established. However, impedance studies of energy sources and sinks face several problems related to their nature as energy devices:

•

Typically, such devices are objects with very low impedances – for small-size batteries, their internal impedance is in the range of tenths of ohms, whereas for large batteries and for power fuel cells and electrolysis, it drops down to milliohms and even microohms.

Methods and Instruments | Electrochemical Impedance Spectroscopy

• •

3

49

Batteries are typically non-stationary systems – during charge or discharge processes, their internal chemistry changes, and as a result, their impedance is also non-stationary. Energy devices are typical extremely complex objects – their electrodes are porous and have a large developed double-layer surface. As a result, the local kinetic activity coefficients are never uniform; the resulting impedance, measured outside, shows blurred signals and corresponds to a system with a wide distribution of the parameters.

Impedance techniques for the study of energy sources

In order to obtain rich and valuable information, the measurement instrumentation and the configuration of the experimental setup have to be selected depending on the specific nature of the energy source and sink as an object under study. The processing of the experimental data also has to be appropriate and to take into account the fuzzy and nonstationary nature of the object. In addition, some data pre-processing techniques for correction of methodical errors are frequently necessary.

4

Measurement instrumentation

The present-day impedance measurement equipment usually combines in one housing a frequency response analyzer (FRA) with an interface to the test object, which can act, among others, as a potentiostat or galvanostat. Such instruments are designed for maximum flexibility together with maximum accuracy to cover impedance applications in a wide field. Optimizing the frequency range and the accuracy over an extremely dynamic range (mHz to 12 MHz, mO to TO) is in technical conflict with the provision of high voltage and currents at the cell terminals. State of the art impedance instruments for universal applications cover therefore a limited voltage and current capability of for instance |vout|  |  32 V| and |iout|  |  4A|. This is fully sufficient for most lab research applications on single electrodes or small-scale full cells and stacks, but may be not sufficient for testing for instance high-power batteries. It is therefore often necessary to add further equipment, capable to provide the necessary power to force the object under test into the appropriate steady state operating point. The authors recommend to check instruments to come into question for the acquisition, if they support an unproblematic backup with some high-power supplies and electronic loads, suited in particular for impedance technique. The origin of an electrochemical object’s non-linearity is found in general in the locally exponential overvoltage-current characteristic of the charge transfer process due to the Butler-Volmer equation. Here, the impact of the overvoltage is scaled by the “temperature voltage” vT ¼ RT zF with the ideal gas constant R, the absolute temperature T, the Faraday constant F and the effective charge number z. vT has a value of about 25.7 mV at room temperature for a one-electron charge transfer number of z ¼ 1. In order to maintain linearity, the effective amplitude of an excitation signal across the electrochemical double layer should be small, compared with vT. As mentioned above, an amplitude of about 5 mV is a good choice for sufficient signal to noise ratio as well as negligible non-linearity interference for most general ECS, evaluated in potentiostatic mode. In this mode the response signal is the current. The dynamic range of impedance instruments for current measurement spans usually many orders of magnitude, which is much more than the dynamic range for voltage. For general ECS, the potentiostatic impedance measurement should be the method of choice, because of accuracy, amplitude and dynamic range considerations. In galvanostatic mode the effective overvoltage amplitude, which determines the non-linearity, is not a constant, but a function of the impedance under test, and therefore frequency dependent, complicating the amplitude considerations regarding linearity. Nevertheless, dealing with electrochemical power devices the galvanostatic mode is often the better choice from the reasons listed in the following:

• • •

5

The impedance change over the frequency of typical electrochemical power devices (except super caps) does not exhibit strong dynamics: an excitation amplitude chosen appropriate for a medium frequency augur to be also an acceptable choice for the whole frequency range of interest. The system’s steady state of an electrochemical power device, for instance a fuel cell, is better characterized by a certain current flow at the operation point than by a certain voltage. Setting a certain excitation voltage like 5 mV in potentiostatic mode may overstrain the instrumentation, if the objects impedance is in the range of mO. The applicability of a certain excitation current in galvanostatic mode is more transparent for the operator.

Prevention of measurement artifacts

During the course of impedance measurement, analysis and interpretation several problems may occur complicating the correct understanding of the evaluated data. Starting with the fact, that the user is interested on the sole impedance of the object under test, while the instrumentation recognizes the whole object’s environment including for instance connection lines and electromagnetic interference from outside. In the fields of high-ohmic systems, stray capacitance and electrostatic interference dominates. Here, using shielded cables and grounded Faraday cages may limit the impact of those interferences sufficiently. The situation is contrary in the fields of electrochemical power sources: the main interference nature is magnetic and, in contradiction to a well spread

50

Methods and Instruments | Electrochemical Impedance Spectroscopy

Fig. 6 Illustrating the “mutual induction” effect. Left: the mutual induction is caused by the magnetic coupling in a transformer’s manner between test current and measured voltage drop. Right: minimize the impact of the mutual inductance interference by means of the twisted pair technique. Zf, Zs: parasitic inductance of current feeding and voltage sensing connection. Z, Zm: impedance object under test.

opinion, electrostatic shielding is useless. Also, the role of the self-inductance of test object and cables is overestimated – the dominating interference source at high frequency is not the inductance itself, but the so-called “mutual induction.” Regarding the self-inductance contributions it is often proposed to cancel them out by means of a preceding calibration procedure. State of the art instruments support an internal calibration including the parasitic reactance of the current sensing shunt, providing maximum precision at the instrument’s terminals. The residual inductance after the strongly recommended optimization of the wiring should be included therefore uncompensated in the measurement, because it can be taken into account more advantageous with higher precision during the process of an equivalent circuit modeling, discussed later. Moreover, the influence of the mutual induction, which dominates the high frequency error strongly, is non-additive and cannot be cancelled out by calibration from principle. The most promising strategy to avoid the interference caused by the connection lines is summed up in the following (see Fig. 6):

• • • • •

6

4-pole (Kelvin scheme) wiring using a separate current-feeding line pair and a separate voltage-sensing line pair is obligatory. “Twisted pair” technique for the current – the line to and the one from the object must be close to each other in order to suppress the emitted magnetic field for both current flow directions, leading to a perfect field cancellation. “Twisted pair” technique also for the voltage – the sensing lines must be close to each other in order to suppress the antenna activity for both voltage signs by perfect compensation. Keep both twisted pairs short and at distance from each other. In order to suppress residual series inductance best case, minimize the area spread out in connection line loops, where magnetic field could build up.

Artifacts in context with the state of an ECS, changing in time

Following the arguments in the introduction, the response function of a monochromatic excitation will be again monochromatic – just one line in the frequency domain at exactly the same frequency position for both excitation and response. In a digital representation of such signals, it appears sufficient to have four sample points in the time domain of a sine wave period to determine exactly amplitude and phase. Until the year 1984, the dominating impedance techniques “lock in” and “digital correlation” used just this minimum definition. With the first fully digital instruments an advanced standard was established: since then, advanced instruments use a High Over Sampling technique with at least 256 sampling points per period in the upper frequency range and many thousands at low frequencies, in the following addressed shortly as “HOS.” HOS allows the precise determination of harmonics (multiples of the “fundamental frequency”) in the response, appearing mostly unwanted besides the regular test signal. The knowledge about the presence and magnitude of harmonics enables an attractive set of features enhancing the IS:

• • •

In the absence of external interference and non-linearities the harmonic content should be zero. From the harmonics present nevertheless, an immediate and realistic measurement error estimation can be evaluated. This can be done for instance by means of the “Weighted Harmonics Autocorrelation” (WHA).1,2 The excitation of additional frequencies, the “multi-sine” technique, is a strategy to save measurement time. HOS and the feature to evaluate the harmonic content is a prerequisite for the analysis of a multi-sine response signal mixture.3 The technique of “Non-Linear Frequency Response Analysis” (NFRA), an enhancement of the IS, performs the intentionally overdrive of the linear range of the system under test by means of a high perturbation amplitude. For the evaluation, the non-linear response in form of the generated harmonic content is carefully analyzed.4 Besides, the NFRA feature, which records the harmonics, is a valuable extension to judge the relevance of the actual non-linearity effects during a regular IS.

Methods and Instruments | Electrochemical Impedance Spectroscopy

•

51

Defeating artifacts caused by a changing state of the system under test during the measurement is one of the most challenging problems in the IS. Such “Time Drift” (TD) due to non-stationarity appears at different time scales. TD, among other side effects discussed later, creates harmonics in the acquisition signal at low frequencies – a sophisticated analysis of those harmonics enables a strategy to cancel out the appearing signal deterioration by compensation. This will be treated in the following section in more detail.

7 The influence of drift on the EIS data during the acquisition of a single low frequency sample: Online drift compensation “ODC” At the lowest frequencies of an EIS the period time of the excitation signal may be long enough, so that deviations of the bias line due to the time drift can no longer be neglected. This is illustrated in Fig. 7. A correction like sketched in Fig. 7 must be performed “online” by the instrument, during the process of data acquisition (ODC), otherwise the information about the component of the TD, contaminating the low frequency samples, is lost. The idea behind ODC is, that for sufficient small drift the effective response signal can be assumed as the sum of the regular (unaffected) response on the excitation signal on the one hand and a linear change of the bias line due to drift on the other hand. If this bias line could be determined by means of an appropriate algorithm during the acquisition, the undisturbed regular response on the excitation signal could be found by subtraction and analyzed instead of the falsified immediate acquisition data.5 The acquisition of a complete EIS takes of course even more time, than just one low-frequency sample. TD during the record of a spectrum may lead to a severe violation of the causality rule by the acquired data: different frequency regions of the same spectrum may belong to different states of the system. This is discussed in the following.

8

The influence of drift on the EIS data during the acquisition of a complete EIS

Significant TD during an EIS acquisition delivers spectral data, which can neither be correctly represented nor understood by means of regular impedance models. A typical example is the EIS record, acquired during charging or discharging of a battery. An approach, standing to reason for handling the upcoming complications, is the idea to perform not just one, but several EIS in series (Fig. 9). The result is a sequence of EIS, every individual one still affected from the TD. But a simple interpolation algorithm allows now to reconstruct a series of EIS from the drift affected data, “as if” measured at once at a certain moment in time. The series represents still the changing state of the ECS from one spectrum to the next, but the individual spectra are widely freed from TD artifacts. Early works of Göhr6 and Stoynov7 demonstrated the efficiency of this approach (Fig. 8). The arguments above suggest, that a method would be highly appreciated, which identifies violations of the causality due to TD, prior to running into problems during analysis (Fig. 9). The Linear Kramers-Kronig (LKK) transform should accomplish this in principle. Unfortunately, the practical application of the LKK runs into conceptional and computational problems. Well-suited alternatives can be derived from the logarithmic form of the KK transform, here addressed as Hilbert Transform of Z-two poles (ZHIT, Z ¼ short cut for impedance)8–11 and by means of indirect methods like the application of the measurement model.12–14 The technical details about those alternatives will be treated in the later chapter “Validation.” The ZHIT transform is able to re-construct causal spectra from TD affected data. This predestines ZHIT as a tool to identify TD contaminations in experimental EIS: simply compare the experimental data with the data, cleaned by re-construction (Fig. 10). Besides, ZHIT can act as a correction

Fig. 7 Illustration of the TD impact on the IS data acquisition during a low frequency sample. (a) the response signal (red) of a non-steady state ECS is superimposed by (in the simplest case) a linear drift signal component. It can be corrected (blue), if the bias drift course (black) is known. (b upper) without correction the frequency spectrum of the response signal is erroneous, due to the contamination by harmonics, assigned to the TD (red). (b lower) the accurate spectrum (blue) after drift correction provides accurate impedance results.

52

Methods and Instruments | Electrochemical Impedance Spectroscopy

Fig. 8 A three-dimensional plot of an EIS series with the z-axis used for the time flow. The spectra appear bended (red), due to the higher time effort necessary in the low frequency range at the left. Drift affected EIS series vs. time (red) can be calculated back to less disturbed ones (green) by interpolating (blue) between the spectra along the time axis.

Fig. 9 Impedance diagrams of an unprocessed EIS series, recorded during the discharge cycle of a 3 mAh Li-Ion battery. The voltage under load was between 4.077 V at the beginning and 3.1352 V at the end of the cycle. The AC amplitude was 2 mA, the discharge current −6 mA. The frequency range was between 100 kHz (Nyquist curve’s left end) and 0.1 Hz (Nyquist curve’s right end).

Fig. 10 Application of the ZHIT transform as a test on causality of an EIS. A late spectrum during discharge from the series displayed in Fig. 9 (at 3.1352 V) exhibits strong deviations in the low frequency range (red circle left). The impedance samples are drawn as blue circle symbols, the violation free impedance course, reconstructed from the phase angle (red) appears in pink.

Methods and Instruments | Electrochemical Impedance Spectroscopy

53

Fig. 11 Comparison of the best CNLS fit quality of the measurement displayed in Fig. 10 between unprocessed (left) and processed data (right). The maximal impedance modulus deviation between experimental and simulated values using the original data is about 9%, but only 0.7% are remaining after data processing by means of time interpolation and ZHIT refinement.

tool to provide accurate data for the analysis, if the experimental data suffer from TD. A two-step procedure, starting with the above sketched time interpolation routine, followed by cancelling out the residual TD artifacts with the help of a ZHIT correction, is an appropriate way to process TD contaminated data very successfully. Fig. 11 illustrates the effect on an example selected from the battery EIS series, displayed in Fig. 9. In the following section some aspects of different analysis strategies will be elucidated.

9

Understanding the results of impedance measurements

In the fields of impedance analysis mainly two different strategies are favored, which are addressed in the following as “modeling approach” and as “Model-Free” (MF) approach. The MF approach starts from the consideration, that the impedance transfer function itself contains a set of valuable information, for instance in form of the distribution and intensity values of the time constant content. Such can be achieved without having recourse to deeper knowledge upon topology, mechanisms and kinetics, present in the object under test. The most popular MF approaches will be treated briefly in the later chapters upon the methods “Distribution of Relaxation Times” (DRT) and “Differential Impedance Analysis” (DIA). In contrast, like anticipated already in the introduction, the classical modeling approach begins with an assembly of facts, well-known about the system under test from own experience and from literature. The goal is to translate those known facts, about for instance processes, topology, mechanisms and kinetics, into an electrical network model an EQC. The strategy to construct an EQC “bottom up,” starting from the known properties instead of interpreting the TF first, will avoid the ambiguity discussed in the section around Fig. 5. In a subsequent repetitive procedure of trial-and-error refinement by cyclic CNLS-fit, expansion and simplification of the EQC, the model is improved, until a sufficient relevance is obtained. Advanced features of the CNLS simulation and fitting program are essential for the success of this process. The program should support:

• • • • • •

a large library of impedance elements, including for instance complex models for porous electrodes and relaxation processes,15 which play an important role in the EQC description of electrochemical power devices; an extension possibility of the library by means of user-defined own mathematical element descriptions, because no library can be complete enough to cover all situations appearing in the field; the series handling of experimental spectra and EQC models including series fit and the ability of time series interpolation performing TD cancellation; the TD identification and TD cancellation by means of for instance ZHIT; advanced impedance instruments provide immediate measurement accuracy tags for each individual impedance sample (WHA). The CNLS program should not only reveal fit errors, but support in addition a consistent error propagation treatment from the experimental impedance uncertainty of the measured impedance to a final accuracy estimation of the evaluated parameters; the EQC, describing for instance a complete battery, may need a set of 10 or more impedance elements, characterized by even more numerical parameters. The more parameters present, the more is it important to find an EQC representation with the adequate count of impedance elements, but avoiding an over-determination. Therefore, the software should provide a significance check for the elements present in an EQC.

54

10

Methods and Instruments | Electrochemical Impedance Spectroscopy

Differential impedance analysis

Differential impedance analysis (DIA) is an improved technique for analyzing impedance data that follows the structural identification approach widely used in engineering cybernetics. The concepts of DIA is inextricably linked to the Bulgarian scientists D. Vladiokova and Z. Stoynov.16 The method increases the objectivity and accuracy of model recognition by eliminating the need for an initial working hypothesis and performing structural and parametric identification directly from the experimental data. The improved analytical performance of DIA is based on the combination of a scanning procedure with a local parametric analysis. The role of the local estimator is performed by a simple local operating model (LOM). The simplest LOM is a first-order inertial system extended by an additive term with impedance ZLOM according to: ZLOM ðjoÞ ¼ R1 +

R2 o  R2  T − j 1 + o2 T 2 1 + o2 T 2

(4)

where R1, C, and R2 are the LOM parameters and T ¼ C R2 is the effective time constant (Fig. 12). The parametric identification of the LOM can be performed in different ways and depends on the window’s width. For the limiting case of a single point width, the experimental data set is extended with two additional terms – the derivatives of the real and imaginary components of the impedance with respect to the frequency. As a result, a new data set (D1(i )) is obtained:   0   00   dZ dZ , (5) D1ðiÞ ¼ oi , Z0i , Z00i , do i do i

bðiÞ , R b1ðiÞ , C b2ðiÞ , Tb1ðiÞ , at every frequency oi. In general, they are It ensures the calculation of the LOM parameter estimates PbLOM R functions of the frequency and are presented in the so-called temporal plot:

   log PbLOM ¼ F log o−1 (6) In the frequency ranges where the LOM corresponds to the object’s behavior, these functions are frequency invariant and exhibit plateaus (Fig. 13). Every plateau recognizes a time-constant sub model. Its position enables the parametric identification. The example of a simulated model presented in Fig. 13 is characterized by two plateaus in the temporal plot, that is, DIA recognizes two time constants.

Fig. 12 Local operating model (LOM) – first order inertial system: equivalent circuit and impedance diagram.

Fig. 13 Differential impedance analysis of a faradaic reaction involving one adsorbed species (the EQC used is similar to the circuit of Fig. 5b), simulated in the b (solid), bt (dashed), b C (dotted). frequency range 104–10−3 Hz: (left) temporal plots and (right) spectral plots R

Methods and Instruments | Electrochemical Impedance Spectroscopy

55

The plateaus are separated by frequency-dependent regions that describe the mixing of two neighboring time constants, as shown in Fig. 13 (left), or more complicated frequency-dependent phenomenon. The plateaus are represented also in the more illustrative spectral form (Fig. 13, (right)). The simplest spectral transform can be regarded as a construction of an ordinary histogram. The number of the spectral peaks defines the number of time constants. The amplitude (height) of every peak is proportional to the frequency range in which the corresponding parameter’s estimates have close values, i.e., where the phenomenon is well pronounced. The spectral transform ensures a robust analysis with respect to noise, including efficient filtration of non-statistical noise, because the outliers introduce only low-intensity lines, located away from the spectral kernel of the basic phenomenon. The spectral peaks also provide objective information on the parameter degree of dispersion, which is unavailable by other methods for analysis. The width of every spectral peak increase with the enhancement of the dispersion. The further development of DIA, known as secondary DIA, ensures model identification in the frequency zones where the LOM parameter estimates demonstrate frequency dependence, i.e., in the regions between the plateaus in the temporal plots. It is based on the differentiation of the temporal functions, given in Eq. (5), and consecutive spectral transform. As a result, the secondary DIA ensures reliable recognition of CPE and of more complicated models with CPE in their structure.

11

Distribution of relaxation times

Another tool for a comprehensive analysis of impedance spectra is the evaluation using the distribution of relaxation times (DRT).17 When plotting an impedance spectrum in a Nyquist diagram, different loss processes which that take place in a cell cannot be readily separated from each other. The more similar the characteristic frequency of the individual processes are, the more their semicircles overlap. The DRT is a technique that allows the processes to be viewed separately. Calculation of the DRT is based on the assumption that any Kramers-Kronig conformal impedance spectrum can be represented by an infinite number of RC elements. An ideal polarization process has a discrete time constant t and its frequency-dependent behavior can be represented by a single RC element with resistance R, capacitance C, and time constant t ¼ RC: ZðjoÞ ¼

R R ¼ 1 + joRC 1 + jot

(7)

Real processes consist of a distribution of time constants, which can be described by the sum of RC elements. Concerning this theory, first introduced by Fuoss and Kirkwood in 1941, the relationship between the impedance Z(o) and the distribution function g(t) is as follows18: Z1 ZðjoÞ ¼ RS + RPol  0

gðtÞ dt, with 1 + jot

Z1 gðtÞ dt ¼ 1

(8)

0

RS is the Ohmic, frequency-independent part of the impedance and RPol is the scaling factor of the frequency-dependent part of the Þ defines the part of the (polarization) resistance at each impedance concerning the particular time constant t. The expression 1 g+ðtjot

time constant t. For practical calculations, a finite number of RC elements, which has to be defined, is used. The DRT is usually plotted logarithmically versus the time constant t. From a theoretical point of view, each peak represents a particular process: the main frequency or time constant of the process corresponds to the peak frequency or time constant, and the resistance of the process corresponds to the area under the peak. DRT can helps to develop EQC, can be used to validated ECQ and to show relative changes for the different processes in one cell at varying conditions.19,20 Fig. 14 shows the simulation of the impedance of a Lithium-ion cell in terms of the DRT.

Fig. 14 Simulated impedance spectrum of a Lithium-ion cell. Top: Nyquist diagram (three time constants + inductance + serial resistance). Bottom: Corresponding DRT-spectrum.

56

Methods and Instruments | Electrochemical Impedance Spectroscopy

In the Nyquist diagram, only two time constants are clearly visible, while the DRT-representation reveals a third time constant. However, in real terms, not every peak of the DRT corresponds to a process. The exact assignment of peaks in the DRT requires systematic parameter variation in detailed investigations. Limiting measurement accuracy, noise from the measurement environment (like water drop formation in fuel cell measurements) and inductive shares from the measurement setup limit add further difficulties in the application of the DRT analysis. Strategies to overcome the challenges for DRT analysis have been recently published.21,22 With DRT, processes can be better separated from each other, but this method also has its limits. For processes with time constants close to each other, the peaks also overlap in the DRT, a separation is then no longer possible.

12

Validity considerations

The validation of experimental IS data is one of the key aspects in EIS data evaluation. Electrochemical IS must comply with the prerequisites of causality, linearity, stability and continuity, what can be controlled in principle by the application of the Kramers-Kronig transform (KK) on the experimental data set. The KK transform in its linear version (LKK) provides a relationship between the frequency curve of the real part and the imaginary part. When calculating, for example, the imaginary from the real part of the spectrum according to Eq. (10) and comparing the result with the measured data set leads to the same result, the data set should be compliant with the LKK and can therefore be considered reliable. RefZðo0 Þg ¼ RefZð0Þg −

2 PV p

Z1 0

2 ImfZðo0 Þg ¼ o0 PV p

Z1 0

o  ImfZðoÞg do o2 − o0 2

RefZðoÞg do o2 − o0 2

(9)

(10)

Despite being reported by some authors as early as the 90s, a reproducible algorithm for a sufficiently accurate numerical solution avoiding pitfalls of the KK, which is applicable to experimental EIS data, has never been published. Both, the Principal Value “PV” integral in the KK formula as well as the integral frequency o limits ranging from 0 to 1, which are not accessible in the experimental data, stand in the way of practicable implementation (“limited bandwidth problem”). Besides, the restrictions imposed by the applicability of the LKK Eqs. (9 and 10) are not very rigid: as an example, the mutual induction widely dominates the high frequency artifacts of IS in low ohmic systems (see above) and are not in contradiction with the KK, meaning that they cannot be detected. Meanwhile, several groups provide alternative test algorithms based on fitting or linear equation solutions, which are applied on a serial chain ladder of several RC elements (resp. time constants). This so-called “measurement model” is finalized by closing elements like a capacity plus an inductance, and is of course compliant with the KK, because they are built from regular impedance elements.9,10 They are often, not quite correctly, referred to as “Linear Kramers-Kronig test.” As an example, the operating principle of the “Linear Kramers-Kronig Validity Test,” available by the KIT, Karlsruhe,11 will be discussed in more detail here.

13

Measurement model

The measurement model for validating impedance spectra is a concept used in electrochemistry and materials sciences to ensure the accuracy and reliability of EIS data. As mentioned above, measurement errors can be caused by various factors, such as noise, time drift, instabilities, discontinuities, and mutual induction artifacts. The validation of an impedance spectrum is crucial to make sure the measured data set is accurate and representative of the system under investigation. The measurement model should consider the nature and impact of these error sources on the data, which may involve analyzing atypical shapes in Nyquist and Bode plots, unwanted peaks, or noise. A general approach of a measurement model for validating impedance spectra consists of a series of components, designed to reproduce all practical cases of TF, appearing in IS. In an approach from the 1990s by Agarwal and Orazem, a circuit is chosen for the practical implementation, which includes a variable number of RC elements (Fig. 15). In addition to this sum of RC elements, a series resistance and an inductive component may be needed at high frequencies to account for the sum of purely Ohmic components or the inductive contributions of the measured object. At low frequencies, an ideal or non-ideal capacitance can also be added to approximate charge processes. Apart from these additional contributions, this

Fig. 15 Graphical representation of an equivalent circuit describing a measurement model of the series connection of n RC elements.

57

Methods and Instruments | Electrochemical Impedance Spectroscopy

Fig. 16 Implementation of a free available test tool after the “measurement model algorithm,” available from KIT, Karlsruhe. The test is performed on the simulated data of a complex EQC of a model, containing inductive components due to relaxation processes at low frequency.

form of the measurement model can be described mathematically as follows, with o as the angular frequency, j as the imaginary unit, tn ¼ Rn  Cn as the nth time constant, and Rn as the corresponding resistance. Z0 represents the sum of the additional contributions. The expression for the calculation of the total impedance is shown in Eq. (11): ZðoÞ ¼ Z0 +

X n

Rn ð1 + jotn Þ

(11)

In the approach displayed in Fig. 16, the number of time constants per frequency-decade is defined and the resulting total number of time constants is calculated in logarithmically equal distances within the frequency range of interest. This results in a linear equation system, in which only the unknown resistance values need to be determined as coefficients. In conclusion, the measurement model for validating impedance spectra may play an important role in ensuring the quality and reliability of IS data. It helps to improve the measurement accuracy and enhances the quality of data interpretation.

14

ZHIT: Minimum phase system conformity test

As mentioned above, using the logarithm of the complex transfer function (1) leads to Eq. (2), which is the base of the Bode graphic representation of IS transfer functions. 2 lnjZðo0 Þj ¼ lnjZð0Þj − o20 PV p

Z1 0

2 ’ðo0 Þ ¼ o0 PV p

Z1 0

’ðoÞ do o  ðo2 − o0 2 Þ

lnjZðoÞj do o2 − o0 2

(12)

(13)

In Eq. (2), ’ denotes the phase angle between excitation and response, here current and voltage. Unlike the expressions (9) and (10) describing the linear transformations, the corresponding logarithmic (Hilbert) transforms Eqs. (12) and (13)15 are not applicable to

58

Methods and Instruments | Electrochemical Impedance Spectroscopy

all causal transfer functions. The restriction is imposed by the application of the ‘logarithmic operator’. Transfer functions, for which Eqs. (12) and (13) are applicable, may not show logarithmic singularities in the positive real half part of the complex frequency plane.23,24 This restriction is equivalent to the exclusion of all-pass contributions to the transfer function.25 Due to the fact that electrochemical systems do not exhibit any all-pass character, these systems may be regarded as ‘minimum phase systems’ (MPS).26 In the two-pole behavior of electrochemical systems, current and potential are correlated without any signal propagation delay between force and response. This means that immediately after the application of a stimulus, a nonzero response must result. As already stated above, a certain interrelationship becomes recognizable, if two-pole transfer functions are plotted in the Bode diagram: the plot of the phase angle resembles the first derivative of the impedance modulus course. It was shown in5,27 that starting from the series expansion of Eq. (12), leading to Eq. (14): 2 lnjZðo0 Þj- lnjZð0Þj ¼  p

Zo0 ’ðoÞ  d ln o + 0

X k1, k odd

gk  ’k ðo0 Þ

(14)

the ZHIT approximation (15) can be derived, explaining the similarity between the slope of the impedance modulus course and the course of the phase angle: lnjZðo0 Þj 

2 p

Zo0 ’ðoÞ  d ln o − oS

p d’ðo0 Þ + const:  6 d ln o

(15)

The ZHIT demands that the impedance modulus course can be calculated from the integral of the phase angle plus some correction terms – in analogy to the reconstruction of the real part of the impedance course from the imaginary one after the KK transform Eq. (9). Different from the application of the LKK, the ZHIT works locally, and no “limited bandwidth problem” exists. The dominating term is the integral of the phase, which implies some noise reducing properties. No PV problem is complicating the numerical computation. The determination of the correction term relies on the derivatives of the phase, easily provided by means of a smoothing spline. The third term is equivalent to an additive shift of the reconstructed logarithm of the impedance modulus course, what is performed by a simple fitting to the experimental impedance data. So far, the ZHIT performance appears comparable with the features of the “measurement model” technique. But the reconstruction from the phase angle comprises a strong advantage of the ZHIT: The course of the phase angle is inherently much less sensitive to violations of steady state conditions, what is easily comprehensible by the following arguments: for example, the phase angle of a double layer capacity does not differ from −90 even if the capacity value changes arbitrarily. The same applies for all impedance elements with a general CPE behavior, such as the charge transfer resistance with a phase angle of zero: it remains at zero, even if the resistance value changes dramatically. This allows to calculate an approximation of the causally connected impedance modulus course from the phase angle even from data, which are affected by time drift. Time drift appears predominantly in the low frequency region. For the determination of the const. Value, the ZHIT implementation weights the experimental impedance modulus samples with emphasis on the middle and higher frequencies. The reconstructed impedance course therefore corresponds well with an experimental result, “as if” measured instantaneous at around the beginning of the measurement. Moreover, ZHIT allows for adjustment of the emphasis frequency and selectivity of the weighting function to different values by the operator. In this way, the impact of high frequency mutual inductance on the determination of the const. Value can be minimized. It must be mentioned that the outstanding reconstruction feature of ZHIT is limited to time drift phenomena. ZHIT can detect all further causality violations reliably, like the high frequency artifacts caused by mutual induction, but ZHIT is unable to remove them. As explained in the section around Fig. 11, for a best-case TD cancelling effect the ZHIT reconstruction feature should be applied to small TD artifacts, for example after a time series interpolation procedure. A simple form of TD interpolation – without tedious setting-up effort for the operator – is available also for a single impedance spectrum by the following strategy: if your data acquisition software allows, perform the frequency scan starting at the high frequency limit, down to the lowest frequency and scan back again to the high frequency end in one run. During the ZHIT flow, the software will automatically interpolate between the double measured frequency samples.

15

Summary/conclusions

EIS is a versatile and powerful technique for probing the electrochemical properties of materials and systems. By analyzing the impedance response across a range of frequencies and applying equivalent circuit models, researchers get valuable insights into the underlying processes governing the behavior of complex systems. This is particularly valid for low-Ohmic systems such as fuel cells or batteries. Algorithms are presented that allow for the detection of artifacts in the spectra. Further important techniques comprise DRT and DIA, which allow for detailed analysis of batteries and fuel cells. In addition, various validation algorithms such as ZHIT and the Measurement Model are presented, with which the spectra can be checked for plausibility. The application of all these

Methods and Instruments | Electrochemical Impedance Spectroscopy

59

techniques and algorithms provides the user with reliable information about fuel cells, battery or similar systems towards their successful improvement and development. Large-scale deployment of electrolysis and fuel cell systems or, likewise, batteries require systematic quality control and fast development of technology with improved efficiency. EIS is a well-established technique for various energy-related topics since several decades and is one of the few in situ techniques available. However, for most of these systems, generally accepted models are not available. It is important to establish standard models for the different technologies. EIS is particularly suitable if operating conditions are well defined and active cell areas are comparably small. One of the most important challenges is the transformation of the lab-scale knowledge and experience of research laboratories into the multi-gigawatt-scale industry.

References 1. Schiller, C. A.; Kaus, R. Consistent Discussion of the Uncertainty of Physical Parameters Evaluated by EIS Based on an Automatic Measurement Error Determination. ECS Trans. 2010, 25, 49–62. 2. Schiller, C. A.; Kaus, R. On-Line Error Determination and Processing for Electrochemical Impedance Spectroscopy Measurement Data Based on Weighted Harmonics Autocorrelation. Bulg. Chem. Commun. 2009, 41, 192–198. 3. iMSine n.d.: https://doc.zahner.de/application_notes/intelligent_multi_sine.pdf. 4. NFRA Technical Note n.d.: https://doc.zahner.de/technical_notes/nfra.pdf. 5. Schiller, C. A.; Richter, F.; Gülzow, E.; Wagner, N. Validation and Evaluation of Electrochemical Impedance Spectra of Systems with States That Change with Time. J. Phys. Chem. Chem. Phys 2001, 3, 374. 6. Göhr, H.; Richter, F. Einfluß löslicher oxidischer Produkte vorausgehender Oxidation von Platin-Oberflächen auf die Elektrosorption von Wasserstoff. Z. Phys. Chem. 1979, 115, 69–88. 7. Stoynov, Z.; Savova-Stoynov, B. S. Impedance Study of Non-stationary Systems: Four-Dimensional Analysis. J. Electroanal. Chem. 1985, 183, 133–144. 8. Göhr, H.; Röseler, B.; Schiller, C. A. Advantages of Hilbert Transform Compared with Kramers-Kronig Rule When Examining the Causality of Experimental Impedance Spectra. In ISE 46th Meeting 1995, Xiamen China, Extended Abstract 3.05; 1995. 9. Ehm, W. Expansions for the Logarithmic Kramers-Kronig Relations; 1998. www.zahner.de/downloads/ehm.pdf. 10. Ehm, W.; Kaus, R.; Göhr, H.; Schiller, C. A. The Evaluation of Electrochemical Impedance Spectra Using a Modified Logarithmic Hilbert Transform. Acta Chim. Hung. 2000, 137, 145. 11. Ehm, W.; Kaus, R.; Schiller, C. A.; Strunz, W. ZHIT—A Simple Relation between Impedance Modulus and Phase Angle, Providing a New Way to the Validation of Electrochemical Impedance Spectra. In Proceedings of the ECS, New Trends in Electrochemical Impedance Spectroscopy (EIS) and Electrochemical Noise Analysis (ENA); Mansfeld, F., Huet, F., Mattos, O. R., Eds.; vols. 2000–24; Electrochemical Society: Pennington, NJ, 2001; p. 1. 12. Agarwal, P.; Orazem, M. E.; Garcıa-Rubio, L. H. Measurement Models for Electrochemical Impedance Spectroscopy: I. Demonstration of Applicability. J. Electrochem. Soc. 1992, 139, 1917. 13. Boukamp, B. A. A Linear Kronig-Kramers Transform Test for Immittance Data Validation. J. Electrochem. Soc. 1995, 142, 1885. 14. Schönleber, M.; Klotz, D.; Ivers-Tiffée, E. A Method for Improving the Robustness of Linear Kramers-Kronig Validity Tests. Electrochim. Acta 2014, 131, 20–27. 15. Schiller, C. A.; Richter, F.; Gülzow, E.; Wagner, N. Relaxation Impedance as a Model for the Deactivation Mechanism of Fuel Cells Due to Carbon Monoxide Poisoning. J. Phys. Chem. Chem. Phys. 2001, 3, 2113. 16. Vladikova, D. Zdravko Stoynov—The Scientist Who Created New Scientific Horizons—Review. J. Electrochem. Sci. Eng. 2020, 10, 65–78. 17. Boukamp, B. A. Distribution (Function) of Relaxation Times, Successor to Complex Nonlinear Least Squares Analysis of Electrochemical Impedance Spectroscopy? J. Phys. Energy 2020, 2, 042001. 18. Fuoss, R. M.; Kirkwood, J. G. Electrical Properties of Solids. VIII. Dipole Moments in Polyvinyl Chloride-Diphenyl Systems. J. Am. Chem. Soc. 1941, 63, 385–394. 19. Schichlein, H.; Müller, A.; Voigts, M.; Krügel, A. Ivers-Tiffée, Deconvolution of Electrochemical Impedance Spectra for the Identification of Electrode Reaction Mechanisms in Solid Oxide Fuel Cells. J. Appl. Electrochem. 2002, 32, 875–882. 20. Dierickx, S.; Weber, A.; Ivers-Tiffée, E. How the Distribution of Relaxation Times Enhances Complex Equivalent Circuit Models for Fuel Cells. Electrochim. Acta 2020, 355, 136764. 21. Boukamp, B. A. Fourier Transform Distribution Function of Relaxation Times: Application and Limitations. Electrochim. Acta 2015, 154, 35–46. 22. Danzer, M. A. Generalized Distribution of Relaxation Times Analysis for the Characterization of Impedance Spectra. Batteries 2019, 5, 31. 23. Papoulis, A. The Fourier Integral and Its Applications; McGraw-Hill Book Co.: New York, 1962. 24. Bode, H. W. Network Analysis and Feedback Amplifier Design; Van Nostrand Company: Toronto, 1945. 25. Schüßler, H. W. Netzwerke, Signale und Systeme; Springer-Verlag: Berlin, 1984. Chapter V2. 26. Schüßler, H. W. Netzwerke, Signale und Systeme; Springer-Verlag: Berlin, 1984. Chapter V3. 27. Ehm, W.; Göhr, H.; Kaus, R.; Röseler, B.; Schiller, C. A. The Evaluation of Electrochemical Impedance Spectra Using a Modified Logarithmic Hilbert Transform. Acta Chim. Hung. 2000, 137, 145.

Methods and Instruments | Electrochemical Quartz Microbalance F Wudy, C Stock, and HJ Gores, Theoretische Chemie der Universität Regensburg, Regensburg, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of F. Wudy, C. Stock, H.J. Gores, MEASUREMENT METHODS | Electrochemical: Quartz Microbalance, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 660–672, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00079-4, with revisions made by the Editor.

1 Introduction 2 Function 2.1 Theory 2.1.1 Piezoelectric effect 2.1.2 Converse piezoelectric effect 2.1.3 Standing wave 2.1.4 Equivalent circuit diagram 2.2 Modeling 2.3 Quality factor 3 Quartz crystal microbalance setups 3.1 Impedance analysis 3.2 Oscillators 3.3 Dissipative analysis 3.4 Fast relative impedance analysis 3.5 Electrochemical quartz crystal microbalance setup 3.6 Calibration 4 Applications 4.1 Introduction 4.2 Lead-acid battery 4.3 Lithium-ion batteries Acknowledgments Further reading

62 62 62 62 62 62 64 66 68 69 69 70 71 71 72 73 73 73 73 74 76 76

Abstract The fundamental principle of the quartz crystal microbalance (QCM) and its use in electrochemistry are explained. Based on the converse piezoelectric effect, these oscillating sensor devices are able to detect mass changes down to the nanogram range. Mechanical networks and electrical equivalent circuits are used to introduce Sauerbrey’s equation. Moreover, model calculations explain the loading parameter R in terms of impedance analysis.

Key points

• •

Introduction of the basics of the electrochemical quartz microbalance techniques. Illustration of applications with emblematic examples related to battery research.

Nomenclature

Symbols and units

A C C0 Cf F f f0 fhw fp

60

Active surface of the quartz Capacitance Static capacitance Calibration factor Force Frequency Resonance frequency Half-width frequency Frequency value at the maximum of RQ

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00313-3

Methods and Instruments | Electrochemical Quartz Microbalance

j k L m M QE, Qt, Qhw, QDn r R RPA RPfb RQ RQmax td U vQ x XQ Z Zmot ZQ Zst dQ h1 l mQ np ns rl rm rQ f v

Current density Spring constant Inductance Mass Molarity Quality factors Coefficient of friction Resistance Resistor for amplification Feedback resistor Effective resistance Maximum resistance Time Cell potential propagation velocity in the quartz Distance in x-direction Reactance Impedance Impedance of the motional branch Quantity impedance Impedance of the static branch Thickness of the quartz crystal Viscosity of the liquid Wavelength Shear modulus of the quartz Parallel resonance frequency Series resonance frequency Density of the liquid Density of the material Density of the quartz Phase shift Angular frequency

Abbreviations and acronyms AAN AC ADC AFM CE CLK CSA CV DAC DC DDS DEC DMC EC EIS EQCM FIS-QCM FTIR ISO LIB LP

Acrylic acid nitrile Alternating current Analog to digital converter Atomic force microscopy Counterelectrode Clock Current sensor amplifier Cyclic voltammogram Digital to analog converter Direct current Direct digital synthesizer Diethylcarbonate Dimethyl carbonate Ethylene carbonate Electrochemical impedance spectroscopy Electrochemical quartz crystal microbalance Fast impedance scanning quartz crystal microbalance Fourier transform infrared Isolation Lithium-ion battery Low pass (filter)

61

62

Methods and Instruments | Electrochemical Quartz Microbalance

PLD PLL QCM QNB REF SEI USB WE XPS XRD

1

Programmable logic device Phase-locked loop Quartz crystal microbalance Quartz crystal nanobalance Reference electrode Solid electrolyte interface Universal serial bus Working electrode X-ray photoelectron spectroscopy X-ray diffraction

Introduction

The quartz crystal microbalance (QCM) is a method to detect minute mass changes on the surface of an electroacoustical sensor device mainly a thin slice of quartz crystal by detecting changes in the electromechanical oscillation behavior. The QCM is commonly used in all sectors of applied research since Sauerbrey made use of quartz crystals as sensor devices in the late 1950s. Therefore, the QCM technique is of great value not only for bioanalytical and life science technologies, but also for materials science and electrochemical field of research. For example, electrochemical metal deposition processes, metal corrosion and prevention, electropolymerization of conducting polymers, intercalation electrodes, and other electrodes are studied in addition to other scientific disciplines such as organic and inorganic chemistry where electrochemistry is currently experiencing a successful comeback. The combination of a potentiostat or a galvanostat with a fast impedance scanning QCM, an easy to integrate setup, is finally presented.

2 2.1

Function Theory

2.1.1 Piezoelectric effect Piezoelectricity was predicted and discovered in 1880 by Pierre and Jacques Curie at several materials including quartz crystals. The word piezoelectric originates from the Greek word ‘piezein’, meaning ‘to press’, and describes the appearance of an electric potential across certain faces of a special material showing this piezoelectric effect when the material is subjected to mechanical pressure. Torsional, compressional, shear, or flexural forces cause a displacement of electrical charge because of the deflection of the lattice. This effect is explained by the displacement of ions in materials that have a nonsymmetrical unit cell. When the material is compressed, the ions in each unit cell are displaced, causing an electric polarization of the unit cell. Because of the regularity of the material’s structure, these effects accumulate, causing the appearance of a measurable electric potential difference at electrodes attached to the faces of the crystal. Materials subject to this effect are synthetic ceramics, such as gallium orthophosphate (GaPO4), lead titanate (PbTiO3), lithium niobate (LiNbO3), and a-quartz (SiO2). The piezoelectric effect is completely reversible; on removing the forces, the electric potential completely disappears.

2.1.2 Converse piezoelectric effect Piezoelectric materials also show the opposite effect. The so-called converse piezoelectric effect occurs when an electric field is applied at two perpendicular faces. This entails a mechanical deformation of the crystal; in the case of an alternating electrical field, the mechanical strain correlates to the frequency of the alternating field. A suitable frequency causes an oscillation of the material at its resonance frequency. The oscillation behavior of the often used quartz crystal oscillators depends particularly on the way the quartz crystal was cut from the crystal (see Fig. 1). The optical Z-axis is an axis of threefold symmetry in quartz. The cuts are usually represented by two-letter names, e.g., AT, in which ‘T’ indicates a temperature-compensated cut. This often-used cut oscillates by performing a thickness shear vibration (see Fig. 2). A typical geometry used for such a quartz crystal sensor device can be seen in Fig. 3.

2.1.3 Standing wave Huygen’s principle describes the reflection of a wave front at the boundary of two materials of different densities, in this case the quartz crystal sensor and the surrounding medium. A standing wave is established by the relationship between the thickness of the quartz crystal oscillator and the wavelength, which reveals the simple postulation that the thickness (dQ) of the quartz crystal including the thin electrodes equals one-half of the wavelength (l) (compare Fig. 4), expressed as the following equation for the first harmonic oscillation: 1 dQ ¼ l 2

(1)

63

Methods and Instruments | Electrochemical Quartz Microbalance

35q15c AT

49q BT

CT 39q

XY

XY X

X

5q 5q

AT

Z CT Y

X

Fig. 1 Left: Quartz crystal, different types of quartz cut. Right: Mechanical behavior depending on the type of cut. Modes of vibration: flexure (XY), extensional (X), thickness shear (AT, BT), face shear (CT). A

B

C

+

+

+

+

+

+

+

–

–

–

–

–

–

–

–

–

–

–

–

–

–

+

+

+

+

+

+

+

U B

A

B

A

A

C

C

t

A C

Fig. 2 Thickness shear deformation (AT-cut) as a result of an external alternating electrical field caused by an alternating voltage U changing with time t at electrodes on the crystal’s surface.

The propagation velocity vQ of the wave front in quartz crystals is well known and is dependent on the resonance frequency f0: uQ ¼ l f 0 ¼ 3340 m s−1

(2)

with f0 ¼

uQ 2dQ

(3)

64

Methods and Instruments | Electrochemical Quartz Microbalance

dQ

Fig. 3 Typical geometry of a quartz crystal sensor device. Metal electrodes (gray) are deposited onto the quartz slice. Metals such as gold, aluminum, and titanium are commonly used. Different sensor diameters are available, typically very cost-effective 0.55-in.-diameter (1 in. ¼ 0.0254 m) and quite expensive 1-in.-diameter crystals.

dQ

Fig. 4 Resonance condition, the thickness (dQ) of the quartz crystal with its electrodes (gray) equals one-half the length of the wave. Table 1 Typical values for the thickness of a quartz crystal sensor and the resulting fundamental frequencies. dQ (mm)

f0 (MHz)

334 278 167 56

5 6 10 30

Typical thickness-frequency value pairs can be found in Table 1. With about 30 MHz and a thickness of 56 mm, the sensor devices would get extremely thin and therefore nearly unmanageable. To achieve higher frequencies, the oscillators are operated at the second, third, or an even higher harmonic frequency.

2.1.4 Equivalent circuit diagram The common model to describe the response of a quartz crystal oscillator is the electrical Butterworth–van-Dyke equivalent circuit, as shown in Fig. 5. The electrical network is represented by the motional branch consisting of a resistance (R), an inductance (L), and a capacitance (C) in serial connection, and by a static branch including a capacitance (C0) in parallel connection with this serial branch, representing accumulated capacitances of the metal electrodes, the cable, connectors, and other measurement equipment. As shown in Fig. 6, electroacoustic devices such as AT-cut quartz resonators can also be mechanically described as a lumped parameter network consisting of a deflectable mass m, a spring showing as acoustic compliance the ability to respond elastically when a force is applied, and a resistance describing the mechanical model of mass m, attached to the spring with a spring constant of 1/k, according to Hooke’s law, and a double-acting piston with a coefficient of friction r. The mechanical model and the electrical serial branch circuit correspond to each other; L represents the inertial component relating to the displaced mass (m) during oscillation, C reveals the energy stored in the oscillation (k), and R represents the energy dissipation factor during oscillation due to internal friction and mechanical and acoustical losses due to the environment (r). Therefore, the following relations hold:

Methods and Instruments | Electrochemical Quartz Microbalance

65

C0

R

L

C

Fig. 5 Basic electrical equivalent circuit according to Butterworth-van-Dyke.

r

k

m

Fig. 6 Basic mechanical model of the quartz crystal oscillation.

F ¼ m€ x + r x_ + kx 1 Q C

€ + RQ_ + U ¼ LQ

(4)

where F is the force and U the voltage. The serial branch is generally referred to as the motional branch. This branch defines the electromechanical characteristics of the quartz crystal resonator.

2.1.4.1 Sensor principle Following the deposition of a—in the best case—rigid and smooth film, for example, by electrodeposition of a rigid metal layer, the effective thickness of the quartz crystal sensor changes. This effect is the general basis for using this device as a very sensitive sensor for reactions in gas and liquid phases. As shown in Fig. 7, by depositing a film onto the sensor device, the thickness changes from dQ to dQ0 by DdQ. Therefore, the oscillation frequency changes from f0 to f0 by Df in an approximately linear manner: DdQ Df ¼− dQ f0

(5) qffiffiffiffiffiffiffiffiffiffiffiffiffiffi Given that f0 ¼ uQ/2dQ and uQ ¼ mQ =rQ , where mQ is the shear modulus and rQ the density of the quartz crystal, it follows that qffiffiffiffi ffi m Q

dQ ¼

rQ

(6)

2f 0

By deposition of mass Dm of a material with a known density rm onto the active surface A of the quartz, it follows that Dm

2f 0 Dm Df Arm qffiffiffiffi ffi ¼ qffiffiffiffi ¼− mQ mQ f0 Ar rQ

m

(7)

rQ

2f 0

Assuming that rm ¼ rQ, known as Sauerbrey’s approximation, the Sauerbrey equation follows: 2f 20 Df ¼ − pffiffiffiffiffiffiffiffiffiffiffi ffi Dm A rQ mQ

(8)

66

Methods and Instruments | Electrochemical Quartz Microbalance

dQ

a

d cQ

Fig. 7 Deposition process (a) of a film (gray) onto the quartz crystal’s electrodes (light gray). This results in an increase of the thickness (dQ), and, therefore, in an increased wavelength of the standing wave and a decrease in resonance frequency.

or even simpler,

Df ¼ Cf Dm

(9)

where Cf represents the calibration factor, which can be easily determined experimentally by combination with Faraday’s law after electrodeposition of silver or copper. Also, Cf is related to the square of the quartz’s resonance frequency f0. If a quartz crystal with a resonance frequency of 5 MHz is chosen, calibration factors of about 175 Hz mg−1 can be found. Using a quartz crystal with a resonance frequency of 6 MHz, calibration factors of about 250 Hz mg−1 can be achieved. With the ability to determine the change in frequency with a resolution of 1 Hz, mass resolutions down to 6 ng for the quartz crystal sensor with 5 MHz or 4 ng for the sensor with 6 MHz fundamental frequency are possible. Because of this, the QCM is often also called quartz-nanobalance. K. K. Kanazawa and J. G. Gordon introduced an enhanced form of the change in resonance frequency taking into account viscosity and density relations of such a QCM sensor in contact with a liquid: rffiffiffiffiffiffiffiffiffiffiffiffiffiffi l rl 3=2 Df ¼ −f 0 (10) prQ mQ where rl is the density of the liquid and l is the viscosity of the liquid.

2.2

Modeling

According to the Butterworth-van-Dyke model in Fig. 5, the electronic behavior of a quartz crystal sensor can be easily modeled. The complex quantity impedance (ZQ) of such a device can be derived as a parallel connection of the impedances of the motional branch (Zmot) and the static branch (Zst): 1 1 1 ¼ + ZQ Zmot Zst

(11)

where Zmot ¼ ZR + ZL + ZC, ZR ¼ R, ZL ¼ ioL, ZC ¼ 1/ioC, Zst ¼ ZC0 ¼ 1/ioC0, and o ¼ 2pf. The reactance is defined as the imaginary part of the impedance, XQ ¼ I(ZQ), and the effective resistance is the real part of the qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi impedance, RQ ¼ ℜ(ZQ). The magnitude of the impedance is defined as |ZQ | ¼ RðZQ Þ2 + IðZQ Þ2 ¼ R2Q + X 2Q . In Table 2, typical values for the electric components in the equivalent circuit are listed. Based on these data, the frequency responses of |ZQ|, XQ, and RQ are calculated and are shown in Fig. 8 for a medium loading of the system, simulated for R ¼ 1000 O. From the three curves in Fig. 8, it is clear that |ZQ| shows a sharp minimum and a sharp maximum. At the minimum of |ZQ|, XQ shows zero-crossing, at the maximum, RQ shows a maximum, too, whereas XQ shows an inflection point. The corresponding frequency at the minimum of |ZQ| is defined as the series resonance frequency (us). The corresponding frequency at the maximum is defined as the parallel resonance frequency (up). Mathematical evaluation of Eq. (11) provides the analytical expressions for these two frequencies, assuming that R ¼ 0: vs ¼

1 1 pffiffiffiffiffiffi and vp ¼ qffiffiffiffiffiffiffiffiffiffiffiffi 0 2p LC 2p L CC C + C0

(12)

According to Fig. 9, it is obvious that an increase in R resulting from an increase of mechanical damping causes a less distinctive change in the |ZQ| signal. Also, the divergence of us and up can be observed clearly, as shown in Fig. 10.

Methods and Instruments | Electrochemical Quartz Microbalance

67

Typical values for the electrical components.

Table 2 Component

Typical values 100–10,000 O 75  10−3H 10−14 F 10−12 F

a

R L C C0 a

It is stressed that R depends on the load of the quartz crystal.

log10 (|ZQ|/:)

XQ,RQ (:)

ZQ

RQ

0

XQ np

ns f (MHZ)

log10 (|ZQ| :–1)

Fig. 8 Simulated behavior of a quartz crystal sensor according to the values specified in Table 2. R is set to 1000 O, a moderate damping value, e.g., to be found in liquid systems.

100 000

10 000

1000 5.80

5.82

5.84

5.86

f (MHZ)

np (MHz)

Fig. 9 Simulated frequency behavior of |ZQ| for various loading parameters R: 1000 O (black), 5000 O (dark gray), 10,000 O (light gray).

5.87 5.86 5.85 5.84

ns (MHz)

5.81 5.80 5.79 5.78 0

10

20

30 R (k:)

Fig. 10 Resonance frequencies as a result of the loading parameter R.

40

50

68

Methods and Instruments | Electrochemical Quartz Microbalance

RQ (k:)

600

400

200

0 5.82

5.84

5.86

f (MHz)

Fig. 11 Simulated frequency behavior of RQ for various loading parameters R: 1000 O (black), 5000 O (dark gray), 10,000 O (light gray).

f hw (HZ)

105 104 103 102

RQmax (:)

107 105

0

10

20

30

40

50

R (k:)

Fig. 12 Change of the half-width frequency fhw and the maximum RQmax of the Lorentzian-type curve with the change of the loading parameter R.

In analogy to |ZQ|, the RQ behavior changes drastically (see Fig. 11). The Lorentzian-type curve increases in its half-width frequency (fhw) and decreases in its maximum resistance (RQmax) with the increase in damping parameter R (detailed values can be found in Fig. 12). The frequency of the maximum changes only very slightly with R.

2.3

Quality factor

The quality factor QE is generally defined as QE ¼ 2p

stored energy in oscillation dissipated energy

(13)

typically having values in the range of 103–106 if the oscillation is not damped by the environment, e.g., in vacuum. Fig. 13 shows the oscillation behavior for an (ideally) nondamped and a damped case. For higher damping ratios, the envelope function is more dominant, and the signal’s amplitude decreases faster. The quality factor Qt is proportional to the time td the signal needs to decrease to 1/e of its original amplitude: Qt ∝t d

(14)

In analogy to the modeling results, the quality factor can also be defined by the half-width frequency (fhw) and the frequency value at the maximum of RQ (fp) (see Fig. 11), as Qhw ¼

fp f hw

(15)

Methods and Instruments | Electrochemical Quartz Microbalance

69

Amplitude

1

0 e–1

–1 td

0

Time

Fig. 13 Decreasing amplitude (black) of the oscillation for a quartz crystal oscillator in contact with a damping medium. Excitation stops at time stamp 0. The ideal completely undamped case is shown by the gray graph.

4 × 104

2 × 104

75

Q hw

Q'n

150

0 0

10

20

30 R (k:)

40

50

Fig. 14 Comparison of the two quality factors Qhw and QDu as a result of a change in the loading parameter R based on the simulation parameters of Table 2.

or as a value described by the divergence of the resonance frequencies (see Fig. 9): vs QDv ¼ vp − vs

(16)

Fig. 14 compares the behavior of these last two quality factors.

3 3.1

Quartz crystal microbalance setups Impedance analysis

By using an impedance analyzer, typical Bode diagrams of the quartz crystal sensor device can be directly evaluated, as shown in Fig. 15. Comparison of Fig. 15 with Figs. 8 and 9 shows that the measured data equal the predicted data for |ZQ|. Fitting algorithms, mostly Levenberg-Marquardt, and simplex methods are used to determine the values of the electrical parts in the Butterworth-van-Dyke equivalent circuit and its variations. Although a detailed information on the sensor state is obtained, the fitting procedure itself is still quite unstable, as these nonlinear methods depend dramatically on the chosen start parameters and the frequency range. As the phase information has to be determined, these network analyzers are quite slow. A sampling rate of one data point in the spectrum per second has to be considered as fast. For fast in situ measurements, this method is barely usable. General-purpose network analyzers for simple sensor applications are barely indicated, as dimension and price remain high.

70

Methods and Instruments | Electrochemical Quartz Microbalance

10 000 A |Z | (:

1000

B

100

90

jq)

45

A

0 B

–45 –90 4.976

4.980

f (MHz)

4.984

4.988

Fig. 15 Typical Bode diagram, showing |ZQ| and the phase shift ’. The measurement was taken in free air (A) and under water (B). Data were acquired using a Solartron 1260 impedance analyzer for use in electrochemistry. The complete acquisition took about 3 h.

RPfb

ICP1 RPA

Q CP1

CP2

p

Fig. 16 Basic quartz crystal oscillator of the Pierce type.

3.2

Oscillators

As crystal oscillator devices were originally used as frequency-determining parts in electronic oscillators for military and broadcast engineering, this kind of setup is the earliest. The quartz crystal sensor acts as an active device in an electric oscillator circuitry. Such a basic oscillator circuit is shown in Fig. 16. The quartz crystal (Q) is supposed to be effectively inductive and in a p-network with CP1 and CP2. This network provides 180 phase shift, necessary to sustain oscillation in combination with the inverting amplifier ICP1 that provides a phase shift of 180 itself, to fulfill the oscillation condition. Both CP1 and CP2 have to be chosen carefully as they rely on the crystal’s parameter, and therefore affect the output frequency; RPA is used to limit the crystal’s drive level by forming a voltage divider. It stabilizes the oscillator against changes in the amplifier’s output resistance, too. As all the parts, such as CP1, CP2, RPA, the crystal oscillator (C0), the amplifier, the cables, and connectors, induce parasitic capacities, the working point of the oscillator has to be tuned manually each time if something changes in the circuit. As the damping factor R can change during measurement, adaptations have to be made for the oscillator circuit to resolve problems with transient oscillation behavior, missing zero-crossing of the phase shift (compare Fig. 15), spurious phase shifts induced by the electric setup, and information loss due to reduced slopes in the phase shift, as the working point of the oscillator is increasingly poorly defined. Therefore, dozens of oscillator circuits are known in literature, from automatic gain control circuits over balanced bridge oscillators to phase-locked loops their; working depends on the fundamental design near the series or the parallel resonance frequency. Frequently, the user has to take care of the oscillation

Methods and Instruments | Electrochemical Quartz Microbalance

71

behavior by manual frequency tuning, even during the measurement, which can result in frequency hops. To use such an oscillator circuit in electrochemistry, setups have to be found, having one of the sensor’s electrodes at ground or at various DC potentials to establish functionality independent of the changes in capacity at one of the electrodes. After establishing a stable and reliable oscillation, the oscillation frequency has to be determined. This can be done simply by counting events (gated by a stable high-frequency reference clock), using reciprocal counters or with appropriate mixing and filtering technologies with subsequent counting. Compared to the general impedance analysis, the oscillator methods are quite fast; however, they cannot provide in-depth information as the impedance methods are able to provide.

3.3

Dissipative analysis

Dissipative (also called decay) methods are based on one of the other methods, as the sensor crystal is excited to oscillation by an oscillator or a variation of an impedance analyzing setup, in most cases a kind of frequency generator. After excitation of the series or parallel resonance frequency of the desired harmonic, the excitation signal is interrupted by opening a switch separating the excitation signal source from the sensor device (see Fig. 17). After opening the switch, the voltage or current signal decays according to an exponentially damped sinusoidal signal (see Fig. 13), which can be expressed mathematically using a common decay function: A ¼ sinð2pft + ’Þe−t=t d A0

(17)

This method yields the quality factor (see Eq. 14) or its change during a process, additional information over the methods forming its base. Therefore, it shows the same characteristics concerning speed and information density as the basing methods.

3.4

Fast relative impedance analysis

Recently developed in Regensburg, the FIS-QCM (FIS, fast relative impedance scanning) represents the advantages of the general impedance analysis gathering a high extent of information and the high speed of the oscillator methods. Not collecting phase shift information and gathering only the change in the magnitude of the impedance, |ZQ|, a sampling rate of over 500,000 data points per second in the spectrum can be achieved, with a maximum possible frequency scan resolution of about 20 mHz. The absolute value for |ZQ| and the phase information are negligible over the fundamental increase of speed. Assuming typical parameters for a slightly loaded quartz crystal, a full spectrum containing serial and parallel resonance frequency, with a bandwidth of 20 kHz, can be scanned within 200 ms at a resolution of 0.2 Hz. This type of QCM was especially developed for electrochemical use. By having direct digital interlink to a potentiostat/galvanostat, the data sets from the QCM and the potentiostat can be paired without detouring over analog inputs and outputs. All digital lines for data transmission are built up DC isolated, so no ground loops can occur. As shown in Fig. 18, the core of this type of QCM is built up by a programmable logic device (PLD). The PLD is connected to a direct digital synthesizer (DDS) that produces a sinusoidal signal, variable in frequency and amplitude. A low-pass filter removes clock feed-through before it is amplified. After this, the signal is coupled by a matching network into the quartz crystal. After coupling out the signal using the same network, the signal is rectified. An analog to digital converter digitizes the gathered signal, and the PLD forms the mean value of at least two measurements. Collected data are transferred to a personal computer (PC), running a runtime optimized program written in C to carry out fitting operations to extract information like the series and parallel resonance frequencies (ns, np), the root mean square deviation, and the reduced quality factor (cf. Eq. 16), among others. As nonlinear fitting algorithms often fail using the transfer function for the Butterworth–van-Dyke equivalent circuit and its adoptions, an alternative linear fittable function based on the following Padé approximant was used: A

A

+ _

1

t

CSA

t

Fig. 17 Decay measurement methods. Left: Setup for working near the parallel resonance frequency. The voltage decay function is determined. Right: Setup for the series resonance frequency. The current over the resistor is determined via the current sense amplifier (CSA). Additionally, C0 can be compensated with this resistor.

72

Methods and Instruments | Electrochemical Quartz Microbalance

DAC Ampl

PLD + PC

DDS

~

LP

AC cpl.

CLK

ADC 14bit

TRMS

ISO

USB

Computer

Fig. 18 Simplified diagram of the fast relative impedance scanning quartz crystal microbalance (QCM). See text for details. DDS, direct digital synthesizer.

y¼

P ðf Þ A + Bf + Cf 2 ¼ 1 + Qðf Þ 1 + Df + Ef 2 + Ff 3

(18)

Linear fitting is not only numerically more stable, but it is also much faster, as it is a noniterative method. A typical spectrum containing 50,000 data points can be fitted within 30 ms on an actual PC. From the fitting parameters A–F, ns, np, and some other information are evaluated analytically. The hardware of the FIS-QCM fits on a single printed-circuit board with a size of 100  160 mm, so the small-sized equipment can be brought very close to the measurement setup.

3.5

Electrochemical quartz crystal microbalance setup

The combination of the QCM with an electrochemical setup, commonly a potentiostat or a galvanostat, forms the typical electrochemical quartz crystal microbalance (EQCM) setup, as shown in Fig. 19. One of the electrodes of the quartz sensor device in contact with an electrolyte acts as the working electrode, and the other electrode is used for the QCM functionality. Usually, the working electrode’s signal is looped electrically through the QCM device. The frequency information of the QCM is typically brought to the potentiostat by converting the frequency to an appropriate voltage information that can be sampled by an auxiliary analog input of the potentiostat. The FIS-QCM is able to pair the electrochemical data and frequency values without analog losses. The combination of these two methods is a powerful approach in numerous fields of applications such as the study of corrosion of metals and corrosion protection, investigation of electrodeposition of various metals, semiconductors, and insulating materials from several electrolytes, examination of current efficiencies of electroplating processes, intercalation studies, investigation of electropolymerization of conducting polymers, utilization of electronic noses and tongues, and finally in complete sensor arrays. CE P/G

REF

U(WE) QCM

WE

Q 10q

Fig. 19 Combination of the quartz crystal microbalance (QCM) with electrochemical equipment, typically a potentiostat or galvanostat (P/G). This results in the typical electrochemical quartz crystal microbalance (EQCM) setup. The working electrode (WE) is formed by one of the quartz’s (Q) electrodes. Counterelectrode (CE) and reference electrode (REF) are connected in a typical manner.

Methods and Instruments | Electrochemical Quartz Microbalance 3.6

73

Calibration

Sensor crystals can be calibrated by electroplating copper onto the surface of the quartz. A typical electrolyte consists of an aqueous solution of 1 mol L−1 H2SO4 and 0.1 mol L−1 CuSO45H2O. A calibration curve can be drawn by gathering a cyclic voltammogram (CV) and detecting simultaneously the frequency changes of the sensor quartz by QCM (see Fig. 20). A standard three-electrode arrangement with a platinum-coated 0.55-in.-diameter (1 in. ¼ 0.0254 m) quartz as working electrode, a platinum foil as counterelectrode, and a saturated mercury-mercurous sulfate reference electrode was used in this example. The deposited mass m is, according to Faraday’s law, directly proportional to the passed charge Q, and Dns plotted against the charge Q results in a linear dependency, as expected. By fitting a linear equation on the experimental data, the calibration factor can be determined from the slope. dDvs ¼ 53:45  0:39 Hz mC−1 dQ Cf

4 4.1

dDvs nF ¼ 162:31  0:12 Hz mg−1 dQ MCu

Applications Introduction

In the field of battery research, the EQCM has found a place among other methods, especially surface analysis methods including Fourier transform infrared (FTIR) spectroscopy, in situ atomic force microscopy (AFM), X-ray photoelectron spectroscopy (XPS), and electrochemical impedance spectroscopy (EIS). Electrochemical quartz crystal microbalance measurements were used to study the corrosion behavior of current collectors for lithium-ion batteries (LIBs), film formation at anode and cathode materials, intercalation, polymerization of additives, alloying, and dealloying, as they yield mass changes at the working electrodes at given electrode potentials. Unfortunately, electrode reactions are often very complex, leading to various products at the electrode. Therefore, only in very advantageous cases, a direct interpretation of mass changes at the QCM electrode is possible. Electrochemical quartz crystal microbalance has some general limitations when applied to novel reaction mechanisms. Therefore, it has to be applied in combination with surface analysis methods. Finally, often deposited films do not follow the Sauerbrey equation and viscoelasticity has to be taken into account. In the following section, some results from our own experiments and from literature for lead–acid batteries and LIBs are given.

4.2

Lead-acid battery

The lead-acid battery as a well-known system was subject to EQCM investigation. A lead electrode was electrodeposited onto one electrode of a 5 MHz platinum sensor quartz, resulting in a mass change of 60 mg, which equals a layer of lead with a thickness of 170 nm. Using an electrolyte of 27% sulfuric acid and a polished lead wire as counterelectrode, a CV experiment was performed in a two-electrode arrangement (see Fig. 21). The cell voltage was linearly varied from +2 to −0.1 V at a scan rate of 25 mV s−1. The electrochemical reduction of lead dioxide (PbO2) to lead sulfate (PbSO4) can be recognized at the region of process I, causing Dns to increase, yielding in reaction II the complete reduction of lead oxide (PbO) and lead sulfate to lead. Similar results were published by M. Taguchi. The maximum of Dns is reached for the pure lead surface, where the molar volume is the lowest with

0

'ns (Hz)

–1000 –2000 –3000 –4000 –5000 0

–20

–40

–60

–80

–100

Q (mC)

Fig. 20 Calibration measurement: Charge Q plotted against the change of the series resonance frequency Dns as received from the cyclic voltammogram cyclic voltammogram (CV) experiment’s first cycle.

74

Methods and Instruments | Electrochemical Quartz Microbalance

0.5 IIIc j (mAcm–2)

IIc

Ic

0.0 II I

–0.5

'ns (Hz)

50

0

Q'n

1289

1288

0.00

0.50

1.00

1.50

2.00

U (V)

Fig. 21 Electrochemical quartz crystal microbalance (EQCM) experiment at lead electrodes. Data were obtained from the fifth cycle of the cyclic voltammogram (CV) experiment. The current density ( j), the change in the series resonance frequency (Dns), and the quality factor (QDu) are shown with respect to the cell potential (U).

18 cm3 mol−1 compared to lead sulfate with 48 cm3 mol−1 or b-lead(IV)oxide with 25 cm3 mol−1. The QDu curve, giving information on the roughness, reaches its minimum after process II. The slow reduction of the oxide layer leads to a smooth lead surface as shown by AFM measurements published by Y. Yamaguchi and coworkers. The flat structure reduces the QDu factor to its lowest level during the cycle. After inverting the polarization direction, lead is dissolved until the solution is saturated. Because of the fast process III0 , the current density j is high and the change in frequency is low. During processes II0 and I0 , the surface is oxidized stepwise from Pb2+ to Pb4+. The effects on the frequency can be seen clearly in Fig. 21 by different slopes of the Dns curve. Fig. 22 shows the change of Dus with respect to the passed charge Q. Here the effects on the electrode material can be observed. As discussed, during the stepwise reduction I and II, the Dns values increase, because of the reduction of the molar volume. It is easy to identify the different processes by the slopes of particular segments, as it is discussed later. The horizontal sections do not contribute to reorganization of the electrode during the polarization.

4.3

Lithium-ion batteries

Current collectors for positive electrode materials in commercial LIBs are made from aluminum foils. Although aluminum is stable against corrosion in air, its use as a current collector in LIBs may limit the lifetime of the battery, as its corrosion depends strongly on the composition of the electrolyte. Whereas severe corrosion was observed for solutions of Li[N(SO2CF3)2] in a mixed solvent composed of ethylene carbonate (EC) and dimethyl carbonate (DMC) under anodic polarization conditions, already a partial replacement of this salt by 20% lithium tetrafluoroborate (LiBF4) resulted in suppression of corrosion owing to the formation of a stable passive layer. Similar results were obtained for LiPF6/EC/DMC. Electrochemical quartz crystal microbalance experiments show that the anodic processes on aluminum in the organic electrolytes consist of the formation of surface films and their dissolution. The lithium salt LiB(C2O4)2 (LiBOB) is a possible alternative for the commonly used salt LiPF6. It also shows the

Methods and Instruments | Electrochemical Quartz Microbalance

75

80

'Q s (Hz)

60 40

II IIc

20

I

0 –20

Ic 0

–1

–2

–3

–4

Q (mC)

Fig. 22 Dns plotted against Q: processes I/I0 with a slope of 11 Hz mC−1 and II/II0 with 85 Hz mC−1 can be identified.

effect of the formation of a protective film when aluminum is polarized above 4.5 V in 1:1 EC/DMC with 1 mol L−1 LiBOB. From EQCM measurements, it was concluded that AlBO3 films are formed, preventing corrosion even in strongly corroding Li[N(SO2CF3)2]-based solutions. The passive film that is formed on aluminum in 1:1 EC/ethylmethyl carbonate with 1.2 mol L−1 lithium hexafluorophosphate (LipF6) and 1:1 EC/DMC with 1.0 mol L−1 LipF6 was studied by EQCM, EIS, and XPS. During anodic polarization of the aluminum, a film of aluminum fluoride (AlF3) forms on the air-formed oxide, creating a two-layered film. The thickness of the AlF3 increases with the applied potential. Independent measurements of film thickness by EQCM and EIS indicate that, at a potential of 5.5 V vs. Li/Li+, the thickness of the AlF3 is approximately 1 nm. Neither pure lithium nor lithium intercalated in various types of carbons, silicon, or tin is stable with the electrolyte. A solid electrolyte interface (SEI) is formed, conducting lithium ions and preventing a further reaction of the anode material with the electrolyte. The properties of the SEI are crucial for both lithium batteries and LIBs. Several film-forming electrolyte additives are well known to create effective SEIs. In several papers, D. Aurbach and coworkers have reported studies of SEI film formation at anode materials. By combination of various in situ methods including in situ FTIR spectroscopy, in situ AFM imaging, and EIS, they were able to study the SEI formed after electrodeposition of lithium at quartz crystals on which thin gold, silver, or nickel electrodes were deposited. From the EQCM experiments, the authors concluded that only a part of the charge is consumed for film formation. In addition, a continuous dissolution of a part of the reduction products is observed until the solution reaches the saturation concentration of soluble reduction products. In a comparative study by FTIR spectroscopy, XPS, EQCM, in situ AFM imaging, and electrochemical methods of LiBOB- and LiPF6-based solutions, LiBOB showed a better passivation of lithium and lithium-graphite electrodes. B. O. Besenhard and coworkers showed that acrylic acid nitrile (AAN) is an effective SEI-forming additive for LIBs even at concentrations of 1%. Experiments performed by in situ methods such as FTIR and EQCM showed an SEI formation mechanism, which proceeded via cathodically induced polymerization of AAN. Carbon as the standard lithium intercalating material in commercial LIBs can be substituted by lithium-alloying materials such as tin and silicon, showing higher specific capacities of 990 mAh g−1 (Li–Sn) and 4000 mAh g−1 (Li–Si) respectively, when compared to 370 mAh g−1 (Li–C). However, Li–Sn exhibits poor cycling ability and a large irreversible capacity in the first charge/discharge cycle. To understand the reactions including alloying/dealloying, solvation/desolvation, and decomposition of the electrolyte, Sn–Li was investigated using EQCM and in situ microscope FTIR spectroscopy. Electrochemical quartz crystal microbalance studies revealed that the mass accumulated per transmitted charge is smaller than the theoretical value for electrolyte decomposition whereas it is higher than the theoretical value during alloying/dealloying. The limited cycle life of the Li–Si electrode is ascribed to severe volume changes of the electrode. In crystalline silicon, during lithium charging, cracking and pulverization of the silicon-lithium alloy is reported whereas little capacity loss is observed for amorphous silicon thin films. Amorphous silicon thin-film electrodes were studied by EQCM and electrochemical techniques. Even before lithium deposition was started, an increase in the mass of the silicon electrode was observed, which was attributed to film formation by reduction products of the 1.3 mol L−1 LiPF6/EC/DEC electrolyte entailing a reduced charge capacity and cycleability. In the presence of additives containing alkoxy functional groups, those negative processes were reduced, showing that the additive is an effective passivating agent for Li–Si anodes. Nanometer thin films of cobalt oxide and antimonide, and metallic cobalt and antimony were prepared by pulsed laser deposition using a stoichiometric target and characterized by X-ray diffraction (XRD) and EQCM measurements. V. Pralong and coworkers state that EQCM may present new opportunities for studying electrode reactions. Lithium cobalt oxide is a cathode material that is widely used in LIBs. The spinel LiMn2O4 is a possible substitute because of its various attractive features, including low cost and low toxicity as well as ease of preparation. However, its performance is affected by

76

Methods and Instruments | Electrochemical Quartz Microbalance

surface films. Lithium manganese oxide films prepared by an electrostatic spray deposition technique were investigated by XRD, scanning electron microscopy, EQCM, and electrochemical experiments including EIS in 1 mol L−1 LiClO4/EC/DEC solutions. It was shown that the surface film was formed after an initial dissolution of a lithium carbonate (Li2CO3) layer. The mass of the surface film has been calculated from the resonant frequency curve to be about 1.2% of the mass of the entire LiMn2O4 film.

Acknowledgments Thanks to D. Herrmann and D. Moosbauer for their contributions.

Further reading 1. Arnau, A. A Review of Interface Electronic Systems for AT Cut Quartz Crystal Microbalance Applications in Solutions. Sensors 2008, 8, 370–411. 2. Aurbach, D. Review of Selected Electrode–Solution Interactions Which Determine the Performance of Li and Li Ion Batteries. J. Power Sources 2000, 89, 206–218. 3. Aurbach, D.; Moshkovich, M.; Cohen, Y.; Schechter, A. The Study of Surface Film Formation on Noble-Metal Electrodes in Alkyl Carbonates/li Salt Solutions, Using Simultaneous In Situ AFM, EQCM, FTIR, and EIS. Langmuir 1999, 15, 2947–2960. 4. Buck, R. P.; Lindner, E.; Kutner, W.; Inzelt, G. Piezoelectric Chemical Sensors (IUPAC Technical Report). Pure Appl. Chem. 2004, 76, 1139–1160. 5. Bund, A.; Schneider, O.; Dehnke, V. Combining AFM and EQCM for the In Situ Investigation of Surface Roughness Effects during Electrochemical Metal Depositions. Phys. Chem. Chem. Phys. 2002, 4, 3552–3554. 6. Buttry, D. A.; Ward, M. D. Measurement of Interfacial Processes at Electrode Surfaces with the Electrochemical Quartz Crystal Microbalance. Chem. Rev. 1992, 92, 1355–1379. 7. Czerwinski, A.; Zelazowska, D.; Grden, M.; et al. Electrochemical Behavior of Lead in Sulfuric Acid Solutions. J. Power Sources 2000, 85, 49. 8. Eichelbaum, F.; Borngraeber, R.; Schroder, J.; Lucklum, R.; Hauptmann, P. Interface Circuits for Quartz-Crystal-Microbalance Sensors. Rev. Sci. Instrum. 1999, 70, 2537–2545. 9. Kanazawa, K. K. Some Basics for Operating and Analyzing Data Using the Thickness Shear Mode Resonator. Analyst 2005, 130, 1459–1464. 10. Kanazawa, K. K.; Gordon, J. G. The Oscillation Frequency of a Quartz Resonator in Contact with Liquid. Anal. Chim. Acta 1985, 175, 99–105. 11. Kankare, J.; Loikas, K.; Salomaki, M. Method for Measuring the Losses and Loading of a Quartz Crystal Microbalance. Anal. Chem. 2006, 78, 1875–1882. 12. Kipling, A. L.; Thompson, M. Network Analysis Method Applied to Liquid-Phase Acoustic Wave Sensors. Anal. Chem. 1990, 62, 1514–1519. 13. Lee, S.; Hinsberg, W. D.; Kanazawa, K. K. Determination of the Viscoelastic Properties of Polymer Films Using a Compensated Phaselocked Oscillator Circuit. Anal. Chem. 2002, 74, 125–131. 14. Martin, S. J.; Spates, J. J.; Wessendorf, K. O.; Schneider, T. W.; Huber, R. J. Resonator/oscillator response to liquid loading. Anal. Chem. 1997, 69, 2050–2054. 15. Moller, K. C.; Santner, H. J.; Kern, W.; Yamaguchi, S.; Besenhard, J. O.; Winter, M. In Situ Characterization of the SEI Formation on Graphite in the Presence of a Vinylene Group Containing Film-Forming Electrolyte Additives. J. Power Sources 2003, 119, 561–566. 16. Morita, M.; Shibata, T.; Yoshimoto, N.; Ishikawa, M. Anodic Behavior of Aluminum in Organic Solutions with Different Electrolytic Salts for lithium Ion Batteries. Electrochim. Acta 2002, 47, 2787–2793. 17. Pralong, V.; Leriche, J.-B.; Beaudoin, B.; Naudin, E.; Morcrette, M.; Tarascon, J.-M. Electrochemical Study of Nanometer Co3O4, Co, CoSb3 and Sb Thin Films Toward Lithium. Solid State Ion. 2004, 166, 295–305. 18. Rodahl, M.; Hook, F.; Kasemo, B. QCM Operation in Liquids: An Explanation of Measured Variations in Frequency and Q Factor With Liquid Conductivity. Anal. Chem. 1996, 68, 2219–2227. 19. Ryu, Y.; Lee, S.; Mah, S.; et al. Electrochemical Behaviors of Silicon Electrode in Lithium Salt Solution Containing Alkoxy Silane Additives. J. Electrochem. Soc. 2008, 155, A583–A589. 20. Sauerbrey, G. Verwendung Von Schwingquarzen Zur Wa¨ gung du¨ nner Schichten und zur Mikrowa¨ gung. Zeitschrift fur Physik A 1959, 155, 206–222. 21. Schroder, J.; Borngraeber, R.; Eichelbaum, F.; Hauptmann, P. Advanced Interface Electronics and Methods for QCM. Sens. Actuators A 2002, 97, 543–547. 22. Shu, D.; Chung, K. Y.; Cho, W. I.; Kim, K. Electrochemical Investigations on Electrostatic Spray Deposited LiMn2O4 Films. J. Power Sources 2003, 114, 253–263. 23. Song, H.; Sung, J.; Jung, Y.; et al. Electrochemical Porosimetry. J. Electrochem. Soc. 2004, 151, E102–E109. 24. Stein, N.; Rocca, E.; Kleim, R.; Lecuire, J. M.; McRae, E. In-Situ Ellipsometric Study of Lead Sulfate Film Electroformation on Lead in a Sulfuric Acid Solution. Electrochim. Acta 1998, 44, 445–454. 25. Taguchi, M.; Sugita, H. Analysis for Electrolytic Oxidation and Reduction of PbSO4/Pb Electrode by Electrochemical QCM Technique. J. Power Sources 2002, 109, 294–300. 26. Vatankhah, G.; Lessard, J.; Jerkiewicz, G.; Zolfaghari, A.; Conway, B. E. Dependence of the Reliability of Electrochemical Quartz-Crystal Nanobalance Mass Responses on the Calibration Constant, Cf: Analysis of Three Procedures for Its Determination. Electrochim. Acta 2003, 48, 1613–1622. 27. Wudy, F.; Multerer, M.; Stock, C.; Schmeer, G.; Gores, H. J. Rapid Impedance Scanning QCM for Electrochemical Applications Based on Miniaturized Hardware and High-Performance Curve Fitting. Electrochim. Acta 2008, 53, 6568–6574. 28. Yamaguchi, Y.; Shiota, M.; Nakayama, Y.; Hirai, N.; Hara, S. Combined In Situ EC-AFM and CV Measurement Study on Lead Electrode for Lead–Acid Batteries. J. Power Sources 2001, 93, 104–111. 29. Zhang, X.; Devine, T. M. Passivation of Aluminum in Lithium-Ion Battery Electrolytes With LIBOB. J. Electrochem. Soc. 2006, 153, B365–B369.

Methods and Instruments | Differential Electrochemical Mass Spectrometry Zenonas Jusysa,b,c and R Jürgen Behma, aInstitute of Theoretical Chemistry, Ulm University, Ulm, Germany; bHelmholtz Institute Ulm Electrochemical Energy Storage (HIU), Ulm, Germany; cKarlsruhe Institute of Technology (KIT), Karlsruhe, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 Introduction 2 DEMS cell designs 2.1 Membrane inlet 2.2 Pinhole inlet systems and hydrodynamic DEMS techniques 3 Challenges arising from solvent permeation, product fragmentation and follow up reactions in the electrolyte 4 DEMS approaches using micro-structured interfaces 5 Non-aqueous electrolytes 5.1 Organic electrolytes 5.1.1 Membrane inlet 5.1.2 Capillary inlet 5.2 Low-vapor-pressure electrolytes 6 Ambient pressure ionization techniques 7 Beyond DEMS – Coupling with other techniques 8 Summary and outlook Acknowledgments References

78 80 80 82 86 87 88 88 88 90 91 92 94 98 98 99

Abstract In this chapter we present an overview over different designs and approaches to couple electrochemical measurements and mass spectrometry for the online monitoring of reaction product and intermediate formation. We cover the method development from the early attempts to the most recent and advanced applications, where hyphenation with additional analytical and spectroscopic techniques allows for detailed kinetic and mechanistic studies of complex processes occurring at the electrochemical interfaces. Applications mainly include studies of electrocatalytic reactions and electrochemical conversion reactions relevant for energy conversion and energy storage devices. We will discuss the related challenges, both in aqueous and organic or solid electrolytes, respectively, as well as the trends for further developments.

Key points

• • • • •

Overview of the development of approaches Overview of designs to couple electrochemical measurements with mass spectrometry for the online analysis of gaseous products Detection of volatile and liquid phase products in aqueous and non-aqueous electrolytes Obstacles and challenges of electrochemical mass spectrometry for specific applications Future trends and perspectives

Abbreviations ASSB ATR-FTIRS BMP-TFSI CE DART-MSD DART-TOF-MS DEMS DESI DMC DMSO EC

all-solid-state batteries attenuated total reflection Fourier transform infrared spectroscopy 1-butyl-1-methylpyrrolidinium-bis(trifluoromethylsulfonyl)imide counter electrode direct analysis in real time mass spectrometry direct analysis in real time – time of flight mass spectrometry differential electrochemical mass spectrometry liquid-sample desorption electrospray ionization dimethyl carbonate dimethyl sulfoxide ethylene carbonate

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00213-9

77

78

Methods and Instruments | Differential Electrochemical Mass Spectrometry

EC-RTMS EC-ESI-MS EMS EQCM ESI-MS FDTS FEP FIA GAME ICP-MS LC LTO MALDI-MS MEA MIMS NMC NMP OEMS OLEMS PBI PC PCTE PEEK PEO PERMS PVDF RE RDE SALVI SCM SDEMS SECM SEMS SFC SIFT-MS SPE TEGDME TOF-SIMS UHV UME WE YSZ

1

electrochemical real-time mass spectrometry electrochemical electrospray ionization mass spectrometry electrochemical mass spectrometry electrochemical quartz crystal microbalance electrospray ionization mass spectrometry perfluorodecyltrichlorosilane perfluoroethylene-propylene flow injection analysis gas accessible membrane electrode inductively coupled plasma mass spectrometry liquid chromatography lithium titanium oxide matrix-assisted laser desorption/ionization mass spectrometry membrane electrode assembly membrane inlet mass spectrometry nickel-manganese-cobalt oxide N-methyl-2-pyrrolidone online electrochemical mass spectrometry online electrochemical mass spectrometry polybenzimidazole polypropylene carbonate polycarbonate track etched polyetheretherketone poly(ethylene oxide) permeable membrane mass spectrometry polyvinylidenefluoride reference electrode rotating disk electrode system for analysis at the liquid − vacuum interface scanning capillary microscope scanning DEMS scanning electrochemical microscopy solid electrochemical mass spectrometry scanning flow cell selected-ion flow-tube mass spectrometry screen-printed electrode tetraethylenglycoldimethylether (tetraglyme) time-of-flight secondary ion mass spectrometry ultrahigh vacuum ultramicroelectrode working electrode yttrium-stabilized zirconia

Introduction

The term ‘Differential Electrochemical Mass Spectrometry (DEMS)’ describes a method that allows the direct mass spectrometric analysis of the gaseous and volatile reaction products formed at the electrochemical interface and the determination of their formation rates. Nowadays, it is a powerful and well-established online technique, for excellent reviews the reader is referred to the references1–12. In general, this method is based on the concept of connecting an electrochemical cell with an ultrahigh vacuum (UHV) chamber housing the mass spectrometer by a semi-permeable interface, which allows gaseous (volatile) products/educts to pass, but hinders permeation of the liquid electrolyte. This way, the formation or consumption of gaseous (volatile) products/ educts, which occurs under ambient conditions, can be monitored continuously and online. In such setups the electrode is often deposited on, or attached to the membrane side facing the electrolyte, or it is in close proximity to the membrane. In a completely different approach, the electrolyte is ionized by thermospray or electrospray and directly injected into the mass spectrometer, which allow also the detection of dissolved species that are non- or little volatile. These different approaches will be discussed in more detail later.

Methods and Instruments | Differential Electrochemical Mass Spectrometry

79

From a historical perspective the combination of electrochemical and mass spectrometric measurements was first introduced in electrochemistry in the early 70s of the last century as ‘Electrochemical Mass Spectrometry (EMS)’, where a hydrophobic teflonized glass frit was used to separate the liquid phase from the UVH system, while still allowing gases to permeate through.13 In this initial setup a layer of Pt powder, which was mechanically embedded in a Teflon film and which was exposed to the electrolyte, served as a working electrode (Fig. 1a). For O2 evolution at constant Faradaic current the authors found a linear increase of the mass spectrometric signal with time, where higher slopes were obtained for higher Faradaic currents. This is characteristic for an integral measurement, where the mass spectrometric signal is proportional to the accumulated ‘integral’ amount of products formed (see Section 2). The authors explained this behavior by transport limitations imposed by the relatively thick interface to the vacuum system. The first attempt of using a porous Teflon membrane as interface to the UHV system, with a Pt sponge electrode directly glued on the electrolyte side of the membrane (Fig. 1b), was reported in 1977 by Grambow and Bruckenstein.14 Applying this configuration for studying the stripping of adsorbed CO, they also found an integral response of the detected CO2 ion current. A similar approach, with a Au grid electrode attached to a silicone membrane (Fig. 1c), was employed for the online mass spectrometric detection of the gaseous products formed during galvanostatic or potentiostatic oxidation of acetate and propionate in aqueous and dimethyl sulfoxide solutions.15 This was referred to as Permeable Membrane Mass Spectrometry (PERMS) by the authors. Aiming at a better time resolution of the measurement, Wolter and Heitbaum prepared very thin electrodes by painting a suspension of Pt powder onto a thin, porous Teflon membrane, which was dried subsequently (Fig. 1d).16 This resulted in highly porous films with thicknesses in the range of 50–100 mm.5 Based on potential step experiments, this configuration allowed for time resolutions in the range of 10 milliseconds in the mass spectrometric signal, where this was determined by the delay of the mass spectrometric signal as compared to the decay of the Faradaic current. In this case, the mass spectrometric signal is proportional to the product formation rate (first derivative of the accumulated amount of product formation) and proportional to the Faradaic current. Therefore, the measurement was referred to as ‘differential electrochemical mass spectrometry (DEMS)’ instead of the former term ‘EMS’16 (for discussion of the term differential see Section 2). A further decrease in the working electrode thickness was achieved by sputtering a metal or oxide layer directly onto the membrane (thickness in the range of 50–80 nm5).17–20 Alternatively, thin sputtered metal layers (Au, Pt or others) were used as conducting substrate for the electrochemical deposition of porous layers of metals and alloys.21–24 In still another approach, in model studies of low-temperature fuel cell catalysts, thin films of realistic carbon-supported nanoparticles were pressed on a

SILVER EPOXY INSULATED WITH EPOXY

(c)

(d)

TO MS SOURCE

ELEKTROLYT

F

E

C

B

4mm Pyrex

EPOXY COVERED Pt WIRE

10 mm

D

TO MASS SPECTROMETER

FINE POROSITY PYREX GLASS FRIT

(b)

PYREX GLASS TUBE

Pt SPONGE + TEFLON - 120

(a)

Gold wire

2 NS

100 mm

A

Silicone membrane

1 NS

Silver Epoxy

Steel screen Gold grid

HV

Fig. 1 Schematic illustrations (cross sections) of early designs to couple electrochemistry and mass spectrometry via a semi-permeable interface, using Pyrex glass cells. Shown is only the interface, the electrochemical cell part is located above (a), (b, d) or below (c) the interface. (a) Pt powder, embedded into a teflonized glass frit, was used as a porous, liquid water repellent working electrode, the open side of the Pyrex glass tube connected to the mass spectrometer chamber.13 (b) Design using a Pt sponge working electrode (top part), which was glued onto a porous Teflon membrane underneath. The membrane was supported by a glass frit and shrunk against the glass tube connection to the mass spectrometer chamber.14 (c) Concept using a Au mini-grid working electrode (bottom part), which was glued with silicone rubber cement to a silicone membrane. This is supported by a stainless steel grid positioned at the one end of a Pyrex glass tube, while the other end connected to the mass spectrometer chamber.15 (d) Design with an ultrathin porous Pt electrode (E), obtained by painting a Pt powder suspension in a conducting lacquer on a porous Teflon membrane (C), which was supported by a glass frit at the end of a Pyrex glass tube (A) connecting to the mass spectrometer chamber (further details see b).16 Panel a: Reproduced with permission of Bruckenstein, S.; Gadde, R.R. Use of a Porous Electrode for In Situ Mass Spectrometric Determination of Volatile Electrode Reaction Product. J. Am. Chem. Soc. 1971, 93, 793–794, American Chemical Society; Panel b: Reproduced with permission of Grambow, L.; Bruckenstein, S. Mass Spectrometric Investigation of the Electrochemical Behavior of Adsorbed Carbon Monoxide at Platinum in 0.2 M Sulphuric Acid. Electrochim. Acta 1977, 22, 377–383, Elsevier; Panel c: Reproduced with permission of Brockman, T.J.; Anderson, L.B. Permeable Membrane Mass Spectrometry of Products of Electrochemical Oxidation of Carboxylate Ions. Anal. Chem. 1984, 56, 207–213, American Chemical Society; Panel d: Reproduced with permission of Wolter, O.; Heitbaum, J. Differential Electrochemical Mass Spectroscopy (DEMS) – A New Method for the Study of Electrode Processes. Ber. Bunsenges. Phys. Chem. 1984, 88, 2–6, Wiley-VCH GmbH and Deutsche Bunsen-Gesellschaft.

80

Methods and Instruments | Differential Electrochemical Mass Spectrometry

metallic current collector grid, e.g., a PdAu grid, and this composite was in turn pressed on a porous Teflon membrane,25,26 or the catalyst ink was pipetted/dried over the Au-sputtered membrane.25,26 Finally, as an alternative approach, different groups introduced designs which did not use a membrane to separate the liquid electrolyte phase from the UHV environment. Kreysa and Breidenbach used an inert carrier gas coupled with a differentially pumped capillary inlet instead of a membrane interface for the continuous transfer of the reaction products from the electrolyte to the mass spectrometric analysis.27 Several groups directly interfaced the exhaust of a realistic low-temperature membrane fuel cell anode compartment to the mass spectrometer, also using a differentially pumped capillary inlet for pressure reduction. This way they online analyzed the reaction products formed at the membrane electrode assembly (MEA) anodes under technically relevant operation conditions (elevated temperature and pressure). In this design, the inert carrier gas is replaced by the gas flow of non-consumed feed gas. This approach was used for a variety of studies of different organic fuels.28–35 In the following we will, after a brief excursion dealing with technical aspects such as differential measurements, time resolution/ time delay, or sensitivity of the measurements, first discuss different designs of the DEMS cell, in particular different inlet designs (Section 2), and challenges arising from solvent permeation, product fragmentation and follow up reactions (Section 3). Here we will also present recent micro-structured designs (Section 4). Though this discussion focuses on aqueous electrolytes, most of the points are also relevant for non-aqueous electrolytes. Section 5 then deals with aspects and designs/applications that are specific for non-aqueous electrolytes. Designs involving soft ionization techniques, in contrast to the commonly used electron impact ionization, and their application are presented and discussed in Section 6. In Section 7 we focus on hyphenated approaches and techniques, coupling DEMS with other spectroscopic techniques. Finally, in Section 8 we provide a brief summary and an outlook on future developments and applications of DEMS from our perspective.

2

DEMS cell designs

Before starting with the descriptions of the different cell designs we will briefly comment on the term ‘differential’ and the difference between differential and integral measurements. In integral measurements the rate for product removal from the cell electrolyte is small compared to that of product formation, which results in a continuously increasing product concentration in the electrolyte in the cell. It is important to realize that product removal can occur in different ways, e.g., with the effluent electrolyte in a flow cell or capillary, or by evaporation through a membrane. The measured concentration can also be affected by the pumping rate in the vacuum chamber, since for low pumping rates the products would tend to accumulate in the vacuum chamber. Also, transport through the electrolyte may play a role. In integral measurements, the formation rate is proportional to the slope (first derivative) of the increasing concentration/partial pressure (increase with time) of the product species measured by the mass spectrometric signal. In differential measurements, in contrast, the rates for product formation and product removal are of comparable magnitude, and in this case the formation rate is proportional to the relative product concentration in the electrolyte, relative to the background concentration. The term ‘differential’ was also used because of the commonly employed two-stage differential pumping of the vacuum chamber, where the majority of the water vapor and also of the gaseous/volatile products is pumped off in the first stage to achieve pressures in the analysis chamber of the mass spectrometer that allow the continuous operation of an electron impact ionization type mass spectrometer.16 It is important to realize that these two cases represent extreme situations, and that there are many cases which are in between these two situations. This has to be tested when aiming at a quantitative determination of rates. Another important aspect to keep in mind is the difference between the time delay between Faradaic current signal and mass spectrometric signal, which can be determined, e.g., in a potential step experiment, and the time resolution of the mass spectrometric measurement, which is given by the decay rate of the mass spectrometric signal for an abrupt stop of the reaction (vertical decay of the Faradaic current). Practically, these two features are mostly closely correlated, but they are not identical. Both the differential character of a measurement and the time delay/time resolution will depend sensitively not only on the design of the cell and the detection scheme, but also on parameters such as porosity of the electrode and membrane, the electrolyte flow rate (in flow cells) and the pumping speed of the vacuum system. Since, as will be discussed in more detail later, these parameters also affect the sensitivity of the measurements and the size of the background signals, optimized measurements require a careful consideration and optimization of these different effects and parameters.

2.1

Membrane inlet

As discussed above, the separation between the vacuum system and the electrolyte is the key component of the DEMS setup. Different approaches for this will be discussed in the following, focusing here on aqueous electrolytes. Specific aspects relevant mainly for non-aqueous electrolytes will be addressed in Section 5. Interestingly, the idea of using a thin membrane as separation or interface between the UHV system and a liquid phase (Membrane Inlet Mass Spectrometry, MIMS, Fig. 2a) for the online sensing of gaseous species was first introduced already in the early 60s in studies on the photosynthesis by algae.36 Apparently, this was overlooked by the electrochemistry community, reporting the application of membrane inlet first in 1977,14 though MIMS itself turned out to be a powerful tool in research of photosynthetic microorganisms37 and other applications involving the quantitative online detection of gaseous/volatile species in liquids.38,39

81

Methods and Instruments | Differential Electrochemical Mass Spectrometry

(a)

(b)

1 2

(c)

a

3 4

c

b d e

f g 5 h vacuum

7 6

(e) e yt ol t An lnle

1 2 3

(f)

An o Ou lyte tle t

(d)

Counter Electrode

4

lon-Conducting Membrane Mounting Screw

Washer Working Electrode

5 6 7

8

te holy Cat tlet Ou

Reference Electrode

Transfer Capillary (1 of 4)

te holy Cat let ln

Pervaporation Membrane

Stainless Steel Frit To MS

1 2 3 4 5 6

Fig. 2 Schematic illustrations of different designs for a bare membrane inlet. (a) Cross-section of the membrane inlet in an early design for membrane inlet mass spectrometry (MIMS), with the liquid cell (not shown) above the membrane (top), which separates the liquid from the connection to the mass spectrometer (bottom) and which is supported by a ceramic frit underneath.36 (b–f ) Different designs for DEMS cells, where the working electrode is separated from the bare membrane inlet. (b) Cross-section of a radially symmetric thin-layer cell for stagnant electrolyte, where the Pt foil working electrode (b) is pressed by an electrode support (a) via a Teflon spacer (c) against the passivated Ti cell body (d–f ) electrolyte outlet/inlet capillaries. The cell body (d) connects via a bare membrane (g) and a metal frit to the UHV chamber (h).46 (c) Cross section of a radially symmetric thin-layer flow cell for a massive working electrode (7), with two flushable counter electrodes and a reference electrode. The connection to the vacuum (top) occurs via a glass tube (2), which ends in a frit supporting the bare interface membrane (3).49 (d) Cross-section of a radially symmetric dual thin-layer flow cell, where one compartment is used for the electrochemical reaction (upper cell) and one for gas sensing (lower cell).50 The two thin-layer compartments are connected by four capillaries located at the inner perimeter of the gaskets, electrolyte inlet and outlet capillaries end in a central opening, leading to a radial electrolyte flow over the working electrode and the membrane pattern.50 (e) Modified dual thin-layer flow cell with the working and counter electrodes in close proximity to reduce Ohmic losses. The working electrode and counter electrode compartments are separated by a ionomer membrane and separately flushed by the electrolyte to avoid electrolyte mixing (further details see figure).78 (f ) Sketch of the cell body of a DEMS flow cell for bead-type single crystals. The electrode (electrode and electrode holder not shown) is located right above a cone shaped glass capillary (1), which in turn is mounted in the center of the Kel-F support (2). The electrolyte forms a hanging droplet between electrode and glass capillary, wetting the former in a hanging meniscus configuration. From there it is sucked via the glass capillary and an inlet opening (3) to a thin-layer compartment with the membrane inlet, which connects to the mass spectrometer chamber (membrane: 4, supporting frit: 5), and finally to the electrolyte outlet (6) (for further details see text).80 Panel a: Reproduced with permission of Hoch, G.; Kok, B. A Mass Spectrometer Inlet System for Sampling Gases Dissolved in Liquid Phases. Arch. Biochem. Biophys. 1963, 101, 160–170, Informa Healthcare; Panel b: Reproduced with permission of Hartung, T.; Baltruschat, H. Differential Electrochemical Mass Spectrometry Using Smooth Electrodes: Adsorption and H/D Exchange Reactions of Benzene on Pt. Langmuir 1990, 6, 953–957, American Chemical Society; Panel c: Reproduced with permission of Vaskelis, A.; Jusys, Z. Differential Electrochemical Mass Spectrometry Study of the Electroless Copper Plating Process Using a Thin-Layer Flow-Through Cell. Anal. Chim. Acta 1995, 305, 227–231, Elsevier Science B.V.; Panel d: Reproduced with permission of Jusys, Z.; Massong, H., Baltruschat, H. A New Approach for Simultaneous DEMS and EQCM: Electrooxidation of Adsorbed CO on Pt and Pt-Ru. J. Electrochem. Soc. 1999, 146, 1093–1098, The Electrochemical Society, Inc.; Panel e: Clark, E.L.; Singh, M.R.; Kwon, Y.; Bell, A.T. Differential Electrochemical Mass Spectrometer Cell Design for Online Quantification of Products Produced During Electrochemical Reduction of CO2. Anal. Chem. 2015, 87, 8013–8920, American Chemical Society; Panel f: Reproduced with permission of Abd-El-Latif, A.E.A.; Xu, J.; Bogolowski, N.; Koenigshoven, P.; Baltruschat, H. New Cell for DEMS Applicable to Different Electrode Sizes. Electrocatalysis 2012, 3, 39–47, Springer.

82

Methods and Instruments | Differential Electrochemical Mass Spectrometry

The evolution of DEMS cell designs was discussed in detail recently in an excellent review.40 In general, these designs can be classified into two major groups, (i) cells with a stagnant electrolyte and (ii) cells with enforced electrolyte convection, which will be discussed in detail below. It should be noted that an enforced mass transport of the electrolyte is crucial not only for the delivery of reactant to the electrode surface, which is particularly important at low concentrations of the educts, as encountered, e.g., in the reduction or oxidation of dissolved gasses, to overcome the diffusion limitations (for details and references see below in this section and Section 2.2). It is also important for the off-transport of reactive intermediates, to avoid or reduce re-adsorption and further reaction of reactive intermediates. The enhanced off-transport of reaction intermediates from the electrode surface region will decrease their ability to re-adsorb and react further towards the final product, which can lead to distinct changes in the selectivity of the respective reaction.41–45 As an example, in the electro-oxidation of methanol the selectivity for CO2 formation was found to decrease significantly with increasing electrolyte flow rate.43 Starting with membrane inlet type cell designs for stagnant electrolytes, the most common approach involved the fabrication of thin, porous gas diffusion electrodes, which, as described briefly in the introduction, were created in different ways directly on the membrane. This reduces the time delay between product formation and detection and thus improves the time resolution. For the structurally better defined polycrystalline or single crystalline massive electrodes, this approach is, however, not possible. For DEMS studies with compact/massive model electrodes a conceptual breakthrough was achieved by locating a bare membrane at a fixed distance from the electrode surface in a thin-layer cell with circular geometry (Fig. 2b).46–48 In that setup, the electrode was separated from the membrane inlet by a narrow gap, forming a so-called symmetrical membrane inlet with an extremely small volume, in the microliter range. This allowed, e.g., for the detection of reaction products formed during the electrochemical desorption (reductive or oxidative, depending of the potential scan direction) of the organic molecules, which were pre-adsorbed at a fixed constant potential, with time constants in the order of 2–3 s and close to 100% collection efficiency.5 While working very well in a stagnant electrolyte, this cell design was not well suited for measurements under continuous electrolyte flow conditions, as the products formed were efficiently removed from the gap before reaching the membrane inlet. In a later study, the absolute signal intensities were improved by increasing the electrode size relative to the membrane inlet, but the loss of products due to off transport in the approximately linear electrolyte flow was still too large (Fig. 2c).49 This problem was resolved in a novel dual thin-layer flow cell design (Fig. 2d),50 which allowed for using not only any type of massive electrodes,51–56 but also thin layers of supported catalysts41,57–69 and nanostructured electrodes,42,44,52,70–72 or even carbon felt flow-through electrodes for DEMS model studies of redox flow batteries.73 In this design the membrane inlet interface is located downstream in a second thin-layer compartment, which is connected with the first compartment by a set of radially arranged thin capillaries, located close to the inner edge of the gasket. In contrast, the inlet and outlet capillaries are located in the center of the wall separating the two compartments (Fig. 2d), leading to a radial flow pattern in each of the two compartments. This configuration results in an efficient product mixing and a homogeneous distribution in the electrolyte in the membrane compartment, and thus in an improved collection efficiency.56,74–76 In a more general sense this cell was also used as a platform for combining electrochemical measurements with other techniques, e.g., by replacing the membrane inlet in the second cell by a collector electrode to detect oxidizable or reducible species formed during reaction at the working electrode.77 A combination between DEMS and an additional electrochemical detector electrode will be discussed further in Section 5.1.1 for organic electrolytes and in Section 6. A slightly modified (continuously flushable counter electrode, electrolyte inlet at the working electrode), but conceptually rather similar version dual thin-layer flow cell design for DEMS application was reported recently by Clark et al. (Fig. 2e).78 Single and dual thin-layer flow cells for DEMS applications are now also commercially available.79 The use of other hydrodynamic techniques allowing DEMS measurements under controlled mass transport conditions will be discussed in Section 2.2. Particularly suited for single crystal electrodes, including small bead crystals, is a flow cell configuration from the Baltruschat group, where the electrode surface is not pressed against a sealing gasket, but operated in a hanging meniscus configuration (Fig. 2f ).80 The electrode is mounted directly opposite to the cone-like end of a thin hydrophilic glass capillary, where it forms a droplet wetting the two. Electrolyte from this droplet is continuously sucked away via the glass capillary, driven by a peristaltic pump, and replenished by a syringe pump. In the cell body the electrolyte is transported to a thin-layer cell, where, due to an elaborate system of connecting capillaries, it passes in a well-defined radial flow pattern over a porous membrane inlet connecting to the mass spectrometer chamber. The pumping speed of the two pumps is adjusted such that a stable hanging meniscus configuration is maintained. The performance of this setup, which shows also a rather short delay time in the mass spectrometric detection, was demonstrated, e.g., in various test measurements on the electrooxidation of organic molecules at different Pt single crystal surfaces.80 Finally, we want to mention a flow cell design where the electrolyte is passed over a commercial screen-printed electrode (SPE).81 Volatile compounds in the effluent electrolyte are detected via a membrane inlet located in the effluent electrolyte stream, which connects via a capillary to the mass spectrometer. The performance of this flow injection analysis (FIA) setup was demonstrated for H2 evolution over Pt. Particular advantage of this FIA-DEMS setup is that the SPE electrodes are easily exchangeable, which is highly attractive for the routine gas product analysis over a number of different electrode materials and in various reactions/electrolytes.

2.2

Pinhole inlet systems and hydrodynamic DEMS techniques

To avoid possible mechanical/structural damage experienced in a thin-layer membrane inlet when pressing the sensitive single-crystal electrodes against the gasket, a modified version of the membrane inlet was developed, in which an ultrathin capillary with a micrometer-sized pinhole covered by a Teflon membrane was used as membrane inlet.82 An example from later work is

Methods and Instruments | Differential Electrochemical Mass Spectrometry (a)

83

(b)

(c)

Fig. 3 (a) Pinhole/capillary inlet DEMS cell design for the analysis of the gaseous products evolved at a single crystal electrode in a hanging meniscus configuration (for details see figure).83 (b) As (a), but for a mesoporous electrode (4) in an immersed configuration (5: electrode support).87 Here the spectrometer inlet tip (2) with the membrane (3) at its end, which is connected to the mass spectrometer chamber by a tube (2), approaches the electrolyte from the top. (c) As (b), but both electrode and membrane covered pinhole are immersed in the electrolyte (further details see figure).88 Panel a: Reproduced with permission of Wonders, A.H.; Housmans, T.H.M.; Rosca, V.; Koper, M.T.M. On-line Mass Spectrometry Systems for Measurements at Single-Crystal Electrodes in Hanging Meniscus Configuration. J. Appl. Electrochem. 2006, 36, 1215–1221, Springer; Panel b: Reproduced with permission of Florez-Montano, J.; Garcia, G.; Guillen-Villafuerte, O.; Rodriguez, J.L.; Planes, G.A.; Pastor, E. Mechanism of Ethanol Electrooxidation on Mesoporous Pt Electrode in Acidic Medium Studied by a Novel Electrochemical Mass Spectrometry Set-Up. Electrochim. Acta 2016, 209, 121–131, Elsevier B.V.; Panel c: Reproduced with permission of Todoroki, N.; Tei, H.; Tsurumaki, H.; Miyakawa, T.; Inoue, T.; Wadayama, T. Surface Atomic Arrangement Dependence of Electrochemical CO2 Reduction on Gold: Online Electrochemical Mass Spectrometric Study on Low-Index Au(hkl) Surfaces. ACS Catal. 2019, 9, 1383–1388, American Chemical Society.

shown in (Fig. 3a).83–85 In these studies, the tip of the capillary was located close to the electrode surface in a hanging meniscus configuration (stagnant electrolyte). Considering that the transport to the mass spectrometer is mainly limited by the size of the (membrane covered) pinhole, this inlet design was also described as pinhole inlet.5,86 In a modified version the electrode was facing upwards in a stagnant electrolyte, and the online mass spectrometry inlet tip approached from the top (Figs. 3b and c).87–89 Essentially, the configurations in Fig. 3 can be referred to as a membrane inlet system, only at much smaller lateral dimension and also capillary – electrode distances. It is important to note that in this configuration gas evolution is probed from a small area of the electrode, which is also partly shielded by the pinhole inlet, while the Faradaic current originates mainly from the freely accessible electrode surface, which may affect conclusions on the reaction activity and selectivity. To overcome the mass transport limitations, DEMS was also combined with a rotating disk electrode (RDE). Tegtmeier et al. reported a porous electrode membrane inlet system in an RDE configuration in the late eighties.90 This was based on a rotating inlet system, where the electrode-covered membrane was located at the end of a tube rotator, which via a rotating feedthrough was connected to the mass spectrometer (Fig. 4a). The performance of this setup was demonstrated in a number of electrocatalytic reactions such as H2 evolution, CO oxidation or oxalic acid oxidation over a porous Pt electrode. Alternatively, Wasmus et al. located the porous membrane inlet located sideways, in the axial plane of the (massive) rotating disk electrode (Fig. 4b).91 They used this configuration for sensing the gaseous reaction products formed during CO2 reduction.91 Subsequently, Fujihira and Noguchi developed a setup where the electrode was stationary, and the electrolyte transport was induced by an inert rotating cylinder, located about 5 mm above the working electrode (rotational flow).92 The working electrode was a thin porous Au film sputtered directly on a membrane, which connected to the mass spectrometer (Fig. 4c). Also for this setup the performance was demonstrated for CO2 electro-reduction.92 More recently, Todoroki et al. combined a standard RDE setup with a pinhole inlet, where a sensing membrane-coated capillary tip approached the rotating electrode from the bottom (Fig. 4d), and employed this setup for studies of the CO2 reduction reaction.93 It is important to realize, however, that in this configuration the sensing capillary itself can disturb the hydrodynamic flux of the electrolyte, which is well defined in a standard RDE configuration.

84

(a)

(d)

Methods and Instruments | Differential Electrochemical Mass Spectrometry

(b)

(c)

(e)

Fig. 4 Schematic illustrations of different designs of hydrodynamic DEMS systems. (a) Cross section of a design featuring a rotating porous electrode membrane inlet, where the electrode is directly deposited on the porous membrane (further details see text and figure).90 (b) Cross section of a design based on a rotating disk electrode, where the gases are detected by a bare membrane inlet system positioned sideways of the rotating disk electrode.91 (c) Modified version of (c), where the rotating disk inducing the electrolyte flow (d) is chemically inert, while the stationary Au working electrode is sputtered on the porous membrane (g) which is located closely underneath the rotating cylinder (further details see text and figure).92 (d) Combination of a rotating disk electrode with a pinhole inlet which is located closely underneath the RDE (further details see text).93 (e) Cross sectional view of a design combining a jet impinging electrode in a wall-jet configuration with a surrounding bare porous membrane ring serving as interface, which results in a convective flow over the electrode to the membrane ring.94 Panel a: Reproduced with permission of Tegtmeyer, D.; Heindrichs, A.; Heitbaum, J. Electrochemical on Line Mass Spectrometry on a Rotating Electrode Inlet System. Ber. Bunsenges. Phys. Chem. 1989, 93, 291–296, VCH Verlagsgesellschaft mbH; Panel b: Reproduced with permission of Wasmus, S.; Cattaneo, E.; Vielstich, W. Reduction of Carbon Dioxide to Methane and Ethene - An On-Line MS Study with Rotating Electrodes. Electrochim. Acta 1990, 35, 771–775, Pergamon Press; Panel c: Reproduced with permission of Fujihira, M.; Noguchi, T. A Novel Differential Electrochemical Mass Spectrometer (DEMS) with a Stationary Gas-Permeable Electrode in a Rotational Flow Produced by a Rotating Rod. J. Electroanal. Chem. 1993, 347, 457–463, Elsevier Sequoia S.A.; Panel d: Reproduced with permission of Todoroki, N.; Tsurumaki, H.; Tei, H.; Mochizuki, T.; Wadayama, T. Online Electrochemical Mass Spectrometry Combined with the Rotating Disk Electrode Method for Direct Observations of Potential-Dependent Molecular Behaviors in the Electrode Surface Vicinity. J. Electrochem. Soc. 2020, 167, 106503(1-5), The Electrochemical Society, Inc.; Panel e: Reproduced with permission of Treufeld, I.; Jebaraj, A.J.J.; Xu, J.; de Godoj, D.M.; Scherson, D. Porous Teflon Ring-Solid Disk Electrode Arrangement for Differential Mass Spectrometry Measurements in the Presence of Convective Flow Generated by a Jet Impinging Electrode in the Wall-Jet Configuration. Anal. Chem. 2012, 84, 5175–5179, American Chemical Society.

As an alternative hydrodynamic approach, Treufeld et al. introduced the combination of a wall-jet electrode configuration with DEMS,94,95 where a porous membrane supported on a highly porous Teflon ring inlet surrounded a solid disk metal electrode located in the center of the ring inlet, which then connects to the mass spectrometer (Fig. 4e). The impinging electrolyte jet arrives from the bottom at the solid Au. This design has the considerable advantage that it allows DEMS studies at massive electrodes at high electrolyte flow rates. In a rather similar approach, Venkatachalam et al. operated a Pt sputter-coated membrane inlet in a wall-jet setup.96 They tested this setup for H2 and O2 evolution in acid electrolyte and in oxalic acid oxidation. While this setup does not allow studies on massive or even single crystal electrodes, it has, according to the authors, the advantage of high time resolution also at high flow rates, due to the rapid permeation of volatile species through the membrane.96 Finally, DEMS measurements were performed in a cyclone flow cell, with a porous electrode layer deposited directly on the membrane.97 Compared to the RDE-DEMS combination this has the advantage of avoiding moving parts, while maintaining the high electrolyte flow rate. Together with the fast transport through the catalyst layer and membrane, this configuration allows DEMS

Methods and Instruments | Differential Electrochemical Mass Spectrometry

85

measurements with a very high time resolution, which was demonstrated in a high time-resolution kinetic study of the CO bulk oxidation reaction on a PtRu/C catalyst layer.97 In a very different approach, aiming at DEMS measurements with spatial resolution, a capillary was scanned over the sample, similar to the scanning electrochemical microscopy (SECM) approach. In a first example, a membrane covered capillary tip was scanned over an array of micro-electrodes of different compositions in a stagnant electrolyte, which allows for combinatorial scanning probe DEMS measurements over these electrode libraries (Fig. 5a and b).98–100 In another approach, which allowed also for enforced and controlled electrolyte transport, DEMS was integrated into a scanning flow cell (SFC) containing the membrane inlet capillary as well as the electrolyte inlet and outlet and the other electrodes. The working electrode was mounted on an x-y-z translational stage. During operation, the electrode was scanned laterally with respect to the cell body (Fig. 5c).101 The performance of this design was demonstrated for H2 evolution and CO2 reduction on Cu electrodes. In a different version of such kind of scanning flow cell, the scanning electrochemical flow cell could be moved vertically and laterally by a stepper motor, while the electrode remained fixed (Fig. 5d).102 In this case, local resolution was obtained by positioning the cell above the desired location of the electrode. The effluent electrolyte then passes a separate membrane inlet compartment, which as usual connects to the mass spectrometer. Also this setup is very useful for high-throughput studies, as demonstrated by the authors in a high-throughput screening study of CO2 reduction catalysts in catalyst libraries. (a)

(b)

lnlet to MS

mass spec. controller

(8)

quadrupole mass (16) spectrometer and vacuum system

(7)

Scan direction

2H+

H2

RF generator

CH3OH

(5)

CO2

(6)

Pt on Au glass 1 mm

(9)

(10) - (13) (bi)potentiostat

(15)

motor controller probe cell

XYZ positioner

20 mm (14)

2.5 mm

(c)

(d)

CE

To MS

RE

force sensor CE

t tle ou ar go n

membrane

int let

PTFE

n go ar

silicone ring

WE

Fig. 5 Schematic illustrations of different designs for DEMS measurements with spatial resolution: (a) Design in which a membrane covered capillary connected to the mass spectrometer chamber is scanned over a structured electrode.98 (b) As (a), but here the setup is integrated in a scanning electrochemical microscope (SECM), where an exchangeable tip allows operation either as SECM or as scanning DEMS (SDEMS) instrument.100 (c) Miniaturized scanning electrochemical flow cell design, where all parts including the micro-structured membrane covered capillary inlet are integrated in the cell body (see figure and text). This remains at a fixed position, while the electrode is scanned (details see text).101 (d) Scanning electrochemical flow cell design (left) for screening libraries of electrocatalysts.102 The principle resembles that in (c), but in this case the entire electrolyte passing over the working electrode (WE) flows from the electrolyte outlet to a separate membrane inlet unit (right). For further details see text. Panel a: Reproduced with permission of Jambunathan, K.; Hillier, A.C. Measuring Electrocatalytic Activity on a Local Scale With Scanning Differential Electrochemical Mass Spectrometry. J. Electrochem. Soc. 2003, 150, E312–E320, The Electrochemical Society; Panel b: Reproduced with permission of Rus, E.D.; Wang, H.; Legard, A.E.; Rizert, N.L.; Van Dover, R.B.; Abruña, H.D. An Exchangeable-Tip Scanning Probe Instrument for the Analysis of Combinatorial Libraries of Electrocatalysts. Rev. Sci. Instr. 2013, 84, 024101(1-8), American Institute of Physics; Panel c: Reproduced with permission of Grote, J.P.; Zeradjanin, A.R.; Cherevko, S.; Mayrhofer, K.J.J. Coupling of a Scanning Flow Cell With Online Electrochemical Mass Spectrometry for Screening of Reaction Selectivity. Rev. Sci. Instr. 2014, 85, 104101(1-5), AIP Publishing LLC; Panel d: Reproduced with permission of Lai, Y.; Jones, R.J.R. Wang, Y.; Zhou, L. Gregoire, J.M. Scanning Electrochemical Flow Cell with Online Mass Spectroscopy for Accelerated Screening of Carbon Dioxide Reduction Electrocatalysts. ACS Comb. Sci. 2019, 21, 692−704, American Chemical Society.

86

3

Methods and Instruments | Differential Electrochemical Mass Spectrometry

Challenges arising from solvent permeation, product fragmentation and follow up reactions in the electrolyte

Specific challenges in DEMS studies arise (i) from the permeation of the solvent vapor through the membrane, as this is mostly present in large excess. This can thus completely dominate the gas composition in the mass spectrometer chamber. Further problems can arise (ii) from the overlap in the fragmentation pattern of different products and/or of the reactant, and (iii) from possible reactions between reaction products and reactant or solvent. In this section we will discuss several approaches to reduce the impact of these effects, focusing on examples reported for aqueous electrolytes, but keeping in mind that many of these approaches may be applicable also for organic electrolytes. This will be discussed further in Section 5.1. The commonly used porous membrane inlet interface, which directly connects to the UHV system, results in a rather high background pressure of the solvent vapor and thus requires efficient differential pumping to keep the solvent partial pressure in the UHV system at a level tolerable for long-term operation of the mass spectrometer. This in turn also reduces the partial pressure of the products and thus their signal intensities. To selectively decrease partial pressure of water vapor, different strategies have been employed, such as positioning a cold trap18,103,104 or a desiccator102 behind the membrane inlet. One should keep in mind, however, that depending on the operating temperature of the trap some products could also be trapped. In general, for a selective removal of the solvent the evaporation temperature of the products of interest should be significantly higher than that of the solvent. Solvent vapor permeation can also be reduced by using a non-porous membrane. Interestingly, non-porous but nevertheless semi-permeable membranes were commonly used in the MIMS approach to reduce the water vapor permeation and the required differential pumping.36 The use of non-porous membranes such as a silicone membrane105 or a Teflon membrane18 was also reported for DEMS applications. In a more recent approach, a non-porous hydrophobic Teflon membrane interface was used for DEMS flow cell studies on the oxidation of C1 and C2 molecules in aqueous electrolytes at elevated temperature and pressure, i.e., under conditions relevant for PEMFC operation.106–110 Its operation principle is schematically illustrated in Fig. 6a.111 Here it is important to realize that not only the porosity, mainly the size/density of the pores, is important, but also the chemical properties of the membrane, in particular its hydrophobicity/hydrophilicity. Lowering the porosity of the membrane would just mean that in most cases also the partial pressures and thus the signal intensities of many product molecules will be reduced. In contrast, a highly hydrophobic membrane would selectively reduce the water vapor partial pressure, while having less or little effect on small non-polar product molecules when keeping the pore sizes. As an example, using a non-porous hydrophobic Teflon membrane interface it was possible to selectively monitor CO2 formation during ethanol oxidation (see previous paragraph), while the more polar incomplete oxidation product acetaldehyde, which also appears at m/z ¼ 44, is hardly able to permeate through the membrane.107,110 This way, one could quantitatively assess the current efficiency for CO2 formation, which is an important parameter in PEMFC operation. The price for this selectivity in detection is of course that certain or even several volatile reaction products are not accessible. (a)

(b)

Glassy carbon Gasket

WE Top

Glass frit

Body

Outlet

CE

15

9

Heating Jacket

10

14

Heating Jacket

H3PO4

Inlet

14

8 4

Inlet

Outlet

1

5

3

to MS

Bottom Gas outlet

2

13

6

Gas inlet

Cu – Heating Plate 12

Fig. 6 Schematic illustration of two different DEMS cell designs for operation at elevated temperatures. (a) Cross section and top view of a thin-film electrode thin-layer channel flow cell for studies of thin-film catalysts at elevated temperature and pressure.111 The channel is formed by an appropriately formed gasket between the cell body and a rectangular glassy carbon plate, which is pressed against the cell body. This results in a well-defined flow pattern (see bottom). The working electrode consists of a Pt spot deposited above the channel in the middle of the glassy carbon plate. The outlet of the cell connects to the separate membrane inlet unit. (b) Design of a beaker-type cell for studies of a proton conductive membrane-electrode assembly, where the gas diffusion electrode representing the working electrode (1), a supporting PEEK sieve (3) connected to the gas inlet, and a PBI membrane (4)are flushed with the feed gas.112 The gas outlet (6) connects to the differentially pumped mass spectrometer (further details see figure). Panel a: Reproduced with permission of Fuhrmann, J.; Zhao, H.; Holzbecher, E.; Langmach, H.; Chojak, M.; Halseid, R.; Jusys, Z.; Behm, R.J. Experimental and Numerical Model Study of the Limiting Current in a Channel Flow Cell With a Circular Electrode. Phys. Chem. Chem. Phys. 2008, 10, 3784–3795, Owner Societies; Panel b: Reproduced with permission of Neither, C.; Rau, M.S.; Cremers, C.; Jones, D.J.; Pinkwart, K.; Tübke, J. Development of a Novel Experimental DEMS Set-Up for Electrocatalyst Characterization Under Working Conditions of High Temperature Polymer Electrolyte Fuel Cells. J. Electroanal. Chem. 2015, 747, 93–107, Elsevier B.V.

Methods and Instruments | Differential Electrochemical Mass Spectrometry

87

Alternative approaches for DEMS model studies at elevated temperatures using realistic membrane electrode assembly (MEA) electrodes and a carrier gas for the delivery of the reaction products to the mass spectrometer were reported in references 112 (Fig. 6b) and 113. A second problem to be considered may result from the fragmentation of the product molecules caused by the electron impact ionization of the different product molecules. This way, fragments of different product molecules may appear at the same mass number, and therefore cannot be separated. To avoid such kind of overlap, isotope labeling techniques can be employed to shift the specific mass number of the solute or of the reaction product and this way discriminate between these signals. As an example, using deuterated ethanol in ethanol oxidation one can discriminate between CO2 and acetaldehyde, which normally would both appear at m/z ¼ 44.23 Alternatively, one can follow the formation of a specific product not via the main mass number, but rather via a unique fragment, which does not interfere with signals from the solute or the fragmentation pattern of other products. As an example, the oxidation of ethanol over a Pt/C catalyst was followed by monitoring the m/z ¼ 29 (CHO+ fragment of acetaldehyde) and m/z ¼ 22 (doubly ionized CO22+ ion) signals.61 In a third approach, one can try to tune the ionization parameters to diminish the relative intensity of the undesired fragments to quantitatively monitor the reaction product (CO) formation, as demonstrated for the detection of the reaction product CO during the electro-reduction of CO2.114 In a more chemical approach one may use a further chemical reaction of the reaction product of interest with another component of the electrolyte. As an example, the incomplete oxidation products of the methanol oxidation reaction, formaldehyde and formic acid, can be monitored online by following the m/z ¼ 76 and m/z ¼ 60 signals, arising from the reaction of the reactant methanol with formaldehyde and formic acid, respectively.115 On the other hand, reactions with the electrolyte can also be a drawback for the detection of the reactant. As an example, CO2 formed during a reaction may react further in alkaline electrolytes to carbonates, which cannot be detected via a bare membrane inlet. In this case, the problem can be overcome, however, by preparing a porous electrode directly onto the membrane. This way, the CO2 formed during reaction can pass through the membrane to the UHV system of the mass spectrometer before reacting with the hydroxyl ions to carbonates.116,117 Finally, we would like to mention that the fragmentation of the molecules upon electron impact ionization can be avoided by using a soft ionization process. This is discussed in more detail in Section 6.

4

DEMS approaches using micro-structured interfaces

A very different way of improving the collection efficiency and possibly reducing the contributions from electrolyte components involves the use of complex, micro-fabricated cell structures. Trimarco et al. introduced recently the so-called ‘sniffer chip’ (Fig. 7a),118,119 which is based on a silicon chip with a micro-machined silicon membrane with arrays of pin-holes. This surface was coated by a hydrophobic self-assembled monolayer of perfluorodecyltrichlorosilane (FDTS) on the electrolyte side. Molecules passing through this ‘membrane’ and through the pinholes in the perforated silicon membrane are collected in channels etched into the silicon chip (sampling volume) and transported by a carrier gas to a capillary through the silicon chip, where they pass to the mass spectrometer UHV system. Since it avoids the pressure drop at the liquid-vacuum interface by the carrier gas, the sniffer chip (b) WE 100

Pr im ar y

lo n

Se co nd ary lon s

(a)

B ea m

1 bar (g) SiN Membrane

To MS

Au WE

Electrolyte Ag/AgCl RE

10–6 mbar (g)

Pt CE Cu Wires

Inlet

PDMS

Outlet

Fig. 7 Schematic illustrations of micro-structured cells developed to improve the performance of DEMS setups (collection efficiency, time resolution etc.). (a) Micro-structured ‘sniffer chip’ DEMS cell design, where the permeation through the membrane between electrolyte (blue) and the pinholes of the micro-structured Si membrane (gray) is reduced by the carrier gas flushed along the channels etched into the Si chip (further details see text and figure).119 (b) Membrane-less design of a micro-structured, where the product species developed at the electrode are detected by time-of-flight secondary ion mass spectrometry (TOF-SIMS), which probes the accessible electrolyte surface in a very small aperture. The aperture is drilled through a thin silicon nitride window with a gold film working electrode beneath by the primary ion beam (further details see text and figure).124 Panel a: Reproduced with permission of Trimarco, D.B.; Scott, S.B.; Thilsted, A.H.; Pan, J.Y.; Pedersen, T.; Hansen, O.; Chorkendorff, I.; Vesborg, P.C.K. Enabling Real-Time Detection of Electrochemical Desorption Phenomena With Sub-Monolayer Sensitivity: Electrochim. Acta 2018, 268, 520–530, Elsevier Ltd; Panel b: Reproduced with permission of Wang, Zh.; Zhang, Y.; Liu, B.; Wu, K.; Thevuthasan, S.; Baer, D.R.; Zhu, Z.; Yu, X.-Y.; Wang, F. In Situ Mass Spectrometric Monitoring of the Dynamic Electrochemical Process at the Electrode - Electrolyte Interface: A SIMS Approach. Anal. Chem. 2017, 89, 960–965, American Chemical Society.

88

Methods and Instruments | Differential Electrochemical Mass Spectrometry

approach benefits from much less solvent and solute evaporation into the analyte gas stream. The device thus successfully combines MIMS and OEMS designs, and was adopted for various electrodes119–122 and also commercialized.123 Another micro-machined, but membrane-free DEMS cell approach was reported recently by Wang et al.124 In their design, the working electrode, an about 30 nm thick Au film, was sputter-deposited on the electrolyte side of a 100 nm thick SiN window, which was part of the top construction of a micro flow cell (Fig. 7b). This cell was transferred into the analysis chamber of a timeof-flight secondary ion mass spectrometer (TOF-SIMS), where the primary ion beam was used to sputter an ultra-small aperture through the thin SiN window and the gold film working electrode, exposing the liquid electrolyte directly to the UHV environment and thus to the primary ion beam. The secondary ions emitted from the electrolyte surface (electrolyte–vacuum interface) are analyzed by TOF-SIMS. This approach was named as ‘System for Analysis at the Liquid − Vacuum Interface (SALVI)’ by the authors.124 Most interesting in this approach is that because of the extremely high sensitivity of SIMS detection and the very short diffusion path lengths it allows for the detection of rather short-living reaction intermediates, which are present only in extremely low concentrations. The performance of this instrument was demonstrated in the electrochemical oxidation of ascorbic acid. Because of the large mass range/mass resolution of TOF-SIMS spectrometers this approach is potentially applicable also in electrochemical studies of complex organic or biological systems.124

5

Non-aqueous electrolytes

After focusing so far on aqueous electrolytes, we will now deal with non-aqueous electrolytes, which in recent years have attracted rapidly increasing interest due to their importance in batteries and battery research. Considering DEMS studies, these differ from aqueous electrolytes in several aspects, in (i) their often either much higher or in some cases much lower vapor pressure, in (ii) their often considerably wider stability range, allowing tests over a much wider potential region, in (iii) their much less well-defined electrochemical stability and decomposition behavior, which may lead to overlap of the electrolyte related mass signals with those of reaction/decomposition products, in (iv) their tendency to fragmentation upon electron impact ionization, which may lead to overlap also with fragments of the solvent, in (v) their generally much lower electrical conductivity, and (vi), important for experimental handling, in their high sensitivity towards traces of water or O2. Also important for experimental handling is that the organic solvents and reagents are often much harder to purify to the ppm level than water.

5.1

Organic electrolytes

5.1.1 Membrane inlet The first attempt to analyze the decomposition/stability of an organic electrolyte by DEMS was reported for an organic Li ion battery type electrolyte (Li perchlorate in polypropylene carbonate (PC)), whose vapor pressure is reasonably low (Fig. 8a).125 In this work the authors were interested in the stability and decomposition of the electrolyte in contact with a porous Pt electrode, which was deposited on a porous Teflon membrane. Half-cell measurements performed in a beaker type DEMS cell filled with the organic electrolyte revealed a cathodic reduction of PC to propene, some methane and H2 from water traces, while CO2 formation was observed at anodic potentials. At high potentials (>1.7 V vs. NHE), oxidation of perchlorate ions yielded primarily ClO2, together with some other chlorine containing products formed in secondary reactions (HCl and chloroprene). These authors also noted that the organic electrolyte should (i) not wet the porous Teflon membrane to prevent permeation of the liquid phase, analogous to the requirement for aqueous electrolytes, and (ii) exhibit a sufficiently low vapor pressure to avoid extensive evaporation into the vacuum chamber. To investigate the stability/decomposition of DMC/EC based electrolytes with different Li salts in contact with carbon (XC 72), Imhof and Novák added a second, thermally treated, less permeable membrane underneath the porous Teflon membrane.126 The electrode consisted of a porous layer of the active material, which was deposited on the electrolyte side of the upper, porous Teflon membrane.126 This study revealed the formation of ethene in the low potential region. In a subsequent, comparable DEMS study, where they investigated the reactive interaction of PC based electrolytes with metal oxide based high voltage electrodes, these authors employed only a single Teflon membrane of unspecified porosity, most likely a non-porous membrane (Fig. 8b).127 In a similar type DEMS study, Wang et al. investigated the gas evolution from technically relevant EC/DMC or EC/DEC electrolytes in contact with high-voltage oxide electrodes consisting of lithium-manganese oxide, lithium-cobalt oxide or mixed lithium-nickel-manganese oxide.128 In their case, a porous layer of active material was deposited on a porous Al foil current collector, which in turn was placed on the porous Teflon membrane with an additional non-porous perfluoroethylene-propylene (FEP) membrane underneath. These examples illustrated the limitation to electrolyte of sufficiently low vapor pressure when using porous membranes. Suitable solvents for such membranes are, e.g., propylene carbonate, dimethyl sulfoxide (DMSO), tetraglyme (TEGDME), and N-methyl-2-pyrrolidone (NMP), which were extensively investigated in model studies.129–135 Bondue et al. improved the existing flow cell configuration such that one could detect also oxidizable/reducible intermediates electrochemically by an additional detection electrode, in addition to the DEMS detection of volatile products (Fig. 8c).136 To reduce the Ohmic drop, they employed separate counter and reference electrodes for the working (generator) electrode and the detector electrode, respectively. This resulted in a dual thin-layer flow cell with six electrodes, where the detector electrode consisted of a thin metal film sputtered on the interface membrane located downstream. Employing this setup in a model study of O2 reduction on Au in Li+-containing DMSO electrolyte, they could simultaneously detect the O2 consumption at the generator electrode and the Faradaic current resulting from the oxidation of the superoxide O−2, which was produced before at the generator electrode.136

Methods and Instruments | Differential Electrochemical Mass Spectrometry

(c)

(b)

(d)

We + membrane + frit

PEEK

Cell

Swagelok

(a)

89

Fig. 8 Schematic illustrations of DEMS cells with membrane inlets designed for studies in organic electrolyte. (a) Early design featuring a porous Pt electrode on a porous Teflon membrane, suitable only for solvents with low vapor pressure such as propylene carbonate electrolyte.125 Here, a Pt wire inside the main cell compartment was used as the counter electrode, to avoid complications due to the high ohmic resistance of the electrolyte (b) Later design where the solvent permeation through the membrane was reduced by using a non-permeable Teflon which carried the active material layer (further details see text and figure).127 (c) Dual thin-layer flow cell with an optional collector electrode sputtered over a porous Teflon membrane for use with organic electrolytes of low volatility.136 (d) DEMS cell consisting of two PEEK cylinders integrated into a Swagelok T-connector cell for studies in a realistic cell environment, including realistic electrodes, realistic organic carbonate based battery electrolytes and realistic electrolyte volumes (see text for further details).138 Panel a: Reproduced with permission of Eggert, G.; Heitbaum, J. Electrochemical Reactions of Propylenecarbonate and Electrolytes Solved Therein - A DEMS Study. Electrochim. Acta 1986, 41, 1443–1448, Pergamon Journals, Ltd; Panel b: Reproduced with permission of Imhof, R.; Novák, P. Oxidative Electrolyte Solvent Degradation in Lithium-Ion Batteries: An In Situ Differential Electrochemical Mass Spectrometry Investigation. J. Electrochem. Soc. 1999; 146, 1702–1706, The Electrochemical Society, Inc; Panel c: Reproduced with permission of Bondue, C.J.; Königshoven, P.; Baltruschat, H. A New 2-Compartment Flow Through Cell for the Simultaneous Detection of Electrochemical Reaction Products by a Detection Electrode and Mass Spectroscopy. Electrochim. Acta 2016; 214, 241–252, Elsevier Ltd; Panel d: Reproduced with permission of Jusys, Z.; Binder, M.; Schnaidt, J.; Behm, R.J. A Novel DEMS Approach for Studying Gas Evolution at Battery-Type Electrode|Electrolyte Interfaces: High-Voltage LiNi0.5Mn1.5O4 Cathode in Ethylene and Dimethyl Carbonate Electrolytes. Electrochim. Acta 2019, 314, 188–201, Elsevier Ltd.

Note that this design is very close to that by Wang et al.,137 which is described in Section 6. The only difference is that in the latter study the second electrode (‘detector’) was close, but nevertheless physically separated from the membrane inlet and was selective for a specific reaction. The high permeation of organic solvent vapors through porous membranes can be reduced by instead using a non-porous membrane. Considering the successful use of a non-porous membrane in DEMS measurements in aqueous electrolytes at elevated temperature and pressure,106–110 we used a non-porous, semi-permeable membrane interface in a DEMS model study in realistic, organic carbonate based electrolytes and realistic metal ion battery related electrode materials (Fig. 8d).138 To specifically allow for DEMS studies in a realistic, battery relevant environment, e.g., with very small amounts of electrolyte, this cell was incorporated into a Swagelok T-connector cell. In that cell, a 50 mm thick non-porous FEP membrane was first covered by a sputter-deposited thin Al film, which served as current collector. Then the actual electrode layer was casted on the collector from a suspension of the active material mixed with graphite and a polyvinylidenefluoride (PVDF) binder. After drying and pressing the coated membrane, the electrode was punched from the coated membrane and installed between two PEEK cylinders, where one of them ends in a frit that supports the membrane at its backside (Fig. 8d). The backside of the frit was directly exposed to the UHV system. The Li (or other metal) counter and reference electrodes were immersed into the cell from the top. With this setup, which resulted in a largely reduced electrolyte vapor pressure in the UHV chamber, we could semi-quantitatively monitor the gas evolution at Co-free Li-rich manganese-nickel oxide high-voltage cathodes in organic carbonate based electrolytes (a standard DMC/EC/LiPF6 battery electrolyte) under potentiodynamic138–140 and galvanostatic141,142 conditions. Comparable measurements were performed also for the potentiodynamic sodiation of tin anodes in glyme electrolyte143 and Zn deposition/stripping in ‘water in salt’ electrolytes,144 demonstrating the versatility and applicability of this approach for DEMS studies of battery-relevant electrolyte and electrode materials. Finally, we want to address a second aspect characteristic for non-aqueous electrolytes. Different from aqueous electrolytes, in organic electrolytes also fragments of the solvent could overlap with those of reaction or decomposition products. Eggert and Heitbaum noted already that in such cases, contributions from the electrolyte can be separated by subtracting the background

90

Methods and Instruments | Differential Electrochemical Mass Spectrometry

spectrum from that measured under reaction conditions, where the background spectrum is recorded under non-reactive conditions.125 Due to the high background levels and the possibly small variations in the respective signals, however, such kind of evaluation may be cumbersome.

5.1.2 Capillary inlet Due to the problems arising from the high vapor pressure of the organic electrolytes in (porous) membrane inlets, designs based on the use of a carrier gas followed by a capillary inlet had been proposed as an attractive alternative (see for example Fig. 9a).145 The gaseous reaction products generated at one of the electrodes accumulate in a space above the electrodes and electrolyte (headspace) and are continuously flushed away by a carrier gas.145–150 Carrier gas and products pass through a capillary to the mass spectrometer

To Mass Spectrometer

(a)

(b)

Calibrated capillary leak (ca 1Pl/min)

PE Sealing Gas Inlet

Quartz Window Sealing Ring

Electrolyte Outlet

(c)

A B C D E

Working Electrode

Counter Electrode

F G H

Electrolyte Titanium Cell Body

Electrolyte lnlet

K M

Li-O2 cell

Screw

Gas outlet

(d)

Air inlet

Lid

HDPE gasket

Ar

cal. 1

cal. 2

S1

S2

S3

Pump

MS

Va

PEEK cylinder

S4

Cup

S5

Spring

Electrode sandwich Current collector

Holder

Vm

Solenoid valve

S6

P sensor

Spring

S7

Vc

Mannual valve

Working electrode Counter electrode Cell bottom

Cell Separator

Temperature controlled

Fig. 9 Different cell designs for online electrochemical mass spectrometry (OEMS) detection of gas evolution into the head space of a battery cell: (a) Highly sensitive design for continuous gas sampling from the headspace over oxide electrodes in EC/DMC and EC/PC electrolytes using a carrier gas (for details see figure and text).146 Cell with an essentially sealed head space that is kept at about constant pressure, gas evolution is detected via a calibrated crimped-capillary leak to the mass spectrometer which limits the gas flow rate to 1 mL/min.151 (c) As (b), but with a two-compartment cell with a Li+-ion conducting but otherwise impermeable separator (Ohara glass) between anode and cathode.155 (d) Schematic representation of a combined electrochemical cell (left) and gas handling/ sampling system (right), which allowed to evacuate the headspace region or fill it with defined gases, and to monitor changes in the gas phase composition by taking samples at defined times, with little effect on the composition of the gas phase in the headspace. Panel a: Reproduced with permission of Wuersig, A.; Scheifele, W.; Novák, P. CO2 Gas Evolution on Cathode Materials for Lithium-Ion Batteries. J. Electrochem. Soc. 2007, 154, A449–A454, The Electrochemical Society; Panel b: Reproduced with permission of Tsiouvaras, N.; Meini, S.; Buchberger, I.; Gasteiger, H.A. A Novel On-Line Mass Spectrometer Design for the Study of Multiple Charging Cycles of a Li-O2 Battery. J. Electrochem. Soc. 2013, 160, A471–A477, The Electrochemical Society; Panel c: Reproduced with permission of Metzger, M.; Marino, C.; Sicklinger, J.; Haering, D.; Gasteiger, H.A. Anodic Oxidation of Conductive Carbon and Ethylene Carbonate in High-Voltage Li-Ion Batteries Quantified by On-Line Electrochemical Mass Spectrometry. J. Electrochem. Soc. 2015, 162, A1123–A1134, The Electrochemical Society; Panel d: Reproduced with permission of Lundström, R.; Berg, E.J. Design and Validation of an Online Partial and Total Pressure Measurement System for Li-Ion Cells. J. Power Sources 2021, 485, 229347(1-8), Elsevier B.V.

Methods and Instruments | Differential Electrochemical Mass Spectrometry

91

chamber. The capillary is generally pumped and only a small fraction of the gas (carrier gas plus analyte) passes to the UHV system via an aperture. In designs close to realistic applications, the OEMS cell often mimics the coin-cell configuration with a two-electrode configuration and an electrolyte soaked separator in between, which contains only a very small amount of electrolyte.145–152 At present, this method (online electrochemical mass spectrometry – OEMS, sometimes OLEMS) is the most popular method for monitoring the gas evolution in battery research (for recent reviews see references 10, 12 and 150). Disadvantage of this design is that the headspace is essentially open, and that significant amounts of reaction products and/or electrolyte can be carried out during the measurement, which limits their suitability for long-term measurements. This problem was addressed in two different approaches. The group around Gasteiger replaced the capillary connection to the mass spectrometer by a calibrated crimped-capillary leak with a very low conductivity. Here the gas transport is driven by the pressure drop over the capillary. This way, they could avoid the use of a carrier gas, with all problems related to that such as the massive dilution, and gas evolution from anodic/cathodic electrolyte decomposition or during charging of Li-O2 battery cathodes could be monitored continuously over approximately 10 h with only minor changes in the gas head-space pressure (Fig. 9b).151 A comparable approach was reported later also by Hahn et al.152 Finally, comparable measurements evaluating the volatile products during operation of a real pouch cell, which was directly connected to a mass spectrometer by a capillary, were reported by Geng et al.153 Also in this case the gas transport was driven by the pressure difference between the two ends of the capillary. Depending on the transport through the capillary and the electrolyte, these capillary inlet measurements can result in (quasi-) differential, semi-integral or integral responses of the mass spectrometric measurement. In a second approach to resolve or at least to significantly reduce effects caused by the open headspace and the continuous flow of carrier gas, this was replaced by a complex gas handling system, which allowed to evacuate the headspace region or to fill it with defined gases, and to monitor the change in the gas phase composition by taking samples at defined times. Because of the small size of these samples they hardly affect the composition of the gas phase in the headspace.147,150 An additional problem of the designs discussed so far is that the detected gases can at least in principle originate from both electrodes.154 To avoid such kind of cross-talking, a Li ion conducting material such as Ohara glass, which is not permeable for other volatile species, can be employed as separator.155 The resulting cell, which otherwise is close to that in Fig. 9b, is illustrated in Fig. 9c. Recently, a commercial OEMS cell of unique design was presented, which features a spiral-shaped flow-field machined into the stainless steel current collector. The working electrode, electrolyte, separator and other electrodes are placed on top of the current collector.156,157 The flow field is continuously purged with a stream of inert carrier gas, which carries away gases generated at the electrode. Because of the direct contact to the working electrode and the very small volume of the flow field, this design shows a very good time resolution. Similar to membrane inlet measurements, efficient transport through the capillary (or membrane) tends to result in a high background level of the electrolyte vapor. This not only drastically increases the solvent peak intensity, as for aqueous electrolyte, but can result also in intense fragment formation upon ionization of the organic molecules, whose signals can overlap with and even obscure those of reaction products appearing at the same mass number. Also in this case contributions from the electrolyte vapor can be reduced by placing a cold trap between inlet and mass spectrometer, with the risk that reaction products may be trapped as well.149 The impact of electrolyte vapor fragmentation on the intensities of decomposition products in operando OEMS measurements was addressed recently.158

5.2

Low-vapor-pressure electrolytes

In contrast to the high volatility of most organic electrolytes, a number of liquid electrolytes, specifically ionic liquid based electrolytes,159 exhibit an extremely low vapor pressure.160 Nevertheless, ionic liquids can pass through the pores of the membrane due to their excellent wetting of typical membrane materials.161 Therefore, using a porous membrane interface to the vacuum system in ionic liquid electrolytes may be problematic, despite of their ultra-low vapor pressure. To avoid any problems, a non-porous Teflon membrane was employed for DEMS studies of ionic liquid (BMP-TFSI) based electrolytes, which were all performed in a flow cell setup.162–164 This membrane is still sufficiently permeable for gases, but prevents passing of the ionic liquid. In addition, the extremely low volatility of ionic liquids offers the unique opportunity to place an open beaker type cell directly into the UHV system of the mass spectrometer.165,166 In this case, the evolution of gaseous decomposition products at the respective electrode can be directly monitored without any additional interface device. This was demonstrated for the first time by our group for the interaction of BMP-TFSI with glassy carbon and Au electrodes (Fig. 10a).165,166 In a comparable approach, the gassing behavior at the cathode of a poly(ethylene oxide) (PEO)-based LiCoO2|PEO-LiTFSI|Li solid polymer electrolyte battery, where the electrolyte was wetted by an ionic liquid, was monitored online via a pinhole in the cathode. The pinhole was directly exposed to the UHV of the mass spectrometer chamber (Fig. 10b).167 A differentially pumped mass spectrometer with a conically shaped open-end (‘sniffer’) was used to monitor the gas formation at a Pt/YSZ (yttrium-stabilized zirconia) solid state electrolyte interface, where the Pt working electrode and a Pt or Au reference electrode were sputter deposited on the two sides of the rectangular shaped YSZ plate.168 Reactant gases were provided via capillaries to the cathodic and anodic chamber of the reactor. Using this setup, which they referred to as Solid Electrochemical Mass Spectrometry (SEMS), the authors could directly monitor the electrochemically promoted CO oxidation reaction at 400  C directly in the UHV chamber. Finally, the gas evolution at a non-volatile solid-state electrolyte with essentially zero vapor pressure in all-solid-state Li ion batteries (ASSB) was explored using a carrier gas driven OEMS design implemented for the ASSB cell geometry (Fig. 10c).169

92

Methods and Instruments | Differential Electrochemical Mass Spectrometry

(a)

(b)

(c)

Fig. 10 DEMS cell designs for low volatility electrolytes. (a) Open beaker type DEMS cell filled with an ionic liquid electrolyte positioned directly in the UHV chamber.165 This design allows to solely detect volatile products developing at the working electrode, as products from the counter electrode are removed by a separate pump. (b) Cell design for measurements at a (poly(ethylene oxide) based) solid polymer electrolyte battery.167 Gas sampling occurs via a pinhole in the LiCoO2 cathode, which is directly exposed to the vacuum of the mass spectrometer chamber. (c) Schematic drawing of the SSB pellet implemented in the DEMS cell for carrier gas driven differential OEMS measurements at all-solid-state-batteries.169 Red and purple arrows indicate potential gas diffusion pathways. Panel a: Reproduced with permission of Alwast, D.; Schnaidt, J.; Law, Y.T.; Behm, R.J. A Novel Approach for Differential Electrochemical Mass Spectrometry Studies on the Decomposition of Ionic Liquids. Electrochim. Acta 2016, 197, 290–299, Elsevier Ltd; Panel b: Reproduced with permission of Nie, K.; Wang, X.; Qiu, J.; Wang, Y.; Yang, Q.; Xu, J.; Yu, X.; Li, H.; Huang, X.; Chen, L. Increasing Poly(ethylene oxide) Stability to 4.5 V by Surface Coating of the Cathode. ACS Energy Lett. 2020, 5, 826–832, American Chemical Society; Panel c: Reproduced with permission of Bartsch, T.; Strauss, F.; Hatsukade, T.; Schiele, A.; Kim, A-Y.; Hartmann, P.; Janek, J.; Brezesinski, T. Gas Evolution in All-Solid-State Battery Cells. ACS Energy Lett. 2018, 3, 2539–2543, American Chemical Society.

A so-called stacked pellet, consisting of the nickel-manganese-cobalt oxide (NMC) cathode, a Li thiophosphate or argyrodite solid-state electrolyte, and an In metal or LTO anode, was mounted in a closed electrochemical cell. The cell was continuously flushed with He or Ar carrier gas, and the carrier gas containing the gaseous reaction products (analyte) was transported to the mass spectrometer chamber via an open-end capillary, where a small fraction of the analyte was introduced into the mass spectrometric chamber, while most of it was passed to the exhaust. This resulted in a differential response, which allowed the determination of gas evolution rates. It should be noted that such kind of ASSB DEMS approach cannot discriminate between gas contributions from the one or the other electrode, while transport of gaseous products from the one to the other electrode is not very likely.

6

Ambient pressure ionization techniques

The different approaches discussed so far all involved electron impact ionization of the product molecules, which may cause problems due to the fragmentation. This is particularly important for organic solvents, but not limited to them. Such kind of fragmentation can be avoided by using non-destructive ‘soft’ ionization methods such as thermospray or electrospray ionization,170 where the analyte is removed and ionized from a nozzle leading to the mass spectrometer chamber either thermally or by an electric field (field desorption), or by chemical ionization. These ionization methods, which can operate at ambient pressure, were typically employed to detect non-volatile (liquid phase) reaction products, which are dominant in the electrochemical oxidation of more complex organic molecules,6,8,11 but also more general for chemical analysis, e.g., by selected-ion flow-tube mass spectrometry (SIFT-MS),171 where gaseous ions such as H3O+, NO+, or O2+ act as soft chemical ionization agents. It should be noted that although the ionization will be performed under ambient conditions, the detection in the mass filter still requires vacuum conditions. Early attempts to couple thermospray ionization mass spectrometry with electrochemistry for studies of electrochemical reactions of organic compounds were addressed in references172,173 (Fig. 11a). Studies coupling electrospray ionization mass spectrometry, which is a common method in analytical chemistry,6 with an electrochemical cell were first reported in references174,175 (see also the review by Herl and Matysik11) (Fig. 11b and c). Electrospray ionization mass spectrometry was demonstrated to be a powerful tool for studying the decomposition of various battery electrolytes and for identifying the resulting products, in particular in combination with other methods such as liquid chromatography or gas chromatography as separation techniques.176 Soft ionization by selected-ion flow-tube mass spectrometry (SIFT-MS) was successfully applied, e.g., for the online monitoring of the CO2 reduction reaction products over a Cu electrode by Lobaccaro et al.177 In their study product ions were created in a flow

Methods and Instruments | Differential Electrochemical Mass Spectrometry

(a)

(b)

(c)

(d)

(e)

(f)

93

Fig. 11 Schematic illustration of different schemes for ambient pressure ionization mass spectrometric detection of volatile and soluble electrochemical reaction products (for details see figures and text): (a) Electrochemical cell containing a smooth working electrode for thermospray ionization mass spectrometry.172 (b) Electrochemical flow cell for electrospray ionization mass spectrometry with a platinum micro-cylinder (2) as working electrode. The electrolyte is infused in port 10, carrier gas to pneumatically assist the electrospray process may be introduced in port 12 of a stainless steel tube (14).174 (c) Thin-layer flow cell for electrospray ionization mass spectrometry with a smooth working electrode.175 (d) Schematic representation of the electrochemical cell (PEEK) used for the real-time product detection in electrochemical CO2 reduction by Selected-Ion Flow-Tube Mass Spectrometry (SIFT-MS).177 (e) Electrochemical thin-layer flow cell coupled to a liquid-sample desorption electrospray ionization (DESI) mass spectrometer. The sample solution is infused via a capillary and subject to DESI ionization once it exits from the tip of the capillary, initiated by the interaction with the charged microdroplets generated from the DESI spray.178 (f ) ‘Water-wheel’ electrode for the DESI-MS detection of short living electrochemical reaction intermediates.179 Panel a: Reproduced with permission of Hambitzer, G.; Heitbaum, J. Electrochemical Thermospray Mass Spectrometry. Anal. Chem. 1988, 58, 1067–1070, American Chemical Society; Panel b: Reproduced with permission of Xu, X.; Lu, W.; Cole, R.B. On-Line Probe for Fast Electrochemistry/Electrospray Mass Spectrometry. Investigation of Polycyclic Aromatic Hydrocarbons. Anal. Chem. 1996, 68, 4244–4253, American Chemical Society; Panel c: Reproduced with permission of Zhou, F.; Van Berkel, G.J. Electrochemistry Combined On-Line with Electrospray Mass Spectrometry. Anal. Chem. 1995; 67, 3643–3649, American Chemical Society; Panel d: Reproduced with permission of Lobaccaro, P.; Mandal, L.; Motapothula, M.R.; Sherburne, M.; Martin, J.; Venkatesan, T.; Ager, J.W. Initial Application of Selected-Ion Flow-Tube Mass Spectrometry to Real-Time Product Detection in Electrochemical CO2 Reduction, Wiley-VCH Verlag GmbH & Co. KGaA; Panel e: Reproduced with permission of Liu, P.; Zheng, Q.; Dewald, H.D.; Zhou, R.; Chen. H. The Study of Electrochemistry with Ambient Mass Spectrometry. Trends Anal. Chem. 2015, 70, 20–30, Elsevier B.V.; Panel f: Reproduced with permission of Brown, T.A.; Chen, H.; Zare, R.N. Identification of Fleeting Electrochemical Reaction Intermediates Using Desorption Electrospray Ionization Mass Spectrometry. J. Am. Chem. Soc. 2015, 137, 7274–7277, American Chemical Society.

tube, employing N2 as a carrier gas and a number of different ions for ionization (Fig. 11d). To exclude contributions from the counter electrode, the authors used a two-compartment electrochemical cell, with the two compartments being separated by an anion conducting membrane. This way they could directly detect the formation of C1–C3 hydrocarbons, alcohols, and aldehydes as reaction products during CO2 reduction. In liquid-sample desorption electrospray ionization (DESI), another soft-ionization approach, charged micro-droplets of a spray solvent, which had been generated separately before by ionization, are sprayed to a region in front of the end of the electrolyte outlet capillary, where they ionize the analyte under ambient conditions (Fig. 11e).178 This approach is particularly suited for coupling with electrochemical measurements, as it physically separates the spray unit, which is operated at high voltage, and the electrochemical cell and related devices, while in common EC-ESI-MS configurations these are at very high or very low potentials. For applications aiming at the detection of short-living reaction intermediates, the jet of sprayed micro-droplets was directly sprayed on the electrode, which was continuously rotated (Fig. 11f ).179 This design results in a very short time between the formation of the reaction products at the working electrode and their ionization/detection, which allows the identification of short-living reaction intermediates. Finally, local resolution of the product formation was achieved with an EC-ESI-MS coupling where a coaxial capillary detected the local formation of products while the sample was scanned laterally (scanning capillary microscope). The coaxial capillary provided the electrolyte via the annulus formed by the inner and outer tube (electrolyte inlet), while the outgoing electrolyte was transported in the inner tube of the coaxial capillary (electrolyte outlet) to the ESI-MS (Fig. 12a and b).180 This setup was employed for screening the electrochemical oxidation of dimethyl-p-phenylenediamine over a structured electrode consisting of 3 Pt stripes on an inert glass (Fig. 12a).

94

Methods and Instruments | Differential Electrochemical Mass Spectrometry

to MS

(a)

15

F0

F1

(b)

Auxiliary stream

r1

H

7b

r2

Z 9

7 Reagents inlet

(c)

7a 11

inlet 12

13

outlet

push-pull probe 5

8

4 6

CE

electrospray ESI-MS inlet pump

bubble/plug formation

RE

air bubble

CE/RE WE track electrolyte droplet

3 1 WE gas

analyte droplet

substrate X 10 2

14

open microchannels

MALDI plate

13

Fig. 12 Scanning mass spectrometry based on ESI detection. (a) Schematic representation of the scanning capillary microscope based on a scanning coaxial impinging inlet/outlet capillary and ESI-MS detection for spatially resolved detection of the local formation of volatile products.180 Main components are the working electrode (1), x-translation stage (2), electrochemical glass cell (3), inner capillary (4), outer capillary (5) and transfer line to the ESI-MS (15). (b) Schematic representation of the miniature flow cell formed at the bottom of the coaxial capillary setup (F1: feed through the annulus of the coaxial capillary set up, F0: product stream to the MS). (c) Schematic view of the general operation principle of a push-pull scanning electrochemical cell in combination with ESI-MS and matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS).181 Electrochemical measurements are carried out within a nanoliter droplet of electrolyte. Panel a and b: Reproduced with permission of Modestov, A.D.; Srebnik, S.; Lev, O.; Gun, J. Scanning Capillary Microscopy/Mass Spectrometry for Mapping Spatial Electrochemical Activity of Electrodes. Anal. Chem. 2001, 73, 4229–4240, American Chemical Society; Panel c: Reproduced with permission of Momotenko, D.; Qiao, L.; Corte’s-Salazar, F.; Lesch, A.; Wittstock, G.; Girault, H.H. Anal. Chem. 2012, 84, 6630–6637, American Chemical Society.

High spatial resolution in the mass spectrometric analysis was achieved also by a powerful combination of a scanning electrochemical microscope (SECM) with a micro-fabricated so-called push-pull probe and an ESI-MS or a matrix-assisted laser desorption/ionization mass spectrometer (MALDI-MS) (Fig. 12c).181 The push-pull probe includes two parallel micro-channels which serve as electrolyte inlet and outlet, respectively, where the outlet channel connects via a capillary to the ESI-MS or the MALDI-MS. With the push-pull probe a nanoliter droplet is created on the electrode surface, which can be considered as local electrochemical reactor. The authors demonstrated the impressive performance (sensitivity, spatial resolution) of this instrument in a number of analytical applications, such as the spatially resolved detection of contamination by explosives or of enzyme activities. In total, these soft ionization techniques are particularly attractive when dealing with more complex electrolytes, where fragmentation and/or transfer to the mass spectrometer pose severe problems.

7

Beyond DEMS – Coupling with other techniques

The DEMS technique, though a powerful approach for the online monitoring of the gaseous and volatile reaction products, is also limited to these species. This excludes the detection of soluble products present in the bulk of the solution or of adsorbed reaction products remaining on the electrode surface, which are not detectable due to their low vapor pressure. Also, short-living reaction intermediates as well as other adsorbed species such as precursors, poisoning species or anionic spectator species cannot be detected by DEMS alone, although they may play a crucial role in the reaction, affecting or even determining the activity and selectivity. Their detection requires the use of other techniques, which are able to detect and identify these species, preferably, in combination with simultaneous (online) DEMS measurements, i.e., at the same electrode and under the same reaction conditions. Such kind of hyphenated techniques are highly valuable when aiming at a detailed kinetic and mechanistic understanding of the ongoing reactions. The dual thin-layer flow cell (Fig. 2d)50 allows not only online DEMS measurements on a variety of electrodes as discussed above, but can serve also as a multipurpose interface for unique combinations of DEMS with other techniques, such as a electrochemical

Methods and Instruments | Differential Electrochemical Mass Spectrometry

95

quartz crystal microbalance EQCM,50,182 attenuated total reflection Fourier transform infrared spectroscopy ATR-FTIRS,183–196 or electrospray ionization mass spectrometry ESI-MS.197,198 This way information on the formation of gaseous products (DEMS), adsorbed species (EQCM, ATR-FTIRS) and/or non-volatile products (ESI-MS) can be obtained online in a single measurement. The combined DEMS/ATR-FTIRS setup184,185 (Fig. 13a) led to an unprecedented amount of spectro-electrochemical kinetic and mechanistic details on a number of complex reactions, in particular on the oxidation of small organic molecules.185–196 These studies provided a solid basis for a detailed molecular scale understanding of the respective reaction and the impact of the adsorbed species on the activity and selectivity of the catalyst. A coupled ultrahigh vacuum (UHV) – electrochemistry setup, allowing electrochemical studies on single-crystal electrode surfaces prepared and characterized under well-defined UHV conditions,199 was combined recently by Schnaidt et al. with a DEMS detection option.200 This instrument, which allowed DEMS measurements of electrochemical processes on structurally and chemically well-defined single-crystal electrodes under controlled electrolyte mass transport conditions, was based on a modified version of the dual thin-layer cell, with a membrane inlet interface to the mass spectrometer (Fig. 13b). Structural characterization of well-defined single crystal electrode surfaces by high-resolution scanning tunneling microscopy (STM) and different spectroscopic techniques under UHV conditions in combination with DEMS measurements provided a detailed, atomic scale understanding of structure-reactivity relations for a variety of mono- and bi-metallic surfaces in different electrocatalytic reactions.120,201–203 Furthermore, electrode characterization after the electrochemical measurements, upon back-transfer of the electrode into the UHV environment, gained information on structural modifications caused during exposure to the electrolyte under the respective reaction conditions.200,203 DEMS measurements turned out to be highly informative also for studies of photoelectrochemical reactions. This was demonstrated first in DEMS measurements of O2 evolution in acidic aqueous solution on thin porous titanium dioxide (anatase) electrodes under illumination by Bogdanoff and Alonso Vante.204 As these measurements were performed in a three-electrode cell with a stagnant electrolyte (Fig. 14a), they were still limited by an ill-defined mass transport. In addition, a relatively thick layer of (a)

(b)

(c)

Fig. 13 (a) Combination of a membrane inlet DEMS with FTIRS in an attenuated total reflection (ATR) configuration for simultaneous monitoring of gaseous/volatile and adsorbed species at thin metal films (details see figure).184,185 (b) Combination of a membrane inlet DEMS with an ultrahigh vacuum (UHV) system (right of position 1) for the controlled preparation/characterization of well-defined single crystalline electrode surfaces under UHV conditions and their electrochemical characterization in the transfer chamber (position 7), without contact to air.200 (c) Detailed schematic representation of the dual thin-layer flow cell setup formed by the working electrode (position 14, hanging from the top) and the electrochemical cell. The latter is moved towards the working electrode such that a droplet forms between them (18, 19: electrolyte inlet/outlet). The effluent electrolyte passes over a membrane inlet (20) in the second compartment of the cell which directly connects to a mass spectrometer. Panel a: Reproduced with permission of Heinen, M.; Chen, Y.X.; Jusys, Z.; Behm, R.J. In Situ ATR-FTIRS Coupled with On-Line DEMS Under Controlled Mass Transport Conditions - A Novel Tool for Electrocatalytic Reaction Studies. Electrochim. Acta 2007, 52, 5634–5643, Elsevier Ltd; Panel b: Reproduced with permission of Schnaidt, J.; Beckord, S.; Engstfeld, A.K.; Klein, J.; Brimaud, S.; Behm, R.J. A Combined UHV-STM-Flow Cell Set-Up for Electrochemical/Electrocatalytic Studies of Structurally Well-Defined UHV Prepared Model Electrodes. Phys. Chem. Chem. Phys. 2017, 19, 4166–4178, The Owner Societies.

96

Methods and Instruments | Differential Electrochemical Mass Spectrometry

1

(a)

(b)

hm

2 1

10

RE

CE Inlet

5

3

3

MS

Outlet to DEMS

4 7 8 9

CE

6

WE

RE

Glass / FTO / TiO2

Inlet

Window

hm

4

Outlet to DEMS

6 Fig. 14 Combination of a membrane inlet DEMS system with a photoelectrochemical cell in (a) a beaker cell,204 and in (b) a thin-layer channel flow cell configuration (top: top view, bottom: side view).205 In the latter setup the membrane inlet to the mass spectrometer follows in a separate cell (not shown). Panel a: Reproduced with permission of Bogdanoff, P.; Alonso-Vante, N. On-line Determination via Differential Electrochemical Mass Spectroscopy (DEMS) of Chemical Products Formed in Photoelectrocatalytical Systems. Ber. Bunsenges. Phys. Chem. 1993, 97, 940–943, VCH Verlagsgesellschaft mbH; Panel b: Reproduced with permission of Reichert, R.; Jusys, Z.; Behm, R.J. A Novel Photoelectrochemical Flow Cell With Online Mass Spectrometric Detection: Oxidation of Formic Acid on a Nanocrystalline TiO2 Electrode. Phys. Chem. Chem. Phys. 2014, 16, 25076–25080, The Owner Societies.

(a)

(b)

Fig. 15 Combination of a membrane inlet DEMS system with an additional electrochemical collector probe. (a) Schematic presentation of a gas accessible membrane electrode (GAME, gray) with an additional local Pt ultramicroelectrode (UME) sensor, which allows for the simultaneous detection of gaseous (DEMS) and liquid (UME) products in a well-defined three-phase boundary system (for details see text and figure).207 (b) Schematic representation of a double-band-electrode channel flow DEMS cell with an additional detector electrode (2) in the flow cell, located downstream from the working electrode (1) and the membrane inlet (5, 10).137 Panel a: Reproduced with permission of Zhang, G.; Cui, Y.; Kucernak, A. Real-Time In Situ Monitoring of CO2 Electroreduction in the Liquid and Gas Phases by Coupled Mass Spectrometry and Localized Electrochemistry. ACS Catal. 2022, 12, 6180–6190, American Chemical Society; Panel b: Reproduced with permission of Wang, H.; Rus, E.; Abruna, H.D. New Double-Band-Electrode Channel Flow Differential Electrochemical Mass Spectrometry Cell: Application for Detecting Product Formation During Methanol Electrooxidation. Anal. Chem. 2010, 82, 4319–4324, American Chemical Society.

aqueous electrolyte on top of the electrode caused significant (photon) absorption losses. To improve the time resolution of the mass spectrometric response and reduce the light absorption by the electrolyte, a novel thin-layer channel photoelectrochemical flow cell was developed, which was interconnected to a separate membrane inlet module (Fig. 14b).205 This way, the authors could discriminate between contributions from photoelectrochemical water splitting and from photo-electrooxidation of formic acid at a nanocrystalline TiO2 (commercial P25) thin-film electrode to the measured overall photocurrent. Similar type measurements allowed the quantitative evaluation of the contributions from photoelectrocatalytic water oxidation and C1 molecule oxidation in sulfuric acid based electrolyte at an electrodeposited WO3 film electrode,206 as well as to identify contributions from electrocatalytic water oxidation, photoelectrocatalytic water oxidation and photocatalytic H2 evolution on planar nanocrystalline Au/TiO2 catalyst film electrodes.104 A DEMS concept based on a gas accessible membrane electrode (GAME), which allows facile gas transport to the electrode and coupling with other techniques, was introduced recently by Zhang et al. (Fig. 15a).207,208 Main component of this is a microstructured polycarbonate track etched (PCTE) membrane, coated on both sides with a hydrophobic Teflon layer. Gases can diffuse through this layer and the PCTE via etched sub-micron sized pinholes. A porous Au or AuCu film sputter-deposited on this membrane on the electrolyte side served as working electrode, which was exposed to the stagnant electrolyte. On the back side of the GAME interface, the gaseous products are gases is collected in the tracks, which is continuously flushed by a carrier gas, and guided via an outlet channel to the mass spectrometer. Depending on whether an inert or a reactive feed gas is used, this purely serves as carrier gas or also as reactant. The performance of this instrument was demonstrated in the above references. This platform can be coupled with other devices such as an ultramicroelectrode for sensing the local pH at the electrode surface (Fig. 15a), with UV irradiation for photoelectrochemical measurements, or with an ATR electrode for in situ FTIR spectroscopy, together with the online monitoring of volatile products.207,208

Methods and Instruments | Differential Electrochemical Mass Spectrometry

97

In another approach, the dual thin-layer DEMS flow cell can be combined with a second electrode, which selectively catalyzes only a specific reaction. This approach is of general relevance, e.g., for determining the selectivity in the oxidation of organic molecules, if the reaction leads to products not directly detectable by DEMS. Using a PtPb alloy,209 which can selectively oxidize formic acid and neither methanol nor formaldehyde, Wang et al. employed this approach recently to quantify the formation of formic acid during the electrooxidation of methanol at a Pt/electrode.137 A special channel flow cell, which accommodated both the membrane inlet for the detection of CO2, and the PtPb/C electrode for the detection of formic acid downstream from the working electrode (Fig. 15b), allowed for a quantitative detection of CO2 and formic acid, hence of two of the three methanol oxidation reaction products. Formaldehyde formation could then be calculated from the difference of the overall oxidation current and the sum of the partial currents for CO2 and formic acid formation.137 Note that this design is very close to that by Bondue et al.,136 which was described in Section 5.1.1. The only difference is that in the latter study the second electrode served as non-specific detector for oxidizable or reducible species. Alternatively, formic acid formation during methanol oxidation on a Pt/C electrode can also be monitored via the formation of methyl ester, its reaction product with methanol, at m/z ¼ 60.41 For the direct analytical detection of volatile and non-volatile reaction products we had combined online DEMS with online ESI-MS detection. Also in this case we used the dual thin-layer flow cell as platform. This was applied in a study of methanol oxidation on a Pt/C catalyst, for the quantitative online detection of the three reaction products CO2, formic acid, and formaldehyde, which is not possible by standard DEMS measurements.198 (Note that methylformate is sometimes accounted as an individual product of methanol oxidation rather than as reaction product of the product formic acid and the reactant methanol.210) In this study the quantitative detection of CO2 via DEMS was combined with online ESI-MS detection of formaldehyde and formic acid (Fig. 16a). This required the development of a specific analytical routine for the online derivatization of formaldehyde.211 To prevent the corrosion of the instrument, the derivatized product as well as formic acid have to be extracted from the mineral acid 6 port valve

(a)

to ESI-MS (liquid phase)

Mobile phase: water nic orga e s pha

Sample loop

organic phase mixing

extraction

separation Organic waste

2,4-DNPH, H2SO4

mixing

Aqueous phase

reaction

EC methanol, H2SO4

Aqueous waste

to EI-MS (gas phase)

(b)

B C F D A

E

Fig. 16 (a) Schematic representation of a membrane inlet DEMS system with a coupled electrospray ionization (ESI) mass spectrometer for the simultaneous online analysis of volatile and liquid reaction products.198 This requires an online derivatization−extraction−separation routine to avoid massive intrusion of the acid electrolyte into the mass spectrometer. (b) Schematic representation of the basic components of a setup for electrochemical real-time mass spectrometry (EC-RTMS).212 (A) Scanning flow cell with extraction capillary near the electrode; (B) gas-liquid separation in a degasser; (C) EI-MS for gas analysis; (D) nebulizer; (E) spray chamber where droplets are removed to waste (bottom) and mist is removed for analysis (top) by a continuous flow of nebulizing gas; (F) DART ion source for the ionization of nebulized liquid products and TOF-MS for the detection of the ionized species. Panel a: Reproduced with permission of Zhao, W.; Jusys, Z.; Behm, R.J. Complete Quantitative Online Detection of Methanol Electrooxidation Products via Electron Impact Mass Spectrometry and Electrospray Ionization Mass Spectrometry. Anal. Chem. 2012, 84, 5479–5483, American Chemical Society; Panel b: Reproduced with permission of Khanipour, P.; Löffler, M.; Reichert, A.M.; Haase, F.T.; Mayrhofer, K.J.J.; Katsounaros, I. Electrochemical Real-Time Mass Spectrometry (EC-RTMS): Monitoring Electrochemical Reaction Products in Real Time. Angew. Chem. Int. Ed. 2019, 58, 7273–7277, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

98

Methods and Instruments | Differential Electrochemical Mass Spectrometry

(0.5 M sulfuric acid supporting electrolyte) before injection into the ESI-MS. Constant potential measurements showed a quantitative agreement between the partial currents (charges) for CO2, formic acid and formaldehyde formation determined this way and the overall current (charge) measured during methanol oxidation, confirming that within the accuracy of the experimental detection these three species are the only methanol oxidation products on Pt/C. The time resolution of this approach is, however, limited, since the detection of the non-volatile components (formaldehyde and formic acid) is significantly delayed although the derivatization−extraction−separation routine was performed online. In another approach to online monitor both volatile and non-volatile organic molecule oxidation products by real time electrochemical mass spectrometry, Khanipour et al. recently developed and applied a method combining a scanning electrochemical cell for local resolution, online detection of volatile species by electron-impact mass spectrometry, and online detection of liquid products by a direct analysis in real time – time of flight mass spectrometer (DART-TOF-MS) with a modified inlet to the mass spectrometer (Fig. 16b).212,213 The analyte was collected by a capillary with its end close to the working electrode, and guided to a Teflon® AF 2400 gas-permeable non-porous membrane interface, connected to the electron impact ionization mass spectrometer, where volatile species are detected. The remaining liquid passes to a nebulizer, and the resulting mist is guided to the DART-MS system for direct analysis in real time. This allowed highly sensitive and rapid detection of all products with a high time resolution in the seconds range. The performance of the instrument was demonstrated for a number of electrocatalytic reactions such as the reduction of CO2 on Cu212 or the potentiodynamic oxidation of C1–C3 primary alcohols on a platinum electrode.213 Finally, aiming at information on the nature and quantity of complex organic or organometallic molecular products, Frensemeier et al. described a powerful combination of an electrochemical flow cell, coupled with liquid chromatography (LC), electrospray ionization mass spectrometry (ESI-MS) and inductively coupled plasma mass spectrometry (ICP-MS).214 They applied this instrument for product analysis resulting from the electrochemical activation of Pt(IV)-based anticancer drugs.214

8

Summary and outlook

As outlined in the preceding sections, the DEMS method or, in a more general sense, online electrochemical mass spectrometry, has developed extensively over the recent half-a-century. Clearly, it has established itself as a powerful research tool, contributing to the understanding of the kinetics and mechanism of complex processes occurring at the electrode |electrolyte interface. Applications involved mostly, but not only, research in electrocatalysis/energy conversion and in energy storage (batteries and capacitors), though applications in electrosynthesis and electroanalysis are important as well. State-of-the-art cell designs represent versatile platforms, allowing to directly couple DEMS with other modern in situ techniques. This results in advanced multi-technique tools for online analysis, which can provide detailed information on the nature and quantity of gaseous, liquid and solid phase reaction products as well as on intermediates adsorbed on the electrode surface. Furthermore, flow cell designs or designs involving other hydrodynamic techniques allow also reaction studies under well-defined transport conditions. Finally, the techniques available for combination with DEMS so far can in principle be further expanded to other more advanced configurations. In these sections we illustrated the large variety of the different technical solutions to separate the liquid phase electrolyte from the UHV environment required for the most commonly employed electron impact ionization mass spectrometry. Design solutions for different types of electrolytes such as aqueous electrolytes, (highly volatile) organic or low vapor pressure electrolytes such as ionic liquids, (wetted) solid polymers or solid electrolytes were discussed. We also discussed the effects of different types of cells and inlets on the relative intensity of the mass spectrometric signal, including the suppression of solvent-related signals, and on the time resolution of the DEMS measurement. This also included a brief discussion of the differences between differential and integral measurements. On the other hand, specific effects of the electrolyte, such as its pH or of organic solvents on the stability of the inlets were elucidated. Finally, a variety of hyphenated approaches, combining DEMS (or OEMS) with other techniques sensitive, e.g., to liquid phase reaction products or to adsorbed intermediates or products, were presented. For the future developments, we expect a rapid further expansion of DEMS application in essentially all fields of electrochemistry due to the recent introduction of commercial instruments, including hardware, electronics and tailored software.79,123,156 This allows standardized DEMS studies on a wide range of electrochemical problems. For specific applications, we expect further development in particular in the area of combined instruments, which have an enormous potential in providing detailed online information on the formation and further reaction of volatile, liquid and adsorbed products and intermediates. Especially promising in the future are to our opinion time-resolved measurements on the dynamics of electrochemical/ -catalytic processes and studies involving isotope labeling techniques as well as studies utilizing kinetic isotope effects. In summary, we foresee a bright future for DEMS applications in electrochemistry and in studies of energy conversion/energy storage systems, in particular in combination with other electrochemical and in situ/operando spectroscopic tools. We hope that this chapter will be helpful for the understanding of the characteristic features of different cell designs and detection approaches and provide a solid basis for the choice of the most suitable design for a given problem.

Acknowledgments This work was supported by the German Federal Ministry of Education and Research (BMBF) in the project 03EK3051C and by the Deutsche Forschungsgemeinschaft (DFG) under project ID390874152 (POLiS Cluster of Excellence, EXC 2154). It contributes to the research performed at CELEST (Center for Electrochemical Energy Storage Ulm-Karlsruhe).

Methods and Instruments | Differential Electrochemical Mass Spectrometry

99

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42.

Iwasita-Vielstich, T. Advances in Electrochemical Science and Engineering; Wiley-VCH Verlag GmbH, 1990; pp. 127–170. Chapter 3. Wasmus, S.; Küver, A. Methanol Oxidation and Direct Methanol Fuel Cells: A Selective Review. J. Electroanal. Chem. 1999, 461, 14–31. Baltruschat, H. Interfacial Electrochemistry - Theory, Experiment and Applications; Marcel Dekker, INC: New York, 1999; pp. 577–597. Vielstich, W. Encyclopedia of Electrochemistry - Interfacial Kinetics and Mass Transport; VCH: Weinheim, 2003. Baltruschat, H. Differential Electrochemical Mass Spectrometry. J. Am. Soc. Mass Spectrom. 2004, 15, 1693–1706. Gun, J.; Bharathi, S.; Gutkin, V.; Rizkov, D.; Voloshenko, A.; Shelkov, R.; Sladkevich, S.; Kyi, N.; Rona, M.; Wolanov, Y.; Rizkov, D.; Koch, M.; Mizrahi, S.; Pridkhochenko, P. V.; Modestov, A.; Lev, O. Highlights in Coupled Electrochemical Flow Cell-Mass Spectrometry, EC/MS. Isr. J. Chem. 2010, 50, 360–373. Abd-El-Latif, A. A.; Bondue, C. J.; Ernst, S.; Hegemann, M.; Kaul, J. K.; Khodayari, M.; Mostafa, E.; Stefanova, A.; Baltruschat, H. Insights into Electrochemical Reactions by Differential Electrochemical Mass Spectrometry. Trends Anal. Chem. 2015, 70, 4–13. Oberbacher, H.; Pitterl, F.; Erb, R.; Plattner, S. Mass Spectrometric Methods for Monitoring Redox Processes in Electrochemical Cells. Mass Spectrom. Rev. 2015, 34, 64–92. Lu, F.; Zhang, Y.; Liu, S.; Lu, D.; Su, D.; Liu, M.; Zhang, Y.; Liu, P.; Wang, J. X.; Adzic, R. R. Surface Proton Transfer Promotes Four-Electron Oxygen Reduction on Gold Nanocrystal Surfaces in Alkaline Solution. J. Am. Chem. Soc. 2017, 139, 7310–7317. Schiele, A.; Sommer, H.; Brezesinski, T.; Janek, J.; Berkes, B. B. Encyclopedia of Interfacial Chemistry; Elsevier: Oxford, 2018; pp. 44–53. Herl, T.; Matysik, F.-M. Recent Developments in Electrochemistry–Mass Spectrometry. ChemElectroChem 2020, 7, 2498–2512. Dreyer, S. L.; Kondrakov, A.; Janek, J.; Brezesinski, T. In Situ Analysis of Gas Evolution in Liquid and Solid-Electrolyte-Based Batteries With Current and Next-Generation Cathode Materials. J. Mater. Res. 2022, 37, 3146–3168. Bruckenstein, S.; Gadde, R. R. Use of a Porous Electrode for In Situ Mass Spectrometric Determination of Volatile Electrode Reaction Products. J. Am. Chem. Soc. 1971, 93, 793–794. Grambow, L.; Bruckenstein, S. Mass Spectrometric Investigation of the Electrochemical Behavior of Adsorbed Carbon Monoxide at Platinum in 0.2 M Sulphuric Acid. Electrochim. Acta 1977, 22, 377–383. Brockman, T. J.; Anderson, L. B. Permeable Membrane Mass Spectrometry of Products of Electrochemical Oxidation of Carboxylate Ions. Anal. Chem. 1984, 56, 207–213. Wolter, O.; Heitbaum, J. Differential Electrochemical Mass Spectroscopy (DEMS) – A New Method for the Study of Electrode Processes. Ber. Bunsenges. Phys. Chem. 1984, 88, 2–6. Wohlfahrt-Mehrens, M.; Heitbaum, J. Oxygen Evolution on Ru and RuO2 Electrodes Studied Using Isotope Labelling and On-Line Mass Spectrometry. J. Electroanal. Chem. Interfacial Electrochem. 1987, 237, 251–260. Jusys, Z.; Liaukonis, J.; Vaskelis, A. The Catalytic Oxidation of Hypophosphite on Nickel Studied by Electrochemical Mass Spectrometry. J. Electroanal. Chem. 1991, 307, 87–97. Ianniello, R.; Schmidt, V. M.; Stimming, U.; Stumper, J.; Wallau, A. CO Adsorption and Oxidation on Pt and Pt-Ru Alloys: Dependence of Substrate Composition. Electrochim. Acta 1994, 39, 1863–1869. Lima, F. B. H.; Pasqualeti, A. M.; Belen Molina Concha, M.; Chatenet, M.; Ticianelli, E. A. Borohydride Electrooxidation on Au and Pt Electrodes. Electrochim. Acta 2012, 84, 202–212. Krausa, M.; Vielstich, W. Study of the Electrocatalytic Influence of Pt/Ru and Ru on the Oxidation of Residues of Small Organic Molecules. J. Electroanal. Chem. 1994, 379, 307–314. Ianniello, R.; Schmidt, V. M. A Simplified DEMS Set Up for Electrocatalytic Studies of Porous PtRu Alloys. Ber. Bunsenges. Phys. Chem. 1995, 99, 83–86. Schmidt, V. M.; Ianniello, R.; Pastor, E.; Gonzalez, S. Electrochemical Reactivity of Ethanol on Porous Pt and PtRu: Oxidation/Reduction Reactions in 1 M HClO4. J. Phys. Chem. 1996, 100, 17901–17908. de Souza, J. P. I.; Queiroz, S. L.; Bergamaski, K.; Gonzalez, E. R.; Nart, F. C. Electro-Oxidation of Ethanol on Pt, Rh, and PtRh Electrodes. A Study Using DEMS and In-Situ FTIR Techniques. J. Phys. Chem. B 2002, 106, 9825–9830. Wasmus, S.; Vielstich, W. Methanol Oxidation at Carbon Supported Pt and PtRu Electrodes: An On-Line MS Study Using Technical Electrodes. J. Appl. Electrochem. 1993, 23, 120–124. Delpeuch, A. B.; Asset, T.; Chatenet, M.; Cremers, C. Electrooxidation of Ethanol at Room Temperature on Carbon-Supported Pt and Rh-Containing Catalysts: A DEMS Study. J. Electrochem. Soc. 2014, 161, F918–F924. Kreysa, G.; Breidenbach, G. Elektrochemische Massenspektrometrie. On-Line Analyse Gasformiger Reaktionsprodukte. Fresenius Z. Anal. Chem 1980, 301, 402–405. Wasmus, S.; Wang, J.-T.; Savinell, R. F. Real-Time Mass Spectrometric Investigation of the Methanol Oxidation in a Direct Methanol Fuel Cell. J. Electrochem. Soc. 1995, 142, 3825–3833. Wang, J.; Wasmus, S.; Savinell, R. F. Evaluation of Ethanol, 1-Propanol and 2-Propanol in a Direct Oxidation Polymer-Electrolyte Fuel Cell. J. Electrochem. Soc. 1995, 142, 4218–4224. Weber, M.; Wang, J.-T.; Wasmus, S.; Savinell, R. F. Formic Acid Oxidation in a Polymer Electrolyte Fuel Cell; A Real-Time Mass-Spectrometry Study. J. Electrochem. Soc. 1996, 143, L158–L160. Fujiwara, N.; Friedrich, K. A.; Stimming, U. Ethanol Oxidation on PtRu Electrodes Studied by Differential Electrochemical Mass Spectrometry. J. Electroanal. Chem. 1999, 472, 120–125. Seiler, T.; Savinova, E. R.; Friedrich, K. A.; Stimming, U. Poisoning of PtRu/C Catalysts in the Anode of a Direct Methanol Fuel Cell: A DEMS Study. Electrochim. Acta 2004, 49, 3927–3936. Rao, V.; Cremers, C.; Stimming, U.; Cao, L.; Sun, S.; Yan, S.; Sun, G.; Xin, Q. Electro-Oxidation of Ethanol at Gas Diffusion Electrodes: A DEMS Study. J. Electrochem. Soc. 2007, 154, B1138–B1147. Wang, Q.; Sun, G. Q.; Jiang, L. H.; Xin, Q.; Sun, S. G.; Jiang, Y. X.; Chen, S. P.; Jusys, Z.; Behm, R. J. Adsorption and Oxidation of Ethanol on Colloid-Based Pt/C, PtRu/C and Pt3Sn/C Catalysts: In Situ FTIR Spectroscopy and On-Line DEMS Studies. Phys. Chem. Chem. Phys. 2007, 9, 2686–2696. Cremers, C.; Jurzinsky, T.; Meier, J.; Schade, A.; Branghofer, M.; Pinkwart, K.; Tübke, J. DEMS and Online Mass Spectrometry Studies of the Carbon Support Corrosion Under Various Polymer Electrolyte Membrane Fuel Cell Operating Conditions. J. Electrochem. Soc. 2018, 165, F3307–F3315. Hoch, G.; Kok, B. A Mass Spectrometer Inlet System for Sampling Gases Dissolved in Liquid Phases. Arch. Biochem. Biophys. 1963, 101, 160–170. Burlacot, A.; Burlacot, F.; Li-Beisson, Y.; Peltier, G. Membrane Inlet Mass Spectrometry: A Powerful Tool for Algal Research. Front. Plant Sci. 2020, 11, 1302-1–1302-15. Johnson, R. C.; Cooks, R. G.; Allen, T. M.; Cisper, M. E.; Hemberger, P. H. Membrane Introduction Mass Spectrometry: Trends and Applications. Mass Spectrom. Rev. 2000, 19, 1–37. Krogh, E. T.; Gill, C. G. Membrane Introduction Mass Spectrometry (MIMS): A Versatile Tool for Direct, Real-Time Chemical Measurements. J. Mass Spectrom. 2014, 49, 1205–1213. Queiroz, A. C.; Souza, M. L.; Camilo, M. R.; Silva, W. O.; Cantane, D. A.; Messias, I.; Pinto, M. R.; Nagao, R.; Lima, F. H. B. Electrochemical Mass Spectrometry: Evolution of the Cell Setup for On-Line Investigation of Products and Screening of Electro-Catalysts for Carbon Dioxide Reduction. ChemElectroChem 2022, 9. e202101408-1–e202101408-15. Jusys, Z.; Behm, R. J. Methanol Oxidation on a Carbon Supported Pt Fuel Cell Catalyst – A Kinetic and Mechanistic Study by Differential Electrochemical Mass Spectrometry. J. Phys. Chem. B 2001, 105, 10874–10883. Lindström, R.; Seidel, Y. E.; Jusys, Z.; Gustavsson, M.; Kasemo, B.; Behm, R. J. Electrocatalysis and Transport Effects on Nanostructured Pt/GC Electrodes. J. Electroanal. Chem. 2010, 644, 90–102.

100

Methods and Instruments | Differential Electrochemical Mass Spectrometry

43. Seidel, Y. E.; Schneider, A.; Jusys, Z.; Wickman, B.; Kasemo, B.; Behm, R. J. Transport Effects in the Electrooxidation of Methanol Studied on Nanostructured Pt/Glassy Carbon Electrodes. Langmuir 2010, 26, 3569–3578. 44. Behm, R. J.; Seidel, Y. E.; Lindström, R. W.; Jusys, Z.; Wiedwald, U.; Ziemann, P.; Wickman, B.; Kasemo, B.; Zhang, D.; Deutschmann, O. Nanotechnology - Fundamentals and Application of Functional Nanostructures; Baden-Württemberg Stiftung: Karlsruhe, 2011; pp. 281–304. Research Series. No. 56. 45. Fuhrmann, J.; Zhao, H.; Langmach, H.; Seidel, Y. E.; Jusys, Z.; Behm, R. J. The Role of Reactive Reaction Intermediates in Two-Step Heterogeneous Electrocatalytic Reactions: A Model Study. Fuel Cells 2011, 11, 501–510. 46. Hartung, T.; Baltruschat, H. Differential Electrochemical Mass Spectrometry Using Smooth Electrodes: Adsorption and H/D Exchange Reactions of Benzene on Pt. Langmuir 1990, 6, 953–957. 47. Hartung, T.; Schmiemann, U.; Kamphausen, I.; Baltruschat, H. Electrodesorption from Single-Crystal Electrodes: Analysis by Differential Electrochemical Mass Spectrometry. Anal. Chem. 1991, 63, 44–48. 48. Baltruschat, H.; Schmiemann, U. The Adsorption of Unsaturated Organic Species at Single-Crystal Electrodes Studied by Differential Electrochemical Mass Spectrometry. Ber. Bunsenges. Phys. Chem. 1993, 97, 452–460. 49. Vaskelis, A.; Jusys, Z. Differential Electrochemical Mass Spectrometry Study of the Electroless Copper Plating Process Using a Thin-Layer Flow-Through Cell. Anal. Chim. Acta 1995, 305, 227–231. 50. Jusys, Z.; Massong, H.; Baltruschat, H. A New Approach for Simultaneous DEMS and EQCM: Electro-Oxidation of Adsorbed CO on Pt and Pt-Ru. J. Electrochem. Soc. 1999, 146, 1093–1098. 51. Vidal-Iglesias, F. J.; Solla-Gullón, J.; Feliu, J.; Baltruschat, H.; Aldaz, A. DEMS Study of Ammonia Oxidation on Platinum Basal Planes. J. Electroanal. Chem. 2005, 588, 331–338. 52. Seidel, Y. E.; Lindström, R.; Jusys, Z.; Cai, J.; Wiedwald, U.; Ziemann, P.; Behm, R. J. Nanostructured Pt/GC Model Electrodes Prepared by Deposition of Metal-Salt Loaded Micelles. Langmuir 2007, 23, 5795–5801. 53. Cuesta, A.; Escudero, M.; Lanova, B.; Baltruschat, H. Cyclic Voltammetry, FTIRS, and DEMS Study of the Electrooxidation of Carbon Monoxide, Formic Acid, and Methanol on Cyanide-Modified Pt(111) Electrodes. Langmuir 2009, 25, 6500–6507. 54. Kapalka, A.; Lanova, B.; Baltruschat, H.; Fóti, G.; Comninellis, C. A DEMS Study of Methanol and Formic Acid Oxidation on Boron-Doped Diamond Electrode. J. Electrochem. Soc. 2009, 156, E149–E153. 55. Abd-El-Latif, A. A.; Mostafa, E.; Huxter, S.; Attard, G.; Baltruschat, H. Electrooxidation of Ethanol at Polycrystalline and Platinum Stepped Single Crystals: A Study by Differential Electrochemical Mass Spectrometry. Electrochim. Acta 2010, 55, 7951–7960. 56. Zhang, D.; Deutschmann, O.; Seidel, Y. E.; Behm, R. J. Interaction of Mass Transport and Electro-Catalytic CO Oxidation in a Thin-Layer Flow Cell. J. Phys. Chem. C 2011, 115, 468–478. 57. Jusys, Z.; Kaiser, J.; Behm, R. J. Electrooxidation of CO and H2/CO Mixtures on a Carbon Supported Pt Catalyst - A Kinetic and Mechanistic Study by Differential Electrochemical Mass Spectrometry. Phys. Chem. Chem. Phys. 2001, 3, 4650–4660. 58. Jusys, Z.; Kaiser, J.; Behm, R. J. Simulated “Air Bleed” Oxidation of Adsorbed CO on Carbon Supported Pt - Part 1: A Differential Electrochemical Mass Spectrometry Study. J. Electroanal. Chem. 2003, 554–555, 427–437. 59. Jusys, Z.; Kaiser, J.; Behm, R. J. Methanol Electrooxidation Over Pt/C Fuel Cell Catalysts - Dependence of Product Yields on Catalyst Loading. Langmuir 2003, 19, 6759–6769. 60. Jusys, Z.; Behm, R. J. Simultaneous Oxygen Reduction and Methanol Oxidation on a Carbon -Supported Pt Catalyst and Mixed Potential Formation - Revisited. Electrochim. Acta 2004, 49, 3891–3900. 61. Wang, H.; Jusys, Z.; Behm, R. J. Ethanol Electrooxidation on a Carbon-Supported Pt Catalyst: Reaction Kinetics and Product Yields. J. Phys. Chem. B 2004, 108, 19413–19424. 62. Wang, H.; Jusys, Z.; Behm, R. J. Electrooxidation of Acetaldehyde on Carbon-Supported Pt, PtRu and Pt3Sn and Unsupported PtRu0.2 Catalysts: A Quantitative DEMS Study. J. Appl. Electrochem. 2006, 36, 1187–1198. 63. Wang, H.; Jusys, Z.; Behm, R. J. Ethanol Electrooxidation on Carbon-Supported Pt, PtRu and Pt3Sn Catalysts - A Quantitative DEMS Study. J. Power Sources 2006, 154, 351–359. 64. Wang, H.; Zhao, Z.; Jusys, Z.; Behm, R. J. Ethylene Glycol Electrooxidation on Carbon Supported Pt, PtRu and Pt3Sn Catalysts - A Comparative DEMS Study. J. Power Sources 2006, 155, 33–46. 65. Wang, H.; Jusys, Z.; Behm, R. J. Electrochemical Oxidation Kinetics and Mechanism of Ethylene Glycol on a Carbon Supported Pt Catalyst: A Quantitative DEMS Study. J. Electroanal. Chem. 2006, 595, 23–36. 66. Jusys, Z.; Behm, R. J. Fuel Cell Catalysis: A Surface Science Approach; Wiley& Sons: Chichester, 2009; pp. 411–464. Chapter 13. 67. Gebauer, C.; Hoffmann, D.; Jusys, Z.; Behm, R. J. Novel, Highly Conductive Pt/TiO2 Thin-Film Model Catalyst Electrodes: The Role of Metal–Support Interactions. ChemElectroChem 2016, 3, 1553–1563. 68. Colmenares, L. C.; Wurth, A.; Jusys, Z.; Behm, R. J. Model Study on the Stability of Carbon Support Materials Under Polymer Electrolyte Fuel Cell Cathode Operation Conditions. J. Power Sources 2009, 190, 14–24. 69. Gebauer, C.; Jusys, Z.; Behm, R. J. On the Role of the Support in Pt Anode Catalyst Degradation under Simulated H2 Fuel Starvation Conditions. J. Electrochem. Soc. 2018, 165, J3342–J3349. 70. Seidel, Y. E.; Müller, M.; Jusys, Z.; Wickman, B.; Hanarp, P.; Kasemo, B.; Hörmann, U.; Kaiser, U.; Behm, R. J. Nanostructured, Glassy Carbon Supported Pt/GC Electrodes: The Presence of Secondary Pt Nanostructures, and How to Avoid Them. J. Electrochem. Soc. 2008, 155, K171–K179. 71. Seidel, Y. E.; Jusys, Z.; Lindström, R.; Stenfeldt, M.; Kasemo, B.; Krischer, K. Oscillatory Behaviour in Galvanostatic Formaldehyde Oxidation on Nanostructured Pt/Glassy Carbon Model Electrodes. ChemPhysChem 2010, 11, 1405–1415. 72. Schwechten, H.; Heinen, M.; Seidel, Y. E.; Jusys, Z.; Wickman, B.; Kasemo, B.; Behm, R. J. Mass Transport Effects in CO Adsorption and Continuous CO Oxidation Over Regular Arrays of Pt Nanostructures on Planar Glassy Carbon Supports. J. Electroanal. Chem. 2011, 662, 157–168. 73. Eifert, L.; Jusys, Z.; Banerjee, R.; Behm, R. J.; Zeis, R. Differential Electrochemical Mass Spectrometry of Carbon Felt Electrodes for Vanadium Redox Flow Batteries. ACS Appl. Energy Mater. 2018, 1, 6714–6718. 74. Fuhrmann, J.; Zhao, H.; Holzbecher, E.; Langmach, H. Flow, Transport, and Reactions in a Thin Layer Flow Cell. J. Fuel Cell Sci. Technol. 2008, 5, 021008-021008-10. 75. Fuhrmann, J.; Linke, A.; Langmach, H.; Baltruschat, H. Numerical Calculation of the Limiting Current for a Cylindrical Thin-Layer Flow Cell. Electrochim. Acta 2009, 55, 430–438. 76. Tschupp, S. A.; Temmel, S. E.; Poyatos Salguero, E.; Herranz, J.; Schmidt, T. J. Numerical Partitioning Model for the Koutecky-Levich Analysis of Electrochemical Flow Cells With a Combined Channel/Wall-Jet Geometry. J. Electrochem. Soc. 2017, 164, E3448–E3456. 77. Jusys, Z.; Kaiser, J.; Behm, R. J. A Novel Thin-Layer Flow Cell Double-Disk Electrode Approach for Kinetic Studies on Supported Catalysts Under Controlled Mass Transport Conditions. Electrochim. Acta 2004, 49, 1297–1305. 78. Clark, E. L.; Singh, M. R.; Kwon, Y.; Bell, A. T. Differential Electrochemical Mass Spectrometer Cell Design for Online Quantification of Products Produced During Electrochemical Reduction of CO2. Anal. Chem. 2015, 87, 8013–8020. 79. Hiden Analytical. Technical Information TI-20022.2; 2022. Technical Information TI-20022.2. 80. Abd-El-Latif, A. E. A.; Xu, J.; Bogolowski, N.; Koenigshoven, P.; Baltruschat, H. New Cell for DEMS Applicable to Different Electrode Sizes. Electrocatalysis 2012, 3, 39–47. 81. Castro-Castillo, C.; Armijo, F.; Isaacs, M.; Pastor, E.; García, G. Flow Injection Analysis Coupled With Differential Electrochemical Mass Spectrometry for Hydrogen Detection and Quantification. Electrochem. Commun. 2020, 118, 106809-1–106809-4.

Methods and Instruments | Differential Electrochemical Mass Spectrometry

101

82. Gao, F.; Dai, C.; Chen, Z.; Huang, G.; Bai, C.; Tao, H.; Yin, B.; Yang, Q.; Zhao, Z. Newly Developed Low-Temperature Scanning Tunneling Microscope and Its Application to the Study of Superconducting Materials. J. Vac. Sci. Technol. B 1994, 12, 1708. 83. Wonders, A. H.; Housmans, T. H. M.; Rosca, V.; Koper, M. T. M. On-Line Mass Spectrometry Systems for Measurements at Single-Crystal Electrodes in Hanging Meniscus Configuration. J. Appl. Electrochem. 2006, 36, 1215–1221. 84. Housmans, T. H. M.; Wonders, A. H.; Koper, M. T. M. Structure Sensitivity of Methanol Electrooxidation Pathways on Platinum: An On-Line Electrochemical Mass Spectrometry Study. J. Phys. Chem. B 2006, 110, 10021–10031. 85. Rizo, R.; Lazaro, J.; Pastor, E.; Koper, M. T. M. Ethanol Oxidation on Sn-Modified Pt Single-Crystal Electrodes: New Mechanistic Insights From On-Line Electrochemical Mass Spectrometry. ChemElectroChem 2016, 3, 2196–2201. 86. Gao, Y.; Tsuji, H.; Hattori, H.; Kita, H. New On-Line Mass Spectrometer System Designed for Platinum- Single Crystal and Electroreduction of Acetylene. J. Electroanal. Chem. 1994, 372, 195–200. 87. Florez-Montano, J.; Garcia, G.; Guillen-Villafuerte, O.; Rodriguez, J. L.; Planes, G. A.; Pastor, E. Mechanism of Ethanol Electrooxidation on Mesoporous Pt Electrode in Acidic Medium Studied by a Novel Electrochemical Mass Spectrometry Set-Up. Electrochim. Acta 2016, 209, 121–131. 88. Todoroki, N.; Tei, H.; Tsurumaki, H.; Miyakawa, T.; Inoue, T.; Wadayama, T. Surface Atomic Arrangement Dependence of Electrochemical CO2 Reduction on Gold: Online Electrochemical Mass Spectrometric Study on Low-Index Au(Hkl) Surfaces. ACS Catal. 2019, 9, 1383–1388. 89. Ju, K.-S.; Pak, S.-N.; Kim, K.-I.; Ri, H.-H.; Ok, C.-S.; Choe, S. A.; Kim, S.-Y.; Han, C.-H.; Kim, D.-I.; Qu, Y.-T. On-Line Electrochemical Mass Spectrometry System to Study Mechanism of the Ethanol Oxidation Reaction. Am. J. Eng. Res. 2019, 8, 161–167. 90. Tegtmeyer, D.; Heindrichs, A.; Heitbaum, J. Electrochemical on Line Mass Spectrometry on a Rotating Electrode Inlet System. Ber. Bunsenges. Phys. Chem. 1989, 93, 201–206. 91. Wasmus, S.; Cattaneo, E.; Vielstich, W. Reduction of Carbon Dioxide to Methane and Ethene - An On-Line MS Study With Rotating Electrodes. Electrochim. Acta 1990, 35, 771–775. 92. Fujihira, M.; Noguchi, T. A Novel Differential Electrochemical Mass Spectrometer (DEMS) With a Stationary Gas-Permeable Electrode In a Rotational Flow Produced by a Rotating Rod. J. Electroanal. Chem. 1993, 347, 457–463. 93. Todoroki, N.; Tsurumaki, H.; Tei, H.; Mochizuki, T.; Wadayama, T. Online Electrochemical Mass Spectrometry Combined With the Rotating Disk Electrode Method for Direct Observations of Potential-Dependent Molecular Behaviors in the Electrode Surface Vicinity. J. Electrochem. Soc. 2020, 167, 106503-1–106503-5. 94. Treufeld, I.; Jebaraj, A. J. J.; Xu, J.; Martins de Godoi, D.; Scherson, D. Porous Teflon Ring-Solid Disk Electrode Arrangement for Differential Mass Spectrometry Measurements in the Presence of Convective Flow Generated by a Jet Impinging Electrode in the Wall-Jet Configuration. Anal. Chem. 2012, 84, 5175–5179. 95. Pozniak, B.; Treufeld, I.; Scherson, D. Hydroxylamine Oxidation on Polycrystalline Gold Electrodes in Aqueous Electrolytes: Quantitative On-Line Mass Spectrometry Under Forced Convection. ChemPhysChem 2019, 20, 3128–3133. 96. Venkatachalam, S.; Angelici, R. J.; Woo, L. K.; Hillier, A. C. High Rate Detection of Volatile Products Using Differential Electrochemical Mass Spectrometry: Combining an Electrode-Coated Membrane With Hydrodynamic Flow in a Wall-Tube Configuration. Anal. Chem. 2013, 85, 6059–6065. 97. Kubannek, F.; Krewer, U. A Cyclone Flow Cell for Quantitative Analysis of Kinetics at Porous Electrodes by Differential Electrochemical Mass Spectrometry. Electrochim. Acta 2016, 210, 862–873. 98. Jambunathan, K.; Shah, B. C.; Hudson, J. L.; Hillier, A. C. Measuring Electrocatalytic Activity on a Local Scale With Scanning Differential Electrochemical Mass Spectrometry (SDEMS) Scanning Electrochemical Microscopy of Hydrogen Electro-Oxidation. Rate Constant Measurements and Carbon Monoxide Poisoning on Platinum. J. Electrochem. Soc. 2001, 150, E312–E320. 99. Jambunathan, K.; Jayaraman, S.; Hillier, A. C. A Multielectrode Electrochemical and Scanning Differential Electrochemical Mass Spectrometry Study of Methanol Oxidation on Electrodeposited PtxRuy. Langmuir 2004, 20, 1856–1863. 100. Rus, E. D.; Wang, H.; Legard, A. E.; Rizert, N. L.; van Dover, R. B.; Abruña, H. D. An Exchangeable-Tip Scanning Probe Instrument for the Analysis of Combinatorial Libraries of Electrocatalysts. Rev. Sci. Instrum. 2013, 84, 024101-1–024101-8. 101. Grote, J. P.; Zeradjanin, A. R.; Cherevko, S.; Mayrhofer, K. J. J. Coupling of a Scanning Flow Cell With Online Electrochemical Mass Spectrometry for Screening of Reaction Selectivity. Rev. Sci. Instrum. 2014, 85, 104101-1–104101-5. 102. Lai, Y.; Jones, R. J. R.; Wang, Y.; Zhou, L.; Gregoire, J. M. Scanning Electrochemical Flow Cell With Online Mass Spectroscopy for Accelerated Screening of Carbon Dioxide Reduction Electrocatalysts. ACS Comb. Sci. 2019, 21, 692–704. 103. Roen, L. M.; Stuve, E. M. Design and Characterization of an On-Line Electrochemical Mass Spectrometry System for Measurement of Multi-Step Reaction Kinetics. ECS Trans. 2006, 3, 969–979. 104. Reichert, R.; Jusys, Z.; Behm, R. J. Au/TiO2 Photo(Electro)Catalysis: The Role of the Au Cocatalyst in Photoelectrochemical Water Splitting and Photocatalytic H2 Evolution. J. Phys. Chem. C 2015, 119, 24750–24759. 105. Skou, E.; Munk, J. A Simplified Set-Up for Electrochemical Mass Spectrometry and the Use of a Gold Substrate for the Investigation of Porous Electrode Materials. J. Electroanal. Chem. 1994, 367, 93–98. 106. Jusys, Z.; Behm, R. J. DEMS Analysis of Small Organic Molecule Electrooxidation: A High-Temperature High-Pressure DEMS Study. ECS Trans. 2008, 16, 1243–1251. 107. Sun, S.; Chojak-Halseid, M.; Heinen, M.; Jusys, Z.; Behm, R. J. Ethanol Electrooxidation on a Carbon-Supported Pt Catalyst at Elevated Temperature and Pressure: A HighTemperature/High-Pressure DEMS Study. J. Power Sources 2009, 190, 2–13. 108. Chojak-Halseid, M.; Jusys, Z.; Behm, R. J. Electrooxidation of Ethylene Glycol on a Carbon-Supported Pt Catalyst at Elevated Temperatures and Pressure: A High-Temperature/ High-Pressure DEMS Study. J. Electroanal. Chem. 2010, 664, 103–109. 109. Chojak-Halseid, M.; Jusys, Z.; Behm, R. J. Methanol Oxidation Over a Pt/C Catalyst at High Temperatures and Pressure: An On-Line Electrochemical Mass Spectrometry Study. J. Phys. Chem. C 2010, 114, 22573–22581. 110. Sun, S.; Heinen, M.; Jusys, Z.; Behm, R. J. Electrooxidation of Acetaldehyde on a Carbon Supported Pt Catalyst at Elevated Temperature/Pressure: An On-Line Differential Electrochemical Mass Spectrometry Study. J. Power Sources 2012, 204, 1–13. 111. Fuhrmann, J.; Zhao, H.; Holzbecher, E.; Langmach, H.; Chojak, M.; Halseid, R.; Jusys, Z.; Behm, R. J. Experimental and Numerical Model Study of the Limiting Current in a Channel Flow Cell With a Circular Electrode. Phys. Chem. Chem. Phys. 2008, 10, 3784–3795. 112. Niether, C.; Rau, M. S.; Cremers, C.; Jones, D. J.; Pinkwart, K.; Tübke, J. Development of a Novel Experimental DEMS Set-Up for Electrocatalyst Characterization Under Working Conditions of High Temperature Polymer Electrolyte Fuel Cells. J. Electroanal. Chem. 2015, 747, 97–103. 113. Bae, S. E.; Gewirth, A. A. In Situ EC-STM Studies of MPS, SPS, and Chloride on Cu(100): Structural Studies of Accelerators for Dual Damascene Electrodeposition. Langmuir 2006, 22, 10315–10321. 114. Bondue, C. J.; Koper, M. T. M. A DEMS Approach for the Direct Detection of CO Formed During Electrochemical CO2 Reduction. J. Electroanal. Chem. 2019, 875, 113842. 115. Iwasita, T.; Vielstich, W. On-Line Mass Spectroscopy of Volatile Products During Methanol Oxidation at Platinum in Acid Solutions. J. Electroanal. Chem. 1986, 201, 403–408. 116. Bayer, D.; Berenger, S.; Joos, M.; Cremers, C.; Tübke, J. Electrochemical Oxidation of C2 Alcohols at Platinum Electrodes in Acidic and Alkaline Environment. Int. J. Hydrog. Energy 2010, 35, 12660–12667. 117. Bayer, D.; Cremers, C.; Baltruschat, H.; Tübke, J. The Electro-Oxidation of Ethanol in Alkaline Medium at Different Catalyst Metals. ECS Trans. 2011, 41, 1669–1680. 118. Trimarco, D. B.; Pedersen, T.; Hansen, O.; Chorkendorff, I.; Vesborg, P. C. K. Fast and Sensitive Method for Detecting Volatile Species in Liquids. Rev. Sci. Instrum. 2015, 86, 075006. 119. Trimarco, D. B.; Scott, S. B.; Thilsted, A. H.; Pan, J. Y.; Pedersen, T.; Hansen, O.; Chorkendorff, I.; Vesborg, P. C. K. Enabling Real-Time Detection of Electrochemical Desorption Phenomena With Sub-Monolayer Sensitivity. Electrochim. Acta 2018, 268, 520–530.

102

Methods and Instruments | Differential Electrochemical Mass Spectrometry

120. Scott, S. B.; Engstfeld, A. K.; Jusys, Z.; Hochfilzer, D.; Knøsgaard, N.; Trimarco, D. B.; Vesborg, P. C. K.; Behm, R. J.; Chorkendorff, I. Anodic Molecular Hydrogen Formation on Ru and Cu Electrodes. Catal. Sci. Technol. 2020, 10, 6870–6877. 121. Tackett, B. M.; Raciti, D.; Hight Walker, A. R.; Moffat, T. P. Surface Hydride Formation on Cu(111) and Its Decomposition to Form H2 in Acid Electrolytes. J. Phys. Chem. Lett. 2021, 12, 10936–10941. 122. Raciti, D.; Moffat, T. P. Quantification of Hydride Coverage on Cu(111) by Electrochemical Mass Spectrometry. J. Phys. Chem. C 2022, 126, 18734–18743. 123. Spectro Inlets. DK-2200 København, Denmark. https://spectroinlets.com/contact/. 124. Wang, Z.; Zhang, Y.; Liu, B.; Wu, K.; Thevuthasan, S.; Baer, D. R.; Zhu, Z.; Yu, X.-Y.; Wang, F. In Situ Mass Spectrometric Monitoring of the Dynamic Electrochemical Process at the Electrode - Electrolyte Interface: A SIMS Approach. Anal. Chem. 2017, 89, 960–965. 125. Eggert, G.; Heitbaum, J. Electrochemical Reactions of Propylenecarbonate and Electrolytes Solved Therein - A DEMS Study. Electrochim. Acta 1986, 31, 1443–1448. 126. Imhof, R.; Novák, P. In Situ Investigation of the Electrochemical Reduction of Carbonate Electrolyte Solutions at Graphite Electrodes. J. Electrochem. Soc. 1998, 145, 1081–1087. 127. Imhof, R.; Novák, P. Oxidative Electrolyte Solvent Degradation in Lithium-Ion Batteries: An In Situ Differential Electrochemical Mass Spectrometry Investigation. J. Electrochem. Soc. 1999, 146, 1702–1706. 128. Wang, H.; Rus, E.; Sakuraba, T.; Kikuchi, J.; Kiya, Y.; Abruña, H. D. CO2 and O2 Evolution at High Voltage Cathode Materials of Li-Ion Batteries: A Differential Electrochemical Mass Spectrometry Study. Anal. Chem. 2014, 86, 6197–6201. 129. Bondue, C. J.; Abd-El-Latif, A. A.; Hegemann, P.; Baltruschat, H. Quantitative Study for Oxygen Reduction and Evolution in Aprotic Organic Electrolytes at Gas Diffusion Electrodes by DEMS. J. Electrochem. Soc. 2015, 162, A479–A487. 130. Khodayari, M.; Reinsberg, P.; Abd-El-Latif, A. A.; Merdon, C.; Fuhrmann, J.; Baltruschat, H. Determining Solubility and Diffusivity by Using a Flow Cell Coupled to a Mass Spectrometer. ChemPhysChem 2016, 17, 1647–1655. 131. Reinsberg, P.; Bondue, C.; Baltruschat, H. Mechanistic Investigation of the Oxygen Reduction in Magnesium Ion-Containing Dimethylsulfoxide. Electrochim. Acta 2016, 200, 214–221. 132. Reinsberg, P.; Bondue, C.; Baltruschat, H. Oxygen Reduction in Ca2+-Containing Electrolytes as a Promising Alternative to Sodium-Oxygen Batteries. J. Phys. Chem. C 2016, 120, 22179–22185. 133. Reinsberg, P. H.; Koellisch, A.; Baltruschat, H. On the Importance of Ion Pair Formation and the Effect of Water in Potassium-Oxygen Batteries. Electrochim. Acta 2019, 313, 223–234. 134. Hegemann, P.; Hegemann, M.; Zan, L.; Baltruschat, H. Stability of Tetraglyme for Reversible Magnesium Deposition From a Magnesium Aluminum Chloride Complex. J. Electrochem. Soc. 2019, 166, A245–A250. 135. Hegemann, M.; Bawol, P. P.; Köllisch-Mirbach, A.; Baltruschat, H. Mixed Lithium and Sodium Ion Aprotic DMSO Electrolytes for Oxygen Reduction on au and Pt Studied by DEMS and RRDE. Electrocatalysis 2021, 12, 564–578. 136. Bondue, C. J.; Königshoven, P.; Baltruschat, H. A New 2-Compartment Flow-Through Cell for the Simultaneous Detection of Electrochemical Reaction Products by a Detection Electrode and Mass Spectroscopy. Electrochim. Acta 2016, 214, 241–252. 137. Wang, H.; Rus, E.; Abruna, H. D. New Double-Band-Electrode Channel Flow Differential Electrochemical Mass Spectrometry Cell: Application for Detecting Product Formation During Methanol Electrooxidation. Anal. Chem. 2010, 82, 4319–4324. 138. Jusys, Z.; Binder, M.; Schnaidt, J.; Behm, R. J. A Novel DEMS Approach for Studying Gas Evolution at Battery-Type Electrode | Electrolyte Interfaces: High-Voltage LiNi0.5Mn1.5O4 Cathode in Ethylene- and Dimethyl-Carbonate Electrolytes. Electrochim. Acta 2019, 314, 188–201. 139. Kazzazi, A.; Bresser, D.; Kuenzel, M.; Hekmatfar, M.; Schnaidt, J.; Jusys, Z.; Diemant, T.; Behm, R. J.; Copley, M.; Maranski, K.; Cookson, J.; de Meatza, I.; Axmann, P.; Wohlfahrt-Mehrens, M.; Passerini, S. Synergistic Electrolyte Additives for Enhancing the Performance of High-Voltage Lithium-Ion Cathodes in Half-Cells and Full-Cells. J. Power Sources 2021, 482, 228975-228975-12. 140. Wu, F.; Kim, G. T.; Diemant, T.; Kuenzel, M.; Schür, A. R.; Gao, X.; Qin, B.; Alwast, D.; Jusys, Z.; Behm, R. J.; Geiger, D.; Kaiser, U.; Passerini, S. Reducing Capacity and Voltage Decay of Co-Free Li1.2Ni0.2Mn0.6O2 as Positive Electrode Material for Lithium Batteries Employing an Ionic Liquid-Based Electrolyte. Adv. Energy Mater. 2020, 10, 2001830. 141. Klein, F.; Pfeifer, C.; Bansmann, J.; Jusys, Z.; Behm, R. J.; Wohlfahrt-Mehrens, M.; Lindén, M.; Axmann, P. Effect of Three-in-One Surface Modification of Spherical, Co-Free Li-Rich Cathode Material for Li-Ion Batteries (Li1.2Mn0.6Ni0.2O2) with Citric Acid. J. Electrochem. Soc. 2022, 169, 120533. 142. Klein, F.; Bansmann, J.; Jusys, Z.; Pfeifer, C.; Scheitenberger, P.; Mundszinger, M.; Geiger, D.; Biskupek, J.; Kaiser, U.; Behm, R. J.; Lindén, M.; Wohlfahrt-Mehrens, M.; Axmann, P. Enhanced Electrochemical Capacity of Spherical Co-Free Li1.2Mn0.6Ni0.2O2 Particles after a Water and Acid Treatment and Its Influence on the Initial Gas Evolution Behavior. ChemSusChem 2022, 15, e202201061. 143. Qin, B.; Schiele, A.; Jusys, Z.; Mariani, A.; Diemant, T.; Liu, X.; Brezesinski, T.; Behm, R. J.; Varzi, A.; Passerini, S. Highly Reversible Sodiation of Tin in Glyme Electrolytes: The Critical Role of the Solid Electrolyte Interphase and Its Formation Mechanism. ACS Appl. Mater. Interfaces 2020, 12, 3697–3708. 144. Han, J.; Mariani, A.; Zarrabeitia, M.; Jusys, Z.; Behm, R. J.; Varzi, A.; Passerini, S. Zinc-Ion Hybrid Supercapacitors Employing Acetate-Based Water-in-Salt Electrolytes. Small 2022, 18, 2201563. 145. Novák, P.; Goers, D.; Hardwick, L.; Holzapfel, M.; Scheifele, W.; Ufheil, J.; Würsig, A. Advanced In Situ Characterization Methods Applied to Carbonaceous Materials. J. Power Sources 2005, 146, 15–20. 146. Wuersig, A.; Scheifele, W.; Novák, P. CO2 Gas Evolution on Cathode Materials for Lithium-Ion Batteries. J. Electrochem. Soc. 2007, 154, A449–A454. 147. McCloskey, B. D.; Bethune, D. S.; Shelby, R. M.; Girishkumar, G.; Luntz, A. C. Solvents’ Critical Role in Nonaqueous Lithium-Oxygen Battery Electrochemistry. J. Phys. Chem. Lett. 2011, 2, 1161–1166. 148. Castel, E.; Berg, E. J.; El Kazzi, M.; Novák, P.; Villevieille, C. Differential Electrochemical Mass Spectrometry Study of the Interface of x Li2MnO3(1-x)LiMO2 (M¼ Ni, Co, and Mn) Material As a Positive Electrode in Li-Ion Batteries. Chem. Mater. 2014, 26, 5051–5057. 149. Berkes, B. B.; Jozwiuk, A.; Vracar, M.; Sommer, H.; Brezesinski, T.; Janek, J. Online Continuous Flow Differential Electrochemical Mass Spectrometry With a Realistic Battery Setup for High-Precision, Long-Term Cycling Tests. Anal. Chem. 2015, 87, 5878–5883. 150. Lundström, R.; Berg, E. J. Design and Validation of an Online Partial and Total Pressure Measurement System for Li-Ion Cells. J. Power Sources 2021, 485, 2293471–229347-8. 151. Tsiouvaras, N.; Meini, S.; Buchberger, I.; Gasteiger, H. A. A Novel On-Line Mass Spectrometer Design for the Study of Multiple Charging Cycles of a Li-O2 Battery. J. Electrochem. Soc. 2013, 160, A471–A477. 152. Hahn, H.; Wagner, R.; Schappacher, F.; Winter, M.; Nowak, S. In Operando X-Shaped Cell Online Electrochemical Mass Spectrometry (OEMS): New Online Analysis Enables Insight into Lab Scale Lithium Ion Batteries During Operation. J. Electroanal. Chem. 2016, 772, 52–57. 153. Geng, L.; Wood, D. L.; Lewis, S. A.; Connatser, R. M.; Li, M.; Jafta, C. J.; Belharouak, I. High Accuracy In-Situ Direct Gas Analysis of Li-Ion Batteries. J. Power Sources 2020, 466, 228211. 154. Michalak, B.; Berkes, B. B.; Sommer, H.; Brezesinski, T.; Janek, J. Electrochemical Cross-Talk Leading to Gas Evolution and Capacity Fade in LiNi0.5Mn1.5O4/Graphite Full-Cells. J. Phys. Chem. C 2017, 121, 211–216. 155. Metzger, M.; Marino, C.; Sicklinger, J.; Haering, D.; Gasteiger, H. A. Anodic Oxidation of Conductive Carbon and Ethylene Carbonate in High-Voltage Li-Ion Batteries Quantified by On-Line Electrochemical Mass Spectrometry. J. Electrochem. Soc. 2015, 162, A1123–A1134. 156. EL-Cell GmbH, Tempowerkring 8, D-21079 Hamburg, https://el-cell.com/products/test-cells/gas-analysis-test-cells/ecc-dems/. 2022. 157. Sim, R.; Langdon, J.; Manthiram, A. Design of an Online Electrochemical Mass Spectrometry System to Study Gas Evolution From Cells With Lean and Volatile Electrolytes. Small Methods 2023, 7, 2201438.

Methods and Instruments | Differential Electrochemical Mass Spectrometry

103

158. Mattinen, U.; Klett, M.; Lindbergh, G.; Lindström, R. W. Gas Evolution in Commercial Li-Ion Battery Cells Measured by on-Line Mass Spectrometry - Effects of C-Rate and Cell Voltage. J. Power Sources 2020, 477, 228968-1–228968-12. 159. Bresser, D.; Paillard, E.; Passerini, S. Ionic Liquid-Based Electrolytes for Li Metal/Air Batteries: A Review of Materials and the New ’LABOHR’ Flow Cell Concept. J. Electrochem. Sci. Technol. 2014, 5, 37–44. 160. Wasserscheid, P. Volatile Times for Ionic Liquids. Nature 2006, 439, 797. 161. Pereira, M. M.; Kurnia, K. A.; Sousa, F. L.; Silva, N. J. O.; Lopes-da-Silva, J. A.; Coutinho, J. A. P.; Freire, M. G. Contact Angles and Wettability of Ionic Liquids on Polar and Non-Polar Surfaces. Phys. Chem. Chem. Phys. 2015, 17, 31653–31661. 162. Schnaidt, J.; Nguyen, T. N.; Jusys, Z.; Behm, R. J. How Many Electrons Are Transferred During the Electrochemical O2 Reduction in a Mg2+-Free/Mg2+-Containing Ionic Liquid? Electrochim. Acta 2019, 299, 372–377. 163. Alwast, D.; Schnaidt, J.; Jusys, Z.; Behm, R. J. Ionic Liquid Electrolytes for Metal-Air Batteries: Interactions Between O2, Zn2+ and H2O Impurities. J. Electrochem. Soc. 2020, 167, 070505. 164. Eckardt, M.; Alwast, D.; Schnaidt, J.; Behm, R. J. Influence of Additives on the Reversible Oxygen Reduction Reaction/Oxygen Evolution Reaction in the Mg2+-Containing Ionic Liquid N-Butyl-N-Methylpyrrolidinium Bis(Trifluoromethanesulfonyl)Imide. ChemSusChem 2020, 13, 3919–3927. 165. Alwast, D.; Schnaidt, J.; Law, Y. T.; Behm, R. J. A Novel Approach for Differential Electrochemical Mass Spectrometry Studies on the Decomposition of Ionic Liquids. Electrochim. Acta 2016, 197, 290–299. 166. Alwast, D.; Schnaidt, J.; Hancock, K.; Yetis, G.; Behm, R. J. Effect of Li+ and Mg2+ on the Electrochemical Decomposition of the Ionic Liquid 1-Butyl-1-Methylpyrrolidinium Bis(Trifluoromethanesulfonyl)Imide and Related Electrolytes. ChemElectroChem 2019, 6, 1–12. 167. Nie, K.; Wang, X.; Qiu, J.; Wang, Y.; Yang, Q.; Xu, J.; Yu, X.; Li, H.; Huang, X.; Chen, L. Increasing Poly(Ethylene Oxide) Stability to 4.5 V by Surface Coating of the Cathode. ACS Energy Lett. 2020, 5, 826–832. 168. Falgairette, C.; Xia, C.; Li, Y.; Harbich, W.; Comninellis, C. Solid Electrochemical Mass Spectrometry (SEMS) for Investigation of Supported Metal Catalysts Under High Vacuum. J. Appl. Electrochem. 2010, 40, 1893–1900. 169. Bartsch, T.; Strauss, F.; Hatsukade, T.; Schiele, A.; Kim, A. Y.; Hartmann, P.; Janek, J.; Brezesinski, T. Gas Evolution in All-Solid-State Battery Cells. ACS Energy Lett. 2018, 3, 2539–2543. 170. Javanshad, R.; Venter, A. R. Ambient Ionization Mass Spectrometry: Real-Time, Proximal Sample Processing and Ionization. Anal. Methods 2017, 9, 4896–4907. 171. Perkins, M. J.; Langford, V. S. Application of Routine Analysis Procedures to a Direct Mass Spectrometry Technique: Selected Ion Flow Tube Mass Spectrometry (SIFT-MS). Rev. Sep. Sci. 2021, 3, e21003-1–e21003-17. 172. Hambitzer, G.; Heitbaum, J. Electrochemical Thermospray Mass Spectrometry. Anal. Chem. 1986, 58, 1067–1070. 173. Regino, M. C.; Brajter-Toth, A. An Electrochemical Cell for On-Line Electrochemistry/Mass Spectrometry. Anal. Chem. 1997, 69, 5067–5072. 174. Xu, X.; Lu, W.; Cole, R. B. On-Line Probe for Fast Electrochemistry/Electrospray Mass Spectrometry. Investigation of Polycyclic Aromatic Hydrocarbons. Anal. Chem. 1996, 68, 4244–4253. 175. Zhou, F.; Van Berkel, G. J. Electrochemistry Combined On-Line With Electrospray Mass Spectrometry. Anal. Chem. 1995, 67, 3643–3649. 176. Nowak, S.; Winter, M. Review - Chemical Analysis for a Better Understanding of Aging and Degradation Mechanisms of Non-Aqueous Electrolytes for Lithium Ion Batteries: Method Development, Application and Lessons Learned. J. Electrochem. Soc. 2015, 162, A2500–A2508. 177. Lobaccaro, P.; Mandal, L.; Motapothula, M. R.; Sherburne, M.; Martin, J.; Venkatesan, T.; Ager, J. W. Initial Application of Selected-Ion Flow-Tube Mass Spectrometry to Real-Time Product Detection in Electrochemical CO2 Reduction. Energy Technol. 2018, 6, 110–121. 178. Liu, P.; Zheng, Q.; Dewald, H. D.; Zhou, R.; Chen, H. The Study of Electrochemistry With Ambient Mass Spectrometry. Trends Anal. Chem. 2015, 70, 20–30. 179. Brown, T. A.; Chen, H.; Zare, R. N. Identification of Fleeting Electrochemical Reaction Intermediates Using Desorption Electrospray Ionization Mass Spectrometry. J. Am. Chem. Soc. 2015, 137, 7274–7277. 180. Modestov, A. D.; Srebnik, S.; Lev, O.; Gun, J. Scanning Capillary Microscopy/Mass Spectrometry for Mapping Spatial Electrochemical Activity of Electrodes. Anal. Chem. 2001, 73, 4229–4240. 181. Momotenko, D.; Qiao, L.; Corte’s-Salazar, F.; Lesch, A.; Wittstock, G.; Girault, H. H. Electrochemical Push-Pull Scanner With Mass Spectrometry Detection. Anal. Chem. 2012, 84, 6630–6637. 182. Stadler, R.; Jusys, Z.; Baltruschat, H. Hydrogen Evolution During the Oxidation of Formaldehyde on Au: The Influence of Single Crystal Structure and Tl-Upd. Electrochim. Acta 2002, 47, 4485–4500. 183. Heinen, M.; Chen, Y.-X.; Jusys, Z.; Behm, R. J. CO Adsorption Kinetics and Adlayer Build-Up Studied by Combined ATR-FTIR Spectroscopy and on-Line DEMS Under Continuous Flow Conditions. Electrochim. Acta 2007, 53, 1279–1289. 184. Heinen, M.; Chen, Y. X.; Jusys, Z.; Behm, R. J. In Situ ATR-FTIRS Coupled With on-Line DEMS Under Controlled Mass Transport Conditions - A Novel Tool for Electrocatalytic Reaction Studies. Electrochim. Acta 2007, 52, 5634–5643. 185. Heinen, M.; Jusys, Z.; Behm, R. J. Handbook of Fuel Cells; John Wiley and Sons: Chichester, 2009; pp. 183–214. Chapter 12. 186. Schnaidt, J.; Heinen, M.; Jusys, Z.; Behm, R. J. Electroxidation of Glycerol Studied by Combined In Situ IR Spectrioscopy and Online Mass Spectrometry Under Continuous Flow Conditions. J. Electroanal. Chem. 2011, 661, 250–264. 187. Schnaidt, J.; Heinen, M.; Jusys, Z.; Behm, R. J. Electro-Oxidation of Ethylene Glycol on a Pt-Film Electrode Studied by Combined In Situ Infrared Spectroscopy and Online Mass Spectrometry. J. Phys. Chem. C 2012, 116, 2872–2883. 188. Schnaidt, J.; Heinen, M.; Jusys, Z.; Behm, R. J. Electrooxidation of 1-Propanol on Pt - Mechanistic Insights From a Spectro-Electrochemical Study Using Isotope Labeling. J. Phys. Chem. C 2012, 116, 25852–25867. 189. Schnaidt, J.; Heinen, M.; Jusys, Z.; Behm, R. J. Mechanistic Aspects of the Electro-Oxidation of Ethylene Glycol on a Pt-Film Electrode: A Combined In Situ IR Spectroscopy and Online Mass Spectrometry Study of Kinetic Isotope Effects. Catal. Today 2013, 202, 154–162. 190. Schnaidt, J.; Heinen, M.; Jusys, Z.; Behm, R. J. Oxidation of the Partly Oxidized Ethylene Glycol Oxidation Products Glycolaldehyde, Glyoxal, Glycolic Acid, Glyoxylic Acid and Oxalic Acid on Pt Electrodes - A Combined ATR-FTIRS and DEMS Spectro-Electrochemical Study. J. Phys. Chem. C 2013, 117, 12689–12701. 191. Schnaidt, J.; Heinen, M.; Jusys, Z.; Behm, R. J. Oxidation of 1-Propanol on a Pt Film Electrode Studied by Combined Electrochemical, In Situ IR Spectroscopy and Online Mass Spectrometry Measurements. Electrochim. Acta 2013, 104, 505–517. 192. Brimaud, S.; Jusys, Z.; Behm, R. J. Shape-Selected Nanocrystals for In Situ Spectro-Electrochemistry Studies on Structurally Well Defined Surfaces Under Controlled Electrolyte Transport: A Combined In Situ ATR-FTIR/Online DEMS Investigation of CO Electrooxidation on Pt. Beilstein J. Nanotechnol. 2014, 5, 735–746. 193. Tao, Q.; Zheng, Y. L.; Jiang, D. C.; Chen, Y. X.; Jusys, Z.; Behm, R. J. Interaction of C1 Molecules With a Pt Electrode at Open Circuit Potential: A Combined Infrared and Mass Spectroscopic Study. J. Phys. Chem. C 2014, 118, 6799–6808. 194. Reichert, R.; Schnaidt, J.; Jusys, Z.; Behm, R. J. The Influence of Reactive Side Products on the Electrooxidation of Methanol – A Combined In Situ Infrared Spectroscopy and Online Mass Spectrometry Study. Phys. Chem. Chem. Phys. 2014, 16, 13780. 195. Buso-Rogero, C.; Brimaud, S.; Solla-Gullon, J.; Vidal-Iglesias, F. J.; Herrero, E.; Behm, R. J.; Feliu, J. M. Ethanol Oxidation on Shape-Controlled Platinum Nanoparticles at Different PHs: A Combined In Situ IR Spectroscopy and Online Mass Spectrometry Study. J. Electroanal. Chem. 2016, 763, 116–124. 196. Jusys, Z.; Behm, R. J. Electrooxidation of Formic Acid on a Polycrystalline Au Film Electrode-A Comparison With Mass Transport Limited Bulk CO Oxidation and Kinetically Limited Oxalic Acid Oxidation. J. Electroanal. Chem. 2017, 800, 60–76. 197. Zhao, W.; Jusys, Z.; Behm, R. J. Quantitative Online Detection of Volatile and Non-Volatile Methanol Electrooxidation Products by Combined Electron Impact Mass Spectrometryand Electrospray Ionization Mass Spectrometry. ECS Trans. 2011, 35, 9–19.

104

Methods and Instruments | Differential Electrochemical Mass Spectrometry

198. Zhao, W.; Jusys, Z.; Behm, R. J. Complete Quantitative Online Detection of Methanol Electrooxidation Products Via Electron Impact Mass Spectrometry and Electrospray Ionization Mass Spectrometry. Anal. Chem. 2012, 84, 5479–5483. 199. Engstfeld, A. K.; Brimaud, S.; Behm, R. J. Potential-Induced Surface Restructuring-The Need for Structural Characterization in Electrocatalysis Research. Angew. Chem. Int. Ed. 2014, 53, 12936–12940. 200. Schnaidt, J.; Beckord, S.; Engstfeld, A. K.; Klein, J.; Brimaud, S.; Behm, R. J. A Combined UHV-STM-Flow Cell Set-Up for Electrochemical/Electrocatalytic Studies of Structurally Well-Defined UHV Prepared Model Electrodes. Phys. Chem. Chem. Phys. 2017, 19, 4166–4178. 201. Klein, J.; Argast, F.; Engstfeld, A. K.; Brimaud, S.; Behm, R. J. Electro-Oxidation of Methanol on Ru-Core Pt-Shell Type Model Electrodes. Electrochim. Acta 2019, 311, 244–254. 202. Engstfeld, A. K.; Weizenegger, S.; Pithan, L.; Beyer, P.; Jusys, Z.; Bansmann, J.; Behm, R. J.; Drnec, J. Ru(0001) Surface Electrochemistry in the Presence of Specifically Adsorbing Anions. Electrochim. Acta 2021, 389. 138350-138350(9). 203. Beckord, S.; Brimaud, S.; Behm, R. J. Stability and ORR Performance of a Well-Defined Bimetallic Ag70Pt30/Pt(111) Monolayer Surface Alloy Electrode - Probing the De-Alloying at an Atomic Scale. Electrochim. Acta 2018, 259, 762–771. 204. Bogdanoff, P.; Alonso-Vante, N. On-Line Determination via Differential Electrochemical Mass Spectroscopy (DEMS) of Chemical Products Formed in Photoelectrocatalytical Systems. Ber. Bunsenges. Phys. Chem. 1993, 97, 940–943. 205. Reichert, R.; Jusys, Z.; Behm, R. J. A Novel Photoelectrochemical Flow Cell With Online Mass Spectrometric Detection: Oxidation of Formic Acid on a Nanocrystalline TiO2 Electrode. Phys. Chem. Chem. Phys. 2014, 16, 25076–25080. 206. Reichert, R.; Zambrzycki, C.; Jusys, Z.; Behm, R. J. Photo-Electrochemical Oxidation of Organic C1 Molecules Over WO3 Films in Aqueous Electrolyte: Competition Between Water Oxidation and C1 Oxidation. ChemSusChem 2015, 8, 3677–3687. 207. Zhang, G.; Kucernak, A. Gas Accessible Membrane Electrode (GAME): A Versatile Platform for Elucidating Electrocatalytic Processes Using Real-Time and in Situ Hyphenated Electrochemical Techniques. ACS Catal. 2020, 10, 9684–9693. 208. Zhang, G.; Cui, Y.; Kucernak, A. Real-Time In Situ Monitoring of CO2 Electroreduction in the Liquid and Gas Phases by Coupled Mass Spectrometry and Localized Electrochemistry. ACS Catal. 2022, 12, 6180–6190. 209. Alden, L. R.; Roychowdhury, C.; Matsumoto, F.; Han, D. K.; Zeldovich, V. B.; Abruña, H. D.; DiSalvo, F. J. Synthesis, Characterization, and Electrocatalytic Activity of PtPb Nanoparticles Prepared by Two Synthesis Approaches. Langmuir 2006, 22, 10465–10471. 210. Abd-El-Latif, A. A.; Baltruschat, H. Formation of Methylformate During Methanol Oxidation Revisited: The Mechanism. J. Electroanal. Chem. 2011, 662, 204–212. 211. Zhao, W.; Jusys, Z.; Behm, R. J. Quantitative On-Line Analysis of Liquid Phase Products of Methanol Oxidation in Aqueous Sulfuric Acid Solutions Using Electrospray Ionization Mass Spectrometry. Anal. Chem. 2010, 82, 2472–2479. 212. Khanipour, P.; Löffler, M.; Reichert, A. M.; Haase, F. T.; Mayrhofer, K. J. J.; Katsounaros, I. Electrochemical Real-Time Mass Spectrometry (EC-RTMS): Monitoring Electrochemical Reaction Products in Real Time. Angew. Chem. Int. Ed. 2019, 58, 7273–7277. 213. Khanipour, P.; Haschke, S.; Bachmann, J.; Mayrhofer, K. J. J.; Katsounaros, I. Electrooxidation of Saturated C1-C3 Primary Alcohols on Platinum: Potential-Resolved Product Analysis With Electrochemical Real-Time Mass Spectrometry (EC-RTMS). Electrochim. Acta 2019, 315, 67–74. 214. Frensemeier, L. M.; Mayr, J.; Koellensperger, G.; Keppler, B. K.; Kowol, C. R.; Karst, U. Structure Elucidation and Quantification of the Reduction Products of Anticancer Pt(IV) Products by Electrochemistry/Mass Spectrometry (EC-MS). Analyst 2018, 143, 1997–2001.

Methods and Instruments | Scanning Electrochemical Microscopy Gunther Wittstocka, Marius Muhlea, and Monika Wilamowska-Zawłockab, aCarl von Ossietzky University of Oldenburg, School of Mathematics and Science, Institute of Chemistry, Oldenburg, Germany; bGdansk University of Technology, Faculty of Chemistry, Department of Energy Conversion and Storage, Gdansk, Poland © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 Introduction 2 Principles 2.1 Feedback mode 2.2 Substrate-generation/tip-collection (SG/TC) mode 2.3 Tip-generation/substrate-collection (TG/SC) mode 2.4 Redox competition mode 2.5 Related techniques 3 Applications 3.1 Spatiotemporal evolution of solid electrolyte interphases 3.2 Ion-transfer reactions at battery materials 3.3 Cathode electrolyte interphase 3.4 Detection of side reaction products in Li+-ion batteries 3.5 Gas diffusion electrodes in fuel cells 3.6 Gas-diffusion electrodes in organic Li-air batteries 3.7 Study of adsorbed intermediates at electrocatalysts 3.8 Screening of new catalyst materials for fuel cells and electrolyzers 3.9 Other applications 3.10 Outlook 4 Summary Acknowledgment References Further reading

106 107 107 108 109 109 109 109 109 113 113 113 113 114 114 116 116 116 116 117 117 118

Abstract Scanning electrochemical microscopy is based on the recording of electrolysis currents (Faradaic currents) at a microelectrode (ME) probe that is scanned over the sample. Different working modes are available to couple the electrolysis at the ME to reactions at the sample. The article explains their principles and provides examples of their application. The feedback mode, the sample-generation/tip collection mode, the redox-competition mode and the surface interrogation mode are most frequently applied to the characterization of interphases and interfaces occurring in electrochemical power sources.

Glossary

Microelectrode A “microelectrode is [an] electrode whose characteristic dimension is, under the given experimental conditions, . . . smaller than the diffusion layer thickness . . . Under these conditions, a steady state . . . is attained.”34

Key points

• • • • • • •

Scanning probe technique based on recording a Faradaic current at a microelectrode probe Applicable to conducting, semiconducting and insulating sample materials Reaction at the probe can be coupled to local reaction at the sample by different modes Quantitative nature of the signal facilitates local kinetic studies for micrometer-sized regions of the sample Feedback mode: Imaging of protecting layers at battery electrodes Generation-collection mode: Sensitive detection of reaction products and side products of electrode reactions Surface interrogation mode: Determination of reactive intermediates at electrocatalyst surfaces

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00208-5

105

106

Methods and Instruments | Scanning Electrochemical Microscopy

Abbreviations AFM Aux CEI DSSC ET FB HER IT LIB ME OCP ORR RC Ref RFB SECCM SECM SEI SFM SG/TC SoC SoH SPECM STM TG/SG TSV WE

1

Atomic force microscopy Auxiliary electrode Cathode electrolyte interphase Dye-sensitized solar cells Electron transfer Feedback (mode) Hydrogen evolution reaction Ion transfer Li+-ion battery Microelectrode Open circuit potential Oxygen reduction reaction Redox-competition (mode) Reference electrode Redox flow battery Scanning electrochemical cell microscopy Scanning electrochemical microscopy Solid electrolyte interphase Scanning force microscopy Substrate-generation/tip collection (mode) State of charge (of a battery) State of health (of a battery) Scanning photoelectrochemical microscopy Scanning tunneling microscopy Tip generation/substrate collection (mode) Tip-sample voltammetry Working electrode

Introduction

Scanning electrochemical microscopy (SECM) is a scanning probe technique that uses the Faradaic current (electrolysis current) at a microelectrode (ME) to generate a reactivity image of the sample (also called “substrate” in SECM literature). There are different ways in which the electrolysis reaction at the ME can be coupled to reactions at the sample. The most relevant modes for the study of interfaces in energy conversion devices will be discussed below and illustrated with examples. The current at the ME can often be precisely calculated by numerically solving the reaction-transport problem using continuum simulations. As a result, equations are available to relate the current measured at the ME to local rate constants or fluxes at the sample. SECM instrumentation is commercially available from different suppliers; in house-made instrumentation is widely used, too. A typical instrument (Fig. 1a) consists of a micropositioning system with three axes and a holder to mount the ME. The horizontal axes (parallel to the sample surface) are commonly denoted with x and y, while the axis perpendicular to the sample is denoted as z.

Fig. 1 (a) Schematic representation of a basic SECM instrument. The measurement cell is shown as cross section. (b) Enlarged cross-sectional view of the microelectrode with indications of the most important quantities to describe SECM experiments. Symbols are depicted in panel (b), iT,1 is given by Eq. (1), D is the diffusion coefficient of the mediator.

Methods and Instruments | Scanning Electrochemical Microscopy

107

The sample is mounted at the bottom of the cell container. The ME as well as auxiliary electrode (Aux) and reference electrode (Ref ) are immersed into the same electrolyte reservoir. The experiment can be operated in a three-electrode setup with the ME as working electrode (WE) together with Aux and Ref connected to a monopotentiostat. The current loop is closed between the ME and Aux facilitating the work on conductive, semiconducting and insulating materials. Alternatively, the sample can be a second working electrode (WE2) connected together with the ME as WE1, Aux and Ref to a bipotentiostat. When discussing the different working modes and applications, quantitative relations are used to relate the electrolysis current at the ME to the local reactivity of the sample. As in any scanning probe technique, the signal also depends on the working distance d and further geometric quantities. They are indicated in Fig. 1b. It is common practice to express the quantitative relations in normalized, dimensionless quantities (Fig. 1b). For instance, the thickness of the insulating sheath of the ME is typically stated as dimensionless ratio RG ¼ rglass [mm] /rT [mm]. The index “T” for quantities of the ME quantities has evolved historically from analogy to scanning tunneling microscopy (STM). However, it is usually beneficial to use a microdisk electrode in SECM rather than a conically shaped probe.1 The index “S” is used for quantities referring to the sample (or substrate). Due to the quantitative nature, SECM instruments are often used as tools for kinetic studies, in which no microscopic image is recorded but the current is evaluated as a function of the working distance at fixed (x, y) coordinates. The possibility to switch between recording images and conducting detailed kinetic experiments at selected locations represent one of the great advantages of SECM. Recommendation of other reviews and an authoritative book on SECM are listed for further study.

2 2.1

Principles Feedback mode

The feedback (FB) mode is perhaps the most important and most widely used among the different working modes because it allows a precise vertical positioning and is therefore part of almost every complex SECM experiment if only to position vertically the ME. The working solution is supplemented with a redox mediator in millimolar concentration c . The mediator is one redox form of a quasi-reversible redox couple. In the following text we use the symbols R and Ox to refer in a generic way to the redox forms of the mediator. Furthermore, we assume that the reduced form R is added to the solution. A potential ET is applied to the ME causing a diffusion-controlled electron transfer (ET) reaction R ! Ox + n e− of the mediator (Fig. 2a). In many cases, an equivalent experiment is possible by adding the oxidized form Ox and conducting a reduction at the ME. The quantitative relation discussed below also hold for this case. In the bulk phase of the solution, the electrolysis current at the ME iT decays within about 1 s to a steady-state current iT,1 given by Eq. (1). iT,1 ¼ gnFDrT c ∗

(1)

The index “1” refers to the quasi-infinite distance of the ME required to achieve iT,1. The factor g in Eq. (1) is a geometry-dependent parameter amounting to exactly 4 if the microdisk electrode is embedded in an infinite insulating sheath (RG ! 1). For electrodes up to RG  10, g  4 represents an sufficiently accurate approximation for most purposes. The value of g rises for decreasing RG reaching g ¼ 4.072 for RG ¼ 10, g ¼ 4.156 for RG ¼ 5 and g ¼ 4.952 for RG ¼ 1.2. Precise values and interpolations as well as

Fig. 2 (a) SECM approach curves (normalized ME current vs. normalized distance) in the feedback mode. The plot shows normalized currents as a function of normalized distances for the ME with RG ¼ 5 and for the normalized dimensionless first order rate constant k for the mediator regeneration of (1) 0; (2) 0.1; (3) 0.5; (4) 1.0; (5) 1.5; (6) 2; (7) 5; (8) 1. The insets show cross-sectional view of the ME (a1) in the bulk solution (d ! 1); (a2) close to an insulating sample (curve 1 for k ¼ 0 at L ¼ 1 or d ¼ rT); (a3) close to a conductive sample at which the mediator regenerated under diffusion-controlled conditions (curve 8 for k ¼ 1 at L ¼ 1 or d ¼ rT); (b) zone diagram for the range of effective heterogeneous rate constants and ME radius, in which approach curves can be used for local kinetic analysis.

108

Methods and Instruments | Scanning Electrochemical Microscopy

Fig. 3 Plots of calculated SECM approach curves (normalized ME current vs. normalized distance) in the feedback mode. (a) Comparison of approach curves to an insulating, impermeable sample with MEs of different RG; (1) 10, (2) 4, (3) 1.5; (b) Comparison of approach curves to a porous insulating substrate; the porosity e is (1) 0; (2) 0.2; (3) 0.4; (4) 0.6, (5) 0.8.

calculated steady-state currents for conically shaped ME are available.1 The precise value of iT,1 is important because the normalized ME current IT ¼ iT(d)/iT,1 is obtained as ratio between the current iT at a specific distance d and iT,1. IT facilitates a generalized quantitative description of the ME current at very different conditions. If the distance d between the ME and the sample is decreased, iT(d) changes in a manner, that is very characteristic for the sample. These data – called “approach curve” – are obtained by recording iT(d) while the distance d is decreased slowly (translation rate  1 mm s−1) to ensure steady state. If the ME approaches an insulating and impermeable sample region, the current decreases with decreasing d (Fig. 2a2 and curve 1) because the sample hinders the diffusion of the reagent R to the ME. This situation is also called “negative feedback” in literature. Approach curves are usually plotted as normalized current IT ¼ iT/iT,1 vs normalized distance L ¼ d/rT because approximate analytical expressions are available for non-linear least square fitting to extract quantitative parameters from experimental approach curves. If the mediator can be regenerated at the sample, the current increases above the values in Fig. 2a, curve 1 (scheme in Fig. 2a2) because the flux of the R from the sample to the ME provides more reagents to the ME than in case of hindered diffusion of R from the bulk solution in Fig. 2a, curve 1. The exact value of iT(d) depends on the heterogeneous rate constant keff [cm s−1] for the reaction Ox + n e− ! R at the sample. To arrive at generalized expression, this first-order heterogeneous rate constant is normalized with the diffusion coefficient D of the mediator and the ME radius to a the normalized rate constant k ¼ keff  rT/D. Fig. 2a shows approach curves calculated for a selection of k-values according to approximate expressions by Cornut and Lefrou.2 If the regeneration reaction at the sample is very fast, i.e., limited by the diffusion of the mediator between the ME and the sample, the maximum values of iT is attained for any given d (Fig. 2a3 and curve 8). This situation is also referred to as “(diffusion-controlled) positive feedback”. The precise shape of the approach curves also depends on RG (Fig. 3a). This influence is already taken into account by the frequently used analytical expressions of Cornut and Lefrou.2 Many electrodes in energy conversion devices are porous electrodes. In such cases, the exact shape of the approach curves depends not only on the reactivity of the sample and the shape of the ME, but also on porosity and tortuosity of the sample (Fig. 3b). Analytical approximation for the currents above porous samples are available for some important cases,3,4 but they have not yet been tested as rigorously as the expressions for flat, impermeable samples. With the current-distance relationship shown in Fig. 2a and 3 two basic experiments can be designed: The ME can be moved at constant vertical distance d in a (x, y) plane. The signal iT(x, y) is transformed to an image representing a mapping of the (normalized) heterogeneous rate constant keff (or k). From such a “map”, interesting regions of the sample can be identified and horizontally approached by the ME. Approach curves can then be recorded at the selected locations. Subsequently, a fit to the analytical expressions for approach curves facilitates the determination of k ¼ keff  rT/D, which defines a range of accessible keff depending also on rT (Fig. 2b). For vertically positioning of MEs or analysis of porous electrodes, it is sometimes desirable to exclude the effect of sample kinetics. This is achieved by selecting an “irreversible mediator” that undergoes a chemically irreversible reaction at the ME (relatively fast ET followed by a homogeneous follow-up reaction to a redox-inactive product). Recommended reactions at the ME are the oxidation of ascorbic acid for aqueous solution4 and the oxidation of anthracene for organic electrolytes.5

2.2

Substrate-generation/tip-collection (SG/TC) mode

In this mode the ME detects compounds that are released from the active area of the sample or are transported through membranes or pores into the working solution (Fig. 4a). The working solution initially contains no compound that can be oxidized/reduced at the ME at the selected potential ET. Therefore, the SG/TC mode is very sensitive to small concentration of the detected compound (analyte). The analyte diffuses from the sample in all three space directions. If the diffusion layer from many microscopic sources overlap (e.g., catalyst particles at gas-diffusion electrodes), one macroscopic planar diffusion layer is formed. In such situations, the current iT at the ME depends on the time after the start of the experiment. Whether quantitative information can be obtained in such

Methods and Instruments | Scanning Electrochemical Microscopy

109

Fig. 4 Schematic representation of (a) the substrate-generation/tip-collection (SG/TC) mode; (b) the tip-generation/sample collection (TG/SC) mode; (c) redox-competition mode. Potential programs can be applied to the sample and the ME, exemplified for the SG/TC mode. The application of a potential sweep to the sample, while holding the ME at a constant potential to detect species generated at the sample is called tip-substrate voltammetry (TSV, panel 4a2).

a situation, depends on the possibility to switch on/off the reaction at the sample by applying a potential program (e.g., Fig. 4a2, a3) or using light in photoelectrochemical reactions. Tip-substrate voltammetry (TSV) refers to an experiment, in which a cyclic voltammogram is recorded at the sample and the ME is used to detect soluble reaction products. If the source of the detected species on the sample is a microstructure itself, a steady-state diffusion layer will develop. The ME can be used to detect the local steady-state concentration. Expressions are available to extract the overall flux and the concentration of the analyte near the sample surface.6

2.3

Tip-generation/substrate-collection (TG/SC) mode

For some irreversible reactions, the detection of the formed product, e.g., water as the product of the oxygen reduction reaction (ORR), is not feasible and therefore the SG/TC mode is not applicable.7 In the TG/SC mode, the ME is used to generate locally the reagent for the reaction at the sample (e.g., ORR) at a constant rate (e.g., galvanostatic formation of O2). The image is constructed by plotting the sample current iS (not the ME current!) vs. the (x,y)-position of the reagent generator, i.e., the ME (Fig. 4b). The signal iS originates from the entire macroscopic sample. In some situations, this current is dominated by the contribution form the region directly below the ME where a much higher flux of the reagent is available. However, the specific local signal is generated on a relatively high background signal, which limits the sensitivity.

2.4

Redox competition mode

The redox-competition (RC) mode can also be applied to the analysis of ORR. In this case O2 is reduced at the active parts of the sample and at the ME (Fig. 4c). Pulse programs have been suggested to overcome the problem that it is hard to distinguish between two relatively active spots.8 Analytical expressions are available that relate the measured current to the activity of the sample.9

2.5

Related techniques

Related electrochemical scanning techniques that are also relevant for study of electrochemical power sources are scanning electrochemical cell microscopy (SECCM) and scanning photoelectrochemical microscopy (SPECM). In SECCM, an electrolyte-filled capillary is used instead of a ME on a sample in a gas atmosphere. The probed area on the sample is restricted to the contact area between the liquid inside the pipette and the sample. The capillary is placed sequentially at different locations on the sample and complex potential programs can be applied to the sample at each contact point. In SPECM, an optical fiber is used instead of the ME to provide local illumination of the sample immersed into an electrolyte to give an image iS(x, y) of the photocurrent iS of the sample as function of the lateral position of the optical fiber.

3 3.1

Applications Spatiotemporal evolution of solid electrolyte interphases

The FB mode of SECM is ideally suited to follow the evolution of solid electrolyte interphases (SEI) on negative electrodes of Li+-ion batteries (LIBs) and other low-potential materials. The SEI is formed upon contact of a charged negative electrode with organic electrolytes. Ideally, the SEI is ion-conducting but electronically insulating. While allowing the passage of Li+-ions, the SEI suppresses the plating of metallic Li and prevents the direct contact between the electrolyte and the negative electrode that would otherwise lead to a continuous consumption of the electrolyte components. The analysis of the SEI is complicated because some structure-sensitive techniques require an artifact-prone transfer of the electrode with the SEI to another instrument or are incapable

110

Methods and Instruments | Scanning Electrochemical Microscopy

of assessing the lateral variation of SEI properties. In-situ or even operando analysis can be conducted using the SECM FB mode with an instrument placed in an Ar-filled glove box (Fig. 5a). Special setups have been designed to additionally prevent the evaporation of the volatile organic solvents inside the glove box. The experiment is conducted in an organic battery electrolyte supplemented with a redox mediator (Fig. 5b). The choice of the mediator is critical because the mediator may or may not partition between the liquid electrolyte and the organic parts of the SEI. A mediator that does not partition into the organic parts of the SEI will tend to exhibit a larger inhibition of the mediator regeneration, a mediator with greater partition into the organic part of the SEI may better simulate the behavior of solvent molecules of the electrolyte (chemically similar to the components of the SEI that are formed by solvent reduction at the negative electrode). Ideally, the SEI should prevent the ET from the negative electrode to these compounds (Fig. 5c, bottom). Tables of applied mediators have been compiled by Rodríguez-López and coworkers in reviews and book chapters listed under Further Reading. The temporal evolution of the SEI could be followed in TSV by redox-cycling a ferrocene mediator between the ME and the sample, while potentiodynamically charging the sample.10 Fig. 6 shows a typical plot of the ME current iT as a function of the sample potential ES. In the negatively going scan, the initial normalized current is IT ¼ 1.8 (Fig. 6a), but starts to decrease at potentials below 0.7 V (Li|Li+) indicating the onset of SEI formation. After reaching 0.0 V (Li|Li+), the potential scan at the sample is inverted but the ME current continues decaying. After delithiation, the normalized ME currents remains below unity indicating that the protecting SEI remains on the sample surface. In a similar experiment, but with additive vinylene carbonate, the SEI formation commences already at 1.2 V (Li|Li+) and reaches stronger blocking of ET reactions at the onset of lithiation (Fig. 6b).10 Potential cycling can be continued to observe the further slow evolution of the SEI. SECM FB experiments also confirmed the assumption that SEIs are inhomogeneous (on different length scales) and change over time (Fig. 7).11 A typical approach uses a pouch bag research cell cycled to a specific state of charge (SoC). The cell is dissected inside

Fig. 5 Schematic representation of a setup for SECM measurements on solid electrolyte interphases (SEIs) of Li+-ion batteries. (a) instrument specifically designed for operation in a glove box; (b) principle of the mediator recycling at a composite electrode and structure of the mediator 2,5-di-tert-butyl-1,4-dimethoxybenzene (DBDMB); (c) mediator recycling at bare and SEI-protected carbon.

Fig. 6 Formation of the solid electrolyte interphase (SEI) followed by tip-substrate voltammetry at fixed ME position. (a) Cyclic voltammogram for lithiation/ delithiation of a graphite paste electrode (black line) and simultaneously measured normalized ME FB current vs. sample potential in 0.02 mol L−1 ferrocene +1 mol L−1 LiClO4 in 50% (m/m) ethylene carbonate and 50% (m/m) diethyl carbonate (red line); (b) Cyclic voltammogram for lithiation/delithiation of graphite paste electrode (black line) and simultaneously measured normalized ME FB current vs. sample potential in the same electrolyte as in (a) but with 5% (m/m) vinylene carbonate (VC) as additive to promote the formation of a stable SEI. The decreasing current at the ME shows increasing blocking of the electron transfer (ET) from the negative electrode to the oxidized form of the mediator by the SEI. The formation of the SEI commences at more positive potential than the lithiation. When using the VC additive, the SEI attains blocking behavior for ET reaction at more positive potentials than without VC. The potential scan rate at the sample was v ¼ 10 mV s−1. Reprinted from Zampardi, G.; La Mantia, F.; Schuhmann, W. In-Operando Evaluation of the Effect of Vinylene Carbonate on the Insulating Character of the Solid Electrolyte Interphase. Electrochem. Commun. 2015, 58, 1–5, Copyright (2015), with permission from Elsevier.

Methods and Instruments | Scanning Electrochemical Microscopy

111

Fig. 7 Series of SECM images recorded above the same region of a graphite composite electrode. (a) 22.7 h after start of the SECM experiment, (b) 26.9 h after the start of the SECM experiment, (c) 28.2 h after the start of the SECM experiment and after decrease of the working distance to cause collision with protruding carbon particles, (d) current vs. location plot of a forward and backward linescan with the determination of points that show significant differences between the line scans, (e) color-coded two-dimensional histogram showing the incidence at each measurement point, at which iT in forward and backward line scans differed significantly from each other. Seven consecutive images were evaluated. Adapted from ref. Bülter, H.; Peters, F.; Schwenzel, J.; Wittstock, G. Spatiotemporal Changes of the Solid Electrolyte Interphase in Lithium-Ion Batteries Detected by Scanning Electrochemical Microscopy. Angew. Chem. Int. Ed. 2014, 53, 10531–10535, Copyright (2014), with permission from Wiley VCH Verlag GmbH & Co KGaA.

an Ar-filled glove box, the negative electrode is recovered and transferred with the adhering electrolyte layer to the SECM cell. The electrolyte is filled up by a minimum volume that is necessary to immerse the ME, Ref and Aux into the electrolyte of the SECM cell. The solution also contains a millimolar concentration of the mediator DBDMB (Fig. 4b). The ME is approached to the sample, where a curve similar to Fig. 3b is recorded that is controlled by the local quality of the SEI and the porosity of the used composite electrode. After finding the vertical position well above the sample, imaging experiments are initiated. An individual SECM FB image, such as that shown in Fig. 7a, bears little information content because the influence of local variation of the kinetics for mediator regeneration overlaps with the influences from topography and porosity of the sample. However, the influence of the local SEI quality, that is of highest interest here, can be clearly identified by recording image series of exactly the same region.11 The comparison of the images in Fig. 7a and b, that are part of a much larger series of sequentially recorded images of the same region, shows that the ME current changes at some regions and remains fairly constant at others. Such changes occur slowly over hours at open circuit potential (OCP). The protruding regions of the composite electrode can be identified by sequentially lowering the ME toward the sample, until a mechanical and electrical contact occurs that is easily identified by locally very high short-circuit currents (Fig. 7c). The identification of the protruding regions enables important extensions to the mere analysis of local currents. It is strongly advised to record the ME currents in the forward and backward scan (Fig. 7d) and to conduct the scans in a comb-like fashion (rather than in the shape of saw teeth or meanders). This procedure facilitates a comparison of iT in the forward and backward scan at identical locations (Fig. 7d). The counting of the instances N(x, y) within a series of images, at which there is a difference between the value larger than the data scatter can be represented as a two-dimensional histogram plot (Fig. 7e), in which N(x, y) is color-coded. It is evident, that the observed local differences between forward and backward line scans are not randomly distributed. There are large parts of the sample, where such differences were not recorded at all, and other, where such a difference always occurred. The regions with high N(x, y) are mostly located close to, but not exactly at, the protruding regions, which are marked by a thick line and were identified by short circuit currents in the subsequently recorded Fig. 7c. The relative location of

112

Methods and Instruments | Scanning Electrochemical Microscopy

protruding points and locations with fluctuating SEI properties suggests that the latter are areas between graphite particles, where the SEI might be damaged by very small relative movement of the particle to release mechanical tension introduced during calendering of the composite electrode. At the regions, where the ME touched the sample, the SEI is locally damaged. Slightly retracting the ME followed by another series of images allows following the re-formation of a protecting SEI over time. During this period, the current tends to show many short-term fluctuations.11 Similar studies about the local quality of SEIs or restoration of protecting properties after a local disturbance have been conducted on various negative electrode materials, among them highly ordered pyrolytic graphite,12 Si13 and hard carbon for Na+-ion batteries.14 While experiments similar to those in Fig. 7 are a powerful tool to ascertain the inhibition of ET reactions at the negative electrode, they are essentially blind to another equally important property of the SEI, i.e., the ion transfer (IT) reactions (see next section). A second limitation is the post mortem character of the experiment shown in Fig. 7. They capture a state of the negative electrode at a preset SoC under OCP conditions. However, it would be very interesting to assess the blocking properties for ET reactions during Li+-ion transfer, i.e., during charging or discharging. Such information is provided by TSV in Fig. 6, albeit without lateral resolution. Furthermore, the presence of the ME close to the sample will diminish the local current density exactly at the area that is interrogated. This situation has stimulated ongoing methodical developments to bring SECM imaging of SEI layers closer to practically relevant operating conditions, especially during formation of the SEI. A break-through became possible by cell construction with a Li counter electrode almost equal in size to the sample and oriented parallel to the sample at a distance of about 1 mm (Fig. 8a).15 A central opening in the counter electrode facilitates the approach of the ME to the sample electrode. This setup was alternatingly operated as symmetrical cell for galvanostatic charging/discharging and for SECM FB imaging with potentiostatic control of the ME. Simulation of the primary current density distribution during charging confirmed that it is indeed necessary to retract the ME during charging and discharging in order to ensure that the local current densities at the studied sample regions are approximately equal to the average current density (Fig. 8b, curve 2). If the ME remains close to the sample during charging and discharging, the local current density is greatly decreased below the average current density

Fig. 8 SECM FB imaging with intermittent galvanostatic charging of a Li metal electrode. (a) Photograph and cross-sectional schematics of the measurement cell containing a macroscopic Li counter electrode (CE) with a central opening; (b) simulation of the local current density distribution at the Li metal sample (1) with and (2) without the ME located at working distance during galvanostatic charging in an electrolyte with a conductivity of 0.01 S cm−1; (c0) histogram plots of the normalized ME currents recorded in images (c1-c6; c1-c6) sequentially recorded SECM FB images of the Li sample in 0.005 mol L−1 DBDMB +1 mol L−1 LiClO4 in propylene carbonate, ET ¼ + 4.2 V (Li|Li+). The charging/discharging operations between the recording of the SECM images is given in the figure as multiples of the lowest geometrical sample current density jS ¼ 0.35 mA cm−2 for galvanostatic charging/discharging, the cycle number and multiples of the initial sample charging duration ts ¼ 100 s. Image c6 was recorded at a larger working distance (+43 mm) because protruding high-surface area Li emerged at two locations after five charging/discharging cycles at 2.8 mA cm−2 for 1250 s. Adapted from Krueger, B.; Balboa, L.; Dohmann, J. F.; Winter, M.; Bieker, P.; Wittstock, G. Solid Electrolyte Interphase Evolution on Lithium Metal Electrodes Followed by Scanning Electrochemical Microscopy Under Realistic Battery Cycling Current Densities. ChemElectroChem 2020, 7, 3590–3596, Copyright (2020) The Authors, reproduced with permission.

Methods and Instruments | Scanning Electrochemical Microscopy

113

(Fig. 8b, curve 1).15 This finding has important implication also for other scanning probe investigations of battery materials, where operando studies must take into account the diminution of the local current density by the presence of the probe close to the sample. Images obtained immediately after galvanostatic charging/discharging showed very clearly, that the electrode tolerated charging/ discharging with current densities below 0.7 mA cm−2 (Fig. 8c1-c5), but started to develop unwanted, protruding high-surface area Li when the current density was increased to 2.8 mA cm−2 (Fig. 8c6). Using approach curves in the FB mode above the protruding Li structure and the unaffected regions also revealed a much higher reactivity of the protruding Li structures compared to the remaining flat regions of the Li metal electrode.15 As discussed above in the context of Fig. 7a, image series rather than an isolated image should be analyzed. They may show (i) appearance and disappearance of regions with particular low or high currents, (ii) change of the mean normalized currents. The use of normalized currents is recommended for evaluation because this compensates for changes of the ME electrode surface that are sometimes unavoidable over prolonged experimentation. Using false color images as in Fig. 8c1-c6, the color scale can be selected to highlights the contrast within an individual image frame. This makes it difficult to see the evolution of the average (normalized) current , that is typically decreasing as the SEI is formed. Scaling the color bar to the lowest and highest current within the entire series, shows nicely such a decrease of , but tends to result in featureless images in which the variation within an image frame are difficult to discern. Plotting series of histograms obtained from individual SECM images as in Fig. 8c0 represents one way of data aggregation that shows both, the decrease of indicated by the center of the histogram and the development of variation within individual images as evident from the shape and the width of the histograms. The width of the histograms increases for images during the incipient phases of SEI images, where iT shows the strongest local variations.

3.2

Ion-transfer reactions at battery materials

Ion-transfer reactions, such as intercalation of the working ions into graphite, alloying with Si or Sn, or conversion reactions are central to the storage mechanisms in modern batteries. They can be imaged by detecting the working ion (e.g., Li+, Na+, K+, Pb2+, or Zn2+) at an Hg amalgam electrode, which can be fabricated with different defined sizes and shapes. Typically, the Hg ME detects the working ion by anodic stripping voltammetry, during which the working ion is first reduced at the Hg ME forming an amalgam. After a preset accumulation time, the amalgam is dissolved in a positively going scan. The peak current iT,p(x,y,d) is proportional to the local concentration of the working ion. If stripping analysis is conducted in the vicinity of a negative battery electrode consuming the working ion during charging, iT,p(x,y,d) will be reduced to the situation when no IT processes occur at the sample (competition mode analogous to Fig. 4c). The opposite holds during deintercalation of the working ion (SG/TC mode analogous to Fig. 4a). This experiment does not require the addition of a redox mediator and can be conducted directly in the battery electrolyte. Due to the need to conduct a potential scan at each grid point of an image, the experiment takes considerably more time than standard imaging with the same number of grid points. Attention must be paid not to oversaturate the small amount of Hg at the ME, for which quantitative guidelines are available. When the working solution contains additionally a redox mediator, local IT reactions as well as local ET reactions can be studied in the same experiments. Such experiments have also been used to study intercalation of Na+-ions into multi-layer graphene after an SEI was formed in a Li+-containing electrolyte.16

3.3

Cathode electrolyte interphase

For high-voltage positive electrodes in LIBs, an ion-conducting protective film of electrolyte decomposition products, called cathode electrolyte interphase (CEI) is formed on the positive electrode. Attempts to image them similar to SEIs is more complicated because carbon-based conductive additives, required to compensate for the low electronic conductivity of positive electrode materials, may sustain a positive feedback in SECM FB experiments even if the transition metal oxides of the active material are covered by an electronically insulating and impermeable CEI.17

3.4

Detection of side reaction products in Li+-ion batteries

Overcharging or deep discharging can cause damage to the positive electrode of LIBs. Additionally, this process may result in O2 evolution and release of transition metal ions to the electrolyte, where they may enter into or catalyze reaction that degrade the cell and compromise their safety. SG/TC experiments (TSV mode) provide the sensitivity for the analysis of the degradation products because the ME collects the low amounts of the species of interest in very close proximity of the positive electrode. As an example, Co2+ and O2 have been detected above LiCoO2 and other positive electrode materials during overcharge and deep discharge.18

3.5

Gas diffusion electrodes in fuel cells

Gas diffusion electrodes (GDEs) in fuel cells and positive electrodes in metal-air batteries are porous electrodes that accommodate three-phase boundaries between a gas phase with fuel or O2, a solid electron conductor and a liquid or polymeric electrolyte. Although such electrode exhibits very different local structures and reactivity, the relevant feature sizes of GDEs are well below the

114

Methods and Instruments | Scanning Electrochemical Microscopy

size of MEs or even nanoelectrodes and, thus, cannot be resolved by SECM. Work on real samples is further complicated by their rough surface and is not easily modeled by flat electrodes. Therefore, SECM experiments using real gas diffusion electrodes remain scarce. Heterogeneity in the activity for the oxygen reduction reaction (ORR) of membrane electrode assembly (MEA) for polymer electrolyte membrane fuel cells has been imaged in the RC mode using Pt ME positioned by shear force.19 TSV was conducted at each point of the image, while the Pt ME was kept at a constant potential for ORR. The image was constructed from the ME currents at a preset sample potential showing the distribution of ORR reactivity in correlation with elemental mappings and details in preparation of the MEA. The SG/TC mode is very well suited to detect undesired side products of the ORR, which has especially been used to detect H2O2 above fuel cell cathodes (conceptionally similar to rotating ring-disk experiment, but with higher collection efficiency).20 Practical catalyst powders can be loaded into a cavity microelectrode (CME, Fig. 11a) without the need to use binders.21 The reactions at fuel cells anodes have been imaged in the SECM FB mode using H2/2H+ as mediator. The topography was characterized with a second mediator, whose kinetics depends only very little on electrode material. Combining the information from both mediators allows to disentangle to some extend the influence of reactivity and topography.22

3.6

Gas-diffusion electrodes in organic Li-air batteries

A suitable setup for SECM experiments at GDEs of LidO2 cells is shown in Fig. 9a. The GDE is placed between two gas reservoirs at equal hydrostatic pressure but different O2 partial pressure driving O2 permeation from the lower gas reservoir into the organic electrolyte solution of the SECM cell. The detection of O2 at the ME is accomplished by pulsed amperometric detection in the SG/TC mode (Fig. 9b) that allows the periodic removal of the reduction product Li2O2 from the Pt ME that would otherwise block the Pt ME.23 The influx of oxygen can be modulated by switching on/off the reduction of O2 at the sample GDE itself.23 During the O2 reduction in Li+-containing electrolytes, Li2O2 is formed and deposited in the void spaces of the GDE. This may lead to covering the electron-conducting parts of the GDE (and thus inhibiting further ET reactions) or obstruct the diffusion path of O2. SECM images of the charged GDE (without deposited Li2O2, Fig. 9c2) and the discharged GDE (with deposited Li2O2, Fig. 9c3) showed O2 permeation for both electrodes, albeit at reduced rate for the discharged GDE.24 This indicates that the end of discharge is determined by the blocking of the electron-conducting phase by Li2O2. The variation of iT within Fig. 9c2 and c3 is likely caused by the different local O2 transport (Fig. 9c1) due to the coarse structure of the Toray paper, onto which the GDE was built. The blocking of the electron conducting parts of the GDE by Li2O2 as the main discharge product complicates the recharging of the electrodes. The use of redox mediators in actual Li-O2 cells represents an approach for Li2O2 oxidation in this situation (Fig. 10a, right). The mediator undergoes an ET transfer reaction at free parts of the GDE and then indirectly oxidizes solid Li2O2 to soluble products. The kinetics of this process has been investigated by recording approach curves toward solid, electronically insulating Li2O2 (Fig. 10b) and fitting the data to theoretical curves as displayed in Fig. 2a for different redox mediators. Neither a correlation was found between the rate constant kapp for the reaction with Li2O2 and the standard heterogeneous rate constant k of the mediator at solid electrodes (Fig. 10c) nor between kapp and the formal potential E0 of the mediator.25 This underlines the need for independent testing procedures for such mediators.

3.7

Study of adsorbed intermediates at electrocatalysts

Adsorbed intermediates at electrocatalysts can be studied by the surface interrogation (SI) mode.26 It uses a colinear arrangement of two MEs facing each other. One of the ME is the interrogator or probe, the other ME is the sample (Fig. 11a). A cavity microelectrode may be used to study conventional catalyst powders (inset in Fig. 11b).27 A redox mediator is converted first at the interrogator

Fig. 9 Study of gas diffusion electrodes for Li-air batteries in the SG/TC mode in Li+-containing organic electrolytes. (a) SECM setup, in which the GDE is placed between two gas atmospheres (Ar + O2 | Ar); (b) potential program applied to the SECM ME for the detection of O2. (c) Examples of SECM images of O2 permeation through (c1) carbon paper, (c2) a charged GDE (open pores) and (c3) a discharged GDE (partially blocked pores). Electrolyte 1 mol L−1 LiClO4 in dimethylsulfoxide. Reprinted from Bülter, H.; Schwager, P.; Fenske, D.; Wittstock, G. Observation of Dynamic Interfacial Layers in Li-Ion and Li-O2 Batteries by Scanning Electrochemical Microscopy. Electrochim. Acta. 2016, 199, 366–379, Copyright (2016), with permission from Elsevier.

Methods and Instruments | Scanning Electrochemical Microscopy

115

Fig. 10 Investigation of redox mediators for Li-air batteries. (a) Schematic cross-sectional comparison of charging a GDE of a Li-O2 cell with and without mediator; (b) Schematic cross-sectional SECM setup; (c) plot of the apparent rate constant kapp between the oxidized mediator and solid Li2O2 and the heterogeneous rate constant of the mediator at a Pt electrode. There is no correlation between the two quantities, thus, the need to determine kapp independently. Panel (c) is reprinted from ref. Chen, Y.; Gao, X.; Johnson, L. R.; Bruce, P. G. Kinetics of Lithium Peroxide Oxidation by Redox Mediators and Consequences for the Lithium-Oxygen Cell. Nat. Commun. 2018, 9, 767. Copyright (2018) with permission from The Authors.

Fig. 11 Principle of the surface interrogation mode. (a) Schematic cross-sectional view of the electrode arrangement and cell connection during the conditioning step of the electrocatalyst surface; (b) cross-sectional view of the electrodes and the cell connection during the titration (interrogation) step, inset shows micrograph of an empty cavity electrode suitable to take up an electrocatalyst powder; (c) schematic depiction of sample voltammograms with indication of potential at which the condition is made in the SI-SECM experiment; (d) potential sequence (schematic) of three subsequent titration experiments with conditioning at different sample potentials; (e) recorded current transient (schematic) after conditioning at different potentials, (1) background curve (conditioning at ES,0 – no redox-active adsorbates/intermediates formed), (2) transient after conditioning at ES,1 with formation of one type of surface species, (3) current transient after the conditioning at ES,2 with formation of two types of surface species having distinctly different kinetics for the reaction with the mediator. The charges obtained by integrating the curves vs the background can be used to determine the amounts of adsorbate/reactive sites of a specific type.

electrode and reacts then with surface species at the sample. Thereby the mediator is recycled. Because there is only a limited amount of surface species available, the current will decay to the transient current for hindered diffusion once all surface species have been converted. In this sense, the experiment can be seen as a coulometric surface titration. By connecting the potentiostat to the sample electrode, the active catalyst surface can be “conditioned,” i.e., different surface species can be formed depending on the potential applied to the sample (Fig. 11a). After the conditioning step, the external power source is disconnected so that all charge must be exchanged across the electrode-solution interface (and not via the external circuit). The ME is then connected to the potentiostat and

116

Methods and Instruments | Scanning Electrochemical Microscopy

starts to convert the mediator and thus initiates the titration (Fig. 11b). The current transient is integrated vs. a background signal obtained at the electrode without active surface species (Fig. 11e, curves 1 and 2). This background can often be obtained by just recording another current transient at the interrogator ME without a previous conditioning pulse under otherwise identical conditions (Fig. 11d). The obtained charge can be converted to the surface concentration of the surface species such as adsorbed reaction intermediates of catalytically active redox centers. The current indicates the rate by which the redox active centers react with the redox mediators. This is illustrated schematically in Fig. 11c-e. The oxidation catalyst is assumed to form two types of redox centers at the potentials ES,1 and ES,2 (Fig. 11c). If conditioning is performed at a potential that is more positive than ES,1 but below ES,2, only one type of redox centers is formed. Their amount is accessible via the charge Q1 in the titration experiment (Fig. 11e, curve 2) and their kinetics is indicated by the plateau current iT,1 in the same curve. If the conditioning step is executed at a potential positive of ES,2, both surface species are formed. The charge Q2 from Fig. 11e, curve 3 represents the sum of the charges for both types of surface species. If the kinetics of both centers differ from each other (k1 6¼ k2), different current plateaus will be attained in the current transient. Inserting a defined delay between the conditioning of the sample and the onset of the titration pulse at the interrogator electrode defines a time, during which chemical follow-up reactions of surface species can occur, e.g., recombination of hydroxyl radicals to H2O2 or reaction of the substrate with a transient redox state in an oxide electrocatalyst.28

3.8

Screening of new catalyst materials for fuel cells and electrolyzers

SECM experiments are not only important for studying processed electrodes such as GDEs but also in the discovery process of new electrocatalyst materials. In this realm, SECM and related or combined procedures (SPECM, scanning droplet cell) have gained enormous importance as screening tools in combinatorial exploration of (photo)electrocatalysts. Material libraries are prepared as layers with compositional gradients or as spot arrays. Thus, the location on the sample is associated with a particular composition. Local structural and functional characterizations are then used to establish structure-reactivity relationships. In this context, the SG/TC mode of SECM is used to detect products such as O2 formed at electrolyzers or photoanodes.29

3.9

Other applications

Detection of side products: The sensitivity of the SG/TC mode can be used to detect soluble side products not only in LIBs but also in Li-O2 cells and supercapacitors.23,30 Heterogeneous kinetics at solid electrodes: Heterogeneous redox kinetics are essential for the reaction of dissolved redox systems in redox-flow batteries (RFBs) and for the dye regeneration in dye-sensitized solar cells (DSSCs). There are many examples, in which the approach curves in the SECM FB mode (similar to Fig. 2) have been used to determine the kinetics of those reactions and to relate them to the functional properties of RFBs31 and DSSCs.32,33

3.10 Outlook Increasing the lateral resolution of SECM would allow to address even more problems in the field of electrochemical power sources. A central problem in this context is the convolution of topographic and reactivity influences on the ME signal, that is usually present on rough electrodes preferred in technical devices. Combined AFM-SECM measurement with cantilevers that contain an integrated electrode as exemplified by ref. 14 represent one promising approach in this direction. Their use will certainly expand as the instrumentation has become commercially available. Another trend is the coupling of SECM with other scanning probe techniques to increase the information content. Ex-situ identical location experiments can also provide structural and compositional information that complement the local reactivity data from SECM provided that especially adapted transfer routines are being used.

4

Summary

Scanning electrochemical microscopy is a commercially available technique that maps local reactivities of interfaces suitable for addressing key problems in the development of electrochemical power sources. For allowing access of the ME to the internal interfaces and interphases in electrochemical power sources, samples have to be prepared in a way, that reflects essential properties of the operating device but at the same time allows SECM operation, thus often requiring specialized setups. Some established application area of the SECM feedback mode include the study of protection layers on low-potential battery electrodes and the analysis of the heterogeneous electron transfer kinetics in RFBs and DSSCs. Imaging experiments in the FB mode require the addition of a redox mediator, the choice of which is important for collecting relevant information and not disturbing the cell chemistry. The substrate-generation/tip-collection mode is very sensitive to detect small and transient concentrations of unwanted side products of reactions in fuel cells, lithium-ion and metal-air batteries as well as in supercapacitors. Imaging experiments also find increasing application in screening material libraries of (photo)electrocatalysts often in combination or as a complement to other imaging techniques.

Methods and Instruments | Scanning Electrochemical Microscopy

117

The quantitative nature of the ME response makes SECM also a very versatile tool for local kinetic studies at fixed lateral ME positions. The recording of approach curves in the FB mode is very broadly applied to determine local heterogeneous rate constants; tip-substrate voltammetry is an excellent tool for detecting soluble species generated/released from a sample during a potential scan. The surface interrogation mode can quantitatively determine surface concentrations and reaction rates of adsorbed intermediates in electrocatalysis relevant to fuel cells, water electrolyzers and photoelectrochemical water splitting.

Acknowledgment Ongoing research of G.W. and M.M. in the area are funded by the Federal Ministry of Education and Research (grant number 03XP0521D). M.W.-Z. acknowledges Gda nsk University of Technology, grant No DEC-12/2022/IDUB/II.1/AMERICIUM.

References Zoski, C. G.; Mirkin, M. V. Steady-State Limiting Currents at Finite Conical Microelectrodes. Anal. Chem. 2002, 74, 1986–1992. Lefrou, C.; Cornut, R. Analytical Expressions for Quantitative Scanning Electrochemical Microscopy (SECM). ChemPhysChem 2010, 11, 547–556. Kuss, C.; Payne, N. A.; Mauzeroll, J. Probing Passivating Porous Films by Scanning Electrochemical Microscopy. J. Electrochem. Soc. 2016, 163, H3066–H3071. Haensch, M.; Balboa, L.; Graf, M.; Silva Olaya, A. R.; Weissmüller, J.; Wittstock, G. Mass Transport in Porous Electrodes Studied by Scanning Electrochemical Microscopy: Example of Nanoporous Gold. ChemElectroChem 2019, 6, 3160–3166. 5. Hossain, M. S.; Stephens, L. I.; Hatami, M.; Ghavidel, M.; Chhin, D.; Dawkins, J. I. G.; Savignac, L.; Mauzeroll, J.; Schougaard, S. B. Effective Mass Transport Properties in Lithium Battery Electrodes. ACS Appl. Energy Mater. 2020, 3, 440–446. 6. Scott, E. R.; White, H. S.; Phipps, J. B. Ionotophoretic Transport through Porous Membranes Using Scanning Electrochemical Microscopy: Application to In Vitro Studies of Ion Fluxes through Skin. Anal. Chem. 1993, 65, 1537–1545. 7. Walsh, D. A.; Fernandez, J. L.; Bard, A. J. Rapid Screening of Bimetallic Electrocatalysts for Oxygen Reduction in Acidic Media by Scanning Electrochemical Microscopy. J. Electrochem. Soc. 2006, 153, E99–E103. 8. Eckhard, K.; Chen, X.; Turcu, F.; Schuhmann, W. Redox Competition Mode of Scanning Electrochemical Microscopy (RC-SECM) for Visualisation of Local Catalytic Activity. Phys. Chem. Chem. Phys. 2006, 8, 5359–5365. 9. Dobrzeniecka, A.; Zeradjanin, A. R.; Masa, J.; Blicharska, M.; Wintrich, D.; Kulesza, P. J.; Schuhmann, W. Evaluation of Kinetic Constants on Porous, Non-Noble Catalyst Layers for Oxygen Reduction—A Comparative Study between SECM and Hydrodynamic Methods. Catal. Today 2016, 262, 74–81. 10. Zampardi, G.; La Mantia, F.; Schuhmann, W. In-Operando Evaluation of the Effect of Vinylene Carbonate on the Insulating Character of the Solid Electrolyte Interphase. Electrochem. Commun. 2015, 58, 1–5. 11 Bülter, H.; Peters, F.; Schwenzel, J.; Wittstock, G. Spatiotemporal Changes of the Solid Electrolyte Interphase in Lithium-Ion Batteries Detected by Scanning Electrochemical Microscopy. Angew. Chem. Int. Ed. 2014, 53, 10531–10535. 12. Bülter, H.; Peters, F.; Wittstock, G. Scanning Electrochemical Microscopy for the In Situ Characterization of Solid-Electrolyte Interphases: Highly Oriented Pyrolytic Graphite Versus Graphite Composite. Energ. Technol. 2016, 4, 1486–1494. 13. Bärmann, P.; Krueger, B.; Casino, S.; Winter, M.; Placke, T.; Wittstock, G. Impact of the Crystalline Li15Si4 Phase on the Self-Discharge Mechanism of Silicon Negative Electrodes in Organic Electrolytes. ACS Appl. Mater. Interfaces 2020, 12, 55903–55912. 14. Daboss, S.; Philipp, T.; Palanisamy, K.; Flowers, J.; Stein, H. S.; Kranz, C. Characterization of the Solid/Electrolyte Interphase at Hard Carbon Anodes Via Scanning (Electrochemical) Probe Microscopy. Electrochim. Acta 2023, 453, 142345. 15. Krueger, B.; Balboa, L.; Dohmann, J. F.; Winter, M.; Bieker, P.; Wittstock, G. Solid Electrolyte Interphase Evolution on Lithium Metal Electrodes Followed by Scanning Electrochemical Microscopy under Realistic Battery Cycling Current Densities. ChemElectroChem 2020, 7, 3590–3596. 16. Sarbapalli, D.; Lin, Y.-H.; Stafford, S.; Son, J.; Mishra, A.; Hui, J.; Nijamudheen, A.; Romo, A. I. B.; Gossage, Z. T.; van der Zande, A. M.; Mendoza-Cortes, J. L.; Rodríguez-López, J. A Surface Modification Strategy Towards Reversible Na-Ion Intercalation on Graphitic Carbon Using Fluorinated Few-Layer Graphene. J. Electrochem. Soc. 2022, 169, 106522. 17. Zampardi, G.; Trocoli, R.; Schuhmann, W.; La Mantia, F. Revealing the Electronic Character of the Positive Electrode/Electrolyte Interface in Lithium-Ion Batteries. Phys. Chem. Chem. Phys. 2017, 19, 28381–28387. 18. Mishra, A.; Sarbapalli, D.; Hossain, M. S.; Gossage, Z. T.; Li, Z.; Urban, A.; Rodríguez-López, J. Highly Sensitive Detection and Mapping of Incipient and Steady-State Oxygen Evolution from Operating Li-Ion Battery Cathodes Via Scanning Electrochemical Microscopy. J. Electrochem. Soc. 2022, 169, 86501. 19. Schulte, W.; Liu, S.; Plettenberg, I.; Kuhri, S.; Lüke, W.; Lehnert, W.; Wittstock, G. Local Evaluation of Processed Membrane Electrode Assemblies by Scanning Electrochemical Microscopy. J. Electrochem. Soc. 2017, 164, F873–F878. 20. Shen, Y.; Träuble, M.; Wittstock, G. Detection of Hydrogen Peroxide Produced During Electrochemical Oxygen Reduction Using Scanning Electrochemical Microscopy. Anal. Chem. 2008, 80, 750–759. 21. Kishi, A.; Shironita, S.; Umeda, M. H2O2 Detection Analysis of Oxygen Reduction Reaction on Cathode and Anode Catalysts for Polymer Electrolyte Fuel Cells. J. Power Sources 2012, 197, 88–92. 22. Nicholson, P. G.; Zhou, S.; Hinds, G.; Wain, A. J.; Turnbull, A. Electrocatalytic Activity Mapping of Model Fuel Cell Catalyst Films Using Scanning Electrochemical Microscopy. Electrochim. Acta 2009, 54, 4525–4533. 23. Schwager, P.; Dongmo, S.; Fenske, D.; Wittstock, G. Reactive Oxygen Species Formed in Organic Lithium-Oxygen Batteries. Phys. Chem. Chem. Phys. 2016, 18, 10774–10780. 24. Bülter, H.; Schwager, P.; Fenske, D.; Wittstock, G. Observation of Dynamic Interfacial Layers in Li-Ion and Li-O2 Batteries by Scanning Electrochemical Microscopy. Electrochim. Acta 2016, 199, 366–379. 25. Chen, Y.; Gao, X.; Johnson, L. R.; Bruce, P. G. Kinetics of Lithium Peroxide Oxidation by Redox Mediators and Consequences for the Lithium-Oxygen Cell. Nat. Commun. 2018, 9, 767. 26. Rodríguez-López, J. Surface Interrogation Mode of Scanning Electrochemical Microscopy (SI-SECM): An Approach to the Study of Adsorption and (Electro)Catalysis at Electrodes. In Electroanalytical Chemistry, Vol. 24; Bard, A. J., Zoski, C., Eds.; Electroanalytical Chemistry: A Series of Advances, Vol. 24; CRC Press: Boca Raton, Fl, USA, 2011; pp. 287–351. 27. Haensch, M.; Behnken, J.; Balboa, L.; Dyck, A.; Wittstock, G. Redox Titration of Gold and Platinum Surface Oxides at Porous Microelectrodes. Phys. Chem. Chem. Phys. 2017, 19, 22915–22925. 28. Park, H. S.; Leonard, K. C.; Bard, A. J. Surface Interrogation Scanning Electrochemical Microscopy (SI-SECM) of Photoelectrochemistry at a W/Mo-BiVO4 Semiconductor: Quantification of Hydroxyl Radicals During Water Oxidation Electrode. J. Phys. Chem. C 2013, 117, 12093–12102. 29. Conzuelo, F.; Sliozberg, K.; Gutkowski, R.; Grützke, S.; Nebel, M.; Schuhmann, W. High-Resolution Analysis of Photoanodes for Water Splitting by Means of Scanning Photoelectrochemical Microscopy. Anal. Chem. 2017, 89, 1222–1228. 1. 2. 3. 4.

118

Methods and Instruments | Scanning Electrochemical Microscopy

30. Liu, X.; Lu, Z.; Pan, H.; Cheng, J.; Dou, J.; Huang, X.; Chen, X. Investigation of Functionalization Effect of Carbon Nanotubes as Supercapacitor Electrode Material on Hydrogen Evolution Side-Reaction by Scanning Electrochemical Microscopy. Electrochim. Acta 2022, 429, 141056. 31. Watkins, T. S.; Sarbapalli, D.; Counihan, M. J.; Danis, A. S.; Zhang, J.; Zhang, L.; Zavadil, K. R.; Rodríguez-López, J. A Combined SECM and Electrochemical AFM Approach to Probe Interfacial Processes Affecting Molecular Reactivity at Redox Flow Battery Electrodes. J. Mater. Chem. A 2020, 8, 15734–15745. 32. Xu, X.; Zhang, B.; Cui, J.; Xiong, D.; Shen, Y.; Chen, W.; Sun, L.; Cheng, Y.; Wang, M. Efficient P-Type Dye-Sensitized Solar Cells Based on Disulfide/Thiolate Electrolytes. Nanoscale 2013, 5, 7963–7969. 33. Ellis, H.; Schmidt, I.; Hagfeldt, A.; Wittstock, G.; Boschloo, G. Influence of Dye Architecture of Triphenylamine Based Organic Dyes on the Kinetics in Dye-Sensitized Solar Cells. J. Phys. Chem. C 2015, 119, 21775–21783. 34. Stulik, K.; Amatore, C. A.; Holub, K.; Marecek, V.; Kutner, W. Microelectrodes. Definitions, Characterization, and Application. Pure Appl. Chem. 2000, 72, 1483–1492.

Further reading 1. Bard, A. J., Mirkin, M. V., Eds. Scanning Electrochemical Microscopy, 3rd ed.; CRC Press Taylor & Francis Group: Boca Raton, FL, 2022. Especially Chapters 6, 13, 16. DOI:19.1201/9781003004592. 2. Mishra, A.; Sarbapalli, D.; Rodríguez, O.; Rodríguez-López, J. Electrochemical Imaging of Interfaces in Energy Storage Via Scanning Probe Methods: Techniques, Applications, and Prospects. Annu. Rev. Anal. Chem. 2023, 15.1–15.23. 3. Danis, L.; Gateman, S. M.; Kuss, C.; Schougaard, S. B.; Mauzeroll, J. Nanoscale Measurements of Lithium-Ion-Battery Materials Using Scanning Probe Techniques. ChemElectroChem 2017, 4, 6–19. 4. Schwager, P.; Bülter, H.; Plettenberg, I.; Wittstock, G. Review of Local in Situ Probing Techniques for the Interfaces of Lithium-Ion and Lithium-Oxygen Batteries. Energ. Technol. 2016, 4, 1472–1485.

Methods and Instruments | Karl-Fischer-Titration Balwant S Chohana and Dan G Sykesb, aSchool of Arts & Sciences, Felician University, Lodi, NJ, United States; bDepartment of Chemistry, The Pennsylvania State University, University Park, PA, United States © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 References

Introduction Karl Fischer titrations Principle of KF titration methods Advantages of KF titrations Disadvantages of KF titrations KF kinetics Sample preparation and selection of KF reagents Reagents for volumetric titration Reagents for diaphragm and diaphragm-less coulometric titration Extraction of water from samples KF endpoint detection KF Apparatus A detailed view of coulometric KF titrators Conclusion

120 120 121 122 122 123 123 124 125 125 126 126 129 133 133

Abstract Karl Fischer (KF) titration is a widely used analytical technique for the quantification of water in various substances including laboratory solvents, petrochemicals, transformer oils, pharmaceuticals, cosmetics, and food products. This method of titration is carried out using iodine as the titrant in the presence of an alcohol-based solvent and a base to control pH. Iodine is added to the sample solution either volumetrically or produced in situ via coulometric generation, where the water content of a sample is quantified based on the number of electrons transferred during titration. KF is a fast, accurate, and precise method which requires minimal sample preparation.

Glossary Anode The electrode where oxidation occurs in an electrochemical cell. It is the positive electrode in an electrolytic cell. Amperometry An electroanalytical technique based upon the measurement of the current flowing through the working electrode of an electrochemical cell, at a controlled applied potential. Anolytes The electrolyte on the anode compartment of an electrochemical cell. Biamperometric A titration that uses two polarizing or indicating ng electrodes to detect the end point of redox reaction. The technique follows changes in the current flowing between two electrodes that are maintained at constant potential difference. Bipotentiometric (bivoltammetric) A method where a constant current is passed through a solution, and the titration is followed by observing changes in the potential between the electrodes. Constant current polarization voltage A detecting method where a minute current is fed to dual platinum electrodes, as a detecting electrode, to measure the polarization voltage across the dual platinum electrodes. Cathode The positive electrode in an electrolytic cell, at which the electrochemical reaction is a reduction. Catholytes The electrolyte on the cathode compartment of an electrochemical cell. Coulometry Electrochemical measurement in which the electric charge required to carry out a known electrochemical reaction is measured. Electrolyte Conducting medium in which the flow of current is accompanied by the movement of ions. Electrolytic conductivity The measure of a material’s ability to carry electrical current. The SI unit of conductivity is siemens per meter (S/m). Emulsified A physical process by which the dispersed phase is broken up into small droplets, often by rapid mixing of the ingredients. Homogenizing A physical process of reducing a substance to extremely small particles and distributing it uniformly throughout a fluid. Hydrophilic A material that attracts water. A hydrophilic substance will bond, on a molecular level with water. Limit of quantitation (LOQ) The minimum measured concentration at which an analyte concentration may be reported. In analytical chemistry a simple method to find LOQ is to calculate the concentration that corresponds to a signal level that equals the baseline plus 10 times the noise.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00199-7

119

120

Methods and Instruments | Karl-Fischer-Titration

Minimum detectable volume (MDV) The minimum measured volume at which an analyte may be reported as being detected in the test portion or sample. It is the lowest level that can be reliably measured under routine conditions. Potentiometry Electrochemical measurement where the potential difference between an indicator electrode and a reference electrode is measured. Solubilizers Surfactants used to solubilize oils; essentially reduces oil molecules to a smaller size allowing it to disperse in water or another hydrophilic medium. Spectrophotometric A quantitative measurement technique based on the optical properties of materials from the ultraviolet to the visible and infrared spectral regions. It involves the measuring how much light a chemical substance absorbs as the beam of light passes through a solution of the sample. Titer the concentration of a solution as determined by titration. Titrant A solution of known concentration that is added to another solution to determine the concentration of a second chemical species. Titration endpoint The endpoint of the titration reaction indicated by a change in color or other property of the solution being titrated and represents the approximate molar equivalence between the titrant and the analyte. In KF a rapid decrease in the indicator electrode voltage marks the endpoint of the titration. Volumetric A chemical analysis based on measuring the volumes of reagents.

Key points The objectives of the article are to provide readers with a better understanding of:

• • • •

1

The chemistry and specificity of KF titrations. The advantages and disadvantages of KF titrations. Instrumental considerations when designing and using KF apparatus. The electronic circuitry to construct and perform bipotentiometric titrations.

Introduction

The determination of the water content in substances is one of the most frequently used methods in analytical chemistry laboratories. For example, Karl Fischer titration is the preferred method for measuring the water content in non-aqueous battery electrolytes since Li (and other metal) salts are reactive towards trace amounts of water, resulting in their degradation and the potential generation of toxic and corrosive products. Further, a knowledge of the water content of materials is a crucial parameter in the pharmaceutical and food industries, among many others. Water content impacts the texture and consistency, binding properties, shelf life, storage, and transport and distribution of pharmaceuticals and food products. For example, moisture-retaining agents are used in cosmetic products to facilitate skin hydration and protect against wrinkling.1 In contrast, high moisture levels in dry pet food may promote mold growth and produce toxins that negatively impact animal health.2 For lubricating oils, the presence of moisture can lead to premature corrosion, an increase in the debris load resulting in diminished lubrication and premature plugging of filters, an impedance in the effect of additives, and undesirable support of deleterious bacterial growth. The choice of water analysis must be efficient and fast to avoid time-consuming, and therefore expensive procedures. For research or manufacturing purposes, several methods for water determination have been developed, and selection of the most appropriate method depends on whether the sample is in solid, liquid, or gas form, time constraints, and available equipment. Water can be physically separated from the sample by distillation,3 chromatography, or oven or microwave drying. As water content influences physical properties of the sample, it can also be determined through densitometric, polarimetric, refractometric, or electrical methods.4–6 NMR, microwave, NIR spectroscopy, and microwave resonator methods focus on identifying specific characteristic properties of the water molecule. Several studies, comparing different water determination methods have been reported, with many recognizing that Karl Fischer (KF) titration has the ability to analyze a wide range of samples with varying complexity, and is the simplest, most accurate, and reproducible, among them.7–18

2

Karl Fischer titrations

Titration is a quantitative technique in which a solution of known concentration, titrant, is used to determine the exact concentration of an unknown compound. The KF method in chemical analysis uses volumetric or coulometric titrations to determine trace amounts of water in a sample. The KF technique was introduced in 1935, for determining small amounts of water in various organic and inorganic solid and liquid samples.19

Methods and Instruments | Karl-Fischer-Titration

121

The high selectivity to water represents the major advantage of the KF titration with respect to the weight-loss technique: only the water content is determined, since iodine selectively reacts with water. The KF titration is fast (1–2 min per titration), accurate, and precise. The wide measuring range (0.1% to 100% v/v) represents an additional strength of this method: samples with different water contents can be titrated volumetrically with a single instrument, and for very low water contents (below 50 ppm), the coulometric technique is applied, all in automated mode. The KF titration can only quantify freely available water. Thus, the release of water from the sample must be complete to obtain accurate results. This is achieved by performing suitable sample preparation methods prior to the KF titration. For instance, water is present in food in different forms each requiring different sample preparative steps to release water:20 Bulk water including water solubilized in other miscible liquids; capillary-bound or entrapped water, e.g., water droplets in cosmetics, emulsions, and water found within biological cells; physically-bound water, e.g., water contained within dry foodstuffs and certain cosmetics; and chemically-bound water, e.g., water of crystallization in hydrates. The water content determination by this method can be performed by two different techniques: Volumetric and Coulometric KF titration. The selection of the appropriate technique is based on the estimated water content in the sample. Volumetric titrations are suitable for samples with high water content; where at least 10 mg of water must be present in the sample for the volumetric technique to produce results with an acceptable level of precision. Generally, this method can determine water content in the 100 ppm to 100% v/v range, and therefore is not suitable if a trace amount of water is present in the sample. Volumetric titrations also involve simpler and cheaper equipment and a wider choice of solvents. In this type of titration, a buret equipped with a motorized piston slowly dispenses an iodine solution directly into the analyte solution until the endpoint is detected. The coulometric KF titration can measure water content as low as 10 mg of water and is useful for detecting free and emulsified water. Generally, this method can determine water content in the 1 ppm to 5% v/v range. However, impractically large amounts of sample are required when the water content is less than 0.05%. Instead of a buret the iodine reagent is generated in situ via electrolysis of an iodide-containing precursor solution. The switch for stopping the electrolysis in effect serves the same function as a buret stopcock.

3

Principle of KF titration methods

The principle behind KF titrations is based on the Bunsen reaction between iodine and sulfur dioxide in an aqueous medium. Karl Fischer obtained his PhD in 1925, and then began working with Lazar Edeleanu on using liquid sulfur dioxide to separate aromatic hydrocarbons from aliphatics. Fischer made use of Robert Bunsen’s21 report from 1854 that stated SO2 could be oxidized to sulfuric acid by iodine in the presence of water, and that the process takes place quantitatively. Fischer went on to find that adding a base like pyridine or aniline shifts the equilibrium for this reaction heavily to the right, as shown in Eq. (1). SO2 + I2 + 2H2 O ! H2 SO4 + 2HI

(1)

In 1937, Fischer proposed a titration using a standard methanolic solution of iodine, sulfur dioxide, and pyridine, where the endpoint was signaled by the persistence of the orange-brown color of iodine. Further, this reaction could be modified for the determination of water in a non-aqueous system containing an excess of sulfur dioxide. The use of methanol (ROH) as the solvent, serves as both a diluent and a reaction medium, and the pyridine base (R’N) as the pH buffering agent. The elementary reactions responsible for water quantification occur in stages (Eqs. 2 and 3): The first involves oxidation of sulfur dioxide with iodine in the presence of a base leading to an alkyl sulfite intermediate. The alkyl sulfite is then oxidized by iodine to an alkyl sulfate, a key process which consumes water from the analyte.22,23 Water and iodine are consumed in a 1:1 stoichiometric ratio according to the following reactions: ROH + SO2 + R ’ N ! ROSO2 − + R ’ NH+

(2)

ROSO2 − + I2 + H2 O + 2R ’ N ! 2R ’ NH+ + ROSO3 − + 2I−

(3)

Iodine is generated in coulometric titrations or added to the solution in volumetric titrations until it is present in excess, marking the endpoint of the titration when all the water is consumed. One of the most important developments in KF titration chemistry was the replacement of pyridine with imidazole.24,25 The noxious odor and toxicity of pyridine was a source of annoyance for many users of the KF method and titrations had to be carried out in a fume hood, which limited its practicality. Pyridine is a weak base (pKa 5.1) and the equilibrium of the reaction is not completely shifted towards the products, resulting in slow reactions and an unstable endpoint. This, in turn, negatively impacts reproducibility of measurements. Imidazole is less toxic, odorless, a stronger base (pKa 7.1), and shifts the reaction equilibrium completely to the right, resulting in significantly faster titration rates and a more stable endpoint. Further, the use of ethanol or propanol instead of methanol can provide for a faster reaction. Fischer originally used pyridine in ethanol for the titrant. In modern KF titrations, patented, pre-mixed reagents are available from most chemical suppliers. Green KF reagents are also available and are based on ethanol and diethylene glycol monoethyl ether. The latter shows low acute toxicity by oral, dermal, and inhalation routes. In volumetric titrations, the KF solution containing dissolved molecular iodine is added as the titrant from a buret until the presence of a trace excess iodine is detected. The amount of water as analyte is simply measured by the iodine required to react with the water content in the sample directly from buret.

122

Methods and Instruments | Karl-Fischer-Titration

In coulometric titrations, the sample is added to an electrolytic solution consisting of iodide ions and other reagents. The coulometric method avoids the need for titer determination and is thus an absolute method. The iodine produced in the reaction is proportional to the quantity of electric charge required for the oxidative conversion of iodide ions to iodine up to the endpoint. The presence of excess iodine is then readily detected using a constant current polarization voltage method utilizing an indicator electrode that sees a sharp voltage drop at the endpoint. Once the titration endpoint is achieved, the current is turned off and the total charge transferred is used to calculate how much iodine was required for the titration. According to Faraday’s law, the iodine produced is in proportion to the quantity of charge transferred (i.e., Coulombs, or C). For instance, to produce 1 mol of iodine from 1 mol iodide requires 96,485C/mol e−  2 mol I−/mole I2 ¼ 192,970C. Because 1 mol of iodine reacts with 1 mol of water, the amount of charge required to cause a reaction with 1 mg of water is: 1 mg H2O ¼ 5.56  10−5 mol H2O ¼ 5.56  10−5 mol I2 for 10.72C (i.e., 1 mg H2O ¼ 10.72C). For 1 mg H2O and constant electrolytic current of 107 mA, the time to reach the endpoint is about 100 s. The water content of a sample is quantified based on the number of electrons transferred during titration. This is determined by the time t necessary to reach the endpoint of an electrolytic titration of water with iodine under constant current conditions (Eq. 4): Moles of e− ¼

I∙t F

(4)

here, I is the constant current (C/s) at which the titration is taking place, t is the time (s) necessary to reach the endpoint, and F is Faraday’s constant (96,485C/mol e−). Thus, a titration system that carries out electrolysis at 300–400 mA, will titrate at the rate of about 30 mg/s. The coulometric method is used when measuring low water content levels or when very accurate water determinations are required.

4

Advantages of KF titrations

The popularity of the KF titrations for water determination is due to its accuracy, speed, and selectivity. KF is selective for water because the titration reaction itself consumes water. The titration does not detect the loss of any other volatile substance unlike other methods, such as heat-induced loss of moisture and reduction in weight of the sample. The coulometric titration is able to detect free water, emulsified water, and dissolved water even at low quantities. It is thus a rapid and robust method that requires minimal sample preparation apart from dissolving the sample in a suitable solvent and pre-titrating the solution and the cell to remove all traces of water before starting the titration. KF reagents have water equivalents which change with time, and therefore they should be regularly calibrated to ensure accuracy of determination. This is conducted using certified water standards with known water equivalents, and so helps avoid calibration with pure water for each titration. The water equivalents for commonly used KF reagents are 2–6 mg H2O/mL of reagent. The KF reaction follows a linear correlation, therefore, single-point calibration using a calibrated 1% water standard is sufficient and no calibration curves are necessary. Variations of the basic titration principle are being constantly developed. KF is ideally suited for automation, where systems control the operating procedures and improve both the reproducibility and accuracy26 Visual detection of a titration endpoint is also possible with colored samples by UV/vis spectrophotometric detection.27

5

Disadvantages of KF titrations

KF titration is a destructive technique. However, the sample quantity is small and is typically limited by the accuracy of weighing. KF titrations do suffer from drift error. The glass walls of the vessel adsorb water, and if any water leaks into the cell, then there is a continuous slow release of water into the titration solution. A consequence of this is that a steady-state condition will be reached at the endpoint of a titration, where the KF reaction consumes water at the same rate as it is entering the cell. This means that the concentration of water, often referred to as unreacted water, will never be zero. Therefore, it is necessary to carefully dry the vessel and conduct a dry-run to calculate the rate of drift. The drift value is then subtracted from the result. KF reagents are not stable and must be protected from light and atmospheric moisture. In volumetric titrations, the reagent must be standardized immediately before use and requires the use of fresh reagents for each determination and hence has a high solvent consumption. There are two main methods of calibration, both based on providing a standard sample with a known water content. The first method makes use of solid sodium tartrate dihydrate crystals that contain 15.66% of water by weight. Sodium tartrate dihydrate (STD) forms a stable complex with water at a known saturation concentration (15.66%) and retains this level even at temperatures over 120  C. Since STD exists as crystals, it is easier to handle and can be weighed with high accuracy using standard analytical balances. Once the crystals are weighed, they are added to the titration vessel. Once dissolved, the KF reagent is added until the endpoint is reached. The titer (water equivalent) is the calculated as follows (Eq. 5): Titer ðWEÞ ¼ Weight of STD ðmgÞ  0:1566=Volume of KF reagent ðmLÞ

(5)

Because STD may not fully dissolve in some reagents, such as ethanol-based reagents, liquid water standards represent a more reliable alternative. The second method uses a water sample dissolved in an organic solvent (standard water-methanol reagent) at a known concentration.

Methods and Instruments | Karl-Fischer-Titration

123

Table 1 Possible side-reactions in KF titrations: (a) ketones and aldehydes, (b) strong acids, and (c) metal carbonates, hydroxides and oxides. (a) R d C(¼O) – R’ + 2 CH3OH $ R d C(OCH3)2 – R’ + H2O R d CH ] O + H2O + SO2 + NR’ $ RCH(OH)(SO3)HNR’ (b) 2 CH3OH + H2SO4 ! (CH3O)2 SO2 + H2O CH3OH + HOOCH ! CH3O − OCH + H2O (c) Na2CO3 + 2 HI ! 2 NaI + CO2 + H2O Ca(OH)2 + H2SO4 ! CaSO4 + 2 H2O MgO + 2 HI ! MgI2 + H2O

KF reagents are highly selective towards water, but chemical substances other than water can interfere with the titration reaction. In such cases, the titration may fail to reach an endpoint, or an abnormal amount of KF reagent may be consumed, leading to negative or positive errors in the analysis value. KF titration depends upon a redox reaction (SO2/I2) and thus any component in the sample that is redox active such as dimethyl sulfoxide may react with the iodine reagent and generate false results. Table 1 lists several possible side-reactions which interfere with KF titrations. Aldehydes and ketones react with methanol to form acetals and ketals respectively, with the concomitant production of water. The water formed undergoes simultaneous titration, resulting in a vanishing endpoint and incorrect high water content. Similarly, aldehyde undergoes a side reaction to form a bisulfite, which consumes water during the reaction, and leading to an incorrectly low water content. Many metal oxides and hydroxides28 react with hydriodic acid and produce water. Boric acid, metal peroxides, silanols, and strong acids are also not suitable for KF titrations without modification,29 as their reaction with the methanol solvent produces water, resulting in a vanishing endpoint and false high-water content. Strong acids also lower the pH of the reaction solution so that the KF reaction becomes kinetically hindered and may not take place. The reaction mixture in this case can be buffered by the addition of a base (e.g., imidazole) and the pH brought back to the optimal range. These analytes require the use of methanol-free reagents that suppress any side-reactions. In some unfortunate cases substances in the sample matrix, or the sample itself, will react with iodine or iodide in the KF reagent. Chlorine and dichromate can for example oxidize iodide to iodine resulting in an underestimation of water. Iodine can be reduced to iodide by substances such as mercaptans, tin salts, hydrazines, or in rare cases add to double bonds in organic compounds, causing an overestimation of water. Some alkali metal carbonates30 and bicarbonates also react with the iodine. Such side-reactions make them unsuitable for KF titrations.

6

KF kinetics

The reaction rate is first order (Eq. 6) with respect to each individual component,31 where k ¼ 3.08  0.08 s−1 d ½I2 =dt ¼ k ½I2  ½SO2  ½H2 O

(6)

The reaction rate is strongly influenced by the concentration of iodide ions. The iodine in the KF solution reacts with iodide to form triiodide which displaces the equilibrium to the right: I2 +I− ) I−3. However, during the oxidation of the alkyl sulfite the free iodine reacts far more quickly than the triiodide. The rate of the KF reaction depends on the pH value of the working medium.22,23,32 The titration proceeds normally when pH is between 5 and 8. However, when the pH is lower than 5, the titration speed is very slow. If pH is below 2, the reaction will not occur. When pH is higher than 8, titration rate is fast due to an interfering side-reaction, which produces water, resulting in a vanishing endpoint. In very basic conditions, iodine is also consumed in side-reactions and leads to false high values. Thus, the optimal pH range for the KF reaction is from 5 to 8, and highly acidic or basic samples need to be buffered accordingly.

7

Sample preparation and selection of KF reagents

KF reagents are carefully chosen with respect to stability, reaction rate, conductivity, side-reactions, and solvation of the sample in the reagent solution. Despite the development of a variety of state-of-the-art improvements to KF instrumentation (e.g., automation by use of such devices as sample changers, ovens, and homogenizers), the key to successful KF analysis remains the fundamental understanding of sample and KF reagent chemistries. Several sample chemistries present challenges to conventional KF methods and reagents. These can be categorized into three general classes:

124

Methods and Instruments | Karl-Fischer-Titration

1. Certain compounds will undergo interfering side reactions either with methanol or iodine, and such side reactions will be detrimental to the accuracy of the titration. For example, the side-reactions of aldehydes and ketones with methanol that produce acetals and ketals, also generate water. Such reactivity issues may be resolved by using specialty reagents, performing titrations at low temperatures, or by employing auxiliary equipment. For example, aldehydes and ketones may be titrated by using special methanol-free coulometric or volumetric reagents. Certain slow-occurring side-reactions may be kinetically frozen-out by performing titrations using a special jacketed titration cell connected to a water bath circulating ice water, brine, or a dry ice/methanol mix. If suppression with formulated KF reagents does not work, then alternate solvents or thermal techniques can be employed using a KF Oven coupled with a KF titrator. Here, moisture is released from the sample in the KF Oven, at temperatures between 100 and 300  C, and is then carried into the titration cell using a dry inert carrier gas, such as nitrogen. 2. Methanol is readily available, it is inexpensive, it is relatively simple to dry, and it is polar and more hydrophilic compared to other alcohols. However, some samples have limited solubility in methanol. For example, atropine sulfate and calcium folinate are only partially soluble in methanol, as are many creams, ointments, and syrups. Solubility issues may be overcome by using different specialty reagents and titration aides, performing titrations at elevated temperatures, or by employing auxiliary equipment mentioned above: KF oven coupled with a KF titrator. Co-solvents such as hexanol, decanol, chloroform, formamide, and xylene are frequently used to increase the solvent capacity of KF reagents in the titration cell. Specifically pre-formulated KF reagents incorporating one or several of these solvents are commercially available, which significantly saves preparation time and expense. This wide range of solvent possibilities allows the user to determine the water content in practically any matrix. It is certainly no longer advantageous to self-prepare such reagents. 3. KF titration is based on a pH-dependent reaction. The optimum reaction rate is observed in the pH range of 5.5 to 8. Samples can change the pH of the KF solvent to either the acidic or basic extremes, and thus require buffering. In the case of acidic samples, imidazole is effective, while in the case of basic samples, buffering using benzoic or salicylic acid is required. The exact amount of buffering required will vary depending on the pH of the initial solution but can be between 5 and 15 g per 60 mL of solvent. For samples that contain components that do not dissolve in any of the available reagents or cause side-reactions that are difficult to prevent, then an oven method is available.33 A drying oven can also be used for samples that only release water at elevated temperatures, such as plastics and resins. The substance under investigation is heated in an oven located upstream from the cell and the evaporated moisture is transferred by a flow of dry carrier gas to the titration cell where it is determined by a KF titration. The oven attachment takes care to prevent the sample material from decomposing into water when heated to release the water.34 Because the moisture is delivered in gas phase and at an elevated temperature, the reagent must be able to dissolve it without boiling. As only the water from the sample enters the titration cell unwanted side-reactions and matrix effects are eliminated. Automated oven sample processors offer the possibility of using temperature gradients. The recorded water release curve can be used to determine the optimum heating temperature for the sample. The curve also allows statements to be made about the kinetics of water release as a function of the temperature. Depending on the water content of the sample the determination can be carried out volumetrically or, in the trace range, coulometrically. All reagents have an unstable titer, and the titer must be determined regularly. The interval between titer determinations depends on the choice of reagent as well as the diffusion of ambient water into the titration cell. Particular attention should be given to the titration cell: water present in the atmosphere can have a significant affect. For example, at a relative humidity of 60%, 0.5 mL of air at 25  C contains approximately 7 mg of water. Clearly, KF instruments must be situated far from sources of moisture or equipment that increases the humidity of the atmosphere. Further, sample handling time must be short as possible to prevent moisture absorption. During the titration, it is advisable to stir and swirl the cell solution from time to time to remove any water adhering to the walls. It is also important that a constant drift value that is as low as possible be achieved. The drift is the amount of KF reagent per unit time that is consumed to keep the titration cell dry. This is particularly important if very small amounts of water are to be determined, or if a long titration time is necessary, the drift value obtained is then subtracted from the added volume. In a coulometric cell, unsteady and high drift values also indicate a change of reagent, and a fresh refill, is necessary.

8

Reagents for volumetric titration

There are four different forms of the KF titration: (1) volumetric titration using a one-component reagent; (2) volumetric titration with a two-component reagent; (3) coulometric titration with a diaphragm; and (4) coulometric titration without a diaphragm. The KF reagents usually come as one-component or two-component types: Single-component reagents contain all reactants, except for water, needed to carry out the KF titration (e.g., iodine, sulfur dioxide, imidazole and diethyleneglycol monoethyl ether). If methanol is replaced with diethyleneglycol, the titer is relatively stable. One-component titration is the most common, due to its simplicity and convenience, and is preferred for titration of different sample matrices and for samples that require the addition of a solubilizer. These titrants are available in different concentrations to determine higher and lower water contents. They also tend to be less expensive than their two-component counterparts; however, they have shorter shelf-lives and lower buffer capacities. Owing to the instability of the one-component reagent, separating35,36 iodine from the base and sulfur dioxide has a clear advantage.

Methods and Instruments | Karl-Fischer-Titration

125

Solubilizers: The addition of solubilizer is often required to enable a direct titration of the sample and avoid pre-dissolution and pre-extraction steps. For example, fats, oils and long-chained hydrocarbons have limited solubility in methanol, which can be greatly improved by the addition of long-chain alcohols (such as propanol) or chloroform to the working medium. Occasionally it is necessary to investigate alcohols of various chain lengths and choose the alcohol based on its solubility towards the sample being investigated; It should be noted that secondary and tertiary alcohols cause increase in titer values. Proteins, carbohydrates, and some inorganic salts are either insoluble or dissolve only slightly in methanol. Therefore, addition of formamide improves the dissolution, and dispersion and the extraction of water from these samples. Because it can also influence the stoichiometry of the reaction, formamide should not be used at concentrations >50% by volume. One-component reagents can be extensively modified, just as long as the basic requisites for the KF titration are maintained (the methanol content should not be less than 35% by volume). Only a limited modification of the solvent component of the two-component system is possible since the solvent contains substrates for the KF reaction system. Any modification of coulometric KF reagents, must retain sufficient electrolytic conductivity within the cell; the minimum conductivity is dictated by the type of instrument used. Two-component reagents disperse the reactants between two separate solutions. The sample solvent is normally a solution of sulfur dioxide and imidazole in methanol, and the titrant is a solution of iodine dissolved in a suitable solvent. Two-component reagents offer several advantages including faster titration speed, exact titer with a high stability (i.e., very sharp endpoint and highly reproducible results), and high buffer capacity, but have solvent capacity restrictions. Methods based on two-component reagents generally work better for the determination of low mass of water since the mass of the iodine titrant is easy to control in small quantities. In both cases, the sample is dissolved in the solvent/working medium and is contained in the titration cell. The iodine-containing titrant is then added, and the water content in the sample is calculated based on the volume of the reagent used to reach the endpoint.

9

Reagents for diaphragm and diaphragm-less coulometric titration

As the coulometric method is an absolute method no titer needs to be determined. Traditionally, there are two types of coulometric KF reagents: Anolytes and Catholytes. For a coulometer equipped with a generator electrode with a diaphragm, the anolyte is found in the titration vessel, whereas the catholyte is in the cathode compartment. For coulometers with a diaphragm-less generator electrode, only one single reagent is required. Typically, there is little difference between anolyte and catholyte solution chemistries. The chemical composition of the anolyte solution changes with use and can impact reproducibility of consecutive measurements. In cells without a diaphragm, the anolyte reagent must be changed more frequently to avoid this issue. For samples with very low moisture content which require large volumes of sample to be added to the reaction vessel it is important to caution that after several measurements, the solution becomes so diluted that current efficiency (i.e., production of iodine) can substantially decrease and lower overall performance. The use of non-polar solvents also increases electrical resistance of the solution which can lead to overheating of the solution especially for long experimental run times. Thus, it is important to monitor the temperature of the cell and periodically allow it to cool to avoid excessive evaporation of the solution.

10

Extraction of water from samples

The role of the solvent is to contribute to the chemical reaction and to ensure the sample is releasing all its moisture content. Sample solubility is extremely important to obtain total water content. One of the key things to consider is whether the sample can be dissolved fast and completely in the Karl Fischer solvent without any additional treatments or if addition of co-solvents (solubilizers), or if any sample preparation is needed to extract the water from the sample. Other solvents can be added to methanol or ethanol in specific proportions to liberate the water more efficiently and completely dissolve the sample. For example, chloroform is a good solvent for fats, and formamide improves the solubility of polar substances. Ethanol or methanol should always be present in the solvent in a proportion not lower than 25%, otherwise unusual stoichiometric ratios will occur for the Karl Fischer reaction, such as the oxidation of bisulfite by hypoiodite results in a stoichiometric deviation towards 2:1 between water and iodine.32,37 Solid samples can bind water in different forms: enclosed water, crystal water, surface water. To measure the total water content of a solid sample, either it must be totally dissolved, or the water completely extracted. Typically, after homogenizing a sample by chopping, grinding, pulverizing, mixing or dispersing, there are different options available to completely release water for titration: (1) Internal extraction is where the sample is added directly into the titration cell to dissolve. Samples that do not release water fast, can be supported by heating, adding co-solvents, homogenization, or extended stirring. Heating is conducted in a specific heated cell at 50  C. Examples are the applications for coffee and starchy products such as wheat, flour or rice. (2) External extraction is where the sample is added in an appropriate solvent outside of the titration cell to release the water, and an exact weighted aliquot is then added into the titration cell. (3) Gas-phase extraction involves heating the sample in a Karl Fischer oven to extract the water and transferring the evaporated water into a KF titration cell. However, the samples need to be thermally stable. A temperature ramp is often carried out first with each sample to determine the exact heating temperature needed. The samples are then heated to this specific temperature, causing the water in them to evaporate, and be transported into the titration vessel by a constant flow of dry air or nitrogen.

126

11

Methods and Instruments | Karl-Fischer-Titration

KF endpoint detection

Because of the intense color it is possible to visually detect the iodine excess that occurs at the end of a titration. The color shift is, however, rather more gradual than distinct and therefore not accurate. At low iodine concentrations around the weak yellow colored tone is usually recognized with the naked eye, and if relatively large amounts of water are titrated then the uncertainty of the endpoint is small enough. The visual determination of the titration endpoint requires a high degree of experience and can normally only be used for uncolored sample solutions. During the KF titration, the red-brown iodine is reduced to colorless iodide when reacting with water. At the endpoint of the titration, the color of the solution gradually turns from colorless to yellow to red-brown (with a large excess of iodine). This lack of a sharp color change and the coloration differences in nonpolar solvents and polar solvents, makes visual determination of the endpoint extremely difficult. Fortunately, iodine and iodide form an electroactive redox couple which can be monitored electrochemically. For this reason, the endpoint of the titration is usually determined using two platinum wire electrodes configured for either biamperometric or bivoltammetric detection, where simultaneous oxidation of iodide and reduction of iodine occurs. Biamperometric indication: A small constant voltage (100 mV) minimize the uncertainties between runs but result in the generation of a greater excess in I2 and longer re-equilibration times between runs. Either the output voltage or the set point are displayed on a digital panel meter. The generator module contains three basic elements (A–C in Fig. 5). Constant current generation is defined and controlled through software via the PIC 16F886 microcontroller which simplifies the overall circuit design (region A). The chip is programmed using PIC Basic Pro software from ME Labs. The generator current is proportional to a pulse width modulated output voltage from pin 13 (RC2) of the microcontroller. The output voltage is limited to values between 0 and 4 V which corresponds to currents between 0 and 400 mA. The microcontroller is programmed for 10, 20, 50, 100, 200 and 400 mA supply currents. The current output is user-selected by depressing the “current” momentary push button (J2–6). Constant current generation occurs in either “Auto” or “Manual” mode. The user defines the mode of operation through a toggle switch on the front panel on the instrument housing (J2–4). In Auto-mode, current generation (and the clock mechanism) begins when the start/stop momentary button (J2–5) is depressed. Current generation stops when the photodiode, which is optically coupled to the detector circuit, is tripped (region C). In Manual-mode, current generation begins and ends by momentarily depressing the start/stop push button (J2–5). The output of the microcontroller is integrated to generate a stable current supply prior to passage through the sample cell (the “Load” in region B) via the generator electrodes. Current generation is regulated by means of a feedback mechanism between the current sink (IRL520) and the microcontroller and is stable to 1%. The LMC 6482 chip (region B) compares the compliance voltage with the voltage necessary to supply the user-defined current. The user is notified if the voltage offset is greater than 0.5 V by activation of the LED and automatic switch-off of the generator. A calibration curve, which plots volume of nanopure water versus experimental run time, is shown in Fig. 6. The magnitudes of the error bars reflect the uncertainty in the volume of injection using the micropipettor (x-axis) and the standard deviation of the four replicate runs relative to the average run time (y-axis). From the linear least-squares fit to the data, a minimum detectable volume (MDV) of (3sy/m¼) 0.26 mL and the lower limit of quantitation (LOQ) of (10sy/m¼) 0.85 mL were calculated. The linear range of response for the custom-built KF apparatus falls between 0.85 and 5.00 mL. The typical limit of detection reported for commercial instruments is 0.1% w/w. We prefer to report a MDV as opposed to a LOD. Accurate and reproducible water concentrations below 0.1% w/w in olive oil can be achieved by simply using sample injection volumes that ensure the water content of the sample is above the MDV.

Methods and Instruments | Karl-Fischer-Titration

Fig. 4 The circuit diagram for the detector module of the KF instrument. (A) power supply, (B) constant current/voltage source, (C) transimpedance amplifier, and (D) comparator or discriminator. Victoria C. D.; Cole R. M.; Matt J.; Doug S.; Rod K.; Dan S.; Benjamin T. W.; Balwant S. C. The Characterization of a Custom-Built Coulometric Karl Fischer Titration Apparatus. J. Chem. Educ. 2010, 87(9), 987–991. https://doi.org/10.1021/ed9000156.

131

132 Methods and Instruments | Karl-Fischer-Titration Fig. 5 The circuit diagram for the generator module of the KF instrument. (A) PIC microcontroller, (B) comparator, (C) detector-generator optocoupler. Victoria C. D.; Cole R. M.; Matt J.; Doug S.; Rod K.; Dan S.; Benjamin T. W.; Balwant S. C. The Characterization of a Custom-Built Coulometric Karl Fischer Titration Apparatus. J. Chem. Educ. 2010, 87(9), 987–991. https://doi.org/10.1021/ed9000156.

Methods and Instruments | Karl-Fischer-Titration

133

Fig. 6 A calibration curve expressing the relationship between volume of nanopure water and the coulometric titration time. Mean average values are shown with standard error bars. The equation of linear least-squares fit is y ¼ 105.2 + 25.2, where r2 ¼ 0.9978. Victoria C. D.; Cole R. M.; Matt J.; Doug S.; Rod K.; Dan S.; Benjamin T. W.; Balwant S. C. The Characterization of a Custom-Built Coulometric Karl Fischer Titration Apparatus. J. Chem. Educ. 2010, 87(9), 987–991. https:// doi.org/10.1021/ed9000156.

14

Conclusion

Karl Fischer titration is an important analytical method to determine the water content in substances. The method is based on a two-step oxidation reaction between sulfur dioxide and iodine in the presence of water. The first step involves oxidation of sulfur dioxide with iodine in the presence of a base leading to an alkyl sulfite intermediate. The alkyl sulfite is then oxidized by iodine to an alkyl sulfate, a key process which consumes water from the sample. Water and iodine are consumed in a 1:1 stoichiometric ratio. The titration can be performed using either a traditional volumetric approach or by the coulometric generation of iodine in situ. The method is fast and accurate with excellent precision. It has a wide range of determination for water from 0.1–100% v/v and can be used to analyze solids, liquids and gases.

References 1. Chen, H. J.; Lee, P. Y.; Chen, C. Y.; Huang, S. L.; Huang, B. W.; Dai, F. J.; Chau, C. F.; Chen, C. S.; Lin, Y. S. Moisture Retention of Glycerin Solutions with Various Concentrations: A Comparative Study. Sci. Rep. 2022, 12, 10232. https://doi.org/10.1038/s41598-022-13452-2. 2. de Brito, C. B. M.; Félix, A. P.; de Jesus, R. M.; de França, M. I.; de Oliveira, S. G.; Krabbe, E. L.; Maiorka, A. Digestibility and Palatability of Dog Foods Containing Different Moisture Levels, and the Inclusion of a Mould Inhibitor. Animal Feed Sci. Tech. 2010, 159, 150–155. https://doi.org/10.1016/j.anifeedsci.2010.06.001. 3. Xylene distillation (DIN ISO 3733): ISO 3733:1999. Petroleum products and bituminous materials. Determination of water. Distillation method. 4. Rubini, M. E.; Wolf, A. V. Refractometric Determination of Total Solids and Water of Serum and Urine. The J. Biol. Chem. 1957, 225, 869–876. 5. Sesta, G.; Lusco, L. Refractometric Determination of Water Content in Royal Jelly. Apidologie 2008, 39, 225–232. 6. Toennies, G.; Elliott, M. A Polarimetric Method for the Determination of Water in Acetic Acid. J. Am. Chem. Soc. 1937, 59, 902–906. 7. Vassileva, E.; Quétel, C. R. Influence of the Correction for Moisture/Water Content on the Quality of the Certification of Cadmium, Copper and Lead Mass Fractions in Rice. Food Chem. 2008, 106, 1485–1490. 8. Hadaruga, N. G.; Hadaruga, D. I.; Isengard, H. D. Water Content of Natural Cyclodextrins and their Essential Oil Complexes: A Comparative Study between Karl Fischer Titration and Thermal Methods. Food Chem. 2012, 132, 1741–1748. 9. Towns, J. K. Moisture Content in Proteins: Its Effects and Measurement. J. Chrom. A 1995, 705, 115–127. 10. Isengard, H. D.; Merkh, G.; Schreib, K.; Labitzke, I.; Dubois, C. The Influence of the Reference Method on the Results of the Secondary Method Via Calibration. Food Chem. 2010, 122, 429–435. 11. Rüegg, M.; Moor, U. Die Bestimmung des Wassergehaltes in Milch und Milchprodukten mit der Karl-Fischer-Methode. V. Die Wasserbestimmung von getrockneten Milchprodukten 1. Mitt. Gebiete Leben. Unter. Hyg. 1987, 78, 309–316. 12. Jensen, R.; Buffangeix, D.; Covi, G. Measuring Water Content of Feces by the Karl Fischer Method. Clin. Chem. 1976, 22, 1351–1354. 13. Mabrouk, P. A.; Castriotta, K. Moisture Analysis in Lotion by Karl Fischer Coulometry. An Experiment for Introductory Analytical Chemistry. J. Chem. Ed. 2001, 78, 1385. 14. Hadaruga, N. G.; Hadaruga, D. I.; Isengard, H. D. Water Content of Natural Cyclodextrins and their Essential Oil Complexes: A Comparative Study between Karl Fischer Titration and Thermal Methods. Food Chem. 2012, 132, 1741–1748. 15. Thiex, N.; Richardson, c.R. Challenges in Measuring Moisture Content of Feeds. J. Anim. Sci. 2003, 81, 3255–3266. 16. El-Sayd, N.; Makawy, M. M. Comparison of Methods for Determination of Moisture in Food. Res. J. Ag. Biol. Sci. 2010, 6, 906–911. 17. Isengard, H. D. Water determination, scientific and economic dimensions. Food Chem. 2008, 106, 1393–1398. 18. De Caro, C. A.; Aichert, A.; Walter, C. M. Efficient, Precise and Fast Water Determination by the Karl Fischer Titration. Food Contr. 2001, 12, 431–436. 19. Fischer, K. Neues Verfahren zur maßanalytischen Bestimmung des Wassergehaltes von Flüssigkeiten und festen Körpern. Angew. Chem. 1935, 48, 394–396. 20. Isengard, H.-D. Water Content, One of the most Important Properties of Food. Food Contr. 2001, 12, 395–400. 21. Ferguson, J. B. The Iodometric Determination of Sulfur Dioxide and the Sulfites. J. Am. Chem. Soc. 1917, 39, 364–373. 22. Verhoef, J. C.; Barendrecht, E. Mechanism and Reaction Rate of the Karl Fischer Titration Reaction. Part V. Analytical Implications. Anal. Chim. Acta 1977, 94, 395–403.

134

Methods and Instruments | Karl-Fischer-Titration

23. Verhoef, J. C.; Barendrecht, E. Mechanism and Reaction Rate of the Karl-Fischer Titration Reaction: Part I. Potentiometric measurements. J. Electro. Chem. Inter. Electro. 1976, 71, 305–315. 24. Meyer, A. S.; Boyd, C. M. Determination of Water by Titration with Coulometrically Generated Karl Fischer Reagent. Anal. Chem. 1959, 31, 215–219. 25. Scholz, E. Karl Fischer Coulometry, the Cathode Reaction. Fres. J. Anal. Chem. 1994, 348, 269–271. 26. Felgner, A.; Schlink, R.; Kirschenbühler, P.; Faas, B.; Isengard, H. D. Automated Karl Fischer Titration for Liquid Samples - Water Determination in Edible Oils. Food Chem. 2008, 106, 1379–1384. 27. Tavcar, E.; Turk, E.; Kreft, S. Simple modification of Karl-Fischer titration method for determination of water content in colored samples. J. Anal. Met. Chem. 2012, https://doi.org/ 10.1155/2012/379724. Article ID 379724. 28. Brumleve, T. R. Determination of Traces of Water, Hydroxide, and Oxide in Metal Halide Salts by Coulometric Karl Fischer Titration. Anal. Chim. Acta 1983, 155, 79–87. 29. Bruttel, P.; Schlink, R. Water determination by Karl Fischer titration. Metrohm Mon. 8.026.5013. 30. Gard, L. N.; Butler, R. C. Determination of moisture in sodium bicarbonate. Anal.Chem. 1954, 26, 1367–1368. 31. Cedergren, A. Determination of Kinetics of the Karl Fischer Reaction Based on Coulometry and True Potentiometry Anal. Chem. 1996, 68, 784–791. 32. Oraedd, C.; Cedergren, A. Reaction Rates of Water in Imidazole-Buffered Methanolic Karl Fischer Reagents. Anal. Chem. 1994, 66, 2603–2607. 33. Themudo, M. E.; Barms, A. A.; Bastos, M. Application of Karl Fischer’s Method to Materials that Only Release Water at High Temperatures. Port. Etec. Acta 2001, 19, 301–311. 34. Isengard, H. D.; Schmitt, K. Karl Fischer Titration at Elevated Temperatures. Mikro. Acta 1995, 120, 329–337. 35. Cedergren, A.; Jonsson, S. Progress in Karl Fischer Coulometry Using Diaphragm-Free Cells. Anal. Chem. 2001, 73, 5611–5615. 36. Johansson, A. Determination of Water by Titration (a Modified Karl Fischer Method). Svensk Papperstidn. 1947, 50, 124–126. 37. Oraedd, C.; Cedergren, A. Coulometric Study of Recovery Rates for Karl Fischer Titration of Water in Aldehydes and Ketones Using Rapidly Reacting Methanolic and 2-Methoxyethanolic Reagents. Anal. Chem. 1996, 67, 999–1004. 38. Spohn, U.; Hahn, M.; Rüttinger, H-H.; Matschiner, H. A New Measuring Device for Rapid Coulometric Karl-Fischer-Titrations. Fres. Z. Anal. Chem. 1989, 333, 39–41. 39. Cedergren, A.; Jonsson, S. Diaphragm-Free Cell for Trace Determination of Water Based on the Karl Fischer Reaction Using Continuous Coulometric Titration. Anal. Chem. 1997, 69, 3100–3108. 40. Hiromasa, K.; Naoko, K.; Toshio, K. Karl Fischer Coulometric Titration Using a Diaphragmless Cell. Bun. Kag. 1999, 48, 711–715. 41. Nordmark, U.; Cedergren, A. Progress in Pulsed-Current Karl Fischer Coulometry Using Diaphragm-Free Cells. Fres. J. Anal. Chem. 2000, 367, 519–524. 42. Lanz, M.; De Caro, C. A.; Ruegg, K.; De Agostini, A. Coulometric Karl Fischer Titration with a Diaphragm-Free Cell: Cell Design and Applications. Food Chem. 2006, 96, 431–435. 43. Dominguez, V. C.; McDonald, C. R.; Johnson, M.; Schunk, D.; Kreuter, R.; Sykes, D.; Wigton, B. T.; Chohan, B. S. The Characterization of a Custom-Built Coulometric Karl Fischer Titration Apparatus. J. Chem. Ed. 2010, 87, 987–991.

Methods and Instruments | IR and Raman Spectroscopy Peter Kurzweil, Technical University of Applied Sciences (OTH), Amberg-Weiden, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is a update of P. Jacobsson, P. Johansson, MEASUREMENT METHODS | Vibrational Properties: Raman and Infra Red, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 802–812, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00076-9.

1 2 2.1 2.2 3 3.1 3.2 4 4.1 4.2 5 5.1 5.2 5.3 5.4 5.5 6 References

Introduction Principles of IR absorption and RAMAN spectroscopy Permitted molecule vibrations and absorption bands IR and RAMAN spectra Instrumentation IR spectroscopy RAMAN scattering and microscopy Ab initio calculations of vibrational spectra Molecular models Computational algorithm Application of vibrational spectroscopy in electrochemistry Charge carriers and structure in electrolytes for lithium batteries Development of lithium battery salts Lithium battery cathodes, anodes, and current collectors Fuel cell membranes and water management RAMAN spectra of carbon materials Conclusion

136 136 136 137 138 138 138 139 139 140 141 141 142 143 145 145 146 146

Abstract Infrared (IR) absorption and RAMAN spectroscopy are powerful analysis techniques for the development of new materials for electrochemical applications in fuel cells, batteries, and supercapacitors. Computational chemistry methods help to better understand the complexity of the highly functionalized materials, to adopt the experimental methods and to analyze the information-rich spectra. This article provides a brief introduction of IR absorption and RAMAN spectroscopy together with ab initio calculations. Practical examples from the field of lithium batteries and polymer electrolyte fuel cells are presented to demonstrate the usefulness of basic analysis conditions and more elaborate approaches of experimental setups.

Key points

• • •

Fundamentals and experimental setup of infrared and RAMAN spectroscopy. Applications in electrochemistry: electrolyte systems and carbon materials. Introduction to ab initio calculations.

Nomenclature

Symbols and units


n A c d E h I q T x a

Wave number (cm−1) Absorbance Speed of light (m s−1), concentration (mol L−1) Optical thickness (m) Potential energy (J), electric field strength (V m−1) Planck constant (J s) Light intensity (W) Displacement coordinate or normal mode Transmittance (1 ¼ 100%) Displacement Polarizability

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00143-2

135

136

Methods and Instruments | IR and Raman Spectroscopy

DE «(l) l m n C

Absorbed radiation energy (J) Molar absorption coefficient (L mol−1 cm−1) Wavelength (m) Dipole moment (C m) Vibrational frequency (Hz) Wave function

Abbreviations and acronyms ATR C6P DMC DMFC DMSO DRIFT EC FT IR LiPATC PC PEO PES RRS SEI SERS TFSI

1

Attenuated total reflectance Calix[6]pyrrole Dimethyl carbonate Direct methanol fuel cell Dimethyl sulfoxide Diffuse reflectance infrared Fourier transform Ethylene carbonate Fourier transform Infrared Lithium-pyrazole-3,4,5-tricarbonitrile Propylene carbonate Poly(ethylene oxide) Potential energy surface Resonance RAMAN spectroscopy Solid electrolyte interphase Surface-enhanced RAMAN spectroscopy Bis(trifluoromethanesulfone)imide

Introduction

Infrared (IR) absorption and RAMAN spectroscopy are vibrational spectroscopy techniques suited for studying inorganic and organic materials for electrochemical applications.1–27 Like a fingerprint, the spectra show the molecular units in the material and also their interactions with the local chemical environment. For example, polymer-based gel electrolytes reveal information about composition, chain structure, degree of branching, conformation of the polymer matrix as well as cation-anion and ion-solvent interactions for the electrolyte counterpart. When applied to electrochemical applications in situ, information about chemical reactions or abundance of certain species is obtained in a non-invasive way. Perturbations such as pressure, temperature, and concentration can easily be applied to the experimental conditions. The obtained vibrational spectra are rich in information and a complete interpretation may be a difficult task. However, the assignments of vibrational features in complex systems have been much advanced by computer-based calculations. In particular, the ab initio techniques and the powerful link to vibrational spectroscopy are discussed further below.

2

Principles of IR absorption and RAMAN spectroscopy

The total energy of a molecule is composed of translational, rotational, vibrational, and electronic energies. Visible and ultraviolet light excite electrons in chemical bonds to higher molecular orbitals, while infrared light triggers chemical bond vibrations. IR spectroscopy works in transmitted light, while RAMAN spectroscopy uses laterally scattered light. IR spectroscopy is mainly used for polar molecules, while the RAMAN spectrum is well suited to image nonpolar bonds such as in hydrocarbons or graphite.

2.1

Permitted molecule vibrations and absorption bands

Each chemical bond in a molecule is a more or less harmonic oscillator. In normal modes, the motion is accomplished by changes in bond lengths, bond angles, or torsions angles. A molecule with N atoms can have 3 N − 6 (3 N − 5 for a linear molecule) vibrational degrees of freedom. For supramolecular compounds and polymers, this results in a large number of possible vibrational modes. Fortunately, the vibrational signature of repeated functional groups or repeated polymer units is almost equivalent so that IR

Methods and Instruments | IR and Raman Spectroscopy

137

spectroscopy is the method of choice to clarify the structure of a molecule and especially the class of substance. Typical vibrational frequencies in IR and RAMAN spectra of functional groups can be found in a tabulated form in handbooks. Commonly, the vibrational frequencies are given in wave numbers (
n ¼ l−1 ¼ n=cÞ in the region 50. . .4000 cm−1 (i.e., n  1011. . .1014 Hz). c is the speed of light. Infrared absorption. A vibrating molecule may interact in two distinctly different ways with electromagnetic radiation (Fig. 1). For a thorough description of the interaction between molecular vibrations and electromagnetic radiation, a quantummechanical treatment is needed. However, a brief description based on classical arguments provides a general understanding. Absorption of electromagnetic radiation in the IR region is possible if the frequency of the radiation corresponds to the frequency of a normal mode of vibration. The region of interest in IR absorption spectroscopy thus corresponds to wavelengths between 2 mm and 25 mm. It should be noted that not all normal modes are IR active. Only an oscillating dipole can absorb electromagnetic radiation of matching frequency. The absorbance A is proportional to the square of the change in dipole moment m during vibration: A
(dm/dq)2, where q is the displacement coordinate of the vibration. The absorbed radiation energy DE ¼ hn, where h is Planck’s constant, will eventually be reemitted or transferred to other molecules in the form of heat energy. IR absorption information is quantitative on an absolute scale according to the Lambert-Beer law, which relates the absorbance A to the sample thickness d, and the concentration of oscillators c, responsible for the absorption at the wavelength l. AðlÞ ¼ −lgT ðlÞ ¼ eðlÞ  c  d Here T is transmittance and e is the frequency-dependent absorptivity (molar absorption coefficient). RAMAN scattering. The second way in which electromagnetic radiation may interact with a vibrating molecule is by scattering of the irradiated light. There will be both elastic scattering (Rayleigh scattering) and inelastic scattering (RAMAN scattering), the latter involving a change in frequency (see Fig. 1) corresponding to the excitation or the deexcitation of a vibrational mode. Only a few photons, approximately 1 in 108–109, undergo RAMAN scattering. The change in the frequency of the RAMAN scattered light is equal to the frequency of a normal mode of vibration. As the strongest RAMAN scattering event, the long-wave Stokes lines have a frequency that is lower than that of the irradiated Rayleigh light (n0 − Dn). Inelastic scattering with a frequency higher than that of the irradiated light, the so-called anti-Stokes lines, is also possible (n0 + Dn). The latter process is weaker in intensity at ambient temperature, owing to the relative population of the ground and excited state levels, but sometimes it is interesting as the ratio between the Stokes and the anti-Stokes intensity can give a measure of the temperature in the scattering volume. As for IR absorption, not all the normal modes are RAMAN active. Only modes that induce a change in the polarizability a, are RAMAN active, that is (da/dq) 6¼ 0 is a condition for the RAMAN process. Accordingly, the selection rules for RAMAN and IR are different and the two techniques provide complementary information.

2.2

IR and RAMAN spectra

The complementarity of the two techniques is exemplified in Fig. 2 where both IR transmittance and RAMAN intensity for 1,1,3,3,5,5-meso-hexamethyl-2,2,4,4,6,6-meso-hexaphenyl calix[6]pyrrole (C6P) is shown. Calix[6]pyrrole belongs to a class of additives suggested for lithium battery application. The additives are tailor-made to form strong supramolecular complexes with anions in order to increase the cation transference number. Loosely speaking, the C6P vibrational modes that are strong in RAMAN are seen to be weak or missing in IR and vice versa. A first coarse understanding of the IR absorption and RAMAN scattering spectra can be obtained from a comparison with tabulated wave numbers for certain functional groups; for example, the stretching

Raman scattering IR absorption

Virtual states

ΔE antiStokes

Rayleigh

–3500

0

2850

400

514

3500

22 900

19 400

Stokes

Excited state Ground state

3500 Raman shift (cm–1) 630

Wavelength (nm)

15 900 Wave number (cm–1)

Fig. 1 The principles of infrared (IR) absorption and RAMAN scattering with transitions between the ground state and the excited state (depicted as arrows) of a normal mode vibration active in both IR and RAMAN.1

138

Methods and Instruments | IR and Raman Spectroscopy

IR spectrum (cm–1) 1000

1500

2000

2500

3000 1.0

0.8

0.8

0.6

N H NH H N

0.4

0.6

N H HN H N

0.4

IR transmittance

Raman intensity (a.u.)

500 1.0

0.2

0.2

0.0

0.0 500

1000

1500

2000

2500

3000

Raman spectrum (cm–1)

Fig. 2 The IR transmittance spectrum and RAMAN intensity for 1,1,3,3,5,5-meso-hexamethyl-2,2,4,4,6,6-meso-hexaphenyl calix[6]pyrrole (C6P). The inset shows the chemical structure of C6P.1

vibrations involving hydrogen (NdH and CdH vibrations) are found at the highest wave numbers, the deformation vibrations of the aromatic rings are found at intermediate wave numbers, and bending vibrations involving larger entities are found in the lowermost region. It is possible to determine the number of RAMAN active and IR active modes from simple considerations of the symmetry of the scattering molecule in its equilibrium configuration by use of character tables presented in standard textbooks on group theory and vibrational spectroscopy. For complex molecules such as C6P, a thorough assignment should however include computational techniques as discussed further below.

3 3.1

Instrumentation IR spectroscopy

Today instruments for IR absorption spectroscopy are of the so-called Fourier transform (FT) type, utilizing a Michelson interferometer to modulate the light and a computer-based fast FT analysis. The light source is polychromatic (e.g., a glowing ceramic element) giving a quasi-continuous spectrum in the frequency region of the molecular vibrations. The modulated light passing the sample compartment is sampled as an interferogram that can be converted to an IR spectrum by an inverse FT. In ordinary IR-transmission spectroscopy, the work is done in sending light through the sample. This setup often demands nontrivial sample preparation, for example, by making solutions that can be spread between two salt crystals or by compacting powders in a KCl matrix. Reflection techniques require less sample preparation than the transmission technique. Attenuated total reflection (ATR) on a diamond or zinc selenide crystal allows samples with very low transmission to be examined. This technique is based on reflection from a sample in good contact with a prism with a high refractive index. Using a critical angle of incidence for the light reaching the sample, total internal reflection is obtained. The raw ATR spectrum differs slightly from the transmission spectrum. However, after transformation with a Kubelka-Munk algorithm, the ATR spectrum can be compared with a transmission spectrum.

3.2

RAMAN scattering and microscopy

In RAMAN spectroscopy, named after the Indian physicist Chandrasekhara V. Raman (1888–1970, Nobel Prize 1930), a monochromatic laser light irradiates a sample in a glass tube, causing electron transitions and resulting in a molecular vibration spectrum. The infrared scattered light, which occurs transversely to the irradiation direction, displays a spectrum of weak lines with longer and shorter wavelengths alongside the reflected light line (Rayleigh scattering). The long-wavelength ‘Stokes lines’ occur as photons release energy to the rotational and vibrational levels upon striking the molecule; this causes a decrease in the frequency of the photon beam. The short wavelength, low intensity anti-Stokes lines arise when the excited molecule imparts rotational and vibrational energy to photons.

Methods and Instruments | IR and Raman Spectroscopy

139

Spectrometer Confocal hole

Notch filter

Laser

Microscope objective W

Sample

Fig. 3 A confocal microscope setup for RAMAN spectroscopy. The spatial resolution of the focal volume indicated is diffraction limited, resulting in about a micron in width and some microns in depth.1

In RAMAN spectroscopy, the light source is a highly monochromatic laser light usually in the ultraviolet to the near-IR region. Inelastic scattering like RAMAN follows the Rayleigh law with an intensity proportional to l−4. Decreasing the wavelength thus increases the scattering intensity but might eventually introduce problems with a strong fluorescence drowning the weak RAMAN signal when the excitation energy becomes comparable with the energy levels of the molecular electronic orbitals. The fluorescence gives rise to a high absorption of energy and a temperature increase in the sample. Sometimes, however, resonance conditions because of an overlap with electronic states give rise to a strong enhancement of the RAMAN signal, the so-called resonance RAMAN spectroscopy (RRS). An even stronger enhancement of the RAMAN signal, surface-enhanced RAMAN spectroscopy (SERS), can be obtained for molecules adsorbed onto a rough metallic surface or metallic nanoparticles. The dispersive elements used in RAMAN spectroscopy can be of grating type or of the Michelson interferometer type. As a rule of thumb, grating instruments are used in the visible to ultraviolet region of the electromagnetic radiation, often in combination with a two-dimensional detector for multichannel detection. In the near-IR region, the FT instrument is the preferred choice because a higher throughput of light partly balances the unfavorable Rayleigh law wavelength dependence. Both RAMAN and IR instruments can be coupled to optical microscopes in order to enhance the spatial selection but here RAMAN is superior for several reasons. First, the spatial resolution obtained in RAMAN is at least 1 order of magnitude higher than in IR as the resolution is directly dependent on the wavelength of the light source used. Second, the unique features of the confocal RAMAN microscope technique give both lateral resolution and depth resolution (see Fig. 3). The confocal hole only allows the RAMAN backscattered light from a chosen sample volume to reach the spectrometer. This constraint lowers the total intensity of the collected light but at the same time also reduces unwanted features. Optimizing the size of the focal hole is therefore a trade-off between optimal spatial resolution, total intensity, and signal to noise ratio.

4

Ab initio calculations of vibrational spectra

Over the last three decades, computational chemistry has developed into an everyday tool, much like spectrometers. The aim in this introductory text is to show how ab initio computational methods can be integrated to simplify and strengthen molecular spectroscopy data analysis.1 It is necessary with a basic description of the algorithms, simplifications, and methods in general used, to understand the possibilities and limitations of the approaches used to compute IR and RAMAN spectra.

4.1

Molecular models

The first step in computational chemistry is to find the relevant molecular (or more complex) equilibrium structure.1 Unfortunately, a systematic approach to this task is simply not possible: consider mapping the potential energy surface (PES) of a molecule of eight atoms, rendering 3 N−6 ¼ 18 degrees of freedom, with an appreciable resolution, 0.01 nm (10 pm) over a rather small range, 0.1 nm; yielding 1018 structures to be evaluated. Thus, even if each calculation would only take 1 s, the total amount of computational time would be billions of years! Therefore, optimization of the structure based on a refinement scheme toward zero energy gradient, thus a stable structure on the PES, and good choices of starting points are required for any system of appreciable size. Optimization routines are standard in computational chemistry programs for molecule-based calculations, but less so for periodic systems such as surfaces and crystals. The finally obtained structure is dependent on the starting guess and thus ‘chemical intuition’ is very valuable to obtain the same equilibrium structure as the spectroscopic measurement is made on. Notably, for comparison with spectra of molecules in solution, it might be necessary to either include the first solvation shell explicitly or simulate it by a continuum model. The former approach has the advantage of obtaining IR and RAMAN spectra of both the solute of interest and the solvent (possibly perturbed by the solute), while having the drawback of being computationally much

140

Methods and Instruments | IR and Raman Spectroscopy

more demanding—computational time approximately scales as N3. The latter approach has the advantage of being computationally much less demanding, at the cost of less data.

4.2

Computational algorithm

In the computation,1 only the energy within the time-independent Schrödinger equation will be considered, in shorthand notation equal to HC ¼ EC. From HC, simplifications can be introduced such as the single determinant C ultimately leading to the Hartree-Fock equations. The Hartree-Fock energy can be considered as a basic level of accuracy. Further simplifications lead to less accuracy and the possibility to treat systems with larger number of atoms N (e.g., semiempirical approaches). The addition of more details leads to solutions closer to the exact Schrödinger equation and thus severely reducing the size of possible N (e.g., electron correlation methods). Much of the following will be based on using the Hartree-Fock level of calculation. For less accurate levels, the modeling of spectra, especially RAMAN intensities, is not of acceptable standard, and more accurate levels are often time- and memory-consuming compared with the accuracy gain. Much of the reasoning below, however, is general for all methods. Combining the energy and minimum energy geometry x0 of a system with N nuclei (and belonging electrons), a second-order Taylor expansion can be made. The energy E(x0) can be chosen to be zero, and the second term, the energy gradient, is also zero in a stationary point like a minimum energy geometry obtained by an optimization routine: EðxÞ  Eðx0 Þ +

∂E 1 ∂2 E ðx − x0 Þ + ðx − x0 Þ 2 ðx − x0 Þ + . . . ∂x 2 ∂x

Therefore, E(x) is 12(x − x0)F(x − x0), where F is a 3 N  3 N matrix, the force constant matrix, containing the second derivatives of the energy with respect to the nuclei coordinates. The F matrix can then be introduced into the Hamiltonian H of the Schrödinger equation. By transforming to mass-dependent coordinates, by means of the G matrix containing the inverse square root of the atomic masses, and then diagonalizing the resulting FG matrix, eigenvalues, ei, and eigenvectors, qi, are obtained. In the resulting q coordinate system, which is the vibrational normal coordinates mentioned above, the Schrödinger equation can be separated into 3 N standard harmonic oscillators. As the eigenvectors are mass-weighted, the vibrational frequencies ni are obtained as follows: pffiffiffiffi ei ni ¼ 2p Only the fundamental vibrational modes, in a harmonic approximation, are obtained this way, but for most cases this is sufficient. For better quantitative comparisons versus experiments, linear scaling factors unique to the applied computational level, derived by using databases of a large number of molecules and vibrations, are applied to the computed frequencies. To obtain the IR and RAMAN intensities for the vibrational modes, the interaction with the electromagnetic radiation must be evaluated. Exact calculations of this kind are costly. Fortunately, most properties of a molecule can be modeled as a response of the wave function to a perturbation. For standard computations of IR and RAMAN intensities, the limiting case when the frequency of the perturbation field F goes to zero is used, a time-independent approach. In this regime, electronic transitions are not possible and thus some processes that in reality may disturb the experimental spectra such as fluorescence and phosphorescence are explicitly not accounted for. As mentioned above, the IR absorption intensity is proportional to the change in dipole moment m with respect to the geometry displacement along a normal coordinate. Similarly, RAMAN intensities are proportional to the change of the polarizability a along a normal coordinate. A Taylor expansion of the energy owing to a homogeneous field perturbation F appropriate at molecular length scales, gives: EðF Þ  Eð0Þ +

∂E 1 ∂2 E F + ... F+ ∂F 2 ∂F2

The energy E can also be written as: E ¼ –mF, and m ¼ m0 + aF + . . . Therefore, the Taylor expansion can be rewritten as: EðFÞ  Eð0Þ − m0 F −

1 2 aF + . . . 2

Hereafter, both IR and RAMAN intensities can be expressed as mixed derivatives of the energy:  2  2 2  2  3 2 ∂m ∂ E ∂a ∂ E IIR

IRAMAN

∂q ∂x∂F ∂q ∂x∂F 2 The derivative techniques for computing IR and RAMAN intensities are today included in most quantum-mechanical computational chemistry software packages. In general, calculations are performed on a set of molecular models and by comparison with experimental spectra the computed spectra serve to identify species or complexes present in the samples. Together with the previous simplifications mentioned, it is therefore natural that emphasis is on relative and qualitative agreement. An approximate upper limit of system size is currently some 100 atoms for which the total computational time is measured in weeks for a standard computer with a sufficiently large amount of internal memory. With the fast development of computer technology, the upper limit

Methods and Instruments | IR and Raman Spectroscopy

141

is extended rapidly. If higher orders are included in the Taylor expansions, it is also possible to address anharmonic frequencies, overtones, and combination bands, but this is much more demanding and difficult.

5

Application of vibrational spectroscopy in electrochemistry

The list of useful applications of vibrational spectroscopy in electrochemical science is almost endless. A few areas that might serve as inspiration to studies and simultaneously be a source of information to the reader on possibilities and limitations are summarized below.1

5.1

Charge carriers and structure in electrolytes for lithium batteries

To be able to use an electrolyte at its optimal conditions, it is important to know in detail the ionic species present as a function of solvent mixture and salt concentration. Furthermore, in order to rationally design new electrolytes, it is essential to know how the molecular structure of the electrolyte is coupled to the macroscopically observed properties. Spectroscopic methods, being non-invasive and applicable also in situ, have been used extensively for identifying these properties by means of shifts and splits in the spectra, primarily related to either the electrolyte matrix or the lithium salt anions. Liquid electrolytes. For nonaqueous liquid lithium battery electrolytes, the analysis situation for the matrix is rather straightforward: the solvent molecules, typically aromatic or linear carbonates such as ethylene carbonate (EC), propylene carbonate (PC), or dimethyl carbonate (DMC), may coordinate the lithium cations by electrostatic interactions—this is the interaction responsible for the salt solvation and thus the electrolyte formation. Here the spectroscopic analysis of the solvent bands contributes to the main solvation characteristics: the number and type of preferred ligands (in mixed solvents) of the lithium cations. For practical usage, the temperature dependence of these interactions can be very important; for example, at lower temperature, the salt solvation mechanism may be weaker and salt precipitation may occur, a phenomenon clearly observed by spectroscopic methods. The cation-anion interaction, however, is preferably studied directly by means of the vibrational bands owing to the anions of the lithium salt. Notably, vibrational analysis is seldom made using the Li-X stretching bands; this is due to a combination of low vibrational frequencies, 1–200 cm−1, a problem for IR spectroscopy, and the lack of RAMAN activity for these modes. Therefore, the analysis more often focuses on the shifts or splits of internal anion bands owing to the perturbation and symmetry reduction caused by an interacting lithium cation. For example, the totally symmetric CldO A1 stretching mode of the perchlorate anion and the symmetric SdO stretching mode of the triflate anion have been monitored to learn about the basic physical chemistry of lithium battery electrolytes. To differentiate between bands originating from anion charge carriers and those due to the formation of cation-anion ion pairs or larger aggregates, an often-used strategy is to start with samples of very low salt concentrations. In such samples, when using the highly dissociative lithium salts of lithium battery electrolytes, free anions can be expected to dominate. However, this also leads to very low concentrations of relevant oscillators in the sample, thus low signal to noise levels and therefore long measuring times. With increasing salt concentration, both the amount of ion pairs and all signals due to the anion should in principle increase. The caveat is that this can be difficult to monitor and that is why proper identification in most cases must be corroborated by ab initio computed spectra of both isolated anions and different models of possible ion pairs. Many studies show that even for very high concentrations, only a small part of the anions in the sample form spectroscopically observable contact ion pairs. Samples with a larger amount of ion pairs can be excellently correlated with a lower ion conductivity compared to electrolytes with presumably weaker cation-anion interactions. Thus, spectroscopy is a tool for comparative analysis of ion conduction of electrolytes. Polymer electrolytes. The changes in the bands from the matrix of liquid electrolytes are rather easy to interpret, whereas the changes in polymer-based electrolytes are more complex. Here also the solvation mechanism is different; the contribution of the matrix to the solvation free energy is largely based on entropy and therefore a common observation for polymer electrolytes is that salt precipitation may occur at elevated temperatures. However, there are also similarities with liquid electrolytes; the lithium salt is solvated by electrostatic interactions of the cation with the electronegative groups of the polymer, for example, the ether oxygen atoms of the polyethylene oxide (PEO) chain. For a polymer such as PEO, being semi-crystalline, the cation-oxygen interactions alter the balance between the amount of crystalline and amorphous regions, something that is reflected, for example, in shifts in bands owing to backbone CdOdC motions. When the cation is fully coordinated by a single chain, the preferred solvation shell of 5–6 oxygen atoms is created, as shown in Fig. 4. The signature of the wrapping is a structure with a strong RAMAN M–On breathing mode at
870 cm−1, analogous to the totally symmetric A1g mode of crown ether alkali metal salt complexes observed at 865–870 cm−1. Quantum-mechanical calculations have revealed the breathing mode as primarily owing to polymer backbone CH2 rocking. Thus, the polymer matrix is changed in at least two ways: it interacts with the cations and it also undergoes a structural change from crystalline to amorphous. The latter also causes local three-dimensional intrachain changes: the conformational equilibrium changes. The conformation of a polymer is described by sequences of the orientations of the nearest neighbors with respect to each other. For a polymer such as PEO, this strongly affects CH2-rocking, wagging, and scissoring vibrational bands, both positions and intensities. The wrapping mentioned earlier is a cation-induced conformational change, but changes can also be due to temperature or other conditions and may strongly affect the physical properties of the electrolyte. As battery electrolytes become more complex, the use of spectroscopy increases, although the wealth of information can be troublesome. Typical extensions are studies of materials based on mixed liquids and polymers as matrices: gels or plasticized systems.

142

Methods and Instruments | IR and Raman Spectroscopy

Raman intensity

(a)

800 (b)

850

900

950

Wave number (cm−1)

Fig. 4 (a) A computational model of a lithium cation wrapped by a poly(ethylene oxide) chain segment. (b) The computed RAMAN spectrum demonstrating the strong M–On breathing mode. After Jacobsson P; Johansson P. Measurement Methods, Vibrational Properties, Raman and Infrared. In Encyclopedia of Electrochemical Power Sources; Garche J, Ed.; Elsevier: Amsterdam, 2009; Vol. 3, pp 769–789.

Here one matrix component or both can contribute also to the salt solvation, an additional advantage apart from the outset macroscopic advantages, for example, mechanical properties. The concentration and temperature behavior of electrolytes, especially gels, can be very intricate. For a successful interpretation of the resulting multicomponent spectra, the support of calculation methods is highly desirable. As a parallel development, there has been interest in simple salt-polymer systems. Some studies have focused on the effect of sample preparation and treatment on the final macroscopic properties. Yet others have focused on the molecular level characterization of polymer electrolytes made by using new lithium salts whose spectral behavior is unknown at the outset.

5.2

Development of lithium battery salts

By integrating the use of both vibrational spectroscopy tools and ab initio calculations, rational ways to design new lithium salts have been proposed. By measuring the vibrational spectra for different concentrations, the ion-pairing properties of new anions with lithium cations can be revealed. With no references, and weak spectroscopic signals to be expected for extremely dilute systems where most anions can be assumed to be ‘free’, the use of ab initio calculations to assist in the analysis is crucial. Several computed spectra of both the anion and various possible ion pairs, ranked in probability by their relative energies, must therefore be produced and compared with experiments. The result of such an analysis is shown in Fig. 5, where the spectral behavior, the directions of the shifts, and the relative energies of the ion pairs all give an unambiguous picture. Ion-pair formation is extensive for the 4 mol L−1 lithium-pyrazole3,4,5-tricarbonitrile (LiPATC) in dimethyl sulfoxide (DMSO) solution, but not for 1.5 mol L−1, and the bidentate coordination to the ring nitrogen atoms (a) is the way these ion pairs are preferentially formed. Further design of anions should thus be targeted

Methods and Instruments | IR and Raman Spectroscopy

143

Verification of calculation by experiments

Relative Raman intensity

a

4.0 mol L–1 1.5 mol L–1

ΔE = 0

+79

b

+74

d

c

+41 2200 2210 2220 2230 2240 2250 2260 2270 Wave number (cm–1) Exp. shifts

a Comp. shifts

b c d

Fig. 5 The RAMAN spectra of lithium-pyrazole-3,4,5-tricarbonitrile (LiPATC) in dimethyl sulfoxide (DMSO) together with ab initio computed ion-pair structures. The relative energies are given in kJ mol−1 and relative RAMAN shifts in cm−1. From Jacobsson P; Johansson P. Measurement Methods, Vibrational Properties, Raman and Infrared. In Encyclopedia of Electrochemical Power Sources; Garche J, Ed.; Elsevier: Amsterdam, 2009; Vol. 3, pp 769–789.

at removing this coordination site—to further reduce ion-pair formation. Computational results promise less than 85% of the binding energy to remain for the same system with a fully substituted ring with no such sites available, that is, N5C10. However, such statements should be verified by spectroscopic studies and preferably also with in situ and ex situ studies of electrolytes and electrode surface in working battery cells.

5.3

Lithium battery cathodes, anodes, and current collectors

To optimize the performance of a lithium battery (cycle life, energy density, and safety), each component must be optimized separately and the compatibility between the materials must be considered, especially with respect to the interface between the electrodes and the electrolyte. Spectroscopy can be made using ex situ or in situ measurements on, for example, an electrode surface in contact with electrolyte, an electrochemical half-cell, or a real working battery. Structural changes in cathode materials. Thorough studies on typical lithium battery cathode materials (LixMn2O4, LixV2O5, LiNiyCo1−yO2) using both IR and RAMAN spectroscopy can identify structural changes, for example, owing to Jahn-Teller distortions of the cubic spinel MO6 octahedron, or creation of new phases upon cycling, for example, the l-MnO2 phase at high voltages, where x in LixMn2O4 approaches zero. Additionally, the effect of phase-impure cathode materials or structurally disordered materials, often the case as the synthesis conditions are difficult, can be monitored by mixing in a second phase in the spectroscopy sample. By in situ studies, surface reactions at the cathode surface can be performed, which may reveal the formation of surface-specific functional groups of compounds, either from the nonaqueous electrolyte or from a combined reaction. By applying different potentials and performing time-resolved measurements, evolution of species can be followed and possible reaction mechanisms can be suggested. The strong crystalline fields cause splits observable in IR spectroscopy (see Fig. 6) that correlate with the size and phase of surface species formed. Quite often the main stable compounds created at the surfaces are carbonate based, for example, lithium carbonate (Li2CO3), but also organic compounds have been detected as the major species. It is important here to stress the sometime difficult conditions for detection and the ambiguous role that different aftertreatments of the surfaces may have. Therefore, ex situ and in situ studies may show very different results and also analyses performed in different laboratories may differ substantially. The use of additional analysis methods is recommended. Surface chemistry of anode materials. For the anode surface, most often a lithium metal or a Li-carbon intercalation compound, the surface chemistry is even more important. For example, different in situ and ex situ IR techniques have been used to reveal a direct correlation between battery performance and anode surface chemistry. These studies are often very applied, for example,

144

Methods and Instruments | IR and Raman Spectroscopy

Absorbance

7 days

Initial

800

1200

1600

Wave number (cm–1)

Fig. 6 The evolution from a weak band at 1430 cm−1 to a double-maximum in the 1400–1600 cm−1 region, owing to crystal field splitting, signals the formation of a polycrystalline layer from the breakdown of a carbonate electrolyte on a cathode material.1

usage of materials of different qualities and always probing the resulting practical potential of the materials, but they are also based on a thorough analysis of complex vibrational spectra arising from species formed during cycling. Lithium metal is not a spectroscopically active material. However, the RAMAN spectrum of a graphite-based anode can be used to characterize the local surface disorder and also to determine the degree of lithium intercalation. Furthermore, the surface chemistry on the anode can be very rich as virtually all solvents can be reduced by lithium metal and the salts used often contribute to the surface species formed. Sometimes, a stable film that hinders further dissolution of the anode and breakdown of the electrolyte, but allows for the critical lithium-ion conduction, is formed. This is often quoted as a solid electrolyte interphase (SEI), which is crucial for battery performance. Originally, the SEI concept was suggested for lithium metal anodes, but further studies have shown it to be valid for even carbonaceous anodes. As for cathode materials, there is no consensus on the chemical composition of the anode SEI, and the situation is not due to different conditions alone. Here even RAMAN spectroscopy can be an invasive technique as dark samples, such as graphite, may absorb the laser light strongly, be locally very hot, and react. In general, special IR techniques have proven more powerful. Diffuse reflectance infrared Fourier transform (DRIFT) spectra have been recorded at different temperatures to identify surface species formed and thereby showing the difference in speed in the build-up of surface films. Typical observations are the formations of lithium formate, lithium carbonate, and lithium-ethylene carbonate reaction products. Additionally, CdF peaks owing to the binder material used on the anode surface can be observed. One practically important observation is that some electrolytes create a stable protective film on the anode surface that remains even when the electrolyte is changed to an inherently unstable one. In practice, this protection mechanism is shown spectroscopically for PC-based electrolytes. By using model systems, lithium cations have been found to act as electrophiles and catalyze surface reactions. Spectroscopic studies were also the first to show, very counterintuitive, that a highly reactive solvent, such as EC, is likely an advantage for carbonaceous anodes. This is explained by the reduction of the solvent on the carbon at high potentials and by the formation of an efficient passive layer before the intercalation processes start. Corrosion of the current collector. Comparatively few studies have been directed toward cell components outside the electrodes. Yet, for example, the current collectors must also be compatible with the electrodes and may even be affected by the electrolyte components. Aluminum is one current collector that was found to corrode by the presence of the LiTFSI salt (TFSI, bis(trifluoromethane sulfone)imide) in the electrolyte. This has been suggested, amongst other techniques, by IR identification of Al(TFSI)OH complexes together with organic products from the electrolytes on aluminum. Spectroscopy of entire cells. Even fewer studies have been made to encompass an entire cell spectroscopically: from the anode via the electrolyte and finally to the cathode. By using confocal RAMAN micro spectrometry, such studies can be performed in situ and with a resolution that allows detection of several important cell characteristics: the identification of new surface species on the anode, the salt concentration gradients formed in the electrolyte upon cycling, and the effects of lithium insertion in the cathode material. The drawback with whole battery studies is that a compromise must be made with respect to materials and cell design, but nevertheless this is probably the best way to obtain real battery data to compare with macroscopic properties of interest.

145

Methods and Instruments | IR and Raman Spectroscopy 5.4

Fuel cell membranes and water management

Vibrational spectroscopy methods are commonly applied also to characterize new membranes and electrode materials for fuel cell technology and to monitor changes in the chemical structure during and after operation of the fuel cell. Both RAMAN and IR along with computational methods have been used to study fuel cell electrocatalysis and chemisorption phenomena on the electrodes in both low- and intermediate-temperature fuel cells. As black body radiation rapidly becomes a problem in IR with increasing temperatures, RAMAN is the preferred choice at elevated temperatures. Considerable research efforts during the last two to three decades have been devoted to the development of proton-conducting polymer membranes for low-temperature fuel cells for mobile applications. In these efforts, the application of vibrational spectroscopy has an important role both for characterizing the membrane and for monitoring the chemical stability. The current polymer electrolyte fuel cell membrane of choice is still based on a fully fluorinated polyelectrolyte (e.g., Nafion®) swollen in water. Challenges for this technology are to a high degree related to complex water management; the proton conductivity is dependent on a high level of hydration and therefore operational temperatures below 100  C at atmospheric pressure. At the same time, the efficiency of the catalyst and its tolerance to impurities in the fuel increase with increasing temperature. Under fuel cell operation, a complex situation develops as the mass transport of different species drives the conditions far from equilibrium. In order to gain knowledge about the performance of electrolytes in real fuel cell environments, electrochemical, microscopic, and spectroscopic techniques have all been used frequently for pre- and post-operation analysis of the electrolytes. However, the most natural and informative way of gaining insight into a process is to analyze the fuel cell in real time. IR-diffuse reflection spectroscopy has been used to probe processes on working anodes and cathodes in fuel cells under operation. In situ confocal RAMAN microscopy has proven to be a useful technique for water management and transport processes in Nafion membranes under operation. Knowledge of water distribution and water content across the Nafion membrane is of importance to avoid conditions such as conductivity loss and membrane degradation. A schematic figure of a fuel cell adopted for a RAMAN confocal microscope is presented in Fig. 7, and the water profile as measured by RAMAN spectroscopy over the membrane crosssection. The cell can be used in hydrogen-oxygen mode or as shown in the results in a hydrogen-hydrogen mode. With the latter setup, the important transport processes, the osmotic drag and the back-diffusion, are still present. With minor modifications, the setup can be used for probing profiles in a direct methanol fuel cell (DMFC) also.

5.5

RAMAN spectra of carbon materials

The Raman spectrum (Fig. 8) allows for distinguishing between carbon allotropes.

• •

The spectrum of diamond exhibits a single absorption band at 1332 cm−1, which corresponds to the regular structure of CdC single bonds in sp3 hybridized carbon tetrahedra. Graphite powder has superior electrical conductivity compared to amorphous carbon (carbon black). The conductivity in the parallel graphene planes of linked benzene rings is high, but it is low between the layers. In graphite, the sp2-hybridized carbon in the planar layers of benzene-like rings is marked by the so-called G band at 1582 cm−1. Other bands suggest that the graphite (a)

(b) 7

Cover glass

O2

H2

in

H2

O2

in

out

out

Membrane–electrode assembly

Relative water content (I(water OH – stretch)/I(nafion reference))

Microscope objective 6 5 4 3 2

H2 flow I = 100 mAcm–2

1

I = 200 mAcm–2 0 Gasbacking, also used as current collectors

200

150

100

50

0

Distance across membrane (μm)

Fig. 7 (a) The experimental setup used for in situ measurements on a fuel cell under operation with a RAMAN confocal microscope. (b) The water profile over a Nafion 117 membrane under operation in hydrogen-hydrogen mode. The electroosmotic drag from the anode side (right) to the cathode side (left) skews the water profile and a drying out effect is observed with increasing current load.1

146

Methods and Instruments | IR and Raman Spectroscopy

Fig. 8 Qualitive shape of RAMAN spectra of carbon materials: diamond, graphene, graphite, nanotubes (SW ¼ single wall, MW ¼ multi-wall, CNT carbon nanotubes). Modified from: Kurzweil, P. Elektrochemische Speicher. Springer Vieweg: Wiesbaden (2018).

• •

6

lattice is less regularly structured than diamond. The intense 2D band or G’ at 2650 cm−1 in a graphene layer extends in the graphite layer stack to 2700 cm−1 and is not present in graphite oxide. The fine structure shows intra-molecular forces between adjacent one to four graphene layers. In activated carbon, the broad D-band appears around 1330 cm−1 due to spatially disordered sp2-hybridized rings. The half-width is a metric of the disorder in the carbon structure. The D-band rises in comparison to the G-band in the series from graphite (100 N/m) diamond probe with high applied force. In this case the scanned surface layer is mechanically removed. Fig. 4 illustrates examples of AFM topography images obtained on samples prepared as described above, supported by schematic illustrations of their setup in AFM. Note that during the ion beam milling/polishing Li+ ions may be sputtered out at higher yield than other elements of the active material, thus leaving the exposed surface of the cross-section Li depleted, which may affect surface sensitive AFM measurements.

Fig. 4 AFM topography images of flat and powder samples and respective schematic illustrations of their setup: (a) HOPG, (b) Ni-rich NMC single crystals pressed into Al foil, (c) cross-section of the positive electrode embedded into epoxy resin and containing NMC111 secondary particles, PVDF binder with conductive additive, and a current collector.

Methods and Instruments | Atomic Force Microscopy

8 8.1

153

In-situ imaging of surface topography In-situ imaging of solid electrolyte interphase (SEI) formation

One of the most common categories of AFM experiments in Li-ion battery research is in-situ/in-operando measurement of surface topography at the solid/liquid interface to study the dynamics of SEI formation. Most often, such experiments are conducted using HOPG as a sample intended to represent the carbonaceous anode. Pristine and atomically flat surface after peeling off the top layer allows for the observation of the dynamics of SEI layer formation and its evolution on the basal plane and terraces with nanometer resolution. In this experiment, HOPG is installed inside a liquid cell assembled in a two or three electrode configuration and connected to an external potentiostat/galvanostat (Fig. 2(b)). HOPG is connected as the working electrode and a piece of metallic Li is connected as the counter and reference electrode. The cell is filled with liquid electrolyte. As already mentioned before, HOPG is a model sample, onto which the SEI formation may not be necessarily representative of the SEI formed on various types of practical graphite. It has been established that the SEI composition and the related irreversible loss of capacity are influenced by the type and quality of carbon, which, among other properties, determine the edge-to-basal plane ratio. The early reduction of the electrolyte salt at the edge plane promotes a higher content of inorganic components in the SEI. In contrast, the preferential reduction of the electrolyte solvent on the basal plane at lower potentials vs Li+/Li results in a greater organic content. The HOPG surface mostly represents the basal plane. Although processes at step edges are observable and give certain understanding of SEI on the edge plane, the final SEI on HOPG has different thickness, composition, mechanical properties, and potential of nucleation as compared with commercially deployed carbonaceous anode materials. As an alternative, in-situ/ in-operando SEI experiments can be performed on cross-sections of composite electrodes embedded in epoxy resin.7 Owing to the extremely delicate nature of SEI, especially during its early stages of formation, measurements are usually performed in the dynamic mode. A cantilever is driven at the resonance frequency, which is reduced in liquid as compared with that in air due to the larger density of liquid and the added effective mass. Strong hydrodynamic interaction of the cantilever with the liquid results in a heavily reduced quality factor, usually by more than an order of magnitude. Fig. 5 illustrates vibration frequency spectra of the same cantilever in argon and in liquid electrolyte. Therefore, for effective implementation of the dynamic mode the cantilever should have a sufficiently high first resonance frequency, usually above 50 kHz. In liquid electrolyte the attractive van der Waals force between the tip and the sample is effectively screened, so the dynamic mode operates in the repulsive regime even with small amplitude of cantilever vibration. Because of that, cantilever should be sufficiently soft (k < 1 N/m), otherwise it may easily damage the growing SEI and even erase it upon scanning. However, in viscous electrolytes in the vicinity of sample surface the cantilever vibrations are additionally heavily damped due to additional hydrodynamic interaction between the cantilever and the liquid caused by the liquid squeezing between the tip and the sample.9 This fact prohibits use of very soft cantilevers (k < 0.1 N/m). Long high aspect ratio tips can reduce the dumping and allow for softer cantilevers, but the final choice of the cantilever largely depends of the electrolyte viscosity. During in-situ SEI imaging, there is a significant change in surface roughness illustrated in Fig. 6. Since the dynamic mode operates in the repulsive regime in liquid, meaning that the tip engages in repulsive interaction with the surface, the initial surface with small roughness should be scanned gently using a small vibration amplitude (a few nm) in order to minimize tip-induced perturbations at the solid-liquid interface. As the SEI grows, roughness increases and the scanning with small amplitude may not accurately trace the surface topography, resulting in imaging artifacts and possible tip-surface collisions. Therefore, the vibration amplitude, feedback gain, and setpoint should be adjusted accordingly in the course of scanning. Upon completion of the in-situ SEI experiment, the SEI thickness can be determined by measuring the step height between the SEI surface and the underlying graphite surface. The graphite surface is exposed by removing the SEI layer. If the removal is performed using the same relatively soft cantilever, it may be sufficient for HOPG and the basal plane SEI, which is primarily organic and soft. However, it may not be sufficient for SEI formed on other carbonaceous materials, where the SEI is more inorganic closer to the carbonaceous material and more organic near the SEI-liquid interface. In order to avoid ambiguity, it is preferable to remove the SEI layer in the conductive AFM mode using a stiff cantilever.7 The onset of current flow unambiguously indicates mechanical contact with the graphite surface.

Fig. 5 Cantilever thermal vibration frequency spectra in argon and in liquid electrolyte. First resonance frequency and the quality factor Q are indicated.

154

Methods and Instruments | Atomic Force Microscopy

Fig. 6 AFM topography image of HOPG acquired during SEI growth and cross-sections of its surface before and after SEI formation.

8.2

In-situ imaging of CEI formation

Not only the anodic SEI but also the cathodic electrolyte interface (CEI) is currently being investigated, particularly in relation to emerging high-voltage materials, where 4.7 V vs. Li+/Li oxidation potential of common organic electrolytes may be exceeded. It has been proposed that CEI formation may occur on surface of cathode active particles at potentials beyond electrolyte stability. In-situ AFM is a suitable tool for direct visualization of CEI formation. However, the main challenges are the absence of a suitable model sample and difficulty of in-situ AFM measurements on powder samples. Consequently, experimental results remain limited. One approach to sample preparation involves creating a thin film via magnetron sputtering or pulsed laser deposition. Recently, particles of active powder were embedded into an Al matrix for measurements.6 Cross-sections of composite electrodes were also utilized for in-situ AFM experiments.7 Once the sample is prepared, the in-situ AFM setup and measurements are the same as those for SEI formation measurements.

9

In-situ strain measurements

A key process in Me-ion batteries is intercalation and deintercalation of mobile ions, which results in molar volume change of active material and in mechanical stress, which may lead to mechanical damage such as fracture. Alloying type anode materials experience particularly large reversible volume change during electrochemical cycling vs Li+/Li, reaching 400% for Si. Of a particular interest are morphological and strain measurements of both the alloying materials and their SEI, which must withstand large cyclic strain without mechanical damage. These measurements can help identify failure moments and mechanisms. In-situ strain measurements technically are the same as in-situ SEI measurements, i.e. rely on topography mapping and its change as a function of cell potential. If SEI is not forming and there is no need to take special care about gentle layer prone to damage during measurements, the tapping mode with large amplitude and stiffer cantilevers can be used. Even the contact mode may be appropriate. Topographical changes due to strain development can be used to quantify the strain. However, it requires a reference point which remains constant, i.e. does not change its topography and spatial position during measurements. Moreover, the measured value is deformation which may be converted to strain only for certain well defined geometries, which require specific patterning of the sample topography. Fig. 7 illustrates an example of in-situ strain measurements with AFM.10 Here the evolution of morphology of silicon nanopillars with different diameters was monitored. It was shown that after the first cycle the pillars displayed a permanent volume expansion, after 5 cycles the pillars of all diameters became rough, and after 27 cycles pillars with diameters above 200 nm cracked. The pillars with diameters below 200 nm became fragile.

10

Electrochemical strain microscopy

The strain mapping resolution can be extended down to the nanometer scale using a method called Electrochemical Strain Microscopy (ESM).11 This is an ex-situ method which operates in the contact mode using an electrically conductive AFM probe, typically under gaseous atmosphere. An AC voltage is applied between the probe and the sample during scanning, and the resulting oscillating surface displacements are detected. Technically this method is equivalent to Piezoresponse Force Microscopy (PFM), but the origin of periodic surface oscillations is not ferroelectric. Upon applying a periodic electric field to the ionic conductor, the local electrochemical potential of mobile ions in a host lattice is altered, resulting in their local redistribution. This leads to expansion and contraction of material in a localized volume beneath the tip-sample contact due to Vegard’s effect.11,12 The oscillating surface displacements are detected by the AFM (Fig. 8(a)). Lateral resolution in ESM is limited by the tip apex diameter and is typically below 50 nm. The vertical resolution of the strain response is of the order of 10 pm owing to the lock-in amplification. As a result, a minute electrochemical response can be detected within individual grains and at grain boundaries of polycrystalline ionic conductors.

Methods and Instruments | Atomic Force Microscopy

155

Fig. 7 (a) Schematic of a-Si pillars on Ni substrate, (b) AFM image showing the as-fabricated array of pillars, (c) current profiles of the various potential holds on the first cycle of the a-Si pillars on Ni substrate, (d) In situ AFM images during lithiation and delithiation of a-Si nanopillars taken at several electrochemical potentials in the first cycle. Reproduced with permission from Ref. Becker, C. R.; Strawhecker, K. E.; McAllister, Q. P.; Lundgren, C. A. In Situ Atomic Force Microscopy of Lithiation and Delithiation of Silicon Nanostructures for Lithium Ion Batteries. ACS Nano 2013, 7, 9173–9182. Copyright © 2013, American Chemical Society.

Fig. 8 (a) Schematic illustration of the surface strain caused by the local change of the lattice parameter due to the change of Li concentration under the tip, and (b) maps of the diffusion coefficient and the ion concentration along with the topography in ionically active particles of LiMn2O4 cathodes. (a) Reproduced from Ref. Luchkin, S. Y.; Romanyuk, K.; Ivanov, M.; Kholkin, A. L. Li Transport in Fresh and Aged LiMn2O4 Cathodes Via Electrochemical Strain Microscopy. J. Appl. Phys. 2015, 118, 072016, with the permission of AIP Publishing. (b) Reproduced from Ref. Alikin, D.; Romanyuk, K.; Slautin, B.; Rosato, D.; Shur, V.; Kholkin, A. Quantitative Characterization of the Ionic Mobility and Concentration in Li-Battery Cathodes Via Low Frequency Electrochemical Strain Microscopy. Nanoscale 2018, 10, 2503–2511 with permission from the Royal Society of Chemistry.

In order to enhance the weak signal-to-noise ratio, additional resonance amplification is used by implementing band excitation or DART techniques. These methods utilize the contact resonance frequency, which is 3–5 times higher than the free resonance frequency of the cantilever and is usually well above 100 kHz for practically used probes. ESM response can potentially be used to quantify local concentration and diffusion coefficient of mobile ions through time spectroscopy, where the ESM response is recorded at a single point as a function of time after a pulse of DC voltage. However, at high frequency besides Vegard’s strain a number of spurious non-Vegard responses such as electrostriction and electrostatics contribute to the measured data, making quantification and interpretation challenging. Recently, low frequency (1–20 kHz) ESM was proposed as the artifact-free and reliable method to quantify ESM response and calculate ionic diffusion and concentration with nanoscale resolution.13 Fig. 8 illustrates results of low-frequency ESM mapping on individual LiMn2O4 particles with quantitative results for Li-concentration and diffusion coefficients.

11

Nanomechanical imaging and spectroscopy

One of the inherent degradation mechanisms in active battery materials is mechanical failure. Reversible Li intercalation and deintercalation during battery operation results in cyclic volume change of host materials and cyclic mechanical stresses. This

156

Methods and Instruments | Atomic Force Microscopy

process leads to particles cracking and fracturing, loss of electrical contact, and exclusion from the reversible electrochemical processes. It is pronounced in polycrystalline particles such as secondary particles of transition metal oxide positive electrodes. Mechanical failure of SEI through cracking and delamination leads to additional irreversible capacity loss. SEI in this regard is especially challenging object to study because it is a thin film with nonuniform composition and mechanical properties. AFM allows qualitative and quantitative measurement of mechanical properties of materials at the nanoscale. This includes Young’s modulus, contact stiffness, storage and loss modulus, adhesive forces, and hardness. Due to the small size of AFM probes and pN force sensitivity, it allows for accurate measurements of nanomechanical properties of individual grains, thin films, and delicate objects with nm spatial resolution. There is a number of AFM based methods that allow mapping of mechanical properties, which can be divided into two groups based on detection principle: (1) deflection techniques and (2) resonance techniques. Deflection techniques, also known as quasistatic techniques, rely on acquisition of force curves. Resonance techniques, also known as dynamic techniques, rely on measurement of the contact resonance shift. Both deflection and resonance techniques can be operated in gaseous and liquid media.

12

Deflection techniques

Among the deflection techniques (force modulation, force spectroscopy, and indentation), force spectroscopy14 is the most common. In this method, an AFM probe positioned at a specific distance above the sample surface is lowered onto the sample and pressed against it at a defined rate (nm/s) until a threshold cantilever deflection value (setpoint) is reached (Fig. 9(a)). Alternatively, the tip-sample distance can be selected as a threshold. Upon reaching the threshold point, the cantilever is retracted to its initial position above the surface. The acquired dependence of the cantilever deflection on the z-sensor displacement is a deflection-distance curve, containing information about the tip-sample interaction forces. To convert the deflection-distance curve into a force-distance curve and extract quantitative values of mechanical properties (Fig. 9(b)), the cantilever spring constant in N/m and deflection sensitivity in nm/V must be determined through a calibration routine. Each cantilever should be calibrated within the AFM before measurements. The deflection sensitivity is determined by measuring the cantilever deflection on a rigid surface, which is presumed to remain undeformed under the applied force. The spring constant is then determined using the thermal noise method. For rectangular cantilevers both the deflection sensitivity and spring constant can be determined by the Sader method. Quantitative values of Young’s modulus can be calculated by fitting the acquired force curves by a theoretical model. There are several theoretical models, among which the most widely used are Hertz/Sheddon, Johnson-Kendall-Roberts (JKR), DerjaguinMüller-Toporov (DMT), and Oliver-Pharr.14 Hertz (spherical indenter)/Sneddon (conical indenter) model assumes purely elastic tip sample interaction and does not consider effects of surface roughness and adhesion. This model is used for small strains within the elastic limit, where the elastically deformed region of the sample is homogeneous and the contact area is significantly smaller that the characteristic size of the interacting bodies. The JKR model extends the Hertz model by incorporating the effects of adhesion forces between the AFM tip and the sample surface in the contact area only. This model is applied when the tip radius is larger than the indentation depth, typically in cases involving relatively soft elastic material of the sample. The DMT model extends the Hertz model by including the effects of adhesion forces between the AFM tip and the sample surface only outside the contact area. This model works better than the JKR on relatively stiff elastic materials with weak but not negligible adhesion. Finally, the Oliver-Pharr model is applied for the case of elastic-plastic deformation. Note that the deflection techniques are capable to quantify elastic properties but are not well-suited for quantifying viscous properties. How accurate are these quantitative nanomechanical properties? The error of spring constant calibration within 5–10% is considered appropriate for most commercial cantilevers with rectangular levers.15 Deflection sensitivity calibration gives another
10% error.

Fig. 9 (a) The deflection - Z-piezo position curve including approaching and retracting parts, and (b) the converted force – distance curve.

Methods and Instruments | Atomic Force Microscopy

157

These are the lower bound estimations. Furthermore, one of the parameters in the theoretical models for force curve fitting is a tip-sample contact geometry, which for nanoscale contacts is a source of significant uncertainty due to sample surface roughness and not ideal tip apex shape. This contact geometry uncertainty increases with decrease of the tip radius of curvature and the tip-sample contact area. In this regard, precise qualitative measurements require additional calibration on a reference sample with known mechanical properties, which ideally should be similar to expected properties of the sample of interest. Tip apex shape and size can be determined using special calibration samples and routines such as blind estimation. In case of AFM nanoindentation, it is important to be aware of the indentation size effect, which manifest itself in increased hardness with decreased impression size or load force, i.e. the nanoscale hardness is typically larger than the macroscale one. Another significant factor for quantification is the selection of an appropriate probe. The repulsive part of the force-distance curve is most sensitive when the cantilever stiffness is close to the tip-sample contact stiffness. Therefore, AFM probes with appropriate stiffness should be used for materials with different elastic properties, which requires an educated guess. With commercially available probes, the force spectroscopy method enables quantitative measurements of Young’s modulus within the range of 103–1011 Pa. Fig. 10(a–i) illustrates an example of force spectroscopy measurements of SEI on metallic Li.16 It demonstrates that SEI comprises of softer organic rich (O) and stiffer inorganic rich (I) parts. Fig. 10 (i) shows that Young’s modulus of different regions vary from single GPa to more than 200 GPa, which highlight the problem of accurate quantification of mechanical properties of heterogeneous thin films. If the cantilever stiffness is optimal for the stiff region, it might not be sensitive enough for the soft regions, and vice versa. An AFM probe with an array of cantilevers with different spring constants on a single chip is a viable solution for highly heterogeneous samples. Besides mechanical properties, force spectroscopy was used to study structure and properties of solid-liquid interfaces such as electrical double layer. Fig. 10(j-l) illustrates an example of electrical double layer structure of an ionic liquid on HOPG measured by force spectroscopy.17 AFM is capable of performing spatial mapping of mechanical properties by acquiring force curves over a grid of points on a sample surface. Thus, spatial distribution of mechanical properties can be obtained. Modern AFMs can quickly acquire force curves with frequency of up to a few kHz in the pulsed force mode. This enables mapping of mechanical properties with resolution and rate typical for AFM topography imaging.

13

Resonance techniques

Resonance methods in AFM nanomechanical mapping are a group of techniques that rely on measuring the contact resonance frequency shift of the AFM cantilever to determine the mechanical properties of materials at the nanoscale. This group of methods can be further divided into two subgroups: (1) contact resonance AFM and (2) bimodal AFM. In the contact resonance AFM, a probe scans the sample surface in contact mode while simultaneously being excited at its contact resonance frequency, which depends on the mechanical properties of the tip-sample contact. By measuring the contact resonance frequency shift and changes of the resonance quality factor, both elastic and viscous mechanical properties can be quantified. It is worth mentioning that a similar approach was used earlier in atomic force acoustic microscopy (AFAM) through excitation of the sample by a piezoelectric transducer and in ultrasonic atomic force microscopy (UAFM) through ultrasonic frequency excitation of the cantilever. However, these early methods allow for the measurement of only elastic properties (Young’s modulus), but not viscous (storage modulus, loss modulus, and loss tangent). In bimodal AFM18 the cantilever is excited simultaneously at two resonance frequencies, typically the first and second eigenmodes. The first frequency operates as the AM tapping mode, obtaining surface topography and loss tangent by acquiring amplitude and phase response. The second frequency is used to obtain elastic properties. In the method called bimodal dual AC (Fig. 11(a)) the amplitude and phase of the second frequency provide enhanced contrast related to mechanical properties as compared with the first frequency, but the results are qualitative. In the method called AM-FM viscoelastic mapping, the amplitude and frequency of the second resonance, operated with frequency modulation (FM) feedback, provide quantitative results for stiffness or elasticity, which are calculated from the second eigenmode frequency shift. AM-FM can measure Young’s modulus in a range from hundreds of kPa to over 100 GPa. However, the AM-FM Young’s modulus depends on several parameters including the second eigenmode frequency and contact area, which necessitates for a calibration on a reference sample. Note that the tip-sample interaction has to be repulsive. In gaseous environment, to enter the repulsive regime, the cantilever driving amplitude must be sufficiently large to overcome the attractive part of the tip-sample interaction potential. In this case, the AFM tip is susceptible to mechanical damage or quick contamination, which may affect measurements especially on stiff or sticky samples. AFM probes with fully diamond or diamond coated tips are less prone to both mechanical damage and contamination, making them more reliable for resonance techniques. When resonance techniques are operated under the liquid environment, there are two important differences with the gaseous environment to consider. First, in liquid media where the attractive force is suppressed, smaller driving amplitudes may be used to work in the repulsive mode, which favors measurements of delicate materials. Second, interpretation of both qualitative and quantitative results may be challenging. For example, for soft cantilevers (k < 1 N/m) working in a tapping mode at the first resonance frequency, the second eigenmode is briefly excited near times of tip-sample contact, which implies multimodal dynamics of soft AFM probes in liquid.19

158 Methods and Instruments | Atomic Force Microscopy Fig. 10 Characteristic deflection-displacement curves acquired on SEI/Li surface being (a) soaked, and polished with (d) single and (g) multiple potential steps in electrochemical stripping-plating (ESP) process; corresponding AFM topography images (b, e, h), and histograms of Young’s modulus (c, f, i). Insets in (c) and (f ) show force-distance curves after conversion from the deflection-displacement curves. (I) stands for inorganic-rich and (O) for organic-rich/mixed SEI layers. Force spectroscopy of electric double layer on HOPG in [Emim+][Tf2N−] ionic liquid: (j) single approach and retract force-tip height curve showing multiple ion layers, (k) single force-separation curve converted from the force-tip height curve in (j), and (l) 2D histogram of 50 consecutively measured force curves. (a-i) Reproduced from Ref. Gu, Y.; Wang, W. W.; Li, Y. J.; Wu, Q. H.; Tang, S.; Yan, J. W.; Zheng, M.S.; Wu, D. Y.; Fan, C. H.; Hu, W. Q.; et al. Designable Ultra-Smooth Ultra-Thin Solid-Electrolyte Interphases of Three Alkali Metal Anodes. Nat. Commun. 2018, 9, 1339 (CC BY 4.0). (j, k, l) Reproduced with permission from Ref. Black, J. M.; Walters, D.; Labuda, A.; Feng, G.; Hillesheim, P. C.; Dai, S.; Cummings, P. T.; Kalinin, S. V.; Proksch, R.; Balke, N. Bias-Dependent Molecular-Level Structure of Electrical Double Layer in Ionic Liquid on Graphite. Nano Lett. 2013, 13, 5954–5960.

Methods and Instruments | Atomic Force Microscopy

159

Fig. 11 (a) Schematic image of bimodal AFM and (b) topography images and Young’s modulus maps of the cross-sections of LiCoO2-based positive electrode. Reproduced with permission from Ref. Sakai, H.; Taniguchi, Y.; Uosaki, K.; Masuda, T. Quantitative Cross-Sectional Mapping of Nanomechanical Properties of Composite Films for Lithium Ion Batteries Using Bimodal Mode Atomic Force Microscopy. J. Power Sources 2019, 413, 29–33.

160

Methods and Instruments | Atomic Force Microscopy

Overall, different resonance techniques are able to obtain either quantitative or qualitative viscoelastic properties with high spatial resolution and at high scanning rate, but their application to energy materials has been very limited. Fig. 11(b) illustrates an example of AFAM application to map Young’s modulus of LiCoO2 particles, binder, and a current collector at different state of charge on cathode cross sections.20

14

Conductive AFM

To maintain charge balance during battery operation and provide electrons to the external load, ionic current between the anode and cathode through the electrolyte inside the battery is balanced by the electronic current between the anode and cathode through the external circuit. Mechanical failure or undesirable phase transition in the active electrode material can hinder electronic conductivity and isolate a fraction of the active material from electrochemical operation, thus reducing the cell capacity. Besides, in electrode materials such as transition metal oxide cathode materials, electronic conductivity varies with Li concentration as electrons (un)occupy d-bands of transition metals in the host lattice. Thus, a local electronic current sensitive method capable of measuring electronic conductivity at the scale of individual particles, grains, and grain boundaries can be a valuable tool for studying the origins of battery degradation and operation mechanisms, complementing macroscopic electrochemical methods. AFM enables such measurements via conductive AFM (c-AFM) and its variations such as Scanning Spreading Resistance Microscopy (SSRM), in which a bias voltage is applied between a conductive AFM probe and a sample mounted on a conductive substrate, and the electric current flowing through the tip-sample junction is measured (Fig. 12). C-AFM operates in a contact mode, simultaneously acquiring topography and current maps. It can also work in combination with the pulsed force method, which may be preferred for delicate samples or high loading forces. In this case, damage to both the sample and the tip is reduced due to neglecting of the shear force. In addition to imaging, c-AFM can be operated in the spectroscopic mode, when the voltage is swept at each point, and current-voltage curves are acquired. C-AFM is typically operated under gaseous environment/ambient atmosphere or under vacuum. For battery materials sensitive to water, c-AFM measurements should not be performed under humid atmosphere. The conductive probe is a key component of c-AFM. These probes are either made of conductive materials or coated with conductive materials. Common types of conductive AFM probes include metal-coated probes (e.g., gold, platinum), doped diamond-coated probes (e.g., boron-doped diamond), conductive carbide and nitride coated probes (e.g., W2C and TiN), PtSi coated probes, fully metallic probes made of Pt or PtIr alloy, and single crystal doped diamond probes. The choice of a probe depends on a number of factors such as the type of electrical conductivity and wear resistance. Spatial resolution in c-AFM is determined by the tip-sample contact area, which depends on the applied force and the tip radius of curvature. The contact area between the AFM tip and the sample is in the nm range, which means the resistance switches into the point contact regime and becomes highly sensitive to properties of the material in the vicinity of the contact, the surface state, and the contact area. In the Ohmic approximation, three models of the point contact resistance can be considered, namely Maxwell, Sharvin, and Landauer–Buttiker. The Maxwell model is valid when the contact area is large compared to the mean free pass of electrons. In this case, the point contact resistance R ¼(rsample/2pa)arctan(2b/a), where rsample is the local resistivity of the sample, a is the contact radius, and b is the sample’s thickness. If b  a, then R ¼rsample/4a. The Sharvin model is valid when the mean free pass of charge carriers is larger than the contact area. In this case R ¼ (4rsamplel)/(3pa2), where l is the mean free pass. The Landauer–Buttiker model is valid in the extreme case of ballistic transport when only a few atoms make the contact. In this case the resistance is proportional to the transmission coefficient of the conducting channel. The situation is even more complicated because the point contact is often rectifying in nature, and the current-voltage dependence is not linear. For battery materials an additional contribution can be associated with the local redistribution of mobile ions and defects under the probe.21 This makes quantitative interpretation challenging, but not impossible if a comparison with a reference sample is performed. In battery research, qualitative results of c-AFM are often sufficient for practical purposes, such as observing electrically disconnected active particles, insulating phases, or changes in conductivity related to changes in concentration of mobile ions.

Fig. 12 Schematic image of CAFM and examples of topography and current images obtained on a cross section of a composite electrode comprising NMC111 active particles.

Methods and Instruments | Atomic Force Microscopy

161

However, errors arising from the contribution of surface states and surface roughness may become intolerable. These errors can be significantly reduced by using higher loading force and higher bias voltage. Under such conditions, metal coated and fully metal probes are prone to rapid degradation. Stiff wear resistant probes such as W2C or conductive diamond coated are preferred. It is important to note that different tip materials may yield different results even on the same sample. For example, a degenerately p-type doped diamond probe will show different results on n-type or p-type doped semiconductors. Consequently, samples measured with different probes may not be directly comparable. The current density through the point contact can be high enough to cause local heating. Additionally, the mN force applied to the nm area results in the GPa pressure, which, when combined with local heating, can induce local phase transitions and plastic deformations. Although these effects are challenging to evaluate quantitatively, they do not appear to significantly impact experimental results on battery materials.

15

Potential sensitive AFM

Besides current-sensitive AFM, electronic properties of battery materials can be measured by potential-sensitive AFM techniques such as electrostatic force microscopy (EFM) and Kelvin Probe Force Microscopy (KPFM). ESM and KPFM are similar in implementation and operation and based on detection of electrostatic force or force gradient between the conductive AFM probe and the sample. Both methods have been applied to study battery materials, but quantitative interpretation of KPFM results is more straightforward and less challenging than that of EFM. Here, implementation of KPFM for battery research will be discussed. KPFM is a non contact dynamic AFM mode in which the conductive probe-sample system constitutes a capacitor connected through a DC bias. Because the AFM probe is a force-sensitive sensor that responds to the electrostatic force, the electrostatic interaction between the conductive AFM probe and the sample is modulated by an external AC voltage, resulting in the modulated vibration of the AFM probe. The force and hence the probe vibrations and nullified by the application of an external DC bias voltage equal and opposite to the local contact potential difference (CPD) between the tip and the local area of the sample surface below the tip. The periodic modulation allows the use of resonance and lock-in amplifications in order to increase the signal-to-noise ratio and achieve high potential and spatial resolution. KPFM can be implemented in gas and liquid media22 in different modes, including a single pass, a contact mode, a pulsed force, but the most frequent implementation is a two pass mode, in which each line in an image is scanned twice (Fig. 13). During the first pass, the surface topography is acquired (usually in the tapping mode). During the second pass the probe, lifted to a certain distance above the surface, follows the recorded topography profile and measures the contact potential difference as described below. Two main modes of KPFM operation are amplitude modulation (AM) and frequency modulation (FM). In the !
2 ! is modulated AM-KPFM during the second pass the probe is not mechanically actuated. The electrostatic force F ¼ −r CV2 by the applied bias voltage V ¼ Vdc + Vac sin(ot), where the modulation frequency o is selected close to the resonance frequency ho0 of the AFM probe in order to utilize resonance i amplification. The resulting electrostatic force V 2ac V 2ac 2 Fz ¼ − 12 ∂C ∂z ðV CPD + V dc Þ + 2 + 2ðV CPD + V dc ÞV ac sinðot Þ − 2 cosð2ot Þ is comprised of three components: static, at the modulation frequency, and at the twice modulation frequency. The force component at the modulation frequency

Fig. 13 Schematic image of KPFM operation in 2 pass AM configuration and examples of obtained topography and surface potential maps on a cross section of a composite electrode comprising NMC111 active particles (the same as in Fig. 12).

162

Methods and Instruments | Atomic Force Microscopy

Fo ¼ − ∂C ∂z ðV CPD + V dc ÞV ac sinðot Þ mediates cantilever oscillations, which are suppressed by nullifying the force through the application of an external DC bias voltage (Vdc) equal and opposite to the local contact potential difference (VCPD) between the tip and the sample. In the FM-KPFM during the second pass the probe is driven mechanically at the first resonance frequency o0. The electrostatic force is modulated by the voltage with low modulation frequency o, usually in the 1–3 kHz range. The small electrostatic force 1 ∂F gradient shifts the oscillation resonance frequency by Do ¼ − 2k ∂z , resulting in the appearance of additional oscillation sidebands adjacent to the resonance frequency. Oscillation amplitudes at o0  o are detected and suppressed by nullifying the force gradient similarly to the AM-KPFM. For metals and semiconductors, the local contact potential difference is related to the local work function of the sample as W −W V CPD ¼ samplee tip , while for insulators it is related to local surface charges. In order to calculate work function of the sample, it is prerequisite to know work function of the tip. The tip is coated with known material which work function can be found in literature. However, work function of metal coatings may notably vary.23 Moreover, it is sensitive to damage and contamination, so calibration on a reference sample is required. A standard reference sample for the calibration is HOPG with work function of the basal plane 4.6 eV. The long range of electrostatic force interaction results in contribution of the remote parts of the cantilever—the cone part of the tip and the cantilever beam—into the measured signal, in addition to the local contribution of the tip apex. The resulting VCPD is the weighted average of the tip apex, tip cone, and the beam contributions. Because the force gradient decays faster with distance than the force, contribution of the remote parts of the probe into the measured signal in FM-KPFM is lower as compared with the AM-KPFM. Therefore, FM-KPFM offers higher lateral resolution and more accurate value of the local contact potential difference. At the same time FM-KPFM suffers from lower sensitivity than AM-KPFM and has lower signal-to-noise ratio, which leads to the need for applying higher AC voltages. For semiconducting samples elevates AC voltages may result in tip induced band bending on the surface, which affect the results. Enhanced potential and spatial resolution can be achieved by increasing the contribution from the tip apex. This can be accomplished by reducing the tip-sample distance, using a blunt tip with a larger apex area, flattening the tip apex,24 and employing probes with long high aspect ratio tips. Electrostatic shielding of the probe’s remote part was proven to be effective, but electrostatically shielded probes are not commercially available. KPFM is less damaging to both the sample and the tip as compared with c-AFM, which results in reduced tip wear. This allows the use of metal coated probes. Typically, force modulation probes with spring constants in the range of 0.5–10 N/m are employed. Fig. 14 illustrates in-situ KPFM measurements on a cross section of all solid state Li-ion battery, showing direct visualization of the change in potential distribution of a composite electrode during battery charging.25

Fig. 14 Results of in situ KPFM of a cross section of a composite electrode: (a) schematic illustration of the experimental setup inside the N2 flow glove box; (b) charging characteristics of the SS-LIB cell; (c) topography image; (d) CPD image before charging (state A); (e) CPD image after charging (state B). Reproduced from Ref. Masuda, H.; Ishida, N.; Ogata, Y.; Ito, D.; Fujita, D. Internal Potential Mapping of Charged Solid-State-Lithium Ion Batteries Using In Situ Kelvin Probe Force Microscopy. Nanoscale 2017, 9, 893–898 (CC BY-NC 3.0).

Methods and Instruments | Atomic Force Microscopy

16

163

Additional remarks

Modern Atomic Force Microscopy offers a number of other methods potentially useful for electrochemical and battery research. Near-field methods such as Tip Enhanced Raman Spectroscopy (TERS) and Infrared AFM can be used for nanoscale chemical imaging. Scanning Microwave Impedance Microscopy (SMIM) can measure sample’s permittivity, conductivity, and capacitance/ differential capacitance. Local electrochemical measurements in liquid media are possible with specially designed conductive isolated probes, which allow to reduce stray current originated from the remote parts of cantilever and measure local electrochemical current by the tip apex. Other well established methods such as magnetic and thermal AFM are less pertinent for battery research and therefore rarely utilized.

17

Conclusion

In conclusion, diverse AFM based techniques are capable of addressing the challenges of characterizing atomic and nanoscale structures and electrochemical functionalities in energy storage materials. By providing localized insights into the microscopic processes during electrochemical operation, AFM contributes to the development of novel materials and the analysis of various properties at the microscopic level.

References 1. Inaba, M.; Siroma, Z.; Funabiki, A.; Ogumi, Z.; Abe, T.; Mizutani, Y.; Asano, M. Electrochemical Scanning Tunneling Microscopy Observation of Highly Oriented Pyrolytic Graphite Surface Reactions in an Ethylene Carbonate-Based Electrolyte Solution. Langmuir 1996, 12, 1535–1540. 2. Giessibl, F. J.; Hembacher, S.; Bielefeldt, H.; Mannhart, J. Subatomic Features on the Silicon (111)-(7x7) Surface Observed by Atomic Force Microscopy. Science (80-.) 2000, (289), 422–425. 3. Voigtländer, B. Scanning Probe Microscopy. In NanoScience and Technology; Springer Berlin Heidelberg: Berlin, Heidelberg, 2016. 4. Meyer, E.; Gyalog, T.; Overney, R. M.; Dransfeld, K. Nanoscience: Friction and Rheology on the Nanometer Scale; World Scientific, 1998. 5. Soppera, O.; Carré, C. Pulsed Force Mode AFM Characterization of photopatterned Polymers Films. In 2005 5th IEEE Conference on Nanotechnology; vol. 2; IEEE Computer Society, 2005; pp. 705–708. 6. Liu, R. R.; Deng, X.; Liu, X. R.; Yan, H. J.; Cao, A. M.; Wang, D. Facet Dependent SEI Formation on the LiNi0.5Mn1.5O4 Cathode Identified by In Situ Single Particle Atomic Force Microscopy. Chem. Commun. 2014, 50, 15756–15759. 7. Luchkin, S. Y.; Lipovskikh, S. A.; Katorova, N. S.; Savina, A. A.; Abakumov, A. M.; Stevenson, K. J. Solid-Electrolyte Interphase Nucleation and Growth on Carbonaceous Negative Electrodes for Li-Ion Batteries Visualized with In Situ Atomic Force Microscopy. Sci. Rep. 2020, 10, 1–10. 8. Park, S. Y.; Baek, W. J.; Lee, S. Y.; Seo, J. A.; Kang, Y. S.; Koh, M.; Kim, S. H. Probing Electrical Degradation of Cathode Materials for Lithium-Ion Batteries with Nanoscale Resolution. Nano Energy 2018, 49, 1–6. 9. Kawakami, M.; Taniguchi, Y.; Hiratsuka, Y.; Shimoike, M.; Smith, D. A. Reduction of the Damping on an AFM Cantilever in Fluid by the Use of Micropillars. Langmuir 2010, 26, 1002–1007. 10. Becker, C. R.; Strawhecker, K. E.; McAllister, Q. P.; Lundgren, C. A. In Situ Atomic Force Microscopy of Lithiation and Delithiation of Silicon Nanostructures for Lithium Ion Batteries. ACS Nano 2013, 7, 9173–9182. 11. Morozovska, A. N.; Eliseev, E. A.; Balke, N.; Kalinin, S. V. Local Probing of Ionic Diffusion by Electrochemical Strain Microscopy: Spatial Resolution and Signal Formation Mechanisms. J. Appl. Phys. 2010, 108, 053712. 12. Luchkin, S. Y.; Romanyuk, K.; Ivanov, M.; Kholkin, A. L. Li Transport in Fresh and Aged LiMn2O4 Cathodes Via Electrochemical Strain Microscopy. J. Appl. Phys. 2015, 118, 072016. 13. Alikin, D.; Romanyuk, K.; Slautin, B.; Rosato, D.; Shur, V.; Kholkin, A. Quantitative Characterization of the Ionic Mobility and Concentration in Li-Battery Cathodes Via Low Frequency Electrochemical Strain Microscopy. Nanoscale 2018, 10, 2503–2511. 14. Butt, H.-J.; Cappella, B.; Kappl, M. Force Measurements with the Atomic Force Microscope: Technique, Interpretation and Applications. Surf. Sci. Rep. 2005, 59, 1–152. 15. Ohler, B. Cantilever Spring Constant Calibration Using Laser Doppler Vibrometry. Rev. Sci. Instrum. 2007, 78. 16. Gu, Y.; Wang, W. W.; Li, Y. J.; Wu, Q. H.; Tang, S.; Yan, J. W.; Zheng, M. S.; Wu, D. Y.; Fan, C. H.; Hu, W. Q.; et al. Designable Ultra-Smooth Ultra-Thin Solid-Electrolyte Interphases of Three Alkali Metal Anodes. Nat. Commun. 2018, 9, 1339. 17. Black, J. M.; Walters, D.; Labuda, A.; Feng, G.; Hillesheim, P. C.; Dai, S.; Cummings, P. T.; Kalinin, S. V.; Proksch, R.; Balke, N. Bias-Dependent Molecular-Level Structure of Electrical Double Layer in Ionic Liquid on Graphite. Nano Lett. 2013, 13, 5954–5960. 18. Garcia, R.; Proksch, R. Nanomechanical Mapping of Soft Matter by Bimodal Force Microscopy. Eur. Polym. J. 2013, 49, 1897–1906. Pergamon. 19. Melcher, J.; Carrasco, C.; Xu, X.; Carrascosa, J. L.; Gómez-Herrero, J.; De Pablo, P. J.; Raman, A. Origins of Phase Contrast in the Atomic Force Microscope in Liquids. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 13655–13660. 20. Sakai, H.; Taniguchi, Y.; Uosaki, K.; Masuda, T. Quantitative Cross-Sectional Mapping of Nanomechanical Properties of Composite Films for Lithium Ion Batteries Using Bimodal Mode Atomic Force Microscopy. J. Power Sources 2019, 413, 29–33. 21. Romanyuk, K. N.; Alikin, D. O.; Slautin, B. N.; Tselev, A.; Shur, V. Y.; Kholkin, A. L. Local Electronic Transport across Probe/Ionic Conductor Interface in Scanning Probe Microscopy. Ultramicroscopy 2021, 220, 113147. 22. Collins, L.; Kilpatrick, J. I.; Kalinin, S. V.; Rodriguez, B. J. Towards Nanoscale Electrical Measurements in Liquid by Advanced KPFM Techniques: A Review. Rep. Prog. Phys. 2018, 81, 086101. 23. Kawano, H. Effective Work Functions of the Elements: Database, Most Probable Value, Previously Recommended Value, Polycrystalline Thermionic Contrast, Change at Critical Temperature, Anisotropic Dependence Sequence, Particle Size Dependence. Prog. Surf. Sci. 2022, 97, 100583. 24. Luchkin, S. Y.; Stevenson, K. J. On the Origin of Extended Resolution in Kelvin Probe Force Microscopy with a Worn Tip Apex. Microsc. Microanal. 2018, 24, 126–131. 25. Masuda, H.; Ishida, N.; Ogata, Y.; Ito, D.; Fujita, D. Internal Potential Mapping of Charged Solid-State-Lithium Ion Batteries Using In Situ Kelvin Probe Force Microscopy. Nanoscale 2017, 9, 893–898.

Further reading 1. Baró, A. M.; Reifenberger, R. G. Atomic Force Microscopy in Liquid: Biological Applications; Wiley-VCH Verlag GmbH & Co: KGaA, 2012. 2. Butt, H.-J.; Cappella, B.; Kappl, M. Force Measurements with the Atomic Force Microscope: Technique, Interpretation and Applications. Surf. Sci. Rep. 2005, 59, 1–152. 3. Lanza, M. Conductive Atomic Force Microscopy: Applications in Nanomaterials; Wiley-VCH Verlag GmbH & Co. KGaA, 2017.

Methods and Instruments | X-Ray and Neutron Diffraction H Dittrich and A Bieniok, University of Salzburg, Salzburg, Austria © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of H. Dittrich, A. Bieniok, MEASUREMENT METHODS | Structural Properties: X-Ray and Neutron Diffraction, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 718–737, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00074-5, with revisions made by the Editor.

1 2 2.1 2.2 2.3 3 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 4 Further reading

Introduction Theory of X-ray and neutron diffraction Kinematical scattering theory Dynamical scattering theory Neutron scattering theory Methodology Single-crystal diffraction Powder diffraction Small-angle scattering Grazing incidence High-temperature diffraction The electrochemical in situ cell Synchrotron Neutron diffraction X-ray diffraction peak profile analysis Multiple whole profile fitting The Rietveld method DIFFaX method Applications

167 168 168 169 169 170 171 171 173 173 174 174 175 176 177 180 180 183 185 188

Abstract X-ray and neutron diffraction techniques constitute a powerful means of analyzing and characterizing materials used in batteries and fuel cells. The range of techniques available allow the positions of the atoms in the unit cell (the crystal structure) to be established, the presence and preponderance of know crystalline materials to be established, together with a variety of microstructural information such as crystallite size, crystallite strain, and preferred orientation. In a brief introduction, the principles of X-ray and neutron diffraction are presented. The sources of X-rays and neutrons are described and the basics of the kinematical scattering theory of X-rays are outlined. The structural information principally extractable from diffraction experiments is compiled, demonstrating the potential of the diffraction methods.

Glossary Atom form factor (scattering factor) The atomic form factor, f, is the ratio of the scattering power of an atom by a photon to that of a free electron. It is an angle-dependent scattering power of atoms with different numbers of electrons in their orbitals. It is important for the evaluation of structural and chemical data from intensity measurements of diffracted X-rays or neutrons. Bragg’s law Refers to the simple equation nl ¼ 2d sin y. It explains why the cleavage faces of crystals appear to reflect X-ray beams at certain angles of incidence (y). The variable d is the distance between atomic layers in a crystal and the variable lambda (l) is the wavelength of the incident X-ray beam; n is an integer. Bragg-Brentano geometry Focusing arrangement of X-ray source, sample, and detector in powder diffraction. Owing to this arrangement, the divergence of the primary beam is transformed into a convergence of the diffracted beam and therefore enhances the resolution of the measurement. Channel type structures Structures with one-dimensional channels (tunnels) as diffusion paths for intercalated ions. Examples: Ramsdellite, Li0.5MnO, Li2+xTi3O7, LiCrTiO4, and so on. Critical angle Angle of incidence below which a total external reflection of X-rays occurs on solid surfaces. This is caused by the fact that the refractive index for X-rays in solid materials is lower than 1. The critical angle could be derived by Snell’s law. Crystal structure determination Extracting structural and chemical data from the interaction of electromagnetic waves or particle rays. Crystallite size Size of crystal regions with homogeneous diffraction of X-rays. The full-width at half-maximum of powder diffraction peaks is correlated with crystallite size (crystallite size contribution to the peak broadening effect).

164

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00314-5

Methods and Instruments | X-Ray and Neutron Diffraction

165

Crystallite strain Crystallite strain shifts the lattice constants in the crystals. It is also responsible for a peak broadening effect in powder diffraction. Using two-dimensional detectors, an ellipsoidal deviation of the circularity of the Debye cones can be measured. Debye cones The statistical orientation of a large number of crystallites in a powder sample is responsible for a diffracted intensity in the shape of a cone with 4y as opening angle (y: Bragg angle). DIFFaX method A general recursion method for calculating diffracted intensities from crystals containing planar faults. Diffraction peak profile analysis Careful interpretation of measured powder pattern peak profiles by refinements with modeled peak profiles. The measured peak broadening is a superposition of instrumental broadening and sample broadening. Dynamical scattering theory Extension of the kinematical scattering theory taking into account effects caused by absorption, extinction, and refraction. Also multiple scattering effects are considered. The dynamical scattering theory has to be considered particularly in the analysis of perfect crystals or in electron diffraction experiments. Electrochemical in situ X-ray diffraction Structural characterization of electrode materials during electrochemical cycling experiments. Eulerian cradle Comprehensive investigations of residual stress and texture—as well as measurements of thin films, wafers, or microstructures—require the sample rotation (phi) and tilt (chi). The Eulerian cradle enables the sample to be moved in these additional degrees of freedom. Full-width at half-maximum Expression of the extent of a function, given by the difference between the two extreme values of the independent variable at which the dependent variable is equal to half of its maximum value. Gracing incidence Uses small incident angles for the incoming X-ray or neutron beam, so that diffraction can be made surface sensitive. It is used to study surfaces and layers because wave penetration is limited. Instrumental broadening All instrumental influences on the measured diffraction peak profiles. Kinematical scattering theory Approximation is that, if we now consider scattering from a crystal, we can add all scattered amplitudes originating from all electrons with the same weight, keeping the calculation easy and straightforward. Lorentz factor For single-crystal rotation techniques, this correction factor is a measure of the relative amount of time different sets of planes spend in the diffraction position or, alternatively, the relative amounts of time the corresponding reciprocal lattice points take in passing through the Ewald sphere. For stationary powder methods, it is simply the fractional number of reciprocal lattice points that lie on the surface of the Ewald sphere. Microdiffraction Diffraction methods using microfocused X-ray beams from several micrometers down to 0.1 mm. Peak profile function Mathematical models for the shape of the experimental peaks, whereas powder diffraction patterns are mainly specified by the positions and intensities of diffraction peaks. The experimental peak profiles are modeled by convolutions of intrinsic peak profiles with deformation caused by the instruments used for the measurement. The convoluted model function enables precise extraction of information about the intrinsic peak profiles, including the integrated intensities, and locations and widths of the peaks, from experimental diffraction data. Phase identification By comparing the positions and intensities of diffraction peaks against a library of known crystalline materials, the target material can be identified. In addition, multiple phases in a sample can be identified and quantified. Even if one of the phases is amorphous, X-ray diffraction can determine the relative amount of each phase. Polarization factor The X-ray beam that exits the tube is unpolarized (analogous to light coming from the sun). Low-angle scattering causes polarization of the beam (analogous to light reflecting off a lake). The polarization factor accounts for increase in scattering at low angles. Scattering intensity (Ip) owing to polarization is proportional to (1 + cos2 y)/2 (Thomson equation). Polymorphism Ability of a solid material to exist in more than one form or crystal structure. Polymorphism can potentially be found in any crystalline material including polymers, minerals, and metals. Polymorphism means the crystallization into two or more chemically identical but crystallographically distinct forms. Polytypism An element or compound is polytypic if it occurs in several structural modifications, each of which can be regarded as built up by stacking layers of (nearly) identical structure and composition, and if the modifications differ only in their stacking sequence. Polytypism is a special case of polymorphism: the two-dimensional translations within the layers are essentially preserved. Preferential orientation Deviations from the statistical orientation of crystallites in a polycrystalline structure. This phenomenon may be the result of significant cleavage planes or characteristic growth forms (habitus). Sodium chloride crystallites for example tend to orient themselves on a flat substrate according to the directions {001}. The diffraction pattern gives a particularly intense line (001) that is disproportionate in relation to the other intensities. Reflection mode The detector is located between the X-ray source and the crystal. In reflection mode, the diffracted beam in low 2y angle is narrower than the diffracted beam in high 2y angle (see transmission mode). Rietveld refinement The Rietveld method uses a least squares approach to refine a theoretical line profile until it matches the measured diffraction profile. The introduction of this technique was a significant step forward in the diffraction analysis of powder samples as it was able to deal reliably with strongly overlapping reflections. Sample broadening The breadth of a Bragg peak is a combination of both instrument- and sample-dependent effects. The sample-dependent effect is called sample broadening. To decouple these contributions, it is first necessary to collect a diffraction pattern from a line broadening standard material such as LaB6 to determine the instrumental contribution to the

166

Methods and Instruments | X-Ray and Neutron Diffraction

line shape. The sample of interest is then studied under identical geometrical and environmental conditions. Profile refinement of the diffraction patterns of both the sample and the line broadening standard is then undertaken; the resulting sample broadening is obtained by a process of mathematical deconvolution of the refined peak shape parameters. Single-crystal diffraction The density of electrons within the crystal is determined from the position and brightness of the various reflections observed as the crystal is gradually rotated in the X-ray beam; this density, together with supplementary data, allows the atomic positions to be inferred. For single crystals of sufficient purity and regularity, X-ray diffraction data can determine the mean chemical bond lengths and angles to within a few thousandths of an angström and to within a few tenths of a degree, respectively. The data also allow the static and dynamic disorder in the atomic positions to be estimated, which is usually less than a few tenths of an angström. Small-angel scattering Small- and wide-angle X-ray scattering is a useful and complementary method for determining the size, size distribution, and structure of a wide range of disordered (noncrystalline or semicrystalline) materials. Examples include polymers, liquid crystals, oils, suspensions, and biological samples like fibers or protein molecules in solution. Soller slit A Soller slit device is provided for collimation of X-rays, and has a low angle of divergence ( @

R-Bragg: 4.80

44

50

1: 0

46

R - F: 3.86

48

50

Fig. 18 Example of a Rietveld refinement of LiCoO2.

•

Lattice constants Zero correction Flat plate terms • Scaling Phase fractions Structural parameters • Atom positions Occupancies Displacement parameters • Preferential orientation Absorption Listing of published Rietveld programs:

• • • • • • • • • • • • • • • •

DBWS (Free Dos Structure Refinement Software). FullProf (Free Dos and Mac Structure Refinement Software). RIQAS (Commercial Dos Quantitative Phase Analysis Software). GSAS (Free Dos Structure Refinement Software). SiroQuant (Commercial MS-Windows Quantitative Phase Analysis Software). Quasar (Commercial MS-Windows Quantitative Phase Analysis Software). FAT-Rietan (Free Dos and Mac Structure Refinement Software). Philips PC-Rietveld (Commercial Dos Structure Refinement Software—No Quantitative Analysis). ARITVE (Free Dos Glass Modeling Software). Riet7/SR5 (Dos Structure Refinement Software—No Quantitative Analysis). LHPM from ANSTO (Structure Refinement for DOS, Windows, and native Win95 version). XND (Structure Refinement for DOS). SIMREF and SIMPRO (Structure Refinement for DOS). Koalariet (Developmental Structure Refinement for Win95). BGMN fundamental parameters Rietveld (Structure Refinement for Win95, DOS, WIN-NT, OS/2, Linux, and Unix systems). XRS-82 Rietveld (Structure Refinement for DOS).

Methods and Instruments | X-Ray and Neutron Diffraction

183

3.12 DIFFaX method Layered structures result from anisotropy of crystal bonding energy in different lattice directions. This is most clearly demonstrated in structures where van der Waals forces are involved along one principal crystallographic axis. Hexagonal symmetries are predominant in such types of structures, including different stacking orders of the hexagonal layers. Because of very small differences in the lattice energy, polytypism occurs very often. Owing to the small difference in lattice energies of the different stacking orders in the polytypes, the order-disorder transition activation energy is also very low and high stacking fault concentrations are observed. Polytypes and stacking fault concentrations influence the physical properties of the layered materials. Particularly, electronic properties (bandgap of semiconductors) and electrochemical properties (electrode capacities and stabilities in batteries) can be strongly influenced and in the end be optimized. Therefore, a powerful method of characterizing stacking sequences in layered structures, whether they be ordered or disordered, is of great importance. The simulation of diffraction patterns of materials including stacking faults dates back to Landau in the year 1937 and several different approaches were developed. In the year 1991, a general recursion method for calculating diffracted intensities from crystals containing coherent planar faults was published by M. M. J. Treacy and coworkers. This algorithm was implemented in a Fortran computer program called diffracted intensities from faulted xtals (DIFFaX). DIFFaX exploits the recurring patterns found in randomized stacking sequences to compute the average interference wave function scattered from each layer type occurring in a faulted crystal. The method of working is easy and straightforward: (1) definition of a crystallographic layer unit cell for any involved layer types; (2) definition of a transition vector r specific for the stacking fault type; and (3) definition of a stacking probability aij correlated to the stacking fault concentration. The result is a synthesis of randomized stacking sequences and their simulated X-ray diffraction (XRD) powder pattern. Fig. 19 shows the application of the DIFFaX method to stacking in a nickel hydroxide (Ni(OH)2) structure. The following data are employed as data inputs for the DIFFaX method: Defined unit cell data of the specific layer in the structure:Lattice constants Atomic positions Point group symmetry Extent of the layer Vectors of the different stacking possibilities Probabilities connected to the stacking vectors Fig. 20 shows a fit obtained by the DIFFaX method for the structure of Ni(OH)2. Another example of the use of the method is shown in Fig. 21. Examples of materials to which the DIFFaX method has been applied include solid solutions between Li2MnO3 and Li [Ni0.5Mn0.5]O2, Ni(OH)2, graphite, [M(II)1–xM0 (III)x(OH)2](An–)x/nyH2O ! layered double hydroxides, and orthorhombic LiMnO2. A summary of the analytical tasks to which diffraction methods may be applied is given in Table 5.

Layer unit cell: Atom Nr. x 1 0 Ni 1 1/3 O 2 2/3 O 1 1/3 H 2 2/3 H

y 0 2/3 1/3 2/3 1/3

z 0.5 0.7221 0.2779 0.9275 0.0725

Layer B

Probability aij

Layer A Translation vector rij

Fig. 19 Data evaluation for Ni(OH)2 stacking order estimation by the DIFFaX algorithm.

184

100

101

Methods and Instruments | X-Ray and Neutron Diffraction

Measured Ni(OH)2-pattern DIFFaX profile simulation with: 86% stacking faults type A 8% stacking faults type B

102

Rel. intensity

6% stacking faults type C

30.00

35.00

40.00

30

35

40

45.00

50.00

55.00

45

50

55

2 Fig. 20 Superposition of DIFFaX simulated peak profile and measured peak profile of a Ni(OH)2 sample.

40 000

35 000

30 000

Internal

25 000

20 000

15 000

10 000

5000

0 10.00

20.00

30.00

40.00

50.00 2

60.00

70.00

Fig. 21 DIFFaX modeling sequence of different statistical stacking orders in Takovite (A100/B00/C00 ! A00/B50/C50).

80.00

90.00

Methods and Instruments | X-Ray and Neutron Diffraction Table 5

185

Compilation of analytical tasks in X-ray and neutron diffraction.

XRD method

Hardware

Measurement

Evaluation

Importance

Phase identification Phase quantification Nonambient

Texture

Eulerian cradle

Peak position and intensity distribution of XRD patterns Peak position and relative intensity distribution of XRD patterns XRD patterns under varying temperatures and atmospheres XRD patterns under varying electrochemical conditions Intensity distribution in XRD patterns as a function of crystallite orientation

Stress

Eulerian cradle

Identification of crystalline phases by matching with known data set Quantification of volume fractions from the ratio of the diffracted intensities Determination of structural parameters as a function of sample environment Determination of structural parameters as a function of electrochemical treatment Determination of the orientation distribution function in polycrystalline materials Determination of the 3D stress and strain

XX

In situ

Standard Bragg-Brentano diffractometer Standard Bragg-Brentano diffractometer High-temperature chambers, gas pressure chambers In situ electrochemical cell

Single crystal

2D CCD detector

Small angle X-ray scattering Microdiffraction

Linear arrangement, nonBragg-Brentano geometry Small focus (a few microns), x-y-z sample stage Small focus (a few microns), x-y-z sample stage Göbel mirror, asymmetrical 4-bounce monochromator

Mapping High-resolution X-ray diffraction Grazing incidence Reflectometry

Göbel mirror, rotary absorber Göbel mirror, rotary absorber

Shifts in XRD patterns as a function of sample orientation Large number of reflections in the reciprocal space Small angle X-ray scattering signals in XRD patterns XRD pattern within micrometer area resolution Automatically collected XRD pattern from multiple locations Rocking curves, radial scans, reciprocal space maps, absolute lattice parameters In-plane reflections with very small incident and exit angle Specular scans, diffuse scans, reciprocal space maps

X XX XXX X X

Indexing the patterns, solving the crystal structure Modeling of the data determination of nano-size structural properties Integration of spotty data extraction of desired sample information Automated extraction and mapping of specific sample properties Simultaneous evaluation of several reflections

X

Determination of lattice parameters and phases Determination of lattice parameters and phases

X

X XX X 0

0

XRD, X-ray diffraction.

4

Applications

Diffraction methods are the most powerful means of analyzing solids in general. In the study of materials for energy storage devices (batteries, etc.), they are extensively used in the development and management of electrode materials and a few examples are mentioned here. The cathodes (positive electrodes) in lithium-ion cells are generally oxides into which lithium ions are able to intercalate. A problem that has arisen with the use of lithium cobalt oxide is that there can be self-discharge in delithiated cobalt oxide owing to the formation of spinel phase, which results in capacity loss. This process has been thoroughly studied by full profile analysis (Fig. 22). Another possibility to use LiMn2O4 is also employed as positive electrode material in lithium-ion cells and, in this case, a complicated sequence of phases can be formed as the lithium-ion content of the oxide changes. These changes have been thoroughly studied by X-ray diffraction techniques. The phase changes during cycling in ex situ measurements of Li0.72Mn2O4: MnO1.93 are as follows. LixMn2O4: reduction mechanism of spinel-related MnO1.93; cubic, aC1 ¼ 8.045 A˚ , at Li 0.27 < x < 0.60 reduction proceeded in two cubic phases aC1 ¼ 8.045 A˚ and aC2 ¼ 8.142 A˚ ; at Li 0.60 < x < 1.00 reduction proceeded in continuous lattice expansion from aC2 ¼ 8.142 A˚ to aC2 ¼ 8.239 A˚ ; at Li 1.00 < x < 2.00 reduction proceeded in two phases: a cubic Li1Mn2O4 (aC2 ¼ 8.239 A˚ ) and a tetragonal Li2Mn2O4 phase (aT ¼ 5.649 A˚ , cT ¼ 9.253 A˚ ). Fig. 23 shows the results of the structural study of one of the phases involved. The negative electrode materials of lithium-ion cells are usually graphitic carbon materials again allowing lithium ions to reversibly intercalate. The form of carbon used is very important and, once again, diffraction methods are vital for a full characterization (Fig. 24). Diffraction methods have been at the center of studies to improve the performance of the negative electrodes of nickel-metal hydride batteries. Fig. 25 shows the simulated powder pattern of LaNi5.

186

Methods and Instruments | X-Ray and Neutron Diffraction

10 360

-UCoO

b

a

LiCoO

c

Li

Li

Li

Li

Space group number Space group Cell choice Lattice parameter Angles Atoms in asymmetric unit Atoms in unit cell Volume of cell Rel. mass of unit cell X-ray density Mass absorpt. coef.

0 Co 0

0

0

0

Li 0 Co Co Co

Co

0 Li

5180 0

0

Name P.No. ion wyck

0 0 Co

Li

Li

x

y

z

SOF B-(iso)

Li 3 Li+ 3a 0.0000 0.0000 0.0000 1.0000 0.2000 Co 27 Co3+ 3b 0.0000 0.0000 0.5000 1.0000 0.2000 O 8 O 6c 0.0000 0.0000 0.7583 1.0000 0.2000

0 Li

:166 : R-3 2/m :1 : 2.8144 2.8144 14.0095 : 90.0000 90.0000 120.0000 : 3 : 12.0, 12 generated pos. : 96.10Å3 : 293.62 : 5.0735 g cm–3 : 203.16 cm2 g–1

Li

0 5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

Fig. 22 Structural study of tendency of lithium cobalt to spinel formation as a function of temperature.

14 138

LiMn2O4 b a

LiMn2O4 Space group number : 227 : F4_1/d-32.1m Space group Cell choice :2 : 8.2476 8.2476 8.2476 Lattice parameter : 90.0000 90.0000 90.0000 Angles Atoms in asymetric unit : 3 : 56.0, 56 generated pos. Atoms in unit cell Volume of cell : 561.03 A3 Rel. mass of unit cell : 1446.52 : 4.2814 g cm3 X-ray density Mass absorpt. coef. : 177.28 cm1/g

111

c

7069

x

Name P. No. ion wyck

z

SOF B-(iso)

3 Li 0a 0.1250 0.1250 0.1250 1.0000 0.0000 25 Mn 16d 0.5000 0.5000 0.5000 1.0000 0.0000 8 0 32e 0.2610 0.2610 0.2610 1.0000 0.0000

642

711

446

623

620

442

492

22.0

222

231

551

511

640

311

400

Li Mn o

y

0 5

10

15

20

Fig. 23 Crystal structure of cubic LiMn2O4.

25

30

35

40

45

50

55

60

65

70

75

80

85

90

187

Methods and Instruments | X-Ray and Neutron Diffraction

1.2u105

Icalc

4

Iobs – Icalc

I / counts

1u10 8u10

0.342 nm

Iobs

5

72% +

6u104

0.446 nm

0.20 nm

4u104

28%

2u104 0 0

20

40 60 2 (degree)

(a)

80

100

Soft carbon (c)

(b )

8u104

0.343 nm

Iobs Icalc Iobs – Icalc

6u104 I / counts

1.2 nm (La / 2)

2000 0 _2000

0.09 nm 45% +

4u104

0.47 nm

0.25 nm 55%

2u104

0

20

(d)

40 60 2 (degree)

80

100

Pore

1.0 nm

1000 0 _1000

0

Hard carbon (e)

(f)

Fig. 24 (a–f ) Structure model and X-ray diffraction (XRD) of soft carbon and hard carbon from Azuma and coworkers (1999).

14872

111

7436

200

101

002

110

202

001 0 5

10

15

20

25

300 102

30

35

40

Fig. 25 Simulated X-ray diffraction (XRD) pattern of LaNi5 by PowderCell 2.3.

45

220

112 211

201 100

301

50

210 55

60

65

103 003 70

75

113 310

311 203

80

85

90

188

Methods and Instruments | X-Ray and Neutron Diffraction

In most battery systems, the electrode materials are solids and are therefore suitable for diffraction solids, whereas the electrolytes are liquids. In high-temperature systems, however, the situation is reversed. In the sodium/sulfur and ZEBRA batteries, the electrode materials are used as liquids and it is the electrolyte (beta alumina) that is the solid member of the cell. Diffraction methods have been used extensively in the study of the beta alumina family of materials and in the study of another sodiumion-conducting solid, ‘Nasicon’.

Further reading 1. 2. 3. 4.

Bish, D. L.; Post, J. E. Modern Powder Diffraction. Reviews in Mineralogy, Vol. 20; Mineralogical Society of America: Washington, DC, 1989. Eberhart, J. P. Structural and Chemical Analysis of Materials; John Wiley & Sons: Chichester/New York/Brisbane/Toronto/Singapore, 1989. Giacovazzo, C. Fundamentals of Crystallography. IUCr Texts on Crystallography, Vol. 7; Oxford University Press: Oxford, 2002. Treacy, M. M. J.; Newsam, J. M.; Deam, M. W. A General Recursion Method for Calculating Diffracted Intensities From Crystal Containing Planar Faults. Proc. R. Soc. Lond. 1991, A433, 499–520. 5. Wilson, A. J. C. International Tables for Crystallography. Mathematical, Physical and Chemical Tables, Vol. C; Kluwer Academic Publishers: Dordrecht, 1995.

Methods and Instruments | Scanning Electronic Microscopy F Nobili and A Staffolani, School of Science and Technology—Chemistry Division, University of Camerino, Via Madonna delle Carceri, Camerino (MC), Italy © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of R. Marassi, F. Nobili, MEASUREMENT METHODS | Structural and Chemical Properties: Scanning Electron Microscopy, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 758-768, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00071-X.

1 2 2.1 2.2 3 3.1 3.2 3.2.1 3.2.2 4 4.1 4.2 5 6 References Further reading

Introduction Instrumentation and operation Electron gun Electron optics Electron beam-target interactions Emission of backscattered electrons Emission of signals from inelastic scattering Emission of secondary electrons Production of X-rays Formation of image Electron detection and image reconstruction X-ray detectors and electron probe micro analysis Applications to the characterization of electrochemical power sources Concluding remarks

190 191 191 193 194 194 194 194 195 195 196 198 199 204 204 205

Abstract Scanning Electron Microscopy (SEM) principles are introduced and some applications to the characterization of electrochemical power sources are presented. The general features of a SEM microscope are briefly described with particular attention to: (i) generation and focusing of the electron beam over the samples; (ii) interactions between electrons and atoms in the sample producing the analytical signals; (iii) detection and recording of secondary or backscattered electrons for image reconstruction; (iv) detector of X-rays and electron probe microanalysis. Some applications of the technique in the field of electrochemical storage devices are summarized, with the aim of shedding light onto the relationships between morphology/composition and electrochemical behavior of electrodes. Finally, an outlook on recent methodologies is given, such as environmental SEM which, operating under moderate pressures, allows in situ monitoring of chemical and morphological transformations related to electrochemical power sources.

Glossary Backscattered electrons Electrons of the incident beam that re-emerge from the target sample as a result of a sequence of elastic scattering events inside the interaction volume. Bremsstrahlung Continuous X-ray radiation due to deceleration of an energetic beam of electrons in the columbic field of the atoms of a sample. Brightness Current (or power) density emitted from a particular area and passing through a unit solid angle. The SI unit of radiance is watts per steradian per square meter (W sr−1 m−2). Characteristic X-rays X-ray radiation emitted at definite energies as a consequence of internal electron transitions, characteristic of each element. Electron gun A stable source of electrons used to form an electron beam characterized by specific energy, current, divergence angle and brightness. Electron probe microanalysis Spectrometry of characteristic X-rays emitted by atoms of a target sample upon impact with a highly energetic electron beam. Electron range Radius of a semicircle centered on the impact point of an electron beam with a target surface, which defines the region below the surface where interactions of incident electrons with the sample occur. Energy dispersive spectrometry X-rays spectrometry that acts simultaneously collecting radiations at different energies and analyzing them through a multichannel solid-state analyzer. Environmental SEM Method of scanning electron microscopy in which the sample is analyzed at moderate pressures, in order to monitor its evolution upon interaction with gaseous atmosphere or solvent.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00212-7

189

190

Methods and Instruments | Scanning Electronic Microscopy

Field emission A process of electron production that utilizes tunneling through an energy barrier in the presence of a very high electric field. Interaction volume Region below the impact point of an electron beam with the surface of a target sample where interaction phenomena of incident electrons with atoms and electrons in the sample are confined. Secondary electrons Electrons ejected from target atoms in the sample as a consequence of inelastic scattering events with incident beam or backscattered electrons. Thermionic emission Thermionic emission is the flow of charged particles called thermions from a charged metal or a charged metal oxide surface, caused by thermal vibrational energy overcoming the electrostatic forces holding electrons to the surface. Wavelength dispersive spectrometry X-rays spectrometry that acts analyzing the monochromatic components of a polychromatic radiation at different angles according to Bragg’s law.

Key points

• • • • • • •

The working principle of scanning electron microscopy is discussed. General information about the instrumentation and its operation is provided. The interactions between the electron beam and the sample are described in terms of elastic and inelastic scattering. The generation of backscattered electrons, secondary electrons and X-rays, and their use as analytical signals, is described. Morphological studies done by SEM of catalysts for energy conversion devices are provided. Examples of use of SEM for the characterization of fresh and post-mortem materials for Li-ion batteries are provided. Examples of operando measurements of energy storage devices by ESEM are given.

Acronyms and abbreviations CRT EDS EG EPMA ESEM SEM STEM TEM VPSEM WDS

1

Cathode Ray Tube Energy Dispersive Spectrometry Electron Gun Electron Probe Microanalysis Environmental Scanning Electron Microscopy Scanning Electron Microscopy Scanning Transmission Electron Microscopy Transmission Electron Microscopy Variable-pressure Scanning Electron Microscopy Wavelength Dispersive Spectrometry

Introduction

Electron microscopies are investigation techniques which rely on the interactions between high-energy electrons and atoms in a solid sample to produce different analytical signals, which can be used to described morphological, structural, and chemical properties of the investigated samples. In the scanning electron microscope, several magnetic lenses, operating by Lorentz force, demagnify an electron beam, produced by a source called ‘electron gun’, and raster it onto the surface of a bulk sample. The beam/ sample interactions produce two main signals above the surface of the sample: electrons, which are collected to reconstruct three-dimensional images; X-rays, which give information about sample composition. The following topics are here briefly discussed: generation, demagnification and focusing of the electron beam; interactions between the electron beam and the sample; generation of contrast and images; electron probe microanalysis. Some applications of the technique to the characterization of electrochemical power sources are discussed. The development of electron microscopies started when, applying the concepts of particle/wave dualism introduced by De Broglie (1924), Knoll and Ruska (1931), Von Ardenne (1938) and Zworkyin et al. (1942) produced the first Transmission Electron Microscope (TEM), Scanning Transmission Electron Microscope (STEM) and Scanning Electron Microscope (SEM), respectively. The electron wavelength l, not taking into account relativistic effects, is given by the De Broglie’s equation: h 1:22 l ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi  1=2 2m0 eV E

(1)

Methods and Instruments | Scanning Electronic Microscopy

191

where h is Planck’s constant, m0 and e are the rest mass and the charge of an electron, respectively, and V is the accelerating voltage. The approximation yields the numeric value of wavelength l (nm) as a function of the energy E (keV), meaning that, for an electron with energy of tens of keV, the wavelength is in the order of 0.01 nm, smaller than the diameter of an atom. This determines, on one side, the smallest distance which, theoretically, can be resolved based on the classical Rayleigh’s criterion for optical microscopy: R¼

0:61l n sin b

(2)

where R is the resolution, l is the wavelength, n and b are the refractive index and the semiangle of collection of the lens. On the other side, the wavelength plays an important role in determining the aberrations and subsequent resolution losses of the “optical” systems. Therefore, even if the resolution given by Rayleigh’s criterion can hardly be reached because of instrumental limitations, electron microscopies exceed by far the performances of optical microscopies and are among the most powerful tools to investigate morphology, composition, and structure of solid samples. In the case of SEM, the surface properties (morphology, topography, and composition) of a bulk sample can be investigated. A very thin beam of highly energetic electrons (up to 50 keV, more often in the order of 10–20 keV) is focused over a sample and sequentially scanned in a raster over the surface. As a result of elastic and inelastic interactions between the electrons and the atoms in the sample, different signals are generated and collected by dedicated detectors, which finally convert them into an image of the sampled area. Among the generated signals, the most relevant ones are secondary electrons, backscattered electrons, characteristic X-rays, which allow to gather information about sample topography and elemental composition.

2

Instrumentation and operation

The Fig. 1 schematizes the main components of a scanning electron microscope: the electron gun (EG), which produces and accelerates the probing electrons; the “optical” system, which demagnifies, focuses and scans the electron beam over the sample surface; the sample holder; the recording system, which collects the emitted signals and converts them into data1. The data can be visualized on a screen, such as the cathodic ray tube (CRT), which in elder instrumentations allowed an easy synchronization between the scanning beams on the sample and on the screen, or, in modern instrumentations, on LCD displays and stored as digital images. The single components are briefly described below.

2.1

Electron gun

The electron gun (EG) provides a stable source of electrons, which is used to form the electron beam. Two main types of EG are commonly used in SEM, according to the electron emission process which is exploited: the thermionic emission and the field emission. In the thermionic emission, the electron source is a filament (W or LaB6) with a V-shaped tip with a radius ranging from 5 to 100 mm, from which the electrons are those that, according to Boltzmann’s distribution law, can access an energy which is higher than the work function EW of the cathode material. The work function, EW, is the energy required to bring an electron from the Fermi level to infinity, and those electrons, which can overcome by thermal agitation this threshold value, are emitted and eventually accelerated by an electric field. The emission current density Jc (A cm−2) obtained by thermionic emission is expressed by the Richardson’s law: Jc ¼ Ac T 2 expð − EW =kT Þ

(3)

where Ac (A cm−2 K−2) is a constant depending on the emitting material, T (K) is the emission temperature and k the Boltzmann’s constant. The electrons leaving the filament are accelerated by a potential difference, in a range from a few hundred to 100 kV, between the filament and an anode. The filament is surrounded by a grid cap, the Wehnelt cylinder, biased at a negative potential (0–2500 V). The electric field generated by the filament, the Wehnelt cylinder and the anode causes the emitted electrons to converge to the crossover point, which is not actually a point but a disk with diameter d0  10–50 mm and divergence angle a0 below the cylinder. The current beam at the filament ib (A), the diameter of the crossover point and the divergence angle concur in defining the brightness b (A cm−2 sr−1), i.e., the current density per unit solid angle (steradiant units, sr), given by b¼

4ib p2 d20 a20

(4)

Briefly, the brightness defines the ability to concentrate an electron beam in a small region, as parallel as possible. It should be noted that, when the electron beam travels its path from the gun through the optics to the sample, its brightness remains constant even when local values of i, d and a change (obviously not independently). Therefore, higher brightness at the crossover point means

192

Methods and Instruments | Scanning Electronic Microscopy

Fig. 1 Operation of a typical Scanning Electron Microscope: (a) global layout; (b) detail of thermionic emission gun; (c) electron beam path through magnetic optics; (d) interaction volume and electron beam/specimen interactions. In panel (c), S’s represent the focusing distances at each sides of the lenses, f’s represent the focal lengths, d’s and a’s respectively represent the spot sizes and the divergence angles of the beam at different heights during the path from electron source to target sample. Adapted from Ref. Goldstein, J.I.; Newbury, D.E.; Echlin, P.; Joy, D.C.; Fiori, C.; Lifshin, E. Scanning Electron Microscopy and X-Ray Microanalysis. Plenum Press: New York (1981)., with permission.

higher brightness at specimen surface, resulting into higher number of parallelly interacting electrons in a smaller spot, and ultimately higher resolution, contrast, and depth of field in the sample image. Maximizing the brightness at electron gun stage means maximizing the quality of the final image. This can be done by adjusting operating instrumental parameters, according to Eq. (5): b¼

J c eE0 pkT

(5)

where Jc is the current density defined by Richardson’s law (Eq. 3), e is the elementary charge, E0 is the acceleration potential of the anode, k is the Boltzmann’s constant, T is the temperature. At the same time, the Wehnelt cylinder plays a key role in adjusting current density and brightness. Its bias voltage changes the constant field lines at the cathode. At low bias the negative field gradient

Methods and Instruments | Scanning Electronic Microscopy

193

is weak, the focusing is ineffective and the electrons see mainly the positive voltage gradient toward the anode, the crossover d0 is large and the brightness low. At high bias the electrons are repelled toward the filament, and the emission current and brightness decrease to zero. Thus, for each instrument there is an optimum bias setting for maximum brightness. The typical operating parameters for a W filament are T ¼ 2700 K, AC ¼ 6  109 A m−2 K−2, Ew ¼ 4.5 eV, resulting, at 100 kV acceleration voltage, into current density of the order of 5 A m−2, a crossover size of at least 105 nm and a brightness of about 1010 A m−2 sr−2. Lanthanum Hexaboride (LaB6) can provide higher brightness at lower temperature than a W filament because of the lower work function. Typical operating parameters are Ew ¼ 2.4 eV, Ac ¼ 4  109 A m−2 K−2, T ¼ 1700 K resulting, at 100 kV acceleration voltage, into a more efficient electron beam characterized by current density of the order of 102 A m−2, a crossover size of 104 nm and a brightness equal to 5  1011 A m−2 sr−2. However, the LaB6 crystal is prone to oxidation upon operation, progressively losing its electron emission ability and thus requiring a much higher vacuum to operate. On the other side, due to the geometry and dimension of the emitting tip, the field emission source is inherently capable to produce higher brightness. A very thin tungsten single crystal (tip diameter of the order of 100 nm) with a chosen orientation (usually h111i axial direction), is held at a negative potential with respect to an anode. The electric field at the tip is very high, allowing electrons to tunnel directly through the energy barrier. At 100 keV accelerating voltage, the cathode current density may range between 105 for thermal field emission (TFE) where the tip is also heated (T ¼ 1700 K, Ew ¼ 3 eV) and 106 A m−2 for cold field emission (CFE) which guarantees a more efficient focusing due to reduced thermal agitation (T ¼ 300 K, Ew ¼ 4.5 eV). This results into effective crossover size and brightness values of 15 nm and 1012 A m−2 sr−1, respectively, for TFE, and 3 nm and 1013 A m−2 sr−1 for CFE. The optical system must be operated at 10 nPa to prevent contamination of the tip, which would result into current fluctuations and reduced focusing ability.

2.2

Electron optics

The electron beam is directed toward the sample through the optical column as shown in Fig. 1c. The role of condenser lenses (which narrow and focus the beam by interacting with the paths of electrons) and apertures (which are metal disks with a central hole selecting the central portion of the beam) is to project a demagnified image of the crossover to the sample surface, progressively adjusting the current density and the convergence of the electron beam; the final probe-forming lens system, composed by scanning coils and objective lens, deflects and raster the electronic beam over X-Y positions of the target surface with a final spot size of the order of 5–200 nm. While classical, optical lenses deflect photons by refraction, electromagnetic condenser lenses focus the electrons making them interact with an applied electromagnetic field, as described by Lorentz force: F ¼ qðE + v  BÞ ¼ −eðv  BÞ

(6)

Where F, E, v, and B are vectors describing the Lorentz force, the electric field, the electrons velocity, and the magnetic field, respectively while q and e are a generic charge and the electron elementary charge. Since there is no electric field applied through the lenses, the electrons are forced uniquely by the magnetic field generated by the lenses to converge toward the Z optical axis through helicoidal paths. The more the electron paths are off-axis, the more they are bent toward the Z axis. In this regard, the behavior of the condenser lenses toward the electrons is analogous with that of glass convergent lenses toward photons, and most of the concepts of classical optics can be translated to electron optics. For instance, electron optics, as classical optics, can be affected by spherical aberration (focusing toward a spherical region rather than a plane), chromatic aberration (bending of electron paths more or less than needed, due to a distribution of energies/wavelengths of the electrons around an average value), astigmatism (unprecise focusing due to inhomogeneous magnetic field along X and Y axis), diffraction (which is due to the short wavelength of electrons—see Eq. (1)—and results in the spread of the current density over a diffraction pattern). Further details of electron optics can be found in the relevant literature (see Further reading). The final spot over the sample determines the resolution (together with the spread of interaction volume as described in Section 3), which is given by the diameter of the crossover d0 divided by the product of the demagnification (M) of each lens in the electron optical system. M is given by S0/Si, where S values are the focusing distances at each side of the different lenses. The distances may easily be changed by tuning the strength of the electromagnetic lenses. The working distance is defined as the distance from the bottom pole piece of the objective lens to the sample surface (S in Fig. 1c). In most instruments S is of the order of 5–25 mm. Such a long distance is necessary to prevent interactions of the emitted electrons or radiations with the magnetic field of the lens. The objective aperture has the function of decreasing the divergence angle of the electron beam and, in combination with the strength of the condenser lens, determines the current in the final probe, according to the brightness Eq. (5). The latter implies that beam current (1 pA to 1 mA), beam diameter (5 nm to 1 mm) and beam convergence a (10−4 to 10−2 sr) cannot be adjusted independently, and improving the resolution inherently means worsening current and contrast, and vice versa.

194

3

Methods and Instruments | Scanning Electronic Microscopy

Electron beam-target interactions

The electrons striking the surface of the sample interact with it in several different ways. For a given beam, characterized by a certain diameter, current and divergence, the diameter of the area sampled should ideally be equal to this diameter. However, because of electron scattering this is not the case. Scattering is defined as an interaction of the incident electrons with electrons and nuclei in the specimen, which causes a change of the electron trajectory and/or energy. The key concepts in describing scattering are the cross section (cm2), that quantifies the probability of a certain event, and the mean free path (cm) that gives the average distance traveled by the electrons between scattering events. Those may be divided in two classes: (i) elastic scattering, in which the collisions of the beam electrons with the sample only affect their trajectories without significant energy transfer; (ii) inelastic scattering, in which part of the beam energy is transferred to the sample, which then dissipates the excess energy by processes leading to further analytical signals. Each of the beam electrons undergoes multiple, subsequent, elastic and inelastic scatterings. When a scattering event occurs, the beam electrons deviate from their original path and diffuse through the specimen. Inelastic scattering progressively reduces the energy of the beam electrons until those either escape from the surface or are captured by the solid. This limits the range of travel of the electrons within the solid. The region over which these interactions occurs is known as interaction volume. The form of the interaction volume is schematically shown in Fig. 1d together with the region from which different signals, caused by different processes, originate. To better define the interaction volume, it is convenient to introduce the concept of electron-range as a measure of the distance traveled by an electron in the solid. Different definitions exist in the literature. The most used definition is the one after Kanaya and Okayama2 given by RKO ¼

0:0276AE1:67 0 Z0:889 r

(7)

where E0 is the beam energy (keV), r is the specimen density (g cm−3), A is the atomic weight (g mol−1) and Z is the atomic number of the target atoms. RKO may be approximated as the radius of a hemispherical interaction volume containing the envelop of the electron trajectories. RKO critically depends on the beam energy, on the sample composition and is inversely proportional to the atomic number and density of the material. Typical values for aluminum and gold at 10 keV are 1.4 mm and 0.26 mm, respectively. As a consequence of the spread of interaction volume below the sample surface, the resolution of SEM is not given by the diameter of the beam, but rather controlled by RKO, i.e. by a combination of beam energy and sample composition.

3.1

Emission of backscattered electrons

A significant fraction of the beam electrons striking the sample subsequently escape: these re-emergent electrons are known as backscattered electrons (BSE). Backscattered electrons can be used as such to gain information on the sample composition and, in addition, contribute significantly to the emission of the so-called secondary electrons. The process of backscattering takes place as result of a sequence of elastic scattering events inside the interaction volume in which the change of direction eventually results into a re-emission of electrons (which is much more probable as a sequence of multiple low-angle scatterings than of few high-angle events). A considerable amount of work, either experimental or using Monte Carlo simulation, has been done to evaluate the behavior of the so-called backscattered electrons coefficient , defined as the ratio between the number of backscattered electrons nBSE and the number of beam electrons striking the target nBE ( ¼ nBSE/nBE).  strongly depends on the atomic number of the atoms in the sample, increasing with increasing atomic number, while it is almost irrelevant with respect to the beam energy (the higher the energy, the longer the electrons can penetrate the sample, but the longer is the distance they have to travel in the backward direction to escape). The backscatter coefficient exhibits a pronounced dependence on the tilt angle y between the surface normal and the beam direction: (y) ¼ (1 + cosy)p, where p ¼ (9/Z)1/2. This means that the more the sample is tilted with respect to the normal to the beam, the higher the number of backscattered electrons produced and the intensity of the signal, providing a “number component” to the topographic contrast. As the backscattered electrons have traveled some distance within the solid, they escape with reduced energy E. The plot of the energy distribution as a function of the ratio W ¼ E/E0 is a peak rather sharp for heavy elements and rather broad for light elements. The peak shifts toward W ¼ 1 as the atomic value increases. Typical energies of backscattered electrons for medium and high atomic number materials are peaked in the range 0.8–0.9E0.

3.2

Emission of signals from inelastic scattering

During inelastic scattering energy is transferred to the nuclei and electrons in the target. Among the several possible processes, the most important from the point of view of scanning electron microscopy are emission of secondary electrons and production of X-rays.

3.2.1 Emission of secondary electrons Secondary electrons (SE) are low-kinetic-energy (0–50 eV) electrons formed in the sample upon inelastic scattering of incident beam electrons (as they enter the target) and backscattered electrons (as they exit). Because of the low kinetic energy with which they are generated, the mean escape depth of secondary electrons is of the order of 5–10 nm for metals and insulators, respectively (Fig. 1d). Commonly, more secondary electrons are generated by backscattered electrons than by incoming beam electrons, for two

Methods and Instruments | Scanning Electronic Microscopy

195

reasons: (i) BSE escape with lower energy, increasing the inelastic scattering cross-section; (ii) BSE escape with transverse direction, increasing the actual path length through the escape depth below the sample surface. The yield of secondary electrons generation d ¼ nSE/nBE (where nSE is the number of secondary electrons emitted from the target bombarded by nBE beam electrons) is relatively insensitive to composition. On the other side, it has a non-monotonous dependence on the energy of the beam: higher energy values allow higher penetration depths (but the escape depth is still controlled by the low SE kinetic energy), but at the same time lower energy values increase the inelastic scattering cross-section. The most important parameter influencing d is the relative orientation between the beam and the sample surface investigated. According to the equation d(y) ¼ d0/cosy, where y is the tilt angle between the surface normal and the beam direction and d0 is the yield at normal incidence, more electrons are emitted from tilted surface elements.

3.2.2 Production of X-rays When a beam electron with sufficient energy causes the ejection of an inner-shell electron, the atom is left in an ionized and highly energetic state. The atom relaxes to its original ground state in about 10−12 s after ionization. In the relaxation process, transitions of electrons occur from outer to inner shells. The most probable result of these transitions is the emission of the excess energy as high-energy photons with an energy equal to the energy difference between the shells involved in the transition (characteristic X-rays). For light elements, the dissipation of the excess energy may also involve the emission of Auger electrons. The emission of characteristic X-rays is superimposed to a continuous background due to the deceleration of the energetic beam electrons in the columbic field of the atoms of the sample (bremsstrahlung radiation). The X radiation produced by electron impact (either as characteristic or continuous X-rays) can also be absorbed by atoms in the sample causing the ejection of a photoelectron from the inner shells (X-ray absorption), with possible subsequent X-ray fluorescence when the involved atom relaxes to the ground state. Since for every element the possible X transitions are dictated by the characteristic electron configuration and by the quantum-mechanics selection rules, characteristic X-rays are the fingerprint of an element. Due to their high energy (the upper limit is the energy of the electron beam), characteristic X-rays are generated over a substantial fraction of the interaction volume as shown in Fig. 1d. The actual sampling depth depends on the density of the sample and the energy of the beam; the internal fluorescence phenomena may extend the depth at which X-rays are generated beyond the interaction volume described by BSE.

4

Formation of image

The electron beam entering the specimen chamber strikes the sample at a certain location, interacting within the interaction volume as previously described. The produced signals, as measured with suitable detectors, are used to determine certain properties of the specimen, e.g., local topography, compositions, etc. The scanning system, schematically shown in Fig. 2, is composed by electromagnetic coils that drive the beam at different locations in the X-Y plane producing a grid pattern. Each of the scanned surface elements emits signals from the interaction volume (BSE, SE, X-rays), and a corresponding X-Y matrix is built, which contains in each cell the intensity of the measured signal(s). In elder instrumentations, the synchronization of the scanning coil with the electron beam scanning a cathode ray tube (CRT) allowed a direct, analogous transfer of the local signal intensity to the brightness of the spots on the CRT without distortion and with a magnification factor M, given by the ratio between the length along the screen in the CRT space (L) and the length scanned on the specimen (l). Nowadays, the signal output from the detector(s) is digitized and subsequently visualized or stored in a multitude of media of different sizes, so that L can change when the visualization media is changed. Therefore, it is convenient overlapping a distance marker on images together with reporting M, which remains related with the performance of the electron optics.

BSE X rays SSD SE/BSE

control / display / recording system

scanning coils

scan control

Fig. 2 Controlled raster process of the electron beam over the sample surface and synchronized emission, detection, recording of the analytical signals emitted upon beam/sample interactions. SSD: solid-state detector; E-T: Everhart-Thornley detector; SE: secondary electrons; BSE: backscattered electrons; EDS: energy-dispersive X-ray spectrometer.

196

Methods and Instruments | Scanning Electronic Microscopy

Other important concepts related to the scanning action are picture element diameter and depth of field. The picture element is the specimen region from which information is transferred to a single spot on the image. Given a certain spot size, the picture element diameter is given by the ratio spot size/M. For instance, if the spot size is 100 mm, the picture element diameter is 10 mm and 1 nm at magnifications 10x and 100,000x, respectively. An image is in a true focus when the area sampled is smaller than the picture element size, and in this case the picture element size defines the resolution of the image. For a given beam size diameter, an increase of magnification above a certain value may result in an overlapping of picture elements and blurring of the image with a loss of information and actual resolution. Depth of field defines the range of distances, above and below the focal plane on the specimen surface, where the image remains sharp without blurring. It is related to the picture element size and is important when dealing with rough specimens, which present features at various working distances. In fact, the depth of field is determined by the distances above and below the optimum focus where the beam diverges in a way that it overlaps several picture elements. From geometrical considerations it turns out that the depth of field is inversely proportional to the product of magnification and final divergence, and can be adjusted by proper selection of the final aperture size and working distance. In practice to observe rough specimen the depth of field must be maximized by selecting the smallest as possible aperture and the longest working distance. Scanning electron microscopy provides much larger depth of field than any optical microscopy technique. This is one of the most relevant strength points of SEM and, together with the sensitivity to sample surface topography, contributes in yielding convincing three-dimensional representations of investigated samples.

4.1

Electron detection and image reconstruction

As the result of scanning the electron beam over the sample, secondary or backscattered electrons leave sequentially the sample surface elements, and a synchronized collection of the emitted electrons, generally by a scintillation detector-photomultiplier system (Fig. 2), is carried out to reconstruct an image. The best-known multi-purpose detector is the Everhart-Thornley detector consisting of scintillator material (doped plastic or glass target, or compound such as europium doped CaF2) which, when struck by energetic electrons, produces photons. The scintillator is covered by a thin aluminum layer biased at high positive potentials (about 10 kV) that accelerate the low energy secondary electrons. A light pipe conducts the photons to a high gain photomultiplier and suitable electronics. The scintillator is surrounded by a Faraday cage that may be biased between −50 and +250 V. The detector is normally placed at an intermediate angle between sample surface and incident beam. Because of their high energy, backscattered electrons, coming from portions of sample surface directly facing it, can be easily collected regardless of the potential of the Faraday cage (Fig. 3a). When the Faraday cage potential is biased to a negative value (about −50 V), all secondary electrons are rejected. A high positive value (about +250 V) of the Faraday cage permits collection of all secondary electrons coming from portions of sample surface oriented in every direction, with current intensities which are controlled by the SE coefficient d, i.e., only depend on angle between impinging beam and sample surface (Fig. 3b). Nowadays, the collection of high-energy backscattered electrons is mostly carried out by dedicated solid-state detectors based on p-n junction, in the form of large rings allowing the passage of the electron beam and guaranteeing a high collection efficiency (see Fig. 2).

(a)

(b)

INCIDENT BEAM

INCIDENT BEAM

S

E-T DETECTOR

E-T DETECTOR S

BS BS

S

BS

BS

BS S

S

-50V

S

BS S

BS

BS

TARGET SAMPLE

TARGET SAMPLE

BACKSCATTERED ELECTRONS DETECTION COMPOSITIONAL INFORMATION

S

S

+250 V

SECONDARY (+BS) ELECTRONS DETECTION TOPOGRAPHIC INFORMATION

Fig. 3 Generation of (a) backscattered electrons (BS) and (b) secondary electrons (S) upon interaction between electron beam and sample, and their collection by E-T detector. Adapted from Ref. Goldstein, J.I.; Newbury, D.E.; Echlin, P.; Joy, D.C.; Fiori, C.; Lifshin, E. Scanning Electron Microscopy and X-Ray Microanalysis. Plenum Press: New York (1981)., with permission.

Methods and Instruments | Scanning Electronic Microscopy

197

Image reconstruction relies on contrast between signals coming from adjacent points, defined as: C ¼ ðSmax − Smin Þ=Smax

(8)

where Smax and Smin represent the signals detected at two adjacent points of the sample surface. Contrast is related to the specimen properties because it is generated by events in the specimen (scattering) or in its immediate vicinity (electric or magnetic fields close to surface). Therefore, contrast carries all the information needed to build an image of the sample under examination. Two basic contrast mechanisms are commonly used in gathering information to describe the sample: (i) atomic number contrast and (ii) topographic contrast. Atomic number contrast (also referred to as compositional or Z contrast) describes a difference in electron currents leaving adjacent portions of the sample with different chemical composition after interaction with electron beam. The main contribution to Z contrast is carried by backscattered electrons since their generation efficiency (BSE coefficient Z) is strictly related to Z by a strong and monotonic function. Secondary electrons have only a weaker and less predictable dependence on atomic number and, hence, produce only minor effects toward compositional contrast. In the image, regions of high average atomic number appear brighter than regions of low atomic number. Atomic number contrast mechanism can be used to collect compositional information of heterogeneous samples. Since backscattered electrons are highly energetic and come from a volume extending in a range of 10–100 nm below sample surface, the information obtained is related to the bulk of the sample rather than strictly to the surface. Topographic contrast takes place because efficiency of generation of backscattered and secondary electrons depends on the tilt angle between the beam and the normal to the sample. This angle varies because of sample roughness and, hence, causes a contrast formation which is related to the physical shape of the sample. In this way, both backscattered and secondary electrons can be used to obtain topographic information. However, the different collection efficiencies of the E-T detector toward SE and BSE signals cannot be neglected when reconstructing the sample surface topography. In fact, when the detector is biased at positive potential, all the SE emitted by each surface element in every direction, with current intensities depending on local d values, can be collected. On the contrary, only those BSE emitted in a direction toward the detector can be collected, thus with a much lower collection efficiency. Because of this, the sample topography (size, shape, tilt angle) is commonly reconstructed mostly relying on a secondary electron image, where a partial contribution of backscattered electrons is overlapped. An example of the different images which can be obtained by collecting secondary (E-T detector) or backscattered electrons (solid-state detector) from the same sample (SnOx/C composite) is reported in Fig. 4. In the SE image (Fig. 4a) the three-dimensional morphology is clearly evidenced, with a main contribution from topographic contrast which renders the orientation of surface elements, especially at the edges of graphite flakes. On the other side, the BSE image (Fig. 4b) mainly relies on Z contrast. Therefore, the morphological information is rendered much flatter, while the SnOx bright spots are better evidenced. The resolution obtainable in a SEM is controlled by several factors: (i) the brightness of the source (the much higher brightness obtained by field emission sources enables much better resolutions than thermionic sources); (ii) instrument’s electron optical performance (limited by aberrations); (iii) intrinsic contrast produced by the sample and revealed by the detector (systems with low contrast or collection efficiency need higher current and thus larger beams); (iv) sampling volume (spreading of interaction volume can produce blurring of image). Any of these aspects determines a distribution di of the signal around an average point, which can be summed in quadrature as errors, yielding the final resolution d: sffiffiffiffiffiffiffiffiffiffiffi X 2ffi (9) di d¼ i

At the same time, it can be demonstrated1 that the minimum detectable contrast C, the beam intensity iB and the frame time tf (i.e., the time necessary to acquire an image) are bound together by the threshold equation

Fig. 4 SEM images of a SnOx/C composite anode material for Li-ion batteries: (a) secondary electron micrography; (b) backscattered electron micrography. (Unpublished results).

198

Methods and Instruments | Scanning Electronic Microscopy

iB >

4∙1012 A eC2 t f

(10)

Where e is the collection efficiency of the detector for the required signal. Briefly, the threshold equation states that, to be able record in a time tf an image with a minimum contrast C, a current iB higher than that defined by Eq. (10) must be applied. Size and intensity of the beam are linked together by brightness (Eq. 4), which is defined by the source type and acceleration potential (Eq. 5). This means that, for a certain source operating at a certain potential, improving resolution (smaller d) means worsening visibility (only higher C values can be visualized) and vice versa, unless longer analysis times are allowed.

4.2

X-ray detectors and electron probe micro analysis

The continuous and characteristic X-rays emitted by the sample can be detected and analyzed to gather average and local information about the bulk elemental composition of the specimen. Historically, the first X-ray detectors relied on wavelength dispersive spectroscopy (WDS). In WDS the wavelengths (l) of the X-rays emitted by the sample are sequentially analyzed using an analyzing crystal. The X-rays that satisfy the Bragg law (nl ¼ 2d siny, where n is an integer, d is the interplanar spacing of the crystal and y is the angle of incidence) are diffracted and detected by a proportional counter. The output is a graph of the X-ray intensity vs. wavelength that may be utilized to detect the presence of specific elements. Nowadays, practically all SEM instrumentation are equipped with detectors allowing the much faster energy dispersive spectrometry (EDS). In EDS, a cooled Si(Li) detector collects single X-ray photons and converts their energy into an amount of charge/hole couples, which is proportional to the radiation energy. The recombination between negative and positive charges is forced to occur through an external circuit, and the resulting current values yield information about the energies of the incoming photons. A spectrum is thus recorded as counts versus energy of photons, which allows multi-element qualitative and, if the instrumentation is properly calibrated, quantitative analysis down to 100 ppm level. Despite the lower intrinsic resolution (which is limited by the number of channels in multi-channel analyzers), EDS is the most used detection/analysis technique for Electron Probe Micro Analysis (EPMA), because of the very high readout speed. In fact, each single photon can be counted in few ms, leading to several thousand of counts per second, and the simultaneous readout at different energies in multi-channel analyzers gives the possibility to easily raise the signal-to-noise ratio in few minutes. The possibility to adjust the size of the beam allows to investigate the elemental composition of the sample at different scales, ranging from average to local compositions down to nanometer scale. Coupled with the very high acquisition rate, this allows the ‘elemental mapping’ application, where the scan of the electron probe over the surface allows to collect high-resolution information on the elemental composition of each of the surface elements investigated. The data are then reported in false-color maps, where the local elemental distribution can be easily visualized, as in Fig. 5 where morphological images and elemental mapping are compared3.

Fig. 5 Secondary electron images at different magnification levels (a and b) and EPMA elemental mapping of O, Al, Mn, Co, Ni in a LiNi0.2Co0.2Al0.1Mn0.45O2 cathode for Li-ion battery. Reproduced from Ref. Minneti, L.; Marangon, V.; Andreotti, P.; Staffolani, A.; Nobili, F.; Hassoun, J. Reciprocal Irreversibility Compensation of LiNi0.2Co0.2Al0.1Mn0.45O2 Cathode and Silicon Oxide Anode in New Li-Ion Battery. El. Acta 2023, 452, 142263., with permission.

Methods and Instruments | Scanning Electronic Microscopy

5

199

Applications to the characterization of electrochemical power sources

SEM can be successfully applied to the morphological and compositional study of electrochemical power sources. Electrode morphologies can be studied at various magnification levels to gain information on homogeneity, compactness and granulometry of the materials used. The morphology of electrodes is, in fact, very important as it influences performances of devices such as batteries, fuel cells or supercapacitors. A basic information that can be obtained from SEM micrographs is the granulometry of electroactive materials, that determines the specific active area, i.e., a crucial parameter in controlling the current that can be sustained by an electrode. In addition, the mean particle size of electroactive materials conditions the electrode/electrolyte interfacial processes, as well as the solid-state diffusion, which is the rate-determining step for intercalation electrodes as graphite or mixed Li-metal oxides commonly used in commercial Li-ion batteries. Size and morphology of electroactive materials are also very important when dealing with electrodes based on intermetallic reactions, such as anodes based on Li-metal electroactive alloys, where the extent of mechanical stress caused by the reversible alloying/dealloying reactions is crucial in determining the electrode long term integrity. Morphology of structures able to sustain and buffer mechanical stress can be easily characterized by SEM and related to electrochemical performances. Figs. 6a and b show the cycling behavior at 0.8C rate and SEM images of a Lithium-ion battery anode composed by a Ni3Sn4 alloy4. The alloy has been electrodeposited onto a nanoarchitectured Cu current collector, in which the Cu nanorods were grown by a template electrodeposition process. The first irreversible activation leads to formation of Li4.4Sn and dispersed nanosize Ni powder. The excellent capacity retention that is obtained in the following cycles involving LidSn alloying can be explained by the observed morphology of the nanostructured electrode: the volume variations upon cycling are buffered by the dispersed Ni powder and by the large free volume between the Cu pillars. These conclusions are validated by the micrograph in Fig. 6c, which shows no substantial variation of morphology after cycling.

Fig. 6 Electrochemical behavior (a) and scanning electron microscopy (SEM) micrographs (top view of nanorods) acquired before (b) and after (c) electrochemical tests of a Ni3Sn4 alloy electrodeposited onto a nanoarchitectured Cu current collector. Adapted from Ref. Hassoun, J.; Panero, S.; Simon, P.; Taberma, P.L.; Scrosati, B. High-Rate, Long-Life Ni-Sn Nanostructured Electrodes for Lithium-Ion Batteries. Adv. Mater. 2007, 19, 1632–1635., with permission.

200

Methods and Instruments | Scanning Electronic Microscopy

This represents an example of post-mortem analysis of active materials and electrodes, for which in recent years SEM has become one of the techniques of choice, with the aim to investigate possible correlations between morphological rearrangements, or the lack of, with cell performance or failure. Fig. 7 evidences the potentiality of SEM as a multi-purpose investigation technique for materials at the micro- and nano-scale. Figs. 7a–g reports SEM images obtained from a series of graphite anodes coated by Sn layers of different thicknesses using the physical vapor deposition (PVD) technique, in order to hinder electrolyte decomposition and enhance low-temperature kinetics5. All the electrodes show uniform dispersion of graphite flakes of similar morphology and size, in the micrometer range. In Figs. 7a–f, the characteristic response of E-T detector to SE and BSE is shown: the edges of the particles are brighter, as predicted by the dependence of both secondary and backscattered electrons coefficients on tilt angle, thus yielding topographic contrast; in addition, bright spots reveal the presence of Sn particles (especially in Fig. 7f ), according to the dependence of BSE coefficient on atomic number and yielding a Z contrast component to the images. Fig. 7g provides the BSE micrography of the same sample investigated in Fig. 7f, by using solid-state detector. Here, the Z contrast is much more noticeable, revealing some dendritic accumulation of higher amounts of Sn, while topographic information is poorer. Finally, Fig. 7h reports the quantitative EPMA determination of relative Sn amounts in the different samples investigated. The relative Sn contents have been determined by integrating the X-ray peaks corresponding to La1 and La2 emission lines of Sn, yielding points in the intensity (arbitrary units) vs. Sn coating thickness (A˚ ) which can be interpolated by a straight line, confirming the effectiveness and reliability of PVD process. The Figs. 8a–c show SEM images and EPMA elemental analysis of a Pd7Cu-SPUTTERED Nafion ® 115 membrane at different magnification levels where the metal coating was applied to reduce methanol crossover in direct methanol fuel cells6. At low magnification (500) the image (Fig. 8a) shows that cracks are present as a result of the differences in expansion coefficients between the Nafion membrane and the sputtered metal films. The image (Fig. 8b) taken at high magnification (20,000x) in cracks-free regions demonstrate that the thin metal layers metal films are uniformly formed on the membrane surface, resulting into an efficient physical barrier for methanol crossover from the anode to the cathode compartment. Protons, on the contrary, can pass through the metal barrier via hydride (PdH) formation at the surface facing the anode followed by oxidative desorption at the surface facing the cathode. EDS analysis (Fig. 8c) reveals chemical composition of coating and, after proper calibration, a Pd:Cu ratio of 54:46 (wt%), quite close to the theoretical one 60:40. Recent advances in the development of commercial variable-pressure (VPSEM) and environmental scanning electron microscopes (ESEM) enabled the study of the synthesis of battery materials, and the in situ observation of batteries during operation, at pressure values up to few kPa (typically up to 2700 Pa ¼ 20 Torr). This capability is guaranteed by the combination of a 10 to 30-keV electron beam and a thin sample chamber having a pressure-limited aperture (diameter of few hundred micrometers) placed immediately above the sample, which minimizes gas leaking toward the rest of the optical column and limits the interactions between gaseous environment and electrons. As an example, at 100 Pa the mean free path of 25-keV electrons is about 1 cm, comparable with the typical working distances from the objective lens to the sample surface. Figs. 9a–h report the sintering behavior at T ¼ 775  C of NASICON-type Li1.3Al0.3Ti1.7(PO4)3 (LATP) solid electrolyte under different environmental pressures7. When observing the morphologies of the samples sintered under different conditions, major differences can be noticed. When the sample is heated under vacuum (Fig. 9b), no visible changes in the morphology are detected. At P ¼ 10 Pa, a considerable grain growth and formation of large pores is detected. Furthermore, the treatment results in a

Fig. 7 (a–f ) Secondary electrons micrographs of electrodes A–F coated by Sn layers of different thicknesses (0, 50, 100, 150, 250, 500 A˚ ). (g) Backscattered electrons image of electrode F. (h) Areas of convolutions of Sn–La2 and Sn–La1 X-ray peaks. Adapted from Ref. Nobili, F.; Mancini, M.; Stallworth, P.E.; Croce, F.; Greenbaum, S.; Marassi, M. Tin-Coated Graphite Electrodes as Composite Anodes for Li-ion Batteries. Effects of Tin Coatings Thickness Toward Intercalation Behavior. J. Power Sources 2012, 198, 243–250., with permission.

Methods and Instruments | Scanning Electronic Microscopy

201

Fig. 8 SEM micrographs of Pd:Cu coatings over Nafion W 115 membrane at (a) 500x; (b) 20,000x magnification levels. (c) EPMA-EDS spectrum of sample surface. Reproduced from Ref. Prabhuram, J.; Zhao, T.S.; Liang, Z.X.; Yang, H.; Wong, C.W. Pd and Pd-Cu Alloy Deposited Nafion Membranes for Reduction of Methanol Crossover in Direct Methanol Fuel Cells. J. Electrochem. Soc. 2005, 152, A1390–A1397., with permission.

morphology different from the rectangular and/or cubic commonly obtained ex-situ at ambient pressure (Fig. 9h). On the other hand, when the sample is treated at P ¼ 750 Pa the product is densified and characterized by the common rectangular shape (Fig. 9f ). Thus, the ESEM observation allows to state that the grain growth can be achieved by environmental pressure higher than 750 Pa, while treatment performed under vacuum and P ¼ 10 Pa leads only to the chemical transformation (from precursors to LATP) without any morphological rearrangement. ESEM can also be used for operando probing interfacial processes between current collectors, electrodes, and electrolytes, as in Fig. 10 where the behavior of a silicon-graphite composite anode in two different ionic liquid electrolytes is monitored8. Specifically, the tested electrolytes are 1.2 M LiTFSI in EMITFSI, and 1.2 M LiFSI in PYR13FSI. The in situ ESEM acquired on the first cycle of cells containing EMITFSI (Figs. 10a–i) and PYR13FSI (Figs. 10a’-i’) electrolytes, at three different environmental temperatures (20, 40, and 60  C), are reported. Temperature- and electrolyte-dependent contraction and expansion phenomena of electrodes, as well as interface roughening and electrode detaching from current collector, are clearly evidenced during lithiation and delithiation, allowing an effective real-time monitoring of the interfaces in operating conditions. A further use of ESEM is given in Fig. 11, where ORR and OER in a solid-state LidO2 battery are studied9, by feeding the electrodes with an atmosphere of O2 at 200 Pa. The microscale battery used in this work consists of Li metal, the native Li2O layer on it, and the super aligned carbon nanotube (SACNT) as the anode, the electrolyte, and the cathode, respectively. The formation of Li2O2 during discharge at the triple-phase boundary is observed, and eventually its decomposition starting from the surface and proceeding into the bulk along a certain direction, rather than from the triple-phase boundary. This in situ observation allowed by ESEM suggests that the decomposition of the Li2O2 particle may be limited from the transport of oxygen rather than ionic or electronic transport.

202

Methods and Instruments | Scanning Electronic Microscopy

Fig. 9 Representative SEM images of the compact electrolyte before heating (a, c, e, g) and after heating at T ¼ 775  C for 2 h: (b) under vacuum (10−2 Pa); (d) P ¼ 10 Pa; (f ) P ¼ 750 Pa; (h) atmospheric pressure (105 Pa). The heating at 10−2, 10, and 750 Pa is performed in situ, and the heating at atmospheric pressure is performed outside the SEM using the same heating device. The scale bar in panel (a) applies to all images. Reproduced from Ref. Camara, O.; Xu, Q.; Park, J.; Yu, S.; Lu, X.; Dzieciol, K.; Schierholz, R.; Tempel, H.; Kungl, H.; George, C.; Mayer, J.; Basak, A.; Eichel, R.-A. Effect of Low Environmental Pressure on Sintering Behavior of NASICON-Type Li1.3Al0.3Ti1.7(PO4)3 Solid Electrolytes: An In Situ ESEM Study. Cryst. Growth Des. 2023, 23, 1522–1529., with permission.

Methods and Instruments | Scanning Electronic Microscopy

203

Fig. 10 In situ SEM images of the Si/C electrode morphological evolution with EMI (a–i) and PYR (a’-i’) as electrolyte during the first cycle, where a-c/a’-c’, d-f/d’-f’, g-i/g’-i’ correspond to operating temperature of 20  C, 40  C and 60  C, respectively, scale bar ¼ 200 mm. Reproduced from Ref. Wu, R.; Liu, X.; Zheng, Y.; Li, Y.; Shi, H.; Cheng, X.; Pfleging, W.; Zhang, Y. Unveiling the Intrinsic Reaction Between Silicon-Graphite Composite Anode and Ionic Liquid Electrolyte in Lithium-Ion Battery. J. Power Sources 2020, 473, 228481., with permission.

204

Methods and Instruments | Scanning Electronic Microscopy

Fig. 11 Discharge and charge processes of the LidO2 battery. (a) Images captured at 0, 500, 1000, and 3000 s show the growth process of a spherical particle, which can grow up to 1.5 mm. −3 V was applied on SACNT vs. Li metal to initiate the discharge process. Yellow arrows indicate that the spherical particle grew up at a CNT-solid state electrolyte-oxygen TPI. Note that the CNT curved probably due to the sample drift. (b) Images captured at 0, 900, 1800, and 3200 s show the decomposition process of the spherical particle. 8 V was applied on SACNT vs. Li metal to initiate the charge process. Red arrows indicate the position where the particle decomposed. Reproduced from Ref. Zheng, H.; Xiao, D.; Li, X.; Liu, Y.; Wu, Y.; Wang, J.; Jiang, K.; Chen, C.; Gu, L.; Wei, X.; Hu, Y.-S.; Chen, Q.; Li, H. New Insight in Understanding Oxygen Reduction and Evolution in Solid-State Lithium–Oxygen Batteries Using an In Situ Environmental Scanning Electron Microscope. Nano Lett. 2014, 14, 4245–4249.

6

Concluding remarks

Scanning electron microscopy is a powerful tool to investigate morphological and compositional properties of electrodes for electrochemical energy storage applications. The use of a scanning electron beam allows better magnification and focal depth respect to optical microscopy, still providing images with high contrast and stereoscopy, due to the configuration of secondary electrons detector. In addition, analysis of backscattered electrons can provide information about composition and homogeneity of samples. The concurrent use of electron probe microanalysis and elemental mapping allows the average and local investigation of the elemental composition of very narrow portions of target sample (at qualitative and quantitative level). As regards the applications to electrochemical power sources, SEM is nowadays the technique of choice to study the correlations between morphology and electrochemical behavior. In this context, ex situ and operando methods have recently gained growing attention for monitoring performance evolution and investigating possible failure phenomena, making SEM a unique tool for scientists studying energy-related materials and devices.

References 1. Goldstein, J. I.; Newbury, D. E.; Echlin, P.; Joy, D. C.; Fiori, C.; Lifshin, E. Scanning Electron Microscopy and X-Ray Microanalysis; Plenum Press: New York, 1981. 2. Kanaya, K.; Okayama, S. Penetration and Energy-Loss Theory of Electrons in Solid Targets. J. Phys. D Appl. Phys. 1972, 5, 43–58. 3. Minneti, L.; Marangon, V.; Andreotti, P.; Staffolani, A.; Nobili, F.; Hassoun, J. Reciprocal Irreversibility Compensation of LiNi0.2Co0.2Al0.1Mn0.45O2 Cathode and Silicon Oxide Anode in New Li-Ion Battery. El. Acta 2023, 452, 142263. 4. Hassoun, J.; Panero, S.; Simon, P.; Taberma, P. L.; Scrosati, B. High-Rate, Long-Life Ni-Sn Nanostructured Electrodes for Lithium-Ion Batteries. Adv. Mater. 2007, 19, 1632–1635. 5. Nobili, F.; Mancini, M.; Stallworth, P. E.; Croce, F.; Greenbaum, S.; Marassi, M. Tin-Coated Graphite Electrodes as Composite Anodes for Li-ion Batteries. Effects of Tin Coatings Thickness Toward Intercalation Behavior. J. Power Sources 2012, 198, 243–250. 6. Prabhuram, J.; Zhao, T. S.; Liang, Z. X.; Yang, H.; Wong, C. W. Pd and Pd-Cu Alloy Deposited Nafion Membranes for Reduction of Methanol Crossover in Direct Methanol Fuel Cells. J. Electrochem. Soc. 2005, 152, A1390–A1397. 7. Camara, O.; Xu, Q.; Park, J.; Yu, S.; Lu, X.; Dzieciol, K.; Schierholz, R.; Tempel, H.; Kungl, H.; George, C.; Mayer, J.; Basak, A.; Eichel, R.-A. Effect of Low Environmental Pressure on Sintering Behavior of NASICON-Type Li1.3Al0.3Ti1.7(PO4)3 Solid Electrolytes: An In Situ ESEM Study. Cryst. Growth Des. 2023, 23, 1522–1529. 8. Wu, R.; Liu, X.; Zheng, Y.; Li, Y.; Shi, H.; Cheng, X.; Pfleging, W.; Zhang, Y. Unveiling the Intrinsic Reaction Between Silicon-Graphite Composite Anode and Ionic Liquid Electrolyte in Lithium-Ion Battery. J. Power Sources 2020, 473, 228481. 9. Zheng, H.; Xiao, D.; Li, X.; Liu, Y.; Wu, Y.; Wang, J.; Jiang, K.; Chen, C.; Gu, L.; Wei, X.; Hu, Y.-S.; Chen, Q.; Li, H. New Insight in Understanding Oxygen Reduction and Evolution in Solid-State Lithium–Oxygen Batteries Using an In Situ Environmental Scanning Electron Microscope. Nano Lett. 2014, 14, 4245–4249.

Methods and Instruments | Scanning Electronic Microscopy

205

Further reading 1. 2. 3. 4. 5. 6. 7. 8.

Carter, C. B.; Williams, D. B. Transmission Electron Microscopy. In A Textbook for Materials Science, 2nd edn; Springer: New York, 2009. Egerton, R. F. Physical Principles of Electron Microscopy: An Introduction to TEM, SEM and AEM; Springer: New York, 2005. Flegler, S. L.; Heckman, J. W.; Klomparens, K. L. Scanning and Transmission Electron Microscopy: An Introduction; Oxford University Press: New York, 1995. Goldstein, G. I.; Newbury, D. E.; Michael, J. R.; Ritchie, N. W. M.; Scott, J. H. J.; Joy, D. C. Scanning Electron Microscopy and X-Ray Microanalysis, 4th edn; Springer: New York, 2018. Hillier, J.; Baker, R. F. Microanalysis by Means of Electrons. J. Appl. Phys. 1944, 15, 663–675. Nobili, F.; Staffolani, A. Structural Analysis: Transmission Electron Microscopy. In Encyclopedia of Electrochemical Power Sources; Garche, J., Ed, 2nd edn; Elsevier: Amsterdam, 2023 (this book). Watt, I. M. The Principles and Practice of Electron Microscopy, 2nd edn; Cambridge University Press: Cambridge, 1997. Zworykin, V. A.; Hillier, J.; Snyder, R. L. A Scanning Electron Microscope. ASTM Bull. 1942, 117, 115–123.

Methods and Instruments | Transmission Electron Microscopy F Nobili and A Staffolani, School of Science and Technology—Chemistry Division, University of Camerino, Via Madonna delle Carceri, Camerino (MC), Italy © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of R. Marassi, F. Nobili, MEASUREMENT METHODS | Structural and Chemical Properties: Transmission Electron Microscopy, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 769–789, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5. 00072-1.

1 2 3 3.1 3.2 4 4.1 4.2 4.3 4.4 5 5.1 5.2 6 7 8 References Further reading

Introduction General principles Scattering of electrons Electron diffraction Inelastic scattering Instrumentation and operation Gun assembly Electron lenses, aberrations and resolution The illuminating system Objective lens and stage: Image formation Contrast mechanisms Amplitude contrast Phase contrast Chemical analysis mode Applications to analysis of electrochemical power sources Concluding remarks

208 208 208 210 212 213 213 213 216 217 219 219 220 221 221 228 229 229

Abstract Transmission Electron Microscope (TEM) and related techniques, such as High-Resolution Transmission Electron Microscopy (HRTEM), Scanning Transmission Electron Microscopy (STEM), principles and applications in the field of electrochemical power sources are briefly described. Instrumentation and operation are introduced together with the basis of electron scattering phenomena that generate contrast both in the real (imaging mode) and reciprocal space (diffraction mode). Applications of TEM to the study of materials in electrochemical devices are discussed demonstrating the capability of this technique in the characterization of electrochemically active materials down to the atomic scale, also by the mean of innovative methods such as operando- and cryo-STEM.

Glossary

Aperture Circular hole in a metallic diaphragm that limits the collection angle b of the lenses. It permits the control of resolution, depth of field, focus and contrast in the image and of angular resolution in the diffraction pattern. Backscattered electrons Electrons of the incident beam that re-emerge from the target sample through an angle >90
as a result of a sequence of elastic scattering events inside the interaction volume. Bright field TEM imaging technique in which direct beam forms the image. Dark field TEM imaging technique in which diffracted beams form the image. Diffraction contrast Contrast generated by coherent electron waves elastically scattered, depending on crystal structure and orientation of the specimen. It allows imaging of any internal interface or crystal defect that affects coupling of direct and diffracted beams. Electron gun A stable source of electrons used to form an electron beam characterized by specific energy, current, divergence angle and brightness. Electron scattering Electrostatic interaction of an electron with an atom (nucleus and/or electron cloud) that causes deflection of its trajectory. Elastic scattering results in no loss of energy for the incident electron. Inelastic scattering results in energy transfer between the incident electron and the target atom. Focal plane Plane when parallel rays are brought to focus. In TEM the focal plane is located after the lens: back focal plane at a distance f (focal length). High-resolution transmission electron microscopy A TEM imaging methodology which relies on phase contrast to project images of the crystallographic structure of a sample at an atomic scale.

206

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00206-1

Methods and Instruments | Transmission Electron Microscopy

207

Image plane Plane containing the image of an object projected by a lens. Mass-thickness contrast Contrast generated by incoherent electron waves inelastically scattered, depending on specimen mass and thickness. It is the main source of contrast for imaging of amorphous specimens. Object plane Plane containing the object to be projected by a lens. Phase contrast Contrast generated by difference in phase of the electron waves scattered by the specimen. Phonons Oscillations of lattice caused by inelastic scattering of high-energy electrons striking atoms in the specimen. Plasmons Collective oscillations of free electrons in the specimen caused by an electron beam passing through the electron gas. Resolution The smallest distance between two adjacent points that can be resolved. It is usually limited by lens aberrations. Scanning transmission electron microscopy Imaging technique that scans a highly convergent beam over the specimen. Bright Field or Dark Field images can be recorded. Scattering cross section Quantity that describes the probability of a particular electron to undergo an interaction with an atom. The total scattering cross section is an extensive quantity that introduces also a dependence on the number of target atoms in the specimen (through thickness and density). Secondary electrons Electrons ejected from target atoms in the specimen as a consequence of inelastic scattering events with an incident electron beam or with backscattered electrons. Structure factor Measure of the amplitude of the electron wave scattered by a unit cell. It depends on the atomic scattering factors of the atoms in the unit cell and on the periodicity that makes up the crystal structure.

Key points

• • • • • • • •

The working principle of transmission electron microscopy is discussed. General information about the instrumentation and its operation is provided. Electron elastic and inelastic scattering phenomena are discussed. The generation of images and of electron diffraction patterns is explained. Examples of use of TEM for the characterization of fresh and post-mortem materials for Li-ion batteries are provided. Morphological studies done by TEM of catalysts for energy conversion devices are provided. Examples of use of STEM for the characterization of battery materials are given. Examples of Operando measurements of energy storage devices by STEM are given.

Abbreviations ABF ADF AEM AES BF CDF CRT DF DP EDS EELS EG FE FSE HAADF HRTEM SA(E)D SEM STEM TEM

Annular bright field Annular dark field Analytical electron microscopy Analytical electron spectroscopy Bright field Centered dark field Cathode ray tube Dark field Diffraction pattern Energy dispersive spectrometry Electron energy loss spectrometry Electron gun Field emission Fast secondary electrons High aperture annular dark field High-resolution transmission electron microscopy Selected area (electron) diffraction Scanning electron microscopy Scanning transmission electron microscopy Transmission electron microscopy

208

1

Methods and Instruments | Transmission Electron Microscopy

Introduction

In Transmission Electron Microscopy (TEM) a high-energy electron beam (acceleration voltage up to 300 keV, but more commonly in the range 100 to 150 keV) is focused on a thin specimen, and the electrons transmitted (directly and/or after diffraction) build an image or a diffraction pattern (with the latter which is an image in the reciprocal space) on a visualization screen. For a specimen to be thin means being “electron transparent”: i.e., enough electrons are transmitted so that the intensity falling on a fluorescent screen, or a suitable electron detector such as a semiconductor detector, is sufficient to give an image with suitable contrast and resolution. This requirement is a function of the electron energy and of the average atomic number and thickness of the specimen. The typical sample thickness is in the order of 100 nm or less, requiring very accurate specimen preparation. However, if the conditions are met, TEM allows a microscopical characterization of the investigated samples which can be pushed down to the nanoscale and, when local diffraction phenomena come into play, down to sub-atomic scale. This introduces the main similarities and differences of TEM with respect to Scanning Electron Microscopy (SEM). In fact, both use an accelerated electron beam to probe a sample and share some similar instrumentation parts such as electron sources, electron lenses, and apertures. However, SEM gathers electrons emitted above a bulk sample to acquire information about its surface topography and composition. On the contrary, TEM records electrons transmitted and diffracted below a thin sample to record its bulk image in direct and reciprocal space. This allows to gain information about sample morphology and, at the same time, about its crystallinity and structure.

2

General principles

The fathers of electron microscopy were Knoll and Ruska,1 and the first commercial TEM was built in 1939 by Siemens. Since then, the theory and the instrumentation have developed and modern TEMs have become a fundamental tool for material science. TEM relies on the particle-wave dual nature of the electrons introduced by De Broglie in 1925 and described by h 1:22 l ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi  1=2 2m0 eV E

(1)

where h is Planck’s constant, m0 and e are the rest mass and the charge of an electron, respectively, and V is the accelerating voltage. The approximation yields the numeric value of wavelength l (nm) as a function of the energy E (keV). This means that, by increasing the accelerating voltage, the wavelength of the electrons decreases. For instance, at 100 keV an electron wavelength l  0.004 nm is obtained. This value is smaller than the diameter of an atom, meaning that, according to the classical Rayleigh’s criterion for optical microscopy (which states that the smallest distance which can be resolved is comparable to the investigated wavelength), electrons can be used as a probe for sub-atomic distances. Practical TEM resolution is limited by aberration of lenses and, in dedicated instruments, can reach values as low as 0.2–0.1 nm. Particularly, in high-resolution TEM (HRTEM) diffracted electrons project the electron density at the atomic scale, making the technique the most powerful microscopy available, both in terms of resolution and contrast. High-energy electrons hitting a thin specimen are scattered by the atoms producing a wide range of signals that may be broadly classified based on the type of interactions, elastic or inelastic, taking place. As a result of the scattering processes the electrons emerging from the exit surface of the specimen are non-uniformly distributed. This non-uniform distribution contains all the structural and chemical information about the specimen. The electron microscope is constructed to display this non-uniform distribution in two different ways: (i) angular distribution resulting in a diffraction pattern (image in reciprocal space) and (ii) spatial distribution of scattering that generates the contrast in the image of the specimen in direct space.

3

Scattering of electrons

Electrons are low mass, negatively charged particles that can be easily deflected by passing close to the electron cloud or the positive nucleus of an atom. These electrostatic interactions cause the electron scattering that makes TEM feasible. Because of the dual nature of electrons, an electron beam can be treated in two different ways: in electron scattering it is a succession of particles while in electron diffraction it is treated by wave theory. If electrons are considered as particles the terms elastic and inelastic are simply descriptions of scattering that results in no loss of energy or some measurable loss of energy, respectively. An alternative description, referring to their wave nature, is to separate scattered electrons into coherent and incoherent. Elastically scattered electrons are usually coherent, while inelastic electrons are usually incoherent. When, as in TEM, a coherent, monochromatic electron wave illuminates a specimen, the coherently scattered electrons are those that remain in phase, while incoherently scattered electrons have no phase relationship after interacting with the specimen. Scattering can be either forward scattering or back scattering, where the terms refer to the angle of scattering with respect to the incident beam and a specimen normal. If an electron is scattered through an angle less than 90
it is forward scattered. If the angle is greater than 90
it is backscattered. Forward scattering causes most of the signals used in TEM. A simple model of the scattering process is shown if Fig. 1. In general, it may be stated that: when the specimen is sufficiently thin and crystalline, elastic scattering is usually coherent and occurs in the forward direction at low angles (1–10
).

Methods and Instruments | Transmission Electron Microscopy

209

Fig. 1 Different kinds of electron scattering phenomena arising from interaction of an accelerated electron beam with a thin target specimen, giving rise below the specimen to the analytical signals used in TEM to generate images in direct and reciprocal space. Adapted from Ref. Carter, C.B.; Williams, D.B. Transmission Electron Microscopy. A Textbook for Materials Science; 2nd ed.; Springer: New York, 2009, with permission.

Higher scattering angles (>10
) decrease coherency. Inelastic scattering is always incoherent, at low angles (3 M) 6 Localized high-concentration electrolytes (LHCEs,
1 M) 7 Low-concentrate electrolyte (LCEs, 3 M) The concentration of electrolyte salt is around 3.0 M in the electrolytes. Localized high-concentration electrolytes (LHCEs,
1 M) By adding cation non-coordinating solvents (also called diluents, usually a polyfluorinated ether) to dilute the parental HCEs, and the as-obtained electrolytes (usually 1–1.2 M) are referred to localized high-concentration electrolytes (LHCEs). Low-concentration electrolyte (LCEs, 1 mS cm−1, low viscosity, wide liquidus range, good thermal stability, etc.). The ionic conductivity is a crucial parameter for the electrolytes. The electrolyte should be a good ionic conductor and electronic insulator, so that ion transport can be facile and self-discharge can be kept to a minimum. Therefore, high ionic conductivity results in fast charging and discharging, which increases the power density of the cell. The conductivity of nonaqueous electrolytes is 1–2 order lower than that of aqueous electrolytes, which may influence the rate performance. The electrolyte should be a wide liquidus range, which ensures all-climate batteries operating across a wide temperature range. Meanwhile, the thermal stability of the electrolyte plays a crucial role in the working efficiency, long-term cycling stability and safety of the cell. (3) Chemical characteristics (inertness to other cell components such as binder, conductive agent, separator and cell packing materials, etc.). It is also important that the electrolyte is inertness to other cell components such as polymeric binder agents, conductive agents, separator and cell packing materials. The electrolyte should not react with moisture and air to form various unwanted products that leads to the degeneration of the electrode materials. Meanwhile, the electrolyte should a good wettability toward separator as well as electrodes. (4) Other properties. The components of the electrolyte should be environmental friendliness, wide source, simple preparation and low cost. Actually, the electrolyte must satisfy the requirement of the electrolyte/electrode interfacial stability, while it is very challenging to find an electrolyte matching perfectly with all these requirements. For this reason, there have been numerous and continuous research efforts into non-aqueous electrolytes in the future.

3

Lithium-ion battery electrolytes

Lithium-ion batteries (LIBs) (see Fig. 1), due to their ever-improving performance and cost reduction, are regarded as one of the most important electrochemical energy storage devices, and widely used in everything from smartphones to electric vehicles. For lithium-ion batteries, the interphase formed by the decomposition of the electrolyte is almost always identified on negative electrode surfaces, where it is defined as a solid electrolyte interphase (SEI (−), see Fig. 1). When interphase occurs on a positive electrode surface, it is also a solid electrolyte interphase (SEI (+), see Fig. 1). Generally, to improve the properties of the electrolytes or strength the stability of both SEIs, small amounts of electrolyte additives are incorporated.5 In 2004 and 2014, Xu Kang has reviewed on current state-of-the-art non-aqueous electrolytes for LIBs.2,6 Although early numerous studies have been conducted to adjust the electrolyte composition (solvent ratios, salt types, and additives),7 the regulation of electrolyte concentration did not arouse much concern for the electrolyte systems. Over the past decade, researchers started to recognize that the electrolyte concentration not only affects the battery cost and manufacturing but also impacts the battery performance through liquid diffusion and interfacial properties.8 The commercial LIB

Fig. 1 The working mechanism of the electrolyte for LIBs.

Cell Components – Electrolytes | Non-Aqueous Liquid Electrolyte

471

Fig. 2 The relationship between the solvation sheath structures of both Li+ and anion and the electrolyte concentration.

electrolyte is still based on the LiPF6/carbonate system, which was initially formulated more than 30 years ago. For the optimal ionic conductivity purpose, the salt concentrations used in nonaqueous electrolytes have remained in the neighborhood of
1.0 M (1 M ¼ 1 mol L−1). Specially in 2015, the groundbreaking effort was made by Wang et al. who reported a water-in-salt electrolyte consisting of 21 M bis(trifluoromethylsulfonyl)imide (LiTFSI) dissolved in water and found the electrochemical stability window of water was expanded from 1.23 to 3.0 V in aqueous LIBs.9 Yamada et al. soon independently reported a similar system termed hydrate melt.10 Both works triggered the research of electrolytes into a new direction that emphasizes the regulation of electrolyte concentration. For the electrolyte with different concentration, the solvation sheath structures of both Li+ and anions are significantly altered with their salt concentration3 (Fig. 2). In the low-concentration electrolytes, the molecular structures are predominantly solvent separated ion pair configurations, that is, the majority of the solvent molecules are in a free state and the solvated Li+ and the anions are uniformly dispersed in these free solvents without formation of contact dimers (CDs) and contact ion pairs (CIPs, anion coordinating with one Li ion). As the salt concentration increases, the population of the free solvent molecules decreases with the simultaneous formation of CIPs and aggregate clusters (AGGs, anion coordinating with two or more Li ions), as shown in Fig. 2. Therefore, changes in the concentration directly affect the solvation of the Li+ ions in the electrolytes, and subsequently corresponding changes are anticipated in the formation of the interphases. In this chapter, we will introduce the recent progress in non-aqueous liquid electrolyte for currently prevailing LIBs, focusing on conventional concentration electrolytes (CCEs,
1 M), high-concentration electrolytes (HCEs, >3 M), localized high-concentration electrolytes (LHCEs,
1 M) and low-concentration electrolyte (LCEs, 85.4%@500th, 1.0C, 2.5–4.4 V >97%@600th, 1.0C, 2.8–4.4 V

1.1 M LiFSI/TEP + TTE (1:3, v/v) 1.0 M LiFSI/TMP + TTE (1.3:2, by mol) 1.44 M LiFSI/TMP + TTE (1.4:3 by mol) 1.2 M LiFSI/TEP + BTFE (1:3 by mol)

15 16 17 18 19 20

a

RP and CNT represents red phosphorus and carbon nanotube, respectively.

carboxylate esters, and ethers are volatile and flammable liquids, which maybe the immediate cause of the spontaneous combustion of the battery, even thermal runaway issues. Extensive attempts13 have been made to develop the flame-retardant and non-flammable electrolyte systems, including flame-retardant electrolytes, ionic liquid-based electrolytes and organosiliconbased electrolytes. Considering their high flame-retardant efficiency and their miscibility with EC-base electrolytes, low-molecular-weight phosphorus-based compounds, including trimethyl phosphate (TMP), triethyl phosphate (TEP), dimethyl methyl phosphonate (DMMP) and diethyl ethyl phosphonate (DEEP), are the most common flame retardants (FRs) used in LIBs (see Fig. 4 and Table 1). It is believed that these phosphorus-based flame-retardant electrolytes will become a main research direction in the foreseeable future.21 However, when the content of FRs increases beyond >20 wt%, the electrolytes become totally non-flammable, while the detrimental impact of FRs on the cycling performance of LIBs cannot be neglected. The reason is that these FRs fail to form a stable SEI film resulting in the Gr exfoliation. Many researchers have reported that the introduction of SEI film-forming additives (see Fig. 3), such as fluoroethylene carbonate (FEC), LiDFOB and vinylene carbonate (VC), can effectively suppress the decomposition of FRs on the Gr surface.14–16 However, the effects were unsatisfactory. Recently, another efficient and effective strategy is blending hydrofluoroether (HFE) co-solvents17–20 (also called dilutes that will be discussed in the following. e.g., bis(2,2,2-trifluoroethyl) ether (BTFE), 1,1,2,2-tetrafluoroethyl 2,2,3,3-tetrafluoropropyl ether (TTE).) with the high salt concentration electrolytes, which exhibit many intriguing properties, such as extended anodic stability and effective SEI formation on Gr and Li electrodes. It has been recently reported that the low-flammable TMP-based electrolyte achieves superior cycle life to the conventional electrolyte in Gr/LiNi0.8Co0.1Mn0.1O2 (NMC811) operated within the voltage range of 2.5–4.4 V.19 In another study, an intrinsic FR electrolyte consisting of 1.1 M LiFSI in a solvent mixture of TEP-TTE (1:3, v/v) was also employed to mitigate shuttle effect of Li-S batteries.17

4.3

Fluorinated electrolytes

Besides the above-mentioned trials to explore the flame-retardant electrolytes, from 2000, great efforts worldwide were also dedicated to the development of fluorinated electrolytes for high energy density LIBs. Generally, thanks to the low polarizability, high electronegativity, and high ionic potential of F atom, compared with common organic compounds, fluorinated compounds

474

Cell Components – Electrolytes | Non-Aqueous Liquid Electrolyte

show unique physicochemical and electrochemical properties, including high oxidative stability, high even no flash point, low melting point, low surface energy, along with easy to form a compact and stable interphase layer between electrode and electrolyte. We have reviewed recent research progress in fluorinated compounds used as additives, co-solvents and diluents for high energy density LIBs.22,23

5

High-concentration electrolytes (HCEs, >3 M)

For the past one decade, the battery field has witnessed the power of high-concentration electrolytes (HCEs, >3 M) in boosting battery performance.24 Due to the very high lithium salt/solvent molar ratio (>1:1, molar/molar), HCEs exhibit unique solvation structures with the predominant CIPs and AGGs and promote the formation of an anion-derived, inorganic-rich, and robust interphase on electrode surfaces. HCEs have various unusual functionalities, including high reductive/oxidative stability, suppression of Al corrosion, low volatility and low flammability. Therefore, much-improved performances are achieved for various material systems, including Gr electrode, Li metals, alloy materials, sulfur (S), and high-voltage LiCoO2 or NMC positive electrodes.25,26 Fig. 5 shows representative research work on the HCEs for LIBs.8 Unusual interfacial properties of HCEs were noticed as far back as 1985 by McKinnon and Dahn, who reported that the solvent co-intercalation into ZrS2 layered materials could be effectively suppressed in a saturated LiAsF6/PC electrolyte (
3 M). However, the earliest serious attempt to break the confinement of the electrolyte concentration ironically happened with polymer electrolytes. Angell et al. in 1993 proposed new ionic conductors— ‘polymer-in-salt’ electrolytes in which lithium salts were mixed with small quantities of the polymers (polypropylene oxide (PPO) and polyethylene oxide (PEO)), remaining good ion conductivities and high electrochemical stability. A decade later, Ogumi and his co-workers demonstrated in 2003 and 2008 that highly concentrated PC electrolytes could both allow for highly reversible Li+ intercalation into the interlayer of the graphite electrode and significantly improve the reversibility of Li metal deposition and stripping, respectively. Such previous results motivated many researchers to propose the salt-superconcentrating strategy for universalizing the Gr electrode reaction in various organic solvents (see Table 2, e.g., dimethyl sulfoxide (DMSO), THF, DME, sulfolane (SL) or acetonitrile (AN)) other than EC. Considerable works indicated that the salt-derived SEI in HCEs could mitigate electrode and electrolyte degradation during cycling. Until 2013, Hu et al. first proposed a new class of “solvent-in-salt” (SIS) electrolyte of 7 M LiFSI/DOL+DMC (1:1, v/v), which showed a high Li+ transference number of 0.73, low lithium polysulfide dissolution, as well as an effective suppression of Li dendrite growth. The high coulombic efficiency (CE) of >99% and excellent high-rate cycling stability of Li metal electrode in the HCE of 4.0 M LiFSI/DME was also achieved. Meanwhile, several reports confirmed that carbonate-based HCEs (4.45 mol Kg−1 LiPF6/PC, 7 M LiFSI/FEC, LiFSI/EC (1:1, by molar ratio)) displayed enhanced oxidative stability of 5 V LiNi0.5Mn1.5O4 (LNMO) batteries and anti-corrosion of Al current collectors. Specially, the HCE of LiFSI/DMC (1:1.3, by molar ratio) has a high anodic stability of 5.5 V vs. Li+/Li. The high-voltage Gr/LNMO batteries with this HCE showed excellent cycling performance, high-rate capacity and improved safety. Similar HCEs based on specific Li salt-glyme equimolar mixture was also named “solvate ionic liquids” because of their various physicochemical features analogous to ionic liquids reported by Watanabe et al. in 2010. It was found that [Li(Glyme)]X (Glyme: G3 or G4; X: TFSI− or ClO−4) could form fairly stable complexes, which exhibited many desirable properties, including thermal stability up to 200  C, the ability to support reversible Li+ intercalation chemistry with Gr without the co-intercalation behavior typically associated with ethers, as well as high oxidative stability of ca. 5.0 V vs. Li+/Li. These unusual properties are attributed to the solvation of high population of Li+ by limited population of glyme molecules, leading to the elimination of free solvent. Afterwards, these solvate ionic liquids could be used as efficient electrolytes for Li-ion, Li-S and Li-O2 batteries. Owing to their unique solution structures at certain high concentrations, HCEs have various unusual properties compared to their dilute counterparts, making it unnecessary to rely on the LiPF6 salt for the passivation of the Al current collector, or on the EC solvent for formation of the SEI film on the graphite surface, and providing much design considerations in future battery technologies. Similarly, HCEs also offer new opportunities for many other battery chemistries, including but not limited to Li-O2 batteries, Li/Na/K-ion batteries, Mg/Al/Zn multivalent cation batteries, and aqueous batteries. Superior battery functions achieved through HCEs are well-summarized in some review articles.24–26

Fig. 5 Representative research work on the HCEs for LIBs.8

Cell Components – Electrolytes | Non-Aqueous Liquid Electrolyte Table 2

Properties of the reported HCEs used in rechargeable batteries.1

Electrolyte

Physical property @30  C

Other property

Battery

Cycling performance

Less PC co-intercalates into LixZrS2

Viscosity/(mPa s)

Conductivity (mS cm−1)

3 M LiAsF6/PC

/

/

/

2.72 M LiBETI/PC

/

/

/

Li‖ LixZrS2 Li‖NG

3.27 mol kg−1 LiBETI/PC 7 M LiFSI/DOL+DMC (1:1, v/v)

/

/

Suppress Li dendrite

Li‖Ni

0.814 @RT 0.8

3.2 M LiTFSI/DMSO

/

/

Tg ¼ −77.3  C; t+ ¼ 0.73 0–4.5 V; t+ ¼ 0.51 NDMSO ¼ 2

Li‖S

2.98 M LiCTFSI/G4 (1:1)

72 cP @RT 132

LiFSI/G4 (1:1, by molar ratio)

/

/


5.0 V

Li‖LCO

3.6 M LiFSI/DME

25.1

7.2

/

Li‖NG

4.2 M LiTFSI/AN

138.3

0.98

/

Li‖NG

5.7

/

Li‖Li

Al anticorrosion 5 V; Al anticorrosion Self-extinguishing

Gr‖LNMO Li‖Cu

4 M LiFSI/DME 10.8 M LiFSI/DMC 7 M LiFSI/FEC

238.9

1.12

3.3 M NaFSI/TMP

72

2.2

3 M LiPF6/EC-EMC-DMC (1:1:1, v/v/v)

19.809 @25  C

5.30 @25  C

6

475

/

Li‖LCO Li‖NG-7

Na‖HC Li‖ LFNMO

355 mAh g−1 @1st CE:
80%@50th 770 mAh g−1, 74%@100th, CE: 100%, 0.2C 108 mAh g−1 @50th, C/10 300 mAh g−1 @1st 100 mAh g−1 @200th, C/8 250 mAh g−1 @ 2C
320 mAh g−1 @ 2C 10 mA cm−2 for more than 6000 cycles, CE
99% 90%@100th, 40  C, C/5 CE
99.64% during the 300th
400th cycles CE
99.4% 95%@1200th, 25  C C/5 CE
99.2% 79%@100th, 0.1C; 94%@500th, 2C; 100%@500th, 5C

Localized high-concentration electrolytes (LHCEs,
1 M)

Even though the HCEs could bring tremendous benefits for battery properties, researchers soon realized the drawbacks of HCEs, such as high viscosity, high cost, sluggish ion transport, and poor wetting. To mitigate these issues, researchers proposed to add Li+ non-coordinating solvents (also called diluents, usually a polyfluorinated ether as shown in Fig. 6) to dilute the parental HCEs, and the as-obtained electrolytes (usually 1–1.2 M) are referred to localized high-concentration electrolytes (LHCEs).23 Notably, LHCEs maintain the favorable solvation structures of HCEs and improve viscosity, wetting, and conductivity simultaneously, showing promise for battery applications. The reported LHCEs and the corresponding properties are summarized in Table 3. To obtain the low solvating ability and high oxidative stability, some HFEs (see Fig. 6), such as, TTE, BTFE, 1, 1, 2, 2-tetrafluoroethyl 2, 2, 2-trifluoroethyl ether (TFEE), 1H, 1H, 5H-octafluoropentyl-1, 1, 2, 2-tetrafluoroethyl ether (OFE), tri(2,2,2-trifluoroethyl) orthoformate (TFEO), and 1,1,1,2,2,3,3,4,4-nonafluoro-4-methoxybutane (M3), as diluents for LHCEs, are widely investigated for high energy-density LIBs. For example, Watanabe’s group27 diluted solvate ionic liquids ([Li(G4)1TFSI]) with TTE or TFEE, and found that the addition of TTE, did not break the solvation of the glyme-Li-salt molten complexes and greatly enhanced the power density of Li-S batteries. Meanwhile, by adding inert fluoroalkyl ether of OFE into LiFSI/DME HCEs, has been developed to simultaneously suppress the Li dendrite formation and minimize the solubility of lithium polysulfides.28 In a separate work, Doi and Inaba et al. investigated TTE to dilute the concentrated LiBF4 in PC for 5-V class LMBs.29 Soon afterwards, they proposed a low viscosity LCE of 1.52 mol kg−1 LiBF4 in DMC/TTE (1:1 by vol.), which delivers a high reversible capacity of NMC532 to ca. 200 mAh g−1 at 4.6 V.30 Zhang and co-workers developed a LHCE of 1.2 M LiFSI in DMC/BTFE (1:2 by mol), which enabled a high Li CE of >99% and excellent capacity retention (>80% after 700 cycles) of Li‖NMC111 batteries.31 Moreover, by adding electrochemically inert dichloromethane (DCM) as a diluent in concentrated EA-based electrolytes, this novel LHCE showed a high ionic conductivity of 0.6 mS cm−1, and enabled rechargeable metallic Li battery with high energy (178 Wh kg−1) and power (2877 W kg−1) at –70  C.32 More recently, some LHCEs consisting of LiFSI in DMC or NaFSI in DME with an electrochemically “inert” and poorly solvating fluorinated ether (i.e., BTFE), remarkably enhanced the electrochemical property of Li or sodium metal batteries, respectively.

476

Cell Components – Electrolytes | Non-Aqueous Liquid Electrolyte

Fig. 6 The chemical structure of Li+ non-coordinating solvents used in LHCEs for rechargeable batteries. Table 3

Properties of the reported LHCEs used in rechargeable batteries23.

Electrolyte

Diluent

Property −1

Electrode

Cycling performance

Li‖S–C (79.1% S)

88.3%@50 cycles, 0.3C 775 mAh g−1@150 cycles, 99.2 CE%, 100 mA g−1, 1–3 V 418 mAh g−1@1000 cycles at 25  C, 0.5C

LiTFSI/G4/TTE (1:1:4 by mol)

TTE

1 M LiFSI in DME/OFE (5:95 by vol.)

OFE

6.81 mS cm , 4.56 mPa s, LSV: 5 V (using stainless steel/Li cells at 0.5 mV s−1) 1.24 mS cm−1, 2.4 mPa s

1.1 M LiFSI in TEP/TTE (1:3 by vol.)

TTE

1.25 mS cm−1

2.50 mol kg−1 LiBF4 in PC/TTE (2:1 by vol.) 1.52 mol kg−1 LiBF4 in DMC/TFEE (1:1 by vol.) 1.2 M LiFSI in DMC/BTFE (1:2 by mol)

TTE

51.7 mPa s

Li‖S (65.02 wt% S) Li‖S (52.6 wt% S) Li‖LNMO

TFEE

3.6 mPa s.

Li‖NMC532

BTFE

Li‖Cu; Li‖NMC111

1.2 M LiFSI in TEP/BTFE (1:3 by mol)

BTFE

1.20 M LiFSI in 3TMS/3TTE

TTE

2.67 mS cm−1, 2.7 mPa s, LSV: 4.6 V (Li/Al coin cells at 0.2 mV s−1) 1.29 mS cm−1, 2.9 mPa s, CV: 5.0 V [Li/Super C65 carbon coin cells at 0.2 mV s−1] 2.03 mS cm−1, 14.1 cP at 25  C, 4.6 V

1.2 M LiFSI in EC/EMC (2:8 by wt.) + BTFE (1:2 by mol.) + 0.15 M LiDFOB

BTFE

1 M LiFSI in DME/TFEO (1.2:3 by mol)

TFEO

4.8 cP at 30  C, LSV: 6 V at 0.1 mV s−1

Li‖Li; Li‖NMC811

1.2 M LiFSI in TEP/EC/BTFE (1:0.3:3 by mol) 1.28 M LiFSI in FEC/FEMC/TTE (1:2:7 by vol.) or 0.7 M LiBETI in FEC/DEC/ M3 (1:5:14 by vol.)

BTFE

1.4 mS cm−1, 2.8 cP at 30  C

Gr‖NMC811

TTE or M3

>0.01 mS cm−1 at –80  C, −125
+70  C, 5.6 V

Li‖NCA

Li‖Li; Li‖NMC622 Li‖Cu; Li‖LNMC cells Li‖Li; Li‖NMC111

95.1%@45 cycles, C/10 ca. 20 mA cm−2 93.4%@50 cycles, 3.0–4.6 V, C/10, ca. 120 mA cm−2 99.3% CE, 0.5 mA cm−2, >80%@700 cycles, C/2 charge/2C discharge rates (1C
2 mA cm−2) 99.2% CE at 0.5 mA cm−2 >97%@600 cycles, 1C (1C ca. 1.6 mA cm−2) 98.8% CE at 0.5 mA cm−2 99.8 CE% after stabilization, 2.7–4.3 V at C/3, 1C
1.5 mA cm−2 98.5 CE% 300 cycles at 1 mA cm−2, 1 mAh cm−2; 98.8% CE at 0.5 mA cm−2, 3.8 mAh cm−2, 84% after 100 cycles Li 99.5 CE%, 80%@300 cycles, 0.5 mA cm−2, 1 mAh cm−2; 2.8–4.4 V at a C/3 rate after 2 formation cycles at C/10, 1C ¼ 1.5 mA cm−2 134.8 mAh g−1, 85.4%@300 cycles, 2.8–4.3 V at C/2 (1C ¼ 200 mAh g−1) 150 mAh g−1@450 cycles, –20  C with 1/3C (1C ¼ 170 mAh g−1); 50% of its room temperature capacity at –85  C

Cell Components – Electrolytes | Non-Aqueous Liquid Electrolyte

477

However, all of these electrolytes are flammable and may pose safety hazards, especially for large-scale applications. To solve this issue, some non-flammable LHCEs based phosphorus-containing flame retardants (i.e., TMP and TEP) were also demonstrated. For instance, an intrinsic flame-retardant LHCE consisting of 1.2 M LiFSI in TEP/EC/BTFE (1:0.3:3 by mol), which show good compatibility with both the Gr negative electrode and the NMC811 positive electrode.33 The Gr‖NMC811 full batteries with this LHCE showed a high CE close to 100% during cycling, and delivered a 134.8 mAh g−1 specific capacity after 300 cycles, corresponding to 85.4% capacity retention. The affinity between HFEs and Li+ is mainly determined by the position of electron-withdrawing fluoroalkyl groups instead of the absolute number of fluorine atoms. It is noted that the introduction of dCF3 group at the b position, the HFEs molecules, e.g., BTFE and TFEO, show an enhanced Li CE (>99%) and cycling stability for Li metal, which may be ascribed to the high affinity of the dCF3 group for the surface of Li electrode, resulting in an ordered interface that allows good Li+ diffusion. The phenomenon can provide guidance in choosing a suitable HEF diluents for HCEs used in further rechargeable batteries. However, it remains questionable whether LHCEs will be affordable for practical applications or not, because the major electrolyte components are organofluoride-based compounds, which are even more difficult to synthesize and far more expensive than LiPF6. To date, the reported LHCEs usually necessitate LiFSI, LiTFSI, fluorinated ethers, and/or fluorinated aromatics. We emphasize that cost is a crucial parameter that determines the feasibility of battery materials. In a conventional 1 M LiPF6/ carbonate electrolyte, the salt cost contribution to the electrolyte is over 70%. With the rapidly expanding LIBs market, especially electric vehicles, the lithium salt price will continue to increase.

7

Low-concentrate electrolyte (LCEs, 3 M), localized high-concentration electrolytes (LHCEs,
1 M) and low-concentration electrolyte (LCEs, 3 M) or localized high-concentration electrolytes (
1 M); however, the expensive lithium salt imposes a major concern. Most recently, ultralow concentration electrolytes (5 V vs Na+/Na). Negligible ASP and charge transfer resistance with Na-based anodes and cathodes. The potential to create thin films with modern ceramic processing technologies while maintaining the necessary chemical, EW and IC. Less complicated to scale up the synthesis, accessible to less expensive fabrication techniques, safe for the environment, and stable chemically under ambient conditions.58

All of the aforementioned conditions are not concurrently met by any of the ceramic Na SSE that currently exists. The known ceramic Na-SSEs that have been created for use in SSBs are covered in the sections that follow. Reported Na SSEs are discussed in the following sections.

4.1

b-Alumina electrolyte

b-Alumina electrolyte was the first Na SSE used in commercial Na-S and Na-metal chloride batteries. b-Alumina is a multilayered compound consisting of conduction planes and spinel blocks alternating in layers (Fig. 8).59 The IC is nearly zero in the vertical direction of the structure and is limited to its conduction planes. There are two crystal structures: one is the b-alumina phase (Na2O.11Al2O3, hexagonal symmetry: P63/mmc); and the other one is b00 - alumina phase (Na2O.5Al2O3, rhombohedral symmetry: R3m). The bigger unit cell (1.5 times longer in c axis over the b-alumina phase) and higher Na ion concentration (in the conduction plane) of the rhombohedral b00 -alumina phase result in elevated IC as compared to b-alumina.60,61 According to reports, b00 -alumina single crystal possesses an extremely high IC of 1 S cm−1 at 300
C, which is 4 times higher than that of its polycrystalline phase (0.22–0.35 S cm−1 at 300
C, but only 2.0  10−3 S cm−1 at RT). However, the fabrication of the highly conductive b00 -alumina single crystal at industrial scale is prohibitively expensive.62 (B)

(A)

C Spinel block

A

conduction slab

B' C A

Spinel block

C

B

A

C

B'

A'

A

B

C

C

B

A

A

B

C'

C'

conduction slab

A

A

Spinel block

B

conduction slab

Spinel block

B

Al3+

Fig. 8 (A) Na+-b-alumina and (B) Na+-b”-alumina structures.59

Na+

O2-

502 4.2

Cell Components – Electrolytes | Overview - Solid Electrolytes NASICON SSEs

To get beyond the limitations of b-alumina electrolyte, NASICON SSE is thought to be the most promising sodium electrolyte with appropriate 3D tunnels for Na ion migration. High thermal and chemical stabilities of NASICON are established, and its strong covalent structure is very important from a practical standpoint. Goodenough and Hong introduced the first ever NASICON compound, Na1+xZr2SixP3-xO12 (0  x  3, abbreviated as NZSP), which was a solid solution between NaZr2P3O12 and Na4Zr2Si3O12.63,64 The chemical formula of the original NASICON compound is Na1+xZr2P3−x SixO12 (0  x  3). It is composed of corner-sharing PO4, SiO4, and ZrO6, which together form a 3D covalent network that allows Na ions to move across interstitial sites. In Na1+xZr2P3−xSixO12, all compositions were found to have a rhombohedral structure (Fig. 9A) with space group R-3c, except for those in the range of 1.8  x  2.2, which is monoclinic with space group C2/c (Fig. 9B). The M2 sites split into Ma2 and Mb2 sites due to monoclinic distortion.65 Monoclinic structured Na3Zr2Si2PO12 exhibits the best Na IC of 6.7  10−4 S cm−1 at RT.65 Huge grain boundary resistance is observed in NASICON materials at RT and therefore several sintering techniques have been used for its minimization. Naqash with his coworkers used a solution-assisted solid-state reaction to prepare Na3Zr2Si2PO12 and obtained the total IC of 1 mS cm−1. Using the molecular precursor technique, a highly conducting NASICON structured compound was prepared that showed an IC of 2.2 mS cm−1 at RT. Several elements can be used to dope the Zr-site and P-site to modify the Na-ion diffusion bottleneck. It increases the amount of Na-ions in the structure and/or reduces the Coulombic interactions between Na+ and the nearby cations which allows the facile migration of Na+ ions. The doping of La3+ by self-forming strategy yields the IC of 3.4 mS cm−1 at RT.66–68

4.3

Chalcogenides

Chalcogenides, which are compounds containing elements from the chalcogen group (sulfur, selenium, and tellurium), have been investigated for their potential use as SSEs in Na ion SSBs. Owing to their superior IC and stability, these materials have gained interest for use in SSE applications. Their flexibility makes it simple for them to use cold pressing to produce close-grained pellets. This sets them apart from other electrolyte systems because it allows for the creation of flexible full-cell assembly geometries. The strength of Na3PO4 is the basis for the design of chalcogenide SSEs. By substituting elements with greater atomic radii for P and/or O in an aliovalent or isovalent manner, it is possible to methodically alter the IC and composition of these materials. SSEs of sodium chalcogenide obtained from the original composition consist of Na3PS4, Na3–xPS4–xClx, Na4P2S6, Na3PSxSe4–x, Na3SbS4, Na11Sn2PS12, Na3PSe4, Na3SbSe4, Na11Sn2SbS12, Na11Sn2PSe12 and the glass–ceramic 94Na3PS4-6Na4SiS4 (Fig. 10).69 The first sulfide superionic electrolyte to be reported was Na3PS4. It demonstrates tetragonal phase (Space group; P–421c) and cubic phase (Space group; I–43m). Na atom is partially populated at 6b and 12d sites in the cubic phase. The low-temperature synthesis of the tetragonal phase exhibits an IC of 10−6–10−5 S cm−1 at RT. In contrast, the cubic Na3PS4 (c-Na3PS4) demonstrates enhanced IC of 2  10−4 S cm−1 at RT. Mechanochemical methods followed by heat treatment at 270
C were used to obtain c-Na3PS4. The c-Na3PS4 synthesized at lower temperatures demonstrated an elevated IC of 4.6  10−4 S cm−1.70–72 Replacing P5+ with Sb5+ could result in the acquisition of a superionic conductor Na3SbS4. Due to good air stability, it is seen as a viable SSE for sodium-ion batteries. Aliovalent substitution of Sb5+ sites in Na3SbS4 by Sn4+ has resulted in novel Na SSEs of the general formula, Na4-xSn1-xSbxS4 (0.02  x  0.33). The stoichiometric system of Na11Sn2SbS12 shows rather strong IC (around 0.2 mS cm−1) with an upper limit Sb5+ substitution. Additionally, it was determined that the EW for the composition of Na11.25Sn2.25Sb0.75S12 ranged between 0.44 and 4.5 V (versus Na/Na+). Similarly, these substances maintain ionic conduction even after being exposed to air for several hours.73,74 (A)

(B)

Fig. 9 NASICON with (A) rhombohedral structure and (B) monoclinic structure.65

Cell Components – Electrolytes | Overview - Solid Electrolytes

Fig. 10 Diagrammatic illustration of the IC of chalcogenide NASICONs and their design techniques.

503

504 4.4

Cell Components – Electrolytes | Overview - Solid Electrolytes Complex hydrides

The conduction of sodium ions in complex hydrides was first observed in 2012. The SSEs of the borohydride class were demonstrated as significant substitutes for liquid electrolytes. Research on complex hydrides like the sodium SSEs was first conducted by Udovic and his colleagues. They synthesized NaAlH4 and Na3AlH6 which demonstrated IC of 2.1  10−10 and 6.4  10−4 S cm−1 at RT, respectively. Complex hydride superionic conductors i.e., Na2B10H10 exhibit a monoclinic phase (space group; P21/c). The ordered phase of Na2B10H10 transformed into a disordered phase at 101.85
C, which exhibited superior IC (10−2 S cm−1) and declined activation energy (0.19 eV). With elevated IC (10−4 S cm−1) at RT, the EW of the dodeca/deca-borate combination pseudo-binary complex hydride against sodium metal is also excellent.

4.5

Sodium perovskites and anti-perovskites SSEs

The high IC of oxide-based perovskite SSEs (10−3 S cm−1 at RT) has made them useful for applications such as SOFC, oxygen sensors, and air separation membranes. Recent reports have suggested that these materials have high lithium/sodium IC. The IC of orthorhombic structured Na0.33La0.55ZrO3 and Sr substituted Na0.33La0.33Sr0.33ZrO3, were found to be 6.89  10−7 and 1.025  10−5 S cm−1 at RT, respectively. These materials also exhibit greater chemical stability with sodium electrodes than that of perovskite-type LLTO containing the fast reducible titanium ion.75 After the discovery of anti-perovskite SSEs (Li3OCl1-xBrx), researchers have been looking for sodium-based analogs of anti-perovskite SSEs. In recent years, there has been a lot of interest in sodium-based anti-perovskites with the composition Na3-OX (where X ¼ Cl, Br, I, and BH4). These cubic structured (space group Pm–3 m, a ¼ 4.564 A˚ ) materials demonstrate high IC. It is proven that certain binary (Li-nonmetal) interphase between the lithium metal anode and SSEs function as efficient passivate layers against lithium and sodium. The halogen/rotational clusters, oxygen, and sodium make up sodium anti-perovskite SSEs. The formation of electrolyte-electrode interphase (ionically conducting but electrically insulating) of NaX (X ¼ halogen) is also expected when the SSE comes into contact with sodium anodes. Sadly, it has been determined that the Na antiperovskite SSEs that have been found thus far are stable solely within the range of 0–1.8 V.76–78 Here, we have taken the average values of many features and produced a radar plot (Fig. 11) that highlights the benefits and drawbacks of each class of materials of Na SSES.

Fig. 11 Radar plots showing the relative performances of the different chemical and physical properties of Na-SSE.

Cell Components – Electrolytes | Overview - Solid Electrolytes

5

505

SSEs for KIBs

The concerns over a sustainable supply of LIBs are also raised by the limited reserves, unequal distribution, and rising cost of cobalt. The enormous reserves of K and Na (which are 2.3 and 1.5 wt%, respectively) make the KIBs and NIBs an alternate solution. Moreover, the prices of potassium and sodium carbonate (K2CO3, Na2CO3) are considerably less expensive than Li2CO3. However, KIBs have the potential to be far more advantageous than NIBs. First, KIBs are expected to offer an elevated working voltage than NIBs. High-energy density is ensured by the K/K+ electrode’s lower redox potential (−2.93 V vs. E0) compared to Na (−2.71 V vs. E0).79 KIBs can achieve higher performance due to excellent K+ ion diffusion coefficient. Furthermore, unlike Li+ and Na+, it was found that K+ reflects the least solvation energy in ethylene carbonate (397.5 kJ mol−1).80 This suggests a higher rate capability. The most common anode material for LIBs, i.e., graphite, could reversibly intercalate or de-intercalate in K+. Given the benefits outlined above, KIBs are the most viable alternatives to NIBs and LIBs.81 Conventional liquid electrolytes that are not aqueous and are utilized in KIBs have high IC and good electrochemical characteristics. However, the practical uses of KIBs have been impeded by the significant safety concerns related to the volatile and flammable organic liquid electrolytes. Also, KIBs pose a more serious safety issue due to the use of highly reactive potassium. Compared to organic liquid electrolytes, K-SSEs have several benefits, including increased protection, high thermal stability, a larger EW, and longer cyclability.82 Ceramic SSEs, in particular, have superior mechanical qualities and high IC; on the other hand, organic-based SSEs are flexible and simple to produce. Consequently, SSEs are thought to be efficient alternatives to traditional non-aqueous electrolytes for resolving the safety concerns with next-generation KIBs. Poor IC and a poor contact between the electrodes and electrolytes, however, are the main obstacles to KIBs. For use in KIBs, solid polymer electrolytes and inorganic electrolytes—a subset of SSEs—have been researched. SSEs prolong the cycle life of KIBs and dramatically reduce the dissolution of organic active components in non-aqueous electrolytes.83 The K-SSEs usually have high IC (>0.1 mS cm−1 at RT), high mechanical strength (>1 GPa for oxides), and broader thermal/EW (>4.0 V).84 After the discovery of b-alumina SSE, a series of compounds, i.e., M2OxAl2O3 (where, M ¼ Na+, K+, Rb+, Ag+, 5 < x < 11) were developed by ion exchange method.85 K- b-alumina SSE was initially assembled for the K-S battery, which could be operated at moderate temperatures (150
C).86 The K-b-alumina SSE’s IC at 300
C was around 0.056 S cm−1. The K-S batteries show 75% retention capacity when operated at operated 1.2C rate, which was significantly better than the C/40 rate. In addition, the K-S battery showed charge-discharge cycling stability over 1000 cycles when operated at a C/4.2 rate. Several K-SSEs that include suitable 1D, 2D, or 3D diffusion path for K+ ions to travel rapidly through examined earlier. K3Sb4O10(BO3) compound with 1D K+ ionic pathways through a-b direction were prepared (Fig. 12A).89 At 400
C, it showed an IC of approximately 1.5  10−4 S cm−1, with an activation energy of 0.325 eV. Afterward, the lamellar structured K-SSEs with the general formula KAO2-BO2, where A ¼ In or Sc, and B ¼ Zr, Sn, or Pb were explored. Above 250
C, the IC of K0.721n0.72Hf0.28O2 surpasses that of sintered b-alumina, however, the reported activation energy was higher.90 Recently, a novel open-framework structure (Fig. 12A and B) of potassium ferrite (K2Fe4O7) compound was prepared via hydrothermal method. It includes relatively low sized 1D 3-ring channels at the c-axis and larger 2D 6-ring channels paralleling the a-b planes.91 Its broad working voltage of 5 V vs. K/K+ and superior IC (5.0 10−2 S cm−1) are mostly attributable to the 2D channel’s reduced migration resistance. Using K2Fe4O7 as the electrolyte and Prussian Blue as the analog cathode, the K metal SSBs could deliver consistent cycle performance for up to 50 cycles with a retention capacity of 78%. Layered honeycomb frameworks (Fig. 12C) were used to generate

(A)

(B)

(C)

Fig. 12 (A) FeO6 octahedra and FeO4 tetrahedra make up the 2D layer, while 1D 3-ring pathways are shown along the c-axis of 1. (B) 3D open framework structure and Intersecting 2D 6-ring channels filled with mobile potassium carriers (C) Polyhedral image along the c-axis of the P2-type-layered crystal structure of K2Ni2TeO6: K ions are brown spheres, O ions are tiny red spheres, and the Ni and Te octahedra are displayed in purple and blue, respectively.87,88

506

Cell Components – Electrolytes | Overview - Solid Electrolytes

tellurates compounds, such as K2M2TeO6 (where M is a transition metal), where K vacancies caused rapid K+ migration throughout the layers.87 With a value of 40 mS cm−1 (at 300
C), K2Mg2TeO6 demonstrated the highest IC among all the tellurates. It is interesting to construct NASICON-like compounds with potassium to develop potassium ion conductors.92

6

SOFC

Fuel cells are attracting the researcher’s interest due to their huge potential of power generation for various applications.93,94 Recently, a large range of fuel cell models with different electrolyte compositions has been produced including the oxide-based Li, Na, and K SSEs discussed above. However, the identical fundamental operational principle of all types of fuel cells is depicted in Fig. 13A. At the anode, a fuel, like hydrogen, undergoes oxidation to produce protons and electrons. Simultaneously, at the cathode, oxygen undergoes reduction to form oxide species, which then combine to create water. Protons or oxide ions, depending on the electrolyte, are carried by an IC but electrically insulating electrolyte and electrons move across an external circuit to produce electricity. Fuel cells, particularly SOFCs, offer a large-scale capacity of distributed power generation (up to hundreds of megawatts). In this setup, the waste heat generated by the SOFC can be utilized to run a gas turbine, leading to additional electricity production and enhancing overall system efficiency up to 80%. This efficiency surpasses what is achievable with conventional electricity generation methods. SOFC is characterized by a solid ceramic electrolyte. Its elements include the cathode, where oxygen ions are obtained by the reduction of oxygen. These ions traverse the SSE under an electrical load, reaching the anode, where they react with fuels like hydrogen, and carbon monoxide, producing water and carbon dioxide, along with heat and electricity. Fig. 13B provides a schematic representation of this process. The anode, cathode, electrolyte, and connection materials must meet strict specifications set by the SOFC. Every component has many functions and needs to fulfill specific requirements. To prevent cracking or delamination during fabrication and operation, each component must have the following qualities: proper conductivity, chemical and physical stability in the appropriate chemical environment (oxidation and/or reduction), chemical compatibility with the other components, and the same thermal expansion coefficients as the other components. Furthermore, the SOFC components must be inexpensive, robust, and simple to assemble. The majority of SOFCs that are currently being developed use YSZ electrolyte, doped lanthanum chromite (LaCrO3) as interconnect, and a cathode made of strontium-doped lanthanum manganite (La1−xSrxMnO3) and a mixed nickel/YSZ cermet anode. The pursuit and examination of alternative SSE materials have been a focal point of active research for an extended period. The two most promising substitute electrolytes for YSZ that have been thoroughly researched recently are lanthanum gallate-based structures and gadolinia-doped ceria in particular.95–97 With one of these electrolytes, SOFCs may be able to operate at temperatures between 500
C and 700
C. Since it has comparable qualities to YSZ but shows higher IC, scandinavia-doped zirconia is also being researched as a potential substitute. However, the cost of this material is more than that of YSZ. High-temperature fuel cells (typically operating above 500
C) and high-temperature electrolyzers with proton-conducting SSEs (operating in the range of 400
C to 800
C) have grasped researchers’ attention for energy conversion/storage. Unlike low-temperature counterparts, these technologies offer benefits like; enhanced performance/fuel flexibility, and lesser impurity susceptibility.98

7

Environmental problems of SSEs

SSEs are a key part of SSBs and fuel cells, however, their production and application cause severe environmental challenges. The environmental effects of SSEs depend on the factors, including the material of use, followed synthesis process, and the disposal practices employed at the end of the product’s life. It is important to conduct a life cycle assessment of SSE to understand their complete ecological impact. It takes into account the extraction, processing, production, usage, and disposal at the end of a product’s life. The environmental challenges associated with SSEs are pointed out below:

• • • •

The mining and processing of raw materials needed for the production of SSE lead to environmental damage, including habitat destruction, soil erosion, and water contamination. The high-temperature calcination/sintering process to produce SSEs is energy-intensive, which can contribute to greenhouse gas emissions and environmental damage. Also, the use of hazardous chemicals/solvents to synthesize SSEs leads to environmental contamination. Handling of waste materials/by-products while SSE synthesis and end-of-life management of devices is important. Improving recycling techniques can lessen their negative effects on the environment. Air pollution and carbon emissions may be caused by the transportation of components and raw materials. Optimizing supply chains or choosing more environmentally friendly modes of transportation can help reduce the negative effects of transportation on the environment.99,100

(A)

Cell Components – Electrolytes | Overview - Solid Electrolytes

Fig. 13 (A) Basic operation of a fuel cell. (B) Basic operation of a SOFC running on natural gas.

(B)

507

508

8

Cell Components – Electrolytes | Overview - Solid Electrolytes

Conclusion

In conclusion, it is observed that SSEs are the key component in developing SSBs which are of great demand in future energy storage applications. In the case of Li-ion batteries, four different types of SSEs have been studied. Oxide-based SSEs i.e. LLZO having garnet type structure show promising ionic conductivity at elevated temperature in cubic phase. However, Ga substituted LLZO i.e. Li6.55Ga0.15La3Zr2O12 showed the highest ionic conductivity of 2.3  10−3 S cm−1 at 42
C. While LISICON shows a very large stable potential window (9 V), the ionic conductivities are not to the mark. On the contrary, recently developed the-LISICONS (Li10GeP2S12) show an IC of 1.2  10−2 S cm−1. The precious element Ge has recently been replaced by Sn which also shows better ionic conductivity. Perovskite-based SSEs show moderate ionic conductivity. Another form of SSE based on Li-argyrodite fast-ion conductor Li6PS5X (X ¼ Cl, Br, I) shows initial RT ionic conductivity values varied from 10−2 to 10−3 S cm−1. In the case of Na ion battery, the poly-crystalline b00 - alumina phase (Na2O.5Al2O3, rhombohedral symmetry: R3m) shows an ionic conductivity of the order of 10−3 S cm−1 at room temperature. NASICON-based SSE for sodium ion battery namely Na3Zr2Si2PO12 has been found to show room temperature ionic conductivity of 2.2 mS cm−1. Doping of La3+ results in enhancement of the RT ionic conductivity to 3.4 mS cm−1. Chalcogenide-based SSEs show moderate RT ionic conductivity. In the case of the K ion battery, the potassium ferrite (K2Fe4O7) compound shows an excellent room temperature ionic conductivity of 5.0 10−2 S cm−1. Layered honeycomb frameworks-based tellurates, K2M2TeO6 show IC of 40 mS cm−1 at elevated temperature. YSZ-based SSEs have been used for SOFC applications. As alternatives to YSZ, recently lanthanum gallate-based structures, gadolinia-doped ceria, and scandinavia-doped zirconia have been studied.

Acknowledgment The authors are grateful to Pandit Deendayal Energy University (PDEU) for providing the necessary facilities for this investigation. Financial support from the Department of Science and Technology (DST) Government of India under project no. DST/TMD/MES/ 2017/32(G) and Solar Research and Development Centre (SRDC), PDEU is deeply acknowledged.

References 1. Li, M.; Wang, C.; Chen, Z.; Xu, K.; Lu, J. New Concepts in Electrolytes. Chem. Rev. 2020. https://doi.org/10.1021/acs.chemrev.9b00531. 2. Mishra, A. K.; Chaliyawala, H. A.; Patel, R.; Paneliya, S.; Vanpariya, A.; Patel, P.; Ray, A.; Pati, R.; Mukhopadhyay, I. Inorganic Solid State Electrolytes: Insights on Current and Future Scope. J. Electrochem. Soc. 2021, 168 (8), 80536. 3. Albertus, P.; Anandan, V.; Ban, C.; Balsara, N.; Belharouak, I.; Buettner-Garrett, J.; Chen, Z.; Daniel, C.; Doeff, M.; Dudney, N. J.; Dunn, B.; Harris, S. J.; Herle, S.; Herbert, E.; Kalnaus, S.; Libera, J. A.; Lu, D.; Martin, S.; McCloskey, B. D.; McDowell, M. T.; Meng, Y. S.; Nanda, J.; Sakamoto, J.; Self, E. C.; Tepavcevic, S.; Wachsman, E.; Wang, C.; Westover, A. S.; Xiao, J.; Yersak, T. Challenges for and Pathways toward Li-Metal-Based All-Solid-State Batteries. ACS Energy Lett. 2021, 6 (4). https://doi.org/10.1021/ acsenergylett.1c00445. 4. Masias, A.; Marcicki, J.; Paxton, W. A. Opportunities and Challenges of Lithium Ion Batteries in Automotive Applications. ACS Energy Lett. 2021. https://doi.org/10.1021/ acsenergylett.0c02584. 5. Sun, Y. K. Promising All-Solid-State Batteries for Future Electric Vehicles. ACS Energy Lett. 2020. https://doi.org/10.1021/acsenergylett.0c01977. 6. Ni, J. E.; Case, E. D.; Sakamoto, J. S.; Rangasamy, E.; Wolfenstine, J. B. Room Temperature Elastic Moduli and Vickers Hardness of Hot-Pressed LLZO Cubic Garnet. J. Mater. Sci. 2012, 47 (23), 7978–7985. 7. Yabuuchi, N.; Kubota, K.; Dahbi, M.; Komaba, S. Research Development on Sodium-Ion Batteries. Chem. Rev. 2014. https://doi.org/10.1021/cr500192f. 8. Vignarooban, K.; Kushagra, R.; Elango, A.; Badami, P.; Mellander, B. E.; Xu, X.; Tucker, T. G.; Nam, C.; Kannan, A. M. Current Trends and Future Challenges of Electrolytes for Sodium-Ion Batteries. Int. J. Hydrogen Energy 2016, 41 (4). https://doi.org/10.1016/j.ijhydene.2015.12.090. 9. Xu, Y.; Titirici, M.; Chen, J.; Cora, F.; Cullen, P. L.; Edge, J. S.; Fan, K.; Fan, L.; Feng, J.; Hosaka, T.; Hu, J.; Huang, W.; Hyde, T. I.; Imtiaz, S.; Kang, F.; Kennedy, T.; Kim, E. J.; Komaba, S.; Lander, L.; Le Pham, P. N.; Liu, P.; Lu, B.; Meng, F.; Mitlin, D.; Monconduit, L.; Palgrave, R. G.; Qin, L.; Ryan, K. M.; Sankar, G.; Scanlon, D. O.; Shi, T.; Stievano, L.; Tinker, H. R.; Wang, C.; Wang, H.; Wang, H.; Wu, Y.; Zhai, D.; Zhang, Q.; Zhou, M.; Zou, J. Roadmap for Potassium-Ion Batteries. J. Phys. Energy 2023, 2023, 5 (2) https://doi. org/10.1088/2515-7655/acbf76. 10. Song, K.; Liu, C.; Mi, L.; Chou, S.; Chen, W.; Shen, C. Recent Progress on the Alloy-Based Anode for Sodium-Ion Batteries and Potassium-Ion Batteries. Small 2021. https://doi. org/10.1002/smll.201903194. 11. Liu, S.; Meyer, L. C.; Medenbach, L.; Balducci, A. Glyoxal-Based Electrolytes for Potassium-Ion Batteries. Energy Storage Mater. 2022, 47. https://doi.org/10.1016/j. ensm.2022.02.041. 12. Zhou, M.; Bai, P.; Ji, X.; Yang, J.; Wang, C.; Xu, Y. Electrolytes and Interphases in Potassium Ion Batteries. Adv. Mater. 2021. https://doi.org/10.1002/adma.202003741. 13. Sequeira, C. A. C.; Santos, D. M. F. Polymer Electrolytes: Fundamentals and Applications; Elsevier, 2010. https://doi.org/10.1533/9781845699772. 14. Ngai, K. S.; Ramesh, S.; Ramesh, K.; Juan, J. C. A Review of Polymer Electrolytes: Fundamental, Approaches and Applications. Ionics 2016. https://doi.org/10.1007/s11581016-1756-4. 15. Xu, K. Nonaqueous Liquid Electrolytes for Lithium-Based Rechargeable Batteries. Chem. Rev. 2004, 104 (10). https://doi.org/10.1021/cr030203g. 16. Famprikis, T.; Canepa, P.; Dawson, J.; materials, M. I.-N. undefined. Fundamentals of Inorganic Solid-State Electrolytes for Batteries; 2019. nature.com. 17. Lu, J.; Chen, J.; He, T.; Zhao, J.; Liu, J.; Huo, Y. Inorganic Solid Electrolytes for the Lithium-Ion Batteries. Progress Chem. 2021, 33 (8). https://doi.org/10.7536/PC200772. 18. Gao, Z.; Sun, H.; Fu, L.; Ye, F.; Zhang, Y.; Luo, W.; Huang, Y. Promises, Challenges, and Recent Progress of Inorganic Solid-State Electrolytes for All-Solid-State Lithium Batteries. Adv. Mater. 2018. https://doi.org/10.1002/adma.201705702. 19. Xiao, Z.; Long, T.; Song, L.; Zheng, Y.; Wang, C. Research Progress of Polymer-Inorganic Filler Solid Composite Electrolyte for Lithium-Ion Batteries. Ionics 2022. https://doi.org/ 10.1007/s11581-021-04340-2. 20. Feng, J.; Wang, L.; Chen, Y.; Wang, P.; Zhang, H.; He, X. PEO Based Polymer-Ceramic Hybrid Solid Electrolytes: A Review. Nano Convergence 2021. https://doi.org/10.1186/ s40580-020-00252-5.

Cell Components – Electrolytes | Overview - Solid Electrolytes

509

21. Sun, Y. Y.; Zhang, Q.; Yan, L.; Wang, T. B.; Hou, P. Y. A Review of Interfaces within Solid-State Electrolytes: Fundamentals, Issues and Advancements. Chem. Eng. J. 2022. https://doi.org/10.1016/j.cej.2022.135179. 22. Liang, J.; Luo, J.; Sun, Q.; Yang, X.; Li, R.; Sun, X. Recent Progress on Solid-State Hybrid Electrolytes for Solid-State Lithium Batteries. Energy Storage Mater. 2019. https://doi. org/10.1016/j.ensm.2019.06.021. 23. Lin, D.; Liu, W.; Liu, Y.; Lee, H. R.; Hsu, P. C.; Liu, K.; Cui, Y. High Ionic Conductivity of Composite Solid Polymer Electrolyte Via In Situ Synthesis of Monodispersed SiO2 Nanospheres in Poly(Ethylene Oxide). Nano Lett. 2016, 16 (1). https://doi.org/10.1021/acs.nanolett.5b04117. 24. Croce, F.; Appetecchi, G. B.; Persi, L.; Scrosati, B. Nanocomposite Polymer Electrolytes for Lithium Batteries. Nature 1998, 394 (6692). https://doi.org/10.1038/28818. 25. Han, L.; Lehmann, M. L.; Zhu, J.; Liu, T.; Zhou, Z.; Tang, X.; Te Heish, C.; Sokolov, A. P.; Cao, P.; Chen, X. C.; Saito, T. Recent Developments and Challenges in Hybrid Solid Electrolytes for Lithium-Ion Batteries. Front. Energy Res. 2020. https://doi.org/10.3389/fenrg.2020.00202. 26. Morales, D. J.; Greenbaum, S. NMR Investigations of Crystalline and Glassy Solid Electrolytes for Lithium Batteries: A Brief Review. Int. J. Mol. Sci. 2020, 21 (9). https://doi.org/ 10.3390/ijms21093402. 27. Arya, A.; Saykar, N. G.; Sharma, A. L. Impact of Shape (Nanofiller vs. Nanorod) of TiO2 Nanoparticle on Free-Standing Solid Polymeric Separator for Energy Storage/Conversion Devices. J. Appl. Polym. Sci. 2019, 136 (16). https://doi.org/10.1002/app.47361. 28. Rama Mohan, K.; Achari, V. B. S.; Rao, V. V. R. N.; Sharma, A. K. Electrical and Optical Properties of (PEMA/PVC) Polymer Blend Electrolyte Doped with NaClO4. Polym. Test. 2011, 30 (8). https://doi.org/10.1016/j.polymertesting.2011.08.010. 29. Chen, L.; Venkatram, S.; Kim, C.; Batra, R.; Chandrasekaran, A.; Ramprasad, R. Electrochemical Stability Window of Polymeric Electrolytes. Chem. Mater. 2019, 31 (12). https:// doi.org/10.1021/acs.chemmater.9b01553. 30. Méry, A.; Rousselot, S.; Lepage, D.; Dollé, M. A Critical Review for an Accurate Electrochemical Stability Window Measurement of Solid Polymer and Composite Electrolytes. Materials 2021. https://doi.org/10.3390/ma14143840. 31. Thomas, J. In Solid State Electrochemistry; Bruce, P. G., Ed.; vol. XVI; Cambridge University Press: Cambridge, 1995; p. 344. Hardcover, £60.00, ISBN 0-521-40007-4. Advanced Materials 1996, 8(4) https://doi.org/10.1002/adma.19960080417. 32. Owens, B. B. Solid State Electrolytes: Overview of Materials and Applications during the Last Third of the Twentieth Century. J. Power Sources 2000. https://doi.org/10.1016/ S0378-7753(00)00436-5. 33. Kamaya, N.; Homma, K.; Yamakawa, Y.; Hirayama, M.; Kanno, R.; Yonemura, M.; Kamiyama, T.; Kato, Y.; Hama, S.; Kawamoto, K.; Mitsui, A. A Lithium Superionic Conductor. Nat. Mater. 2011, 10 (9), 682–686. https://doi.org/10.1038/nmat3066. 34. Hayashi, A.; Masuzawa, N.; Yubuchi, S.; Tsuji, F.; Hotehama, C.; Sakuda, A.; Tatsumisago, M. A Sodium-Ion Sulfide Solid Electrolyte with Unprecedented Conductivity at Room Temperature. Nat. Commun. 2019, 10 (1). https://doi.org/10.1038/s41467-019-13178-2. 35. Zhang, B.; Tan, R.; Yang, L.; Zheng, J.; Zhang, K.; Mo, S.; Lin, Z.; Pan, F. Mechanisms and Properties of Ion-Transport in Inorganic Solid Electrolytes. Energy Storage Mater. 2018. https://doi.org/10.1016/j.ensm.2017.08.015. 36. Sun, B.; Mindemark, J.; Morozov, E. V.; Costa, L. T.; Bergman, M.; Johansson, P.; Fang, Y.; Furó, I.; Brandell, D. Ion Transport in Polycarbonate Based Solid Polymer Electrolytes: Experimental and Computational Investigations. Phys. Chem. Chem. Phys. 2016, 18 (14). https://doi.org/10.1039/c6cp00757k. 37. Reddy, M. V.; Julien, C. M.; Mauger, A.; Zaghib, K. Sulfide and Oxide Inorganic Solid Electrolytes for All-Solid-State Li Batteries: A Review. Nanomaterials 2020. https://doi.org/ 10.3390/nano10081606. 38. Murugan, R.; Thangadurai, V.; Weppner, W. Fast Lithium Ion Conduction in Garnet-Type Li7La3Zr2O12. Angew. Chem. Int. Ed. 2007, 46 (41), 7778–7781. 39. Awaka, J.; Kijima, N.; Hayakawa, H.; Akimoto, J. Synthesis and Structure Analysis of Tetragonal Li7La3Zr2O12 with the Garnet-Related Type Structure. J. Solid State Chem. 2009, 182 (8), 2046–2052. https://doi.org/10.1016/j.jssc.2009.05.020. 40. Awaka, J.; Takashima, A.; Kataoka, K.; Kijima, N.; Idemoto, Y.; Akimoto, J. ChemInform Abstract: Crystal Structure of Fast Lithium-Ion-Conducting Cubic Li7La3Zr2O12. ChemInform 2011, 42 (18). https://doi.org/10.1002/chin.201118004. 41. Ong, K. P.; Goh, T. W.; Xu, Q.; Huan, A. Mechanical Origin of the Structural Phase Transition in Methylammonium Lead Iodide CH3NH3PbI3. J. Phys. Chem. Lett. 2015, 6 (4). https://doi.org/10.1021/jz502740d. 42. Geiger, C. A.; Alekseev, E.; Lazic, B.; Fisch, M.; Armbruster, T.; Langner, R.; Fechtelkord, M.; Kim, N.; Pettke, T.; Weppner, W. ChemInform Abstract: Crystal Chemistry and Stability of “Li7La3Zr2O12” Garnet: A Fast Lithium-Ion Conductor. ChemInform 2011, 42 (13). https://doi.org/10.1002/chin.201113005. 43. Han, G.; Kinzer, B.; Garcia-mendez, R.; Choe, H. Correlating the Effect of Dopant Type (Al, Ga, Ta) on the Mechanical and Electrical Properties of Hot-Pressed Li-Garnet Electrolyte. J. Eur. Ceram. Soc. 2019, (September). https://doi.org/10.1016/j.jeurceramsoc.2019.12.054. 44. Stramare, S.; Thangadurai, V.; Weppner, W. Lithium Lanthanum Titanates: A Review. Chem. Mater. 2003. https://doi.org/10.1021/cm0300516. 45. Inaguma, Y.; Liquan, C.; Itoh, M.; Nakamura, T.; Uchida, T.; Ikuta, H.; Wakihara, M. High Ionic Conductivity in Lithium Lanthanum Titanate. Solid State Commun. 1993, 86 (10). https://doi.org/10.1016/0038-1098(93)90841-A. 46. Kanno, R.; Murayama, M. Lithium Ionic Conductor Thio-LISICON: The Li2SGeS2P2S5 System. J. Electrochem. Soc. 2001, 148 (7), A742. 47. Zheng, F.; Kotobuki, M.; Song, S.; Lai, M. O.; Lu, L. Review on Solid Electrolytes for All-Solid-State Lithium-Ion Batteries. J. Power Sources 2018, 389 (February), 198–213. https://doi.org/10.1016/j.jpowsour.2018.04.022. 48. Lian, P. J.; Zhao, B. S.; Zhang, L. Q.; Xu, N.; Wu, M. T.; Gao, X. P. Inorganic Sulfide Solid Electrolytes for All-Solid-State Lithium Secondary Batteries. J. Mater. Chem. A Mater. 2019, 7 (36). https://doi.org/10.1039/c9ta04555d. 49. Deiseroth, H. J.; Kong, S. T.; Eckert, H.; Vannahme, J.; Reiner, C.; Zaiß, T.; Schlosser, M. Li6PS5X: A Class of Crystalline Li-Rich Solids with an Unusually High Li+ Mobility. Angew. Chem. Int. Ed. 2008, 47 (4). https://doi.org/10.1002/anie.200703900. 50. De Klerk, N. J. J.; Rosłon, I.; Wagemaker, M. Diffusion Mechanism of Li Argyrodite Solid Electrolytes for Li-Ion Batteries and Prediction of Optimized Halogen Doping: The Effect of Li Vacancies, Halogens, and Halogen Disorder. Chem. Mater. 2016, 28 (21). https://doi.org/10.1021/acs.chemmater.6b03630. 51. Minafra, N.; Culver, S. P.; Krauskopf, T.; Senyshyn, A.; Zeier, W. G. Effect of Si Substitution on the Structural and Transport Properties of Superionic Li-Argyrodites. J. Mater. Chem. A Mater. 2018, 6 (2). https://doi.org/10.1039/c7ta08581h. 52. Kraft, M. A.; Culver, S. P.; Calderon, M.; Böcher, F.; Krauskopf, T.; Senyshyn, A.; Dietrich, C.; Zevalkink, A.; Janek, J.; Zeier, W. G. Influence of Lattice Polarizability on the Ionic Conductivity in the Lithium Superionic Argyrodites Li6PS5X (X ¼ Cl, Br, I). J. Am. Chem. Soc. 2017, 139 (31). https://doi.org/10.1021/jacs.7b06327. 53. Sicolo, S.; Kalcher, C.; Sedlmaier, S. J.; Janek, J.; Albe, K. Diffusion Mechanism in the Superionic Conductor Li4PS4I Studied by First-Principles Calculations. Solid State Ion. 2018, 319. https://doi.org/10.1016/j.ssi.2018.01.046. 54. Holzmann, T.; Schoop, L. M.; Ali, M. N.; Moudrakovski, I.; Gregori, G.; Maier, J.; Cava, R. J.; Lotsch, B. V. Li0.6[Li0.2Sn0.8S2]-a Layered Lithium Superionic Conductor. Energ. Environ. Sci. 2016, 9 (8). https://doi.org/10.1039/c6ee00633g. 55. Brant, J. A.; Massi, D. M.; Holzwarth, N. A. W.; Macneil, J. H.; Douvalis, A. P.; Bakas, T.; Martin, S. W.; Gross, M. D.; Aitken, J. A. Fast Lithium Ion Conduction in Li2SnS3: Synthesis, Physicochemical Characterization, and Electronic Structure. Chem. Mater. 2015, 27 (1). https://doi.org/10.1021/cm5037524. 56. Kuhn, A.; Holzmann, T.; Nuss, J.; Lotsch, B. V. A Facile Wet Chemistry Approach towards Unilamellar Tin Sulfide Nanosheets from Li4xSn1-XS2 Solid Solutions. J. Mater. Chem. A Mater. 2014, 2 (17). https://doi.org/10.1039/c3ta14190j. 57. Tang, B.; Jaschin, P. W.; Li, X.; Bo, S. H.; Zhou, Z. Critical Interface between Inorganic Solid-State Electrolyte and Sodium Metal. Mater. Today 2020. https://doi.org/10.1016/ j.mattod.2020.08.016. 58. Zhou, C.; Bag, S.; Thangadurai, V. Engineering Materials for Progressive All-Solid-State Na Batteries. ACS Energy Lett. 2018. https://doi.org/10.1021/acsenergylett.8b00948. 59. Lee, D. H.; Lee, S. T.; Kim, J. S.; Lim, S. K. Analysis of Properties of Partially Stabilized Zirconia-Doped Na+-Beta-Alumina Prepared by Calcining-Cum-Sintering Process. Mater. Res. Bull. 2017, 96. https://doi.org/10.1016/j.materresbull.2017.05.003.

510

Cell Components – Electrolytes | Overview - Solid Electrolytes

60. Youngblood, G. E.; Miller, G. R.; Gordon, R. S. Relative Effects of Phase Conversion and Grain Size on Sodium Ion Conduction in Polycrystalline, Lithia-Stabilized B-Alumina. J. Am. Ceram. Soc. 1978, 61 (1–2). https://doi.org/10.1111/j.1151-2916.1978.tb09238.x. 61. Lee, S. T.; Lee, D. H.; Lee, S. M.; Han, S. S.; Lee, S. H.; Lim, S. K. Effects of Calcium Impurity on Phase Relationship, Ionic Conductivity and Microstructure of Na+−b/B00 -Alumina Solid Electrolyte. Bull. Mater. Sci. 2016, 39 (3). https://doi.org/10.1007/s12034-016-1199-6. 62. Yang, Z.; Zhang, J.; Kintner-Meyer, M. C. W.; Lu, X.; Choi, D.; Lemmon, J. P.; Liu, J. Electrochemical Energy Storage for Green Grid. Chem. Rev. 2011. https://doi.org/10.1021/ cr100290v. 63. Hong, H. Y. P. Crystal Structures and Crystal Chemistry in the System Na1+xZr2SixP3-XO12. Mater. Res. Bull. 1976, 11 (2), 173–182. https://doi.org/10.1016/0025-5408(76) 90073-8. 64. Goodenough, J. B.; Hong, H. Y. P.; Kafalas, J. A. Fast Na+-Ion Transport in Skeleton Structures. Mater. Res. Bull. 1976, 11 (2). https://doi.org/10.1016/0025-5408(76) 90077-5. 65. Samiee, M.; Radhakrishnan, B.; Rice, Z.; Deng, Z.; Meng, Y. S.; Ong, S. P.; Luo, J. Divalent-Doped Na3Zr2Si2PO12 Natrium Superionic Conductor: Improving the Ionic Conductivity via Simultaneously Optimizing the Phase and Chemistry of the Primary and Secondary Phases. J. Power Sources 2017, 347. https://doi.org/10.1016/j. jpowsour.2017.02.042. 66. Zhang, Z.; Zhang, Q.; Shi, J.; Chu, Y. S.; Yu, X.; Xu, K.; Ge, M.; Yan, H.; Li, W.; Gu, L.; Hu, Y. S.; Li, H.; Yang, X. Q.; Chen, L.; Huang, X. A Self-Forming Composite Electrolyte for Solid-State Sodium Battery with Ultralong Cycle Life. Adv Energy Mater. 2017, 7 (4). https://doi.org/10.1002/aenm.201601196. 67. Porkodi, P.; Yegnaraman, V.; Kamaraj, P.; Kalyanavalli, V.; Jeyakumar, D. Synthesis of NASICON - A Molecular Precursor-Based Approach. Chem. Mater. 2008, 20 (20). https://doi.org/10.1021/cm800208k. 68. Naqash, S.; Ma, Q.; Tietz, F.; Guillon, O. Na3Zr2(SiO4)2(PO4) Prepared by a Solution-Assisted Solid State Reaction. Solid State Ion. 2017, 302. https://doi.org/10.1016/ j.ssi.2016.11.004. 69. Park, K. H.; Bai, Q.; Kim, D. H.; Oh, D. Y.; Zhu, Y.; Mo, Y.; Jung, Y. S. Design Strategies, Practical Considerations, and New Solution Processes of Sulfide Solid Electrolytes for All-Solid-State Batteries. Adv Energy Mater. 2018, 8 (18). https://doi.org/10.1002/aenm.201800035. 70. Hayashi, A.; Noi, K.; Tanibata, N.; Nagao, M.; Tatsumisago, M. High Sodium Ion Conductivity of Glass-Ceramic Electrolytes with Cubic Na3PS4. J. Power Sources 2014, 258. https://doi.org/10.1016/j.jpowsour.2014.02.054. 71. Hayashi, A.; Noi, K.; Sakuda, A.; Tatsumisago, M. Superionic Glass-Ceramic Electrolytes for Room-Temperature Rechargeable Sodium Batteries. Nat. Commun. 2012, 3. https://doi.org/10.1038/ncomms1843. 72. Jansen, M.; Henseler, U. Synthesis, Structure Determination, and Ionic Conductivity of Sodium Tetrathiophosphate. J. Solid State Chem. 1992, 99 (1). https://doi.org/ 10.1016/0022-4596(92)90295-7. 73. Heo, J. W.; Banerjee, A.; Park, K. H.; Jung, Y. S.; Hong, S. T. New Na-Ion Solid Electrolytes Na4−xSn1−xSbxS4 (0.02  x  0.33) for All-Solid-State Na-Ion Batteries. Adv Energy Mater. 2018, 8 (11). https://doi.org/10.1002/aenm.201702716. 74. Wang, H.; Schultze, T. M.; Kang, L.; Dewees, R.; Hood, Z. D. Effect of Halide Doping on the Structure and Properties of Na3SbS4 Solid Electrolyte. ECS Meeting Abstracts 2018, MA2018-01 (3). https://doi.org/10.1149/ma2018-01/3/451. 75. Zhao, Y.; Liu, Z.; Xu, J.; Zhang, T.; Zhang, F.; Zhang, X. Synthesis and Characterization of a New Perovskite-Type Solid-State Electrolyte of Na1/3La1/3Sr1/3ZrO3 for All-Solid-State Sodium-Ion Batteries. J. Alloys Compd. 2019, 783. https://doi.org/10.1016/j.jallcom.2018.12.289. 76. Richards, W. D.; Miara, L. J.; Wang, Y.; Kim, J. C.; Ceder, G. Interface Stability in Solid-State Batteries. Chem. Mater. 2016, 28 (1). https://doi.org/10.1021/acs. chemmater.5b04082. 77. Choi, S. J.; Choi, S. H.; Bui, A. D.; Lee, Y. J.; Lee, S. M.; Shin, H. C.; Ha, Y. C. LiI-Doped Sulfide Solid Electrolyte: Enabling a High-Capacity Slurry-Cast Electrode by Low-Temperature Post-Sintering for Practical All-Solid-State Lithium Batteries. ACS Appl. Mater. Interfaces 2018, 10 (37). https://doi.org/10.1021/acsami.8b11244. 78. Lü, X.; Wu, G.; Howard, J. W.; Chen, A.; Zhao, Y.; Daemen, L. L.; Jia, Q. Li-Rich Anti-Perovskite Li3OCl Films with Enhanced Ionic Conductivity. Chem. Commun. 2014, 50 (78). https://doi.org/10.1039/c4cc05372a. 79. Matsuura, N.; Umemoto, K.; Takeda, Y.; Sasaki, A. Formation Constants of Dibenzo-18-Crown-6 Complexes with Alkali Metal Ions in DMSO, DMF, and PC at 25
C. Bull. Chem. Soc. Jpn. 1976, 49 (5). https://doi.org/10.1246/bcsj.49.1246. 80. Okoshi, M.; Yamada, Y.; Komaba, S.; Yamada, A.; Nakai, H. Theoretical Analysis of Interactions between Potassium Ions and Organic Electrolyte Solvents: A Comparison with Lithium, Sodium, and Magnesium Ions. J. Electrochem. Soc. 2017, 164 (2). https://doi.org/10.1149/2.0211702jes. 81. Li, Y.; Lu, Y.; Adelhelm, P.; Titirici, M. M.; Hu, Y. S. Intercalation Chemistry of Graphite: Alkali Metal Ions and beyond. Chem. Soc. Rev. 2019. https://doi.org/10.1039/ c9cs00162j. 82. Jiang, C.; Li, H.; Wang, C. Recent Progress in Solid-State Electrolytes for Alkali-Ion Batteries. Sci. Bull. 2017. https://doi.org/10.1016/j.scib.2017.10.011. 83. Rajagopalan, R.; Tang, Y.; Ji, X.; Jia, C.; Wang, H. Advancements and Challenges in Potassium Ion Batteries: A Comprehensive Review. Adv. Funct. Mater. 2020. https://doi.org/ 10.1002/adfm.201909486. 84. Zhao, Q.; Stalin, S.; Zhao, C. Z.; Archer, L. A. Designing Solid-State Electrolytes for Safe, Energy-Dense Batteries. Nature Rev. Mater. 2020. https://doi.org/10.1038/s41578019-0165-5. 85. Yung-Fang, Y.; Yao; Kummer, J. T. Ion Exchange Properties of and Rates of Ionic Diffusion in Beta-Alumina. J. Inorg. Nucl. Chem. 1967, 29 (9). https://doi.org/10.1016/00221902(67)80301-4. 86. Lu, X.; Bowden, M. E.; Sprenkle, V. L.; Liu, J. A Low Cost, High Energy Density, and Long Cycle Life Potassium-Sulfur Battery for Grid-Scale Energy Storage. Adv. Mater. 2015, 27 (39). https://doi.org/10.1002/adma.201502343. 87. Masese, T.; Yoshii, K.; Yamaguchi, Y.; Okumura, T.; Huang, Z. D.; Kato, M.; Kubota, K.; Furutani, J.; Orikasa, Y.; Senoh, H.; Sakaebe, H.; Shikano, M. Rechargeable Potassium-Ion Batteries with Honeycomb-Layered Tellurates as High Voltage Cathodes and Fast Potassium-Ion Conductors. Nat. Commun. 2018, 9 (1). https://doi.org/10.1038/ s41467-018-06343-6. 88. Yuan, H.; Li, H.; Zhang, T.; Li, G.; He, T.; Du, F.; Feng, S. A K2Fe4O7 Superionic Conductor for all-Solid-State Potassium Metal Batteries. J. Mater. Chem. A Mater. 2018, 6 (18). https://doi.org/10.1039/c8ta01418c. 89. Ni, L.; Xu, G.; Li, C.; Cui, G. Electrolyte Formulation Strategies for Potassium-Based Batteries. Exploration 2022, 2 (2). https://doi.org/10.1002/exp.20210239. 90. Delmas, C.; Maazaz, A.; Fouassier, C.; Réau, J. M.; Hagenmuller, P. Influence de l’environnement de l’ion Alcalin Sur Sa Mobilite Dans Les Structures a Feuillets Ax(LxM1-x)O2. Mater. Res. Bull. 1979, 14 (3). https://doi.org/10.1016/0025-5408(79)90097-7. 91. Albrecht, R.; Hunger, J.; Block, T.; Pöttgen, R.; Senyshyn, A.; Doert, T.; Ruck, M. Oxo-Hydroxoferrate K2−xFe4O7−x(OH)x: Hydroflux Synthesis, Chemical and Thermal Instability, Crystal and Magnetic Structures. ChemistryOpen 2019, 8 (1). https://doi.org/10.1002/open.201800229. 92. Guth, U.; Löscher, B.; Schmidt, P.; Wulff, H.; Möbius, H. H. Structure and Electrical Conductivity of Polycrystalline K2YZr(PO4)3. Solid State Ion. 1992, 51, 3–4. https://doi.org/ 10.1016/0167-2738(92)90197-W. 93. Corigliano, O.; Pagnotta, L.; Fragiacomo, P. On the Technology of Solid Oxide Fuel Cell (SOFC) Energy Systems for Stationary Power Generation: A Review. Sustainability (Switzerland) 2022. https://doi.org/10.3390/su142215276. 94. Vinchhi, P.; Khandla, M.; Chaudhary, K.; Pati, R. Recent Advances on Electrolyte Materials for SOFC: A Review. Inorg. Chem. Commun. 2023, 152. https://doi.org/10.1016/j. inoche.2023.110724. 95. Steele, B. C. H.; Heinzel, A. Materials for Fuel-Cell Technologies. Nature 2001. https://doi.org/10.1038/35104620. 96. Steele, B. C. H. Appraisal of Ce1-YGdyO2-Y/2 Electrolytes for IT-SOFC Operation at 500
C. Solid State Ion. 2000, 129 (1). https://doi.org/10.1016/S0167-2738(99)00319-7.

Cell Components – Electrolytes | Overview - Solid Electrolytes

511

97. Mogensen, M.; Sammes, N. M.; Tompsett, G. A. Physical, Chemical and Electrochemical Properties of Pure and Doped Ceria. Solid State Ion. 2000, 129 (1). https://doi.org/ 10.1016/S0167-2738(99)00318-5. 98. Laguna-Bercero, M. A. Recent Advances in High Temperature Electrolysis Using Solid Oxide Fuel Cells: A Review. J. Power Sources 2012. https://doi.org/10.1016/j. jpowsour.2011.12.019. 99. Salado, M.; Lizundia, E. Advances, Challenges, and Environmental Impacts in Metal–Air Battery Electrolytes. Mater. Today Energy 2022. https://doi.org/10.1016/j. mtener.2022.101064. 100. Larrabide, A.; Rey, I.; Lizundia, E. Environmental Impact Assessment of Solid Polymer Electrolytes for Solid-State Lithium Batteries. Adv. Energy Sustainability Res. 2022, 3 (10). https://doi.org/10.1002/aesr.202200079.

Cell Components – Separators | Overview Dan Yang, School of Chemical and Environmental Engineering, RMIT University, Melbourne, VIC, Australia © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of Andrzej Kulczycki, Maternal Mortality and Morbidity, Edited by Stella R. Quah, International Encyclopedia of Public Health (Second Edition), Academic Press, 2017, Pages 553–564, ISBN 9780128037089, https://doi.org/10.1016/B978-0-12-803678-5.00269-1.

1 2 3 3.1 3.2 3.3 3.4 4 5 6 6.1 6.2 6.3 7 8 References

Introduction of separators Key parameters to evaluate a separator Categories of separators Microporous polymer membranes Nonwoven membranes Cellulose-based membrane Solid electrolyte membranes Separators for different types of batteries Manufacturing of separators Modification strategies to improve functions of novel separators Regulating dendrites formation Suppressing the polysulfide shuttling Enhanced thermal stability and tensile strength Data-driven material discovery Summary

513 513 514 514 515 515 515 516 516 518 518 518 519 520 520 521

Abstract Separators are one of the key components inside a battery. Presence of separators prevents internal short-circuit and ensures safety operation of a battery. The performance of the batteries is also determined by the properties of materials used to fabricate the separators. In this chapter, properties, selection criteria, categories and manufacturing process of separators are introduced. Applications of separators in different type of batteries have been discussed. In addition, novel strategies that have been developed to modify the properties of separators and improve performance of batteries, e.g., suppress dendrite growth and polysulfide shuttling, and enhance thermal/mechanical stability, have been overviewed. Recent progress of utilizing machine learning in screening and optimizing solid state electrolyte membranes are also briefly discussed.

Glossary Dendrites Projections of metal that can build up on the meal surface and penetrate into the solid electrolyte Dimensional stability The property of material maintains their original shape and dimensions throughout the manufacturing process, storage, and use. Gel polymer electrolyte A polymeric membrane onto which a minimum amount of classic “salt–solvent” combination is added. Gurley number The time required for a specified volume of gas to traverse a given area of the separator under certain pressure conditions. Machine learning Using data and algorithms to enable artificial intelligence to imitate the way that humans learn, gradually improving its accuracy. Nafion A sulfonated tetrafluoroethylene-based fluoropolymer-copolymer film. Phase inversion Remove the solvent from a liquid-polymer solution, leaving a porous, solid membrane. Polysulfide shuttle Sulfur species reach the negative electrode surface and undergo chemical reduction. Tensile strength The maximum load that a material can support without fracture when being stretched.

Key points

• • • •

512

Separators are selected depending on the specific type of batteries. Properties of separators can be modified to improve performance of batteries. Machine learning is a powerful tool for electrolyte membrane optimization. Advance of separators is the key to lead to high-performance batteries.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00238-3

Cell Components – Separators | Overview

1

513

Introduction of separators

Separator is a key component of a battery. They are usually porous membranes placed in between the positive and negative electrodes (as shown in Fig. 1A). Separators were firstly introduced into battery system in 1859 accompanied by the invention of lead-acid batteries. The use of separators that can transport oxygen, e.g., absorbed glass mats (AGM) enabled manufacturing of sealed batteries with more compact cell designs, e.g., the sealed nickel-cadmium in 1947 and the maintenance-free lead acid in the 1970s, which effectively suppressed the leakage and dry out of electrolytes. The main function of a separator is to keep the electrodes apart to prevent electrical short-circuit, but in the meantime, allow rapid transport of ionic charge carriers to deliver the current in an electrochemical cell. The performance of a battery is closely related with the properties of separators, e.g., porosity, thickness, ionic conductivity, electrolyte wettability, thermal/chemical stability and cost etc. An ideal separator should have high mechanical strength, high ionic conductivity, excellent electrolyte wettability and low cost (as shown in Fig. 1B).

2

Key parameters to evaluate a separator

Separators play a key role in determining the performance and ensuring safety of batteries. Several parameters have been used to evaluate performance of a separator, such as porosity, thermal stability, chemical resistance towards the corrosive electrolyte, and mechanical strength etc. These parameters are interconnected with each other and often balanced to satisfy requirements of one specific battery. For example, thin separators with rich porosity are seen as ideal for high energy-density batteries, while their inferior mechanical strength makes them problematic in some tough applications such as electric vehicles (EVs) or hybrid electric vehicles (HEVs). So far, it is still challenging to achieve a “perfect” separator that can fulfill requirements of all types of batteries. For a more comprehensive view, the typical parameters used for evaluation separators are listed here and Table 1.1 2.1. Porosity and pore sizes. Porous structures are desirable for separators to accommodate sufficient electrolyte and maintain the ionic conductivity. However, the pore size and porosity need to be tailored for optimum performance. Low porosity will result in higher internal resistance, while a high porosity will affect the mechanical strength and result in crossover of electrodes, leading to internal short-circuit. Separators for commercial lithium-ion batteries (LIBs) have a porosity around 30–40%.

Fig. 1 (A) Structural illustration of a battery. (B) Desirable features for battery separators.

Table 1

Typical parameters and requirements of separators.

Parameter

Requirement

Chemical and electrochemical stabilities Wettability Mechanical property Thickness Pore size Porosity Permeability (Gurley) Dimensional stability Thermal stability

Long-term stability Penetrate thoroughly and quickly >1000 kg cm−1 (98.06 MPa) 20–25 mm

> A ¼ , > > ROm ROm > > >   < expð−Dt=t1 Þ expð−Dt=t2 Þ B¼ , expð−Dt=tm Þ expð−Dt=tm Þ > > >   > > > R R > : C ¼ TH1 , TH2 RTHm RTHm

(6)

where Dt is the sampling time, s; t1 ¼ RTH1  CTH1, t2 ¼ RTH2  CTH2, and tm ¼ RTHm  CTHm, which are the time constants of battery polarization processes for Cell 1, Cell 2, and the mean cell. For the battery pack, during charge and discharge, the mean parameters can be easily identified online with the recursive least squares (RLS) based algorithm. Then, the parameters of each individual cell can be obtained by combining the ratio vectors and the online identified mean parameters by the following equation. Therefore, the determination of the ratio vectors is critical in this method. 8 RO1 ¼ A½1ROm > > > > > RO2 ¼ A½2ROm > > > < expð−Dt=t Þ ¼ B½1 expð−Dt=t Þ 1 m (7) > exp −Dt=t ð Þ ¼ B ½ 2  exp ð −Dt=t 2 mÞ > > > > > RTH1 ¼ C½1RTHm > > : RTH2 ¼ C½2RTHm

4.2.2 Details of the ratio vector-based parameter identification for series-connected battery 4.2.2.1 Determination of ratio vector (1) Determination of ratio vector A When the battery pack works under a same current, the relationship between the ohmic resistance of the two cells satisfies: RO1 =RO2 ¼ U O1 =U O2 where UO1 and UO2 are the voltages of RO1 and RO2 under current IB. Then, theoretically, vector A can be determined with

(8)

572

Cell and Battery Design – Batteries | Cell Connections  A¼

   RO1 RO2 U O1 U O2 ¼ , , ROm ROm U Om U Om

(9)

where UOm is the voltage of ROm under current IB. In real applications, it is difficult to obtain UO1, UO2 and UOm online because they cannot be directly measured, thus it is difficult to get the value of A with Eq. (9). However, it can be found that, if the current changes suddenly, RO will cause a sudden voltage change. Based on this analysis, ratio vector A can be determined with the sudden voltage changes of each individual cell and the mean cell, as shown in Eq. (15), where DUs1, DUs2 and DUsm are the sudden voltage changes of the cells respectively.     R R DU s1 DU s2 (10) , A ¼ O1 , O2 ¼ ROm ROm DU sm DU sm (2) Determination of ratio vector B The polarization voltages of the cells satisfy: U TH1,k U TH2,k




 Dt Dt U TH1,k−1 + 1 − exp − RTH1 IB,k exp − t1 t1


 ¼ Dt Dt U TH2,k−1 + 1 − exp − RTH2 IB,k exp − t2 t2

(11)

where UTH is the battery polarization voltage. Normally, the time constant of the RTHCTH network is much larger than sampling period (generally several or tens-of
 Dt milliseconds), thus, 1 − exp − ! 0. Hence, Eq. (11) can be approximately transferred to: tTH
 Dt U TH1,k−1 exp − t1 U TH1,k
  (12) U TH2,k Dt U TH2,k−1 exp − t2 If defined 8 > < K t1 ¼ U TH1,k =U TH1,k−1 K t2 ¼ U TH2,k =U TH2,k−1 > : K tm ¼ U THm,k =U THm,k−1

(13)

Then the ratio vector B can be converted to  B¼

   expð−Dt=t1 Þ expð−Dt=t2 Þ K t1 K t2 , ¼ , K tm K tm expð−Dt=tm Þ expð−Dt=tm Þ

During battery charge and discharge, the polarization voltages can also be calculated as  U TH1 ¼ U d1 − IB RO1 U TH2 ¼ U d2 − IB RO2

(14)

(15)

in which Ud is the dynamic voltage response caused by the impedance, and can be calculated by Ud ¼ UB − UOC(SOC). After measuring the terminal voltages of each cell and the pack, the dynamic voltage Ud caused by the impedance can be calculated. Then, with Eq. (15) and the identified ohmic resistances of cell1, cell2 and the mean cell, the polarization voltages can be obtained. Finally, the ratio vector reflecting the difference of polarization time constants, B can be calculated with Eq. (14). (3) Determination of ratio vector C The battery polarization resistance RTH can be obtained through the following equation: 8
 Dt > > U TH1,k−1 U TH1,k − exp − > > > t1 > >

  R ¼ > TH1 > > Dt > > 1 − exp − < t1
 > Dt > > U TH2,k−1 U TH2,k − exp − > > t2 > >

  > R ¼ > TH2 > Dt > > 1 − exp − : t2 If defined

(16)

Cell and Battery Design – Batteries | Cell Connections
 8 Dt > > U TH1,k−1 U TH1,k − exp − > > t1 > >

 > > K RTH1 ¼ > Dt > > 1 − exp − > > t1 > > >
 > > > Dt > > U TH2,k−1 U TH2,k − exp − < t2

 K RTH2 ¼ > Dt > > 1 − exp − > > t2 > > >
 > > Dt > > U THm,k−1 U THm,k − exp − > > tm > > >

 K RTHm ¼ > > Dt > > : 1 − exp − tm

573

(17)

Then, the ratio vector C can be determined as  C¼

   RTH1 RTH2 K RTH1 K RTH2 ¼ , , RTHm RTHm K RTHm K RTHm

(18)

With vector B, and the identified time constant of the mean cell tm, t1 and t2 can be obtained with Eq. (14), then with Eq. (16) and Eq. (17), vector C can be finally determined. 4.2.2.2 Online update of the ratio vectors For a battery system composed of N series-connected battery cells, the ratio vectors reflecting the characteristics differences among the cells are   8 DU s1 DU s2 DU sn > > , , . . . , A ¼ > > DU DUsm > >  sm DU sm < K t1 K t2 K tn , , ..., B¼ (19) K tm K tm K tm > >   > > > K RTH1 K RTH2 K > :C ¼ , , . . . , RTHn K RTHm K RTHm K RTHm When using the ratio vectors to determine the parameters for each individual cell, to avoid the possible fluctuations and errors, an adaptive filtering is further designed. According to Eq. (10), at the sampling step k, the sudden voltage change vector of all the individual cells can be estimated with pre

DU s,k ¼ Ak−1 DU sm,k

(20)

If the ratio vector A carries some errors, then, the estimated sudden voltage changes of the individual cells should be different from the true values, then vector A can be adjusted with the errors as   pre (21) Ak ¼ Ak−1 + gA DU s,k − DU s,k in which, gA is an adjustment gain for vector A. Similarly, the filters for vectors B and C are designed as shown below 8 pre < K t,k ¼ K tm,k Bk−1   : Bk ¼ Bk−1 + gB K t,k − K pre t,k 8 pre < K RTH,k ¼ K RTHm,k Ck−1   : Ck ¼ Ck−1 + gC K RTH,k − K pre RTH,k

(22)

(23)

in which, gB and gC are the adjustment gains for vector B and C respectively.

4.2.3 Experimental results of the ratio vector-based method A small battery pack composed of 10 series-connected cells is constructed to validate the ratio vector-based parameter identification method. The battery’s nominal capacity is 80 Ah, and the charge/discharge cut-off voltages are 4.2 V and 2.8 V, respectively. Experimental tests with the New European Driving Cycle (NEDC) current profile are designed to investigate the method’s performance thoroughly. The battery pack is put in an environmental chamber during the tests, and the temperature is set to 25  C. The actual battery current and voltage and the voltages of the 10 cells during the tests are all simultaneously measured by the BMS. With all the measured current and voltage, the parameters of each cell are obtained by the repetitive implementation of the

574

Cell and Battery Design – Batteries | Cell Connections

RLS-based identification, and the identification results are considered as the reference values of the parameters. On the other hand, the cell parameters are also identified by the ratio vector-based method. Fig. 6 shows the parameter identification results obtained by the repetitive RLS algorithm and the ratio vector-based method during the tests. The parameters obtained by the ratio vector-based method are very close to the reference values identified by the repetitive RLS algorithm, especially after the algorithm’s convergence. Because of the design of the adaptive filtering of the ratio vectors, the convergence of the ratio vector-based method is slower than the repetitive RLS algorithm. This can be seen more clearly by looking closely at the identified results from 0 s to 200 s. This is because the ratio vector-based algorithm should first use the results identified by traditional RLS as the mean value of the parameters and then determine the parameters for each cell with the ratio vectors. Moreover, the adaptive filters are further designed to avoid possible fluctuations and errors, further slowing down the algorithm’s convergence. The identification results at the end of the tests are zoomed in, as shown in Fig. 6. The parameters

Fig. 6 Identified parameters under NEDC cycle test13: (a) RO identified by the repetitive RLS; (b) RO identified by the ratio vector-based method; (c) RTH identified by the repetitive RLS; (d) RTH identified by the ratio vector-based method; (e) CTH identified by the repetitive RLS; (f ) CTH identified by the ratio vector-based method.

Cell and Battery Design – Batteries | Cell Connections

575

identified by the two methods respectively are almost the same. An apparent convergence of the ratio vectors can be found in the identification results. The computation time cost of the two methods is also calculated. The computation time of the repetitive RLS is 2.74 s, while the computation time of the ratio vector-based method is 1.28 s, which is reduced by 53.2%. The ratio vector-based method can achieve the same identification accuracy while reducing the computation cost.

4.3

State estimation for series connected battery systems

4.3.1 Basic idea As mentioned above, the acquisition of the battery state of each cell in the series-connected battery module is essential but challenging, owing to the difference between the cells. To handle this problem, Dai et al.11 first studied an online cell SOC estimation method of Li-ion battery packs using dual time-scale Kalman filtering for EV applications. With an equivalent circuit-based “averaged cell” model, this method estimates the battery pack’s average SOC first, then incorporates the performance divergences between the “averaged cell” and each cell to generate the SOC estimations for all cells. Considering that the battery capacity changes during battery degradation, it is necessary to realize a joint estimation of the SOC and capacity for series-connected battery packs. Jiang et al.14 present a cell-to-pack state estimation extension method based on a multilayer difference model for series-connected battery packs, which can efficiently realize accurate SOC and capacity estimation for a battery pack based on the existing estimation algorithms for a single cell.

4.3.2 Details of the State estimation for series connected battery systems 4.3.2.1 SOC difference estimation based on cell voltage After acquisition of model parameters, the terminal voltage of ith cell can be expressed as follows:

U iB,k ¼ U OC SOCik − U iTH,k − Rio,k IB,k

(24)

where the superscript i represents the ith cell in the battery pack. Supposing that the reference cell’s SOC and capacity in the battery pack have been estimated accurately, the estimated reference SOCrk can be used to rewrite the above equation. Defining dSOCik ¼ SOCik − SOCrk as the difference between the target ith cell and the reference rth cell, Eq. (24) can be further expressed as a nonlinear function of the SOC difference, as shown below:

U iB,k ¼ U OC SOCrk + dSOCik − U iTH,k − Rio,k IB,k (25) Supposing that dSOCik is a slowly varying quantity and the updating time interval is nDt, the SOC difference can be updated as follows: dSOCik ¼ SOCik − SOCrk ! ! Pk Pk i r k−n IB Dt k−n IB Dt ¼ SOCk−n − − SOCk−n − Crm Cim dCi Xk ¼ dSOCik−n + i mr I Dt k−n B Cm Cm

(26)

where dCim ¼ Cim − Crm, dCim is defined as the capacity difference between the target cell and the reference cell. Considering that dCim is P small, and the charge accumulation ( kk−nIBDt) within the time interval nDt will not be too large, Eq. (26) can be approximated to i i dSOCk  dSOCk−n. This equation can be considered as the state transfer function of state difference dSOCi. Combined with the measurement function shown in Eq. (25), the state-space function of SOC difference system can be established. Further, the optimal solution of dSOCi can be realized through some adaptive algorithms. For SOC estimation of cells in the series-connected battery pack, the SOC difference system for one cell should be extended T to a higher dimensional system. Defining a system state as xdSOC,k ¼ [dSOC1k , . . ., dSOCik, . . ., dSOCN k ] and a system output as T 1 i N ydSOC,k ¼ [UB, k, . . ., UB, k, . . ., UB, k] , which are both (N-1)-dimensional vectors, Eqs. (26) and (25) can be converted to the following function: 8 xdSOC,k  xdSOC,k−n > >

> 3 2 > > U OC SOCrk + xdSOC,k ½1 − U 1TH,k − R1o,k IB,k > > > > 7 6 > < 7 6 ⋯ 7 6 (27) 7 6

> y ¼ g ð x Þ ¼ 7 6 r i i > dSOC,k dSOC dSOC,k SOC + x ½ i  − U − R I U > OC dSOC,k 6 k TH,k o,k B,k 7 > > 7 6 > ... > 5 4 > > >

: r N N U OC SOCk + xdSOC,k ½N − U TH,k − Ro,k IB,k where gdSOC() is the multi-dimensional nonlinear function. It is evident that, based on above SOC difference system, some adaptive filters can be employed to estimate the optimal value of xdSOC, k, and the estimated SOC of ith cell can be expressed as:

576

Cell and Battery Design – Batteries | Cell Connections b i ¼ dSOC b i + SOCr SOC k k k

(28)

4.3.2.2 Capacity difference estimation based on SOC difference The acquired all cells’ SOC can be further used for the capacity difference system. For the reference cell and the target cell, the following equations are satisfied (k2 > k1): 8 Pk¼k2 > > k¼k1 IB,k Dt r r > > SOC ¼ SOC − < k2 k1 Crm (29) Pk¼k2 > > k¼k1 IB,k Dt > i i > SOC ¼ SOC − : k2 k1 Cim P For series-connected battery packs, the current loaded on all cells is consistent; hence k¼k1k¼k2IB, kDt for all cells is same. With the i i help of the defined dSOCk and dCm, the following equation can be deduced:   h   i

SOCrk2 − SOCrk1  Crm ¼ SOCrk2 + dSOCik2 − SOCrk1 + dSOCik1  Crm + dCim (30) After simplification, it can be found that Eq. (30) consists of (dSOCik2 − dSOCik1)  dCim, which is composed of the change in SOC difference dSOCk2i − dSOCk1i and the capacity difference dCim. Both these values are small, and their product is smaller and can be ignored. Hence, Eq. (30) can be approximately simplified as follows:     dSOCik2 − dSOCik1  Crm  SOCrk1 − SOCrk2  dCim (31) In the above equation, only the capacity difference dCim is unknown. Eq. (31) can also be extended from a single cell to the T series-connected battery pack. A (N-1)-dimensional parameter vector can be expressed as udCm ¼ [dC1m, . . ., dCim, . . ., dCN m] , and r 1 1 i i N N T an output vector zdCm ¼ Cm  [(dSOCk2 − dSOCk1), . . ., (dSOCk2 − dSOCk1 ), . . ., (dSOCk2 − dSOCk1 )] with same dimension is also defined. Therefore, the following equation with least squares form is further deduced: zdCm ¼ hdCm udCm

(32)

where hdCm ¼ (SOCrk1 − SOCrk2) is the data vector, and Eq. (32) is known as the capacity difference equation. It is accessible that the acquisition of udCm can be realized through the RLS algorithm, which is a preferred approach to acquire unknown parameters online, with high efficiency and accuracy. Finally, the estimated capacity of ith cell can be expressed as: bi + Cr bi ¼ dC C m m m

(33)

4.3.2.3 State determination of series-connected battery pack The battery pack SOC and capacity can be further derived with the knowledge of all cells’ SOC and capacity. For the series-connected battery module, the charging stop condition of the module is that any cell reaches the charging stop condition, and the discharge stop condition of the module is that any cell reaches the discharge stop condition. Therefore, the calculation of the battery pack SOC and capacity depends on the battery equalization mode (Detailed information about battery equalization can be found in Section 5.3). Without cell equalization, the battery pack’s maximum available capacity can be recognized as the sum of the minimum remaining cell capacity and the minimum chargeable cell capacity15, as formulated in the following equation. Further, the battery pack SOC is the ratio of minimum remaining cell capacity to battery pack capacity. 8 pack  

Cm ¼ min SOCi  Cim + min 1 − SOCj  Cjm > > > 1 i N 1 j N <  (34) min SOCi  Cim > 1 i N > pack > SOC ¼ : Cpack m When the battery module has passive balancing, the battery with a smaller rechargeable capacity will trigger the balancing switch during charging. At this time, part of the power passing through the battery will be dissipated in the form of “heat production.” Theoretically, the batteries within the series module can reach the charging stop condition at the same time. Therefore, the battery module capacity is equivalent to the minimum cell capacity with passive balancing:  ¼ min Cim (35) Cpack m 1 i N

Active balancing can realize energy transfer between batteries, which means that the single cells in the series module can theoretically reach the charging and discharging stop conditions simultaneously, and the utilization of the single-cell capacity will be maximized. At this time, the module capacity is the average capacity of all single cells: Cpack ¼

N 1 X i C N i¼1 m

(36)

Cell and Battery Design – Batteries | Cell Connections

577

4.3.3 Experiments results of the state estimation extension method An 18650-type cell with a nominal capacity of 2.9 Ah is used. The cut-off voltages for charging and discharging are 4.2 V and 2.5 V, respectively. A small battery module consisting of four cells performs the battery pack experiments. Two dynamic condition cycles are employed to verify the state estimation extension method. One is the Compound cycle test, which consists of the NEDC and the Urban Dynamometer Driving Schedule, and another is the Worldwide Harmonized Light Vehicles Test Procedure (WLTP). Figs. 7 and 8 show the cell SOC and capacity estimation results under the Compound cycle test and WLTP test, in which Cell 4 is chosen as the reference cell. Apparently, under different dynamic conditions, the estimated cell SOC shown in Fig. 7 through the state estimation extension method accurately follows the actual SOC. Under the Compound cycle test, the maximum absolute errors (MAEs) of each cell are 1.63%, 1.61%, 1.65%, and 1.39%, respectively, and cell SOC root mean squared errors (RMSEs) are all within 1%. The WLTP test results show similar performance: the cell SOC estimation MAEs and RMSE are less than 2% and 1%, respectively.

Fig. 7 Cell SOC estimation under Compound cycle test and WLTP test.14

Fig. 8 Cell capacity estimation under Compound cycle test and WLTP test.14

578

Cell and Battery Design – Batteries | Cell Connections

Fig. 8(d) shows the capacity estimation of the reference cell. The estimated capacity under different dynamic conditions converges to the 3% error bounds within 2000 s, and after that, the estimation is stable. In contrast, the capacity estimation of the remaining cells is slightly slower and has some fluctuations. Overall, the RLS estimator obtains accurate capacity differences, and most estimated capacity values are within the 3% error bounds. Owing to the convergence of estimation algorithms for capacity and capacity difference, the state estimation extension method takes 50–60 min (about 20% SOC variation) to obtain stable estimates for cell capacity. This means that a 20% SOC variation is necessary for this method. For capacity estimation under the Compound cycle test, the convergence time is 3500 s, while the WLTP test uses 3000 s as the convergence time. After convergence, Cell 3 presents the largest capacity estimation deviation among four cells: under the Compound cycle test, the RMSE is 1.08%; under the WLTP test, the RMSE is 0.71%. The state estimation extension method shows ideal results and accurately estimates cell SOC and capacity.

5

Battery system management

As discussed above, the characteristics of the series/parallel-connected battery modules depend on the cell’s characteristics and the inconsistency between cells. Inconsistency of the cells’ performance, i.e., capacity and internal resistance, is initially formed during production, and then the inconsistency evolves in the lifespan. This section investigates the performance and inconsistency evolution of the series and parallel modules for the design and management guide. Nevertheless, the performance difference of single cells exists in battery pack applications. This problem can only be reduced but cannot be eliminated. In view of the consistency problem of battery cells, research has been mainly carried out from two aspects when applied in groups: one is sorting single cells; the other is alleviating the impact of inconsistency through reasonable battery management. The main methods include battery balance, battery thermal management, modelling and state estimation considering battery cell differences, etc.

5.1

Battery inconsistency

5.1.1 Cause of inconsistency issues in lithium-ion batteries The manufacturing process of lithium-ion power batteries generally includes the steps of mixing, coating, assembly and formation. At each of these steps, there may be inconsistencies between cells due to materials and manufacturing accuracy. For example, in the manufacturing process of LIBs, the coating of positive and negative active materials needs very high precision, which is the key to ensuring reliable and consistent battery performance. The mixed stock is placed in a sealed container, and the amount of stock entering the coating is controlled by a high-precision pump. However, in actual manufacturing, both the sealing of the container and the control of entering amount of stock may be deviate, thus affecting the thickness of the coating, resulting in differences in battery performance. The differences in the performance of cells caused by the manufacturing process will result in the inconsistent degradation rate of the battery performance due to coupling usage conditions during use and lead to the gradual amplification of the difference of cells. Generally, differences in initial performance parameters and the difficulty in ensuring consistency with external usage conditions can lead to differences in battery operating conditions, further exacerbating the differences in battery performance changes. The coupling effect between the two differences is shown in Fig. 9. The research on battery life shows that the main factors affecting battery performance attenuation are temperature, depth of discharge (DOD) and current ratio.16–18 For the cells in the battery pack, under the same current condition, inconsistent capacity will bring the difference in the actual current ratio and DOD. Due to the difference in internal resistance, the cells heat differently. If the temperature field design is reasonable, the temperature of each cell will be consistent. In addition, for the parallel battery system, the actual cell current will be different due to the differences in the internal resistance and connection of the cells. The above factors are coupled with each other during the actual operation of the battery, which ultimately leads to different performance attenuation rates for each cell in the battery pack. Due to the difference in performance decay rate, the difference in battery capacity and internal resistance is further aggravated, which forms a positive feedback effect.

5.1.2 Inconsistency evolution of lithium-ion battery modules 5.1.2.1 Inconsistency quantification of battery modules Cells’ performance inconsistency strongly affects the module’s performance. To quantify the inconsistency of a dataset, range, variance, standard deviation, and coefficient of variation (CV) are usually used. The standard deviation varies with the magnitude of the samples in the dataset and still has dimensions. Only historical data generated by a specific system and with the same dimension can be compared. To broaden the scope of applications, the normalized CV, as shown in Eqs. (37) and (38), can be used to quantify the inconsistency of the cell capacity and internal resistance in the module. The higher the CV is, the more inconsistent the cell parameters are. rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 1 N (37) dC,CV ¼ S ðC − Cmean Þ2  100% Cmean N i¼1 i

Cell and Battery Design – Batteries | Cell Connections

579

Fig. 9 Internal and external factors affecting battery consistency and their coupling relationships.

dR,CV

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 N ¼ S ðR − Rmean Þ2  100% Rmean N i¼1 i 1

(38)

where dC, CV and dR, CV are the battery capacity and internal resistance CVs; the subscript “mean” represents the mean value of battery cell capacity and internal resistance.

5.1.2.2 Evolution of Series Battery Module Performance and Inconsistency Twelve fresh battery cells (the nominal capacity is 2.5 Ah) from Samsung SDI are connected in series to form a module. The module will be aged with a 1C-rate constant current (CC) charge/discharge condition at 25  C. Notably, the capacity and the internal resistance vary with different temperatures and C-rates.19 Here, the capacity is measured at 25  C with 1C-rate discharge current. The internal resistance is obtained under a 2.5C-rate discharge current pulse lasting 18 s at 25  C and 50% SOC. In the aging experiment, the capacity and internal resistance are calibrated every 50 equivalent full cycles (EFCs). An EFC means that the total discharge electricity is equal to the nominal capacity.20 Since the cells used in the experiment are all fresh, the parameter difference between the cells in the module is quite small. Thus, dC,CV and dR,CV shown in Fig. 10 can be maintained below 1% before 400 cycles. However, they continue to evolve and have clear upward trends after 500 cycles. It means that the inconsistency between the cells worsens quickly. The increase in CV indicates that the cell performance in the module becomes more extreme. Even though the cells are all fresh and strictly sorted, the performance and inconsistency worsen as the ageing process continues. The inconsistency of the cell parameters shows a noticeable accelerated upward trend in the later stage of ageing. Relying only on the characteristics of the series module itself cannot reverse the trend of increasing inconsistency. The above results illustrate the importance of maintaining the cell working range’s consistency for improving the performance and consistency of series modules in the lifespan. Because series modules tend to deteriorate in performance spontaneously, external intervention such as cell equalization is necessary.

5.1.2.3 Evolution of parallel battery module performance and inconsistency Eight fresh cells (the nominal capacity is 2.9 Ah) from Panasonic were used in the experiments. As discussed above, the terminal voltage and the operating range of the cells in the parallel module are usually the same. The main inconsistency lies in the performance parameters, i.e., capacity and internal resistance. To construct a significant difference in the parameters, pre-ageing cycles are performed on half of the cells. After artificially causing parameter differences, these cells are grouped in two to form four parallel modules. Each module contains a fresh cell and a pre-aged cell. The modules are aged under 1C-rate CC charge/discharge at 25  C. As shown in Fig. 11(a) and (c), the cells’ capacity and internal resistance have large differences at the beginning of the ageing processes due to the pre-ageing cycles. Fig. 11(b) shows that the capacity inconsistency of Module II (Cell #3 and #4) is a little worse than that of Module I (Cell #1 and #2) at the beginning. The consistency of Module II is improved faster than Module I. Then, a clear trend of convergence appears as the ageing processes continue. The capacity and internal resistance almost reach the same values at the end of the ageing process. Due to the parallel module properties, every cell has the same number of charge and discharge cycles. However, the capacity fading and the internal resistance growth rate of Cell 3 in Module II are slightly faster than Cell 1 in Module I. It leads to a worse inconsistency of the cells in Module II than those in Module I. The good-performance cells bear a larger current in a parallel module

580

Cell and Battery Design – Batteries | Cell Connections

Fig. 10 Evolution of performance and inconsistency of the series module with inconsistent parameters,21 (a) cell capacity, (b) capacity inconsistency, (c) cell internal resistance, (d) internal resistance inconsistency.

with high inconsistency. When other conditions are the same, a larger current leads to a faster ageing rate.22,23 Thus, a cell with better performance ages faster, and the module inconsistency finally decreases. Different from a series module, self-balance of performance and inconsistency are essential characteristics of parallel modules. It makes the cells in the parallel module converge spontaneously. Therefore, for parallel modules, balance during use is not needed. It facilitates the management of parallel modules. However, Wang et al.21 also found that if there is a relatively large parameter inconsistency between the cells in the module, the cells with better performance will be subjected to additional circulation. A large current will worsen this cell’s performance, making the parallel module’s overall performance decay faster. Therefore, it is still necessary to ensure consistent cell performance during sorting and grouping.

5.2

Battery sorting

The purpose of battery sorting is to improve the consistency of individual batteries’ initial performance and status when grouped into modules to reduce the impact of initial differences during battery use. At present, battery sorting often uses the charge and discharge voltage curves as an indicator of battery difference. Nevertheless, multidimensional sorting indices are needed to evaluate the battery state comprehensively. For example, classifying LIBs based solely on the capacity may result in high-capacity batteries with different internal resistances placed in the same category, which is unreasonable. The commonly used and practical sorting indexes are summarized as follows: (1) Capacity and internal resistance: These two indexes refer to the ability of the battery energy and power output. Battery capacity represents the number of active ions embedded into the electrode or dis-embedded from the electrode in the charge/discharge. The internal resistance is usually considered the sum of ohmic and polarization resistance. These two indexes are widely used in current sorting techniques. (2) Electrochemical impedance spectroscopy (EIS): The electrochemical impedance is closely related to the internal physical and chemical processes of LIBs, and the impedance is frequently reported as a powerful tool used in LIB state estimation and diagnosis. The impedance spectroscopy can reflect the battery characteristic at different frequency points, providing more information for battery sorting. Nevertheless, the practical measurement method of the EIS is still a challenge.

Cell and Battery Design – Batteries | Cell Connections

581

Fig. 11 Performance and parameter evolution of parallel modules under 1C-rate CC/CC aging condition21, (a) cell capacity, (b) capacity inconsistency, (c) cell internal resistance, and (d) internal resistance inconsistency.

(3) Voltage Curve: The voltage curve measured during the charge/discharge process reflects the battery’s dynamic performance. The charging/discharge condition, especially the CC condition, is stable and suitable for sorting feature construction. Lai et al.19 proposed a rapid sorting and regrouping of LIB modules based on partial charging curves, in which the charging curves of cells in a module are translated and supplemented to extract the capacity characteristics without disassembling the modules. (4) Dynamic parameters: The battery models can describe the battery characteristics. Commonly used battery equivalent circuit models include the Rint model, Thevenin model, PNGV model, Second-order Randle model, etc. Depending on the selected model and its dynamic parameters identification algorithm, these models and parameters can also be used for the battery’s characteristic division and battery sorting. (5) Thermal behavior: The battery’s thermal behavior is seriously related to the battery’s ageing rate and the user’s safety. The Battery’s surface temperature (Tcell) can reflect thermal behavior but is determined by both internal characteristics and the experimental environment. The temperature rise of the battery during charge/discharge mainly consists of irreversible heat rise, reversible heat rises and heat exchange. The corresponding thermal parameters can also be used for battery sorting. Battery sorting can only ensure the consistency of the initial performance of the cells before the batteries are grouped. In actual use, the consistency of the cells will deteriorate due to differences in usage conditions and environments. Therefore, to further alleviate the problems caused by consistency problem, it is also necessary to cooperate with some of the following technical measures, including battery equalization, thermal management, etc.

5.3

Battery equalization

Battery equalization refers to techniques that improve the difference of a battery pack with multiple cells (usually in series) and increase each cell’s performance. The individual cells in a battery pack naturally have somewhat different capacities throughout the charge and discharge cycles, may be at different SOCs. Variations in capacity are due to manufacturing variances, assembly variances, cell ageing, impurities, or environmental exposure. Balancing a multi-cell pack helps maximize the pack’s capacity and service life by working to maintain the equivalent SOC of every cell. In today’s rapidly evolving battery technology landscape, the role of the BMS in cell equalization is becoming increasingly sophisticated. Advanced algorithms and real-time monitoring capabilities within the BMS continually refine the equalization process, optimizing battery performance and prolonging its operational life. As the demand for energy storage and electric mobility

582

Cell and Battery Design – Batteries | Cell Connections

continues to grow, the BMS remains an indispensable component, orchestrating the delicate strategy of cell equalization to meet the stringent demands of modern applications.

5.3.1 Equalization circuits Balancing circuits can be passive or active and are called passive and active equalization methods. Passive equalization equalizes the SOC at some fixed point – usually either “top balanced,” with all cells reaching 100% SOC simultaneously, or “bottom balanced,” with all cells reaching minimum SOC simultaneously. This can be accomplished by bleeding energy from the cells with higher SOC (e.g., a controlled short through a resistor or transistor, as shown in Fig. 12(a)) or shunting energy through a path in parallel with a cell during the charge cycle so that less of the (typically regulated constant) current is consumed by the cell. Passive equalization is inherently wasteful, with some of the pack’s energy spent as heat for the sake of equalizing the SOC between cells. The build-up of waste heat may also limit the rate at which equalization can occur. Besides, passive equalization has lower efficiency for large-capacity batteries, making it challenging to meet the demand for large amounts of equalization. In active equalization, energy is drawn from the most charged cell and transferred to the least charged cells. Active equalization attempts to redistribute energy from cells at full charge to those with a lower SOC. Energy can be bled from a cell at higher SOC by switching a reservoir capacitor in-circuit with the cell, then disconnecting the capacitor and reconnecting it to a cell with lower SOC or through a DC-to-DC converter connected across the entire pack. Due to inefficiencies, some energy is still wasted as heat, but not to the same degree. Active equalization utilizes external energy storage devices (such as capacitors, inductors, and DC-DC converters) to shuttle the energy among cells to balance the cells. According to the different types of energy storage devices, active equalization methods can be divided into capacitor-based, inductor-based or converters-based. Fig. 12(b) shows the capacitor-based active equalization topology. Despite the apparent advantages, an active equalization topology’s additional cost and complexity can be substantial and sometimes make sense depending on the application.

5.3.2 Equalization strategy Methods of battery equalization can also be divided according to the controlled variables, including voltage-based, SOC-based or capacity-based. The voltage-based method refers to the utilization of the battery’s terminal voltage during charge and discharge cycles as the basis for determining whether the battery needs equalization operation. The advantage of the voltage-based method is that the voltage value is easy to sample, and can achieve real-time, high-speed and accurate sampling. However, in practical applications, the voltage sampling value fluctuates significantly with operating conditions or current pulses, which can easily cause misbalance and over balance of battery voltage. The capacity-based method is to control the maximum available capacity of the cell to be consistent. Compared with the voltage-based method, capacity-based equalization can more directly reflect the essence of balance, reducing the inconsistency of battery capacity. The problem restricting capacity-based equalization is the dynamic estimation of battery capacity. Under the condition of charge and discharge, the battery capacity is often affected by temperature, electrolyte concentration, charge and discharge current and other factors. It is difficult to estimate the actual capacity and ensure that the capacity of the battery tends to be consistent. Accurate estimation of SOC is also the difficulty of the SOC-based method, which greatly limits its practicability.

Fig. 12 The diagram of typical battery equalization, (a) passive equalization using resistors and (b) active equalization using a capacitor.

Cell and Battery Design – Batteries | Cell Connections 5.4

583

Battery thermal management

Thermal management aims to reduce the influence of temperature field distribution on battery performance and attenuation from the outside and to prevent the consistency of battery cells from worsening caused by temperature differences. The temperature difference in the battery is due to the non-uniformity of heat dissipation and the production of cells. In the case of harsh environmental conditions, improper module layout, and poor thermal conductivity, the temperature difference inside the battery pack will increase. If the battery pack works in this uneven temperature field for a long time, the difference between the performance of the battery modules and the cells will be further aggravated. For battery systems, reasonable thermal management design is the prerequisite for battery pack temperature consistency. Currently, the research on the thermal management of battery packs mainly focuses on the heat dissipation of batteries. From the perspective of cooling media, cooling schemes generally include air cooling, liquid cooling and phase change material colling.24 The control of the temperature rise of the battery pack is more accessible than the control of the temperature difference between cells. The optimal design of the battery pack structure is needed to solve the problem of temperature difference control. A systematic design method should be adopted to design a battery thermal management system with good performance. The detailed steps of the battery pack thermal management system’s design can be summarized as follows. (1) Determine design objectives and constraints. This includes determining the type of battery, the acceptable temperature range and degree of change, and the battery packaging required by the vehicle. (2) Obtain the thermal characteristics of the battery (such as heat generation rate and capacity, etc.). These factors affect the size of the cooling and heating system and the speed at which the battery responds to thermal fluctuations. (3) Complete preliminary estimation of module and cooling effect. Preliminary analysis should be conducted to determine transient- and steady-state thermal responses of modules and battery packs. Therefore, selecting the heat transfer medium and flow path is necessary. (4) Predict the thermal behavior of the battery and analyze the influence of various parameters under different driving conditions. (5) Design a preliminary battery cooling system. Determine the parameters of the system based on the expected performance. (6) Build and test battery cooling systems. After the construction of the cooling system, the relevant tests should be carried out on the test bench and the vehicle under different loads and working conditions. (7) Improve and enhance the system. Based on test data and analysis, carry out detailed adjustments and improvements.

6

Conclusions

Series and parallel connections are an important aspect of the battery system. Owing to the special characteristic of series and parallel connections, battery system design, modelling, and management should be thoroughly discussed. This article summarizes the battery system design, modelling and management and provides some typical cases. The experimental results prove that these algorithms are practical. Nevertheless, it must be said that with the introduction of new battery technologies, such as the large-capacity battery and cell-to-pack architecture, more effective techniques need to be studied to achieve advanced battery management technologies.

References 1. Duan, J.; Tang, X.; Dai, H.; Yang, Y.; Wu, W.; Wei, X.; et al. Building Safe Lithium-Ion Batteries for Electric Vehicles: A Review. Electrochem. Energy Rev. 2019, 1–42. 2. Dai, H.; Jiang, B.; Hu, X.; Lin, X.; Wei, X.; Pecht, M. Advanced Battery Management Strategies for a Sustainable Energy Future: Multilayer Design Concepts and Research Trends. Renew. Sustain. Energy Rev. 2021, 138. 3. Wang, Y.; Tian, J.; Sun, Z.; Wang, L.; Xu, R.; Li, M.; et al. A Comprehensive Review of Battery Modeling and State Estimation Approaches for Advanced Battery Management Systems. Renew. Sustain. Energy Rev. 2020, 131, 110015. 4. Dubarry, M.; Pastor-Fernández, C.; Baure, G.; Yu, T. F.; Widanage, W. D.; Marco, J. Battery Energy Storage System Modeling: Investigation of Intrinsic Cell-to-Cell Variations. J. Energy Storage 2019, 23, 19–28. 5. Su, L.; Wang, Z.; Ren, Y. A Novel Two-Steps Method for Estimation of the Capacity Imbalance among in-Pack Cells. J. Energy Storage 2019, 26. 6. Leithoff, R.; Frohlich, A.; Droder, K. Investigation of the Influence of Deposition Accuracy of Electrodes on the Electrochemical Properties of Lithium-Ion Batteries. Energ. Technol. 2020, 8, 8. 7. Hua, Y.; Zhou, S. D.; Cui, H. G.; Liu, X. H.; Zhang, C.; Xu, X. W.; et al. A Comprehensive Review on Inconsistency and Equalization Technology of Lithium-Ion Battery for Electric Vehicles. Int. J. Energy Res. 2020, 44, 11059–11087. 8. Li, J.; Greye, B.; Buchholz, M.; Danzer, M. A. Interval Method for an Efficient State of Charge and Capacity Estimation of Multicell Batteries. J. Energy Storage 2017, 13, 1–9. 9. Dong, G.; Wei, J. Determination of the Load Capability for a Lithium-Ion Battery Pack Using Two Time-Scale Filtering. J. Power Sources 2020, 480. 10. Plett, G. L. Efficient Battery Pack State Estimation Using Bar-Delta Filtering. In EVS24 International Battery, Hybrid and Fuel Cell Electric Vehicle Symposium; 2009; pp. 1–8. 11. Dai, H.; Wei, X.; Sun, Z.; Wang, J.; Gu, W. Online Cell SOC Estimation of Li-Ion Battery Packs Using a Dual Time-Scale Kalman Filtering for EV Applications. Appl. Energy 2012, 95, 227–237. 12. Zheng, Y.; Ouyang, M.; Lu, L.; Li, J.; Han, X.; Xu, L.; et al. Cell State-of-Charge Inconsistency Estimation for LiFePO4 Battery Pack in Hybrid Electric Vehicles Using Mean-Difference Model. Appl. Energy 2013, 111, 571–580. 13. Jiang, B.; Dai, H.; Wei, X.; Zhu, L.; Sun, Z. Online Reliable Peak Charge/Discharge Power Estimation of Series-Connected Lithium-Ion Battery Packs. Energies 2017, 10, 10030390.

584

Cell and Battery Design – Batteries | Cell Connections

14. Jiang, B.; Dai, H.; Wei, X. A Cell-to-Pack State Estimation Extension Method Based on a Multilayer Difference Model for Series-Connected Battery Packs. IEEE Trans. Transp. Electrif. 2022, 8, 2037–2049. 15. Zheng, Y.; Lu, L.; Han, X.; Li, J.; Ouyang, M. LiFePO4 Battery Pack Capacity Estimation for Electric Vehicles Based on Charging Cell Voltage Curve Transformation. J. Power Sources 2013, 226, 33–41. 16. Zhang, Y.; Wang, C.-Y.; Tang, X. Cycling Degradation of an Automotive LiFePO4 Lithium-Ion Battery. J. Power Sources 2011, 196, 1513–1520. 17. Zhu, J.; Knapp, M.; Sørensen, D. R.; Heere, M.; Darma, M. S. D.; Müller, M.; et al. Investigation of Capacity Fade for 18650-Type Lithium-Ion Batteries Cycled in Different State of Charge (SoC) Ranges. J. Power Sources 2021, 489, 229422. 18. Xie, W.; He, R.; Gao, X.; Li, X.; Wang, H.; Liu, X.; et al. Degradation Identification of LiNi0.8Co0.1Mn0.1O2/Graphite Lithium-Ion Batteries under Fast Charging Conditions. Electrochim. Acta 2021, 392, 138979. 19. Liaw, B. Y.; Roth, E. P.; Jungst, R. G.; Nagasubramanian, G.; Case, H. L.; Doughty, D. H. Correlation of Arrhenius Behaviors in Power and Capacity Fades with Cell Impedance and Heat Generation in Cylindrical Lithium-Ion Cells. J. Power Sources 2003, 119-121, 874–886. 20. Braco, E.; San Martin, I.; Berrueta, A.; Sanchis, P.; Ursua, A. Experimental Assessment of Cycling Ageing of Lithium-Ion Second-Life Batteries from Electric Vehicles. J. Energy Storage 2020, 32. 21. Wang, X.; Fang, Q.; Dai, H.; Chen, Q.; Wei, X. Investigation on Cell Performance and Inconsistency Evolution of Series and Parallel Lithium-Ion Battery Modules. Energy Technol. 2021, 9 (7), 2100072. 22. Wang, J.; Liu, P.; Hicks-Garner, J.; Sherman, E.; Soukiazian, S.; Verbrugge, M.; et al. Cycle-Life Model for Graphite-LiFePO4 Cells. J. Power Sources 2011, 196, 3942–3948. 23. Epding, B.; Rumberg, B.; Mense, M.; Jahnke, H.; Kwade, A. Aging-Optimized Fast Charging of Lithium Ion Cells Based on Three-Electrode Cell Measurements. Energy Technol. 2020, 8. 24. Chen, S.; Wei, X.; Dai, H. Liquid Cooling/Heating-Based Battery Thermal Management. In Handbook of Thermal Management Systems; 2023; pp. 255–293.

Cell and Battery Design – Batteries | Bipolar Plates and Batteries Mareike Partsch, Fraunhofer Institute for Ceramic Technologies and Systems, Dresden, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 3 3.1 3.2 3.3 3.4 4 References

Introduction Design principles of bipolar batteries Application of bipolar concept in different cell chemistries Bipolar lead-acid batteries Bipolar nickel-metal hydride batteries Bipolar lithium-ion and sodium-ion batteries Bipolar solid-state batteries Summary and outlook

586 586 588 588 589 589 591 593 593

Abstract Bipolar batteries are a special packaging concept for multi-cell battery systems. The series connection of individual cells that share the current collector eliminates the need for external connections and complex housings. This concept allows a significant increase in energy and power density at the system level. The bipolar design has been studied for a wide range of battery chemistries. With the introduction of solid electrolyte batteries, it now has a high potential for implementation. The practical realization poses challenges for the individual battery chemistries, which are examined in detail.

Glossary Battery active components Electrochemical active components of the battery cathode and anode. Battery capacity Battery capacity is the measure of the amount of energy a battery can store. It is measured in units of Ah or Wh. Battery cell balancing Cell balancing is the process of equalizing the voltages and SoC between the individual battery cells in a stack or module. The difference in cell voltages can be corrected by active or passive cell balancing. Battery energy density Energy density is the measure of how much energy a battery contains in proportion to its weight or volume. It is measured in units of Wh/kg or Wh/L. Battery inactive components All components of the battery which are electrochemically not active. Battery power density Power density is the measure of available power in proportion to battery weight or volume. It is measured in units of W/kg or W/L. Calendaring Calendering is the compression of dried electrodes resulting from the coating and drying of electrode slurry to reduce porosity and improve particle contacts. Joule heat Joule heating describes the conversion of electric energy into thermal energy by the resistance in an electric circuit. Lead-acid battery A lead-acid battery is a secondary cell containing lead and lead(IV) oxide in a sulfuric acid solution. The lead (IV) oxide oxidizes the lead, producing electrical current. Sealed lead-acid batteries are batteries in which the sulfuric acid is immobilized as a gel. Lithium-ion battery A lithium-ion battery is a type of rechargeable battery that comprises two lithium insertion materials. One is used as the positive electrode, and the other as the negative electrode. During charge and discharge, lithium ions shuttle between the electrodes. Nickel-metal hydride battery A nickel metal hydride battery is a type of rechargeable battery. The chemical reaction at the positive electrode uses nickel oxide hydroxide (NiOOH), the negative electrodes use a hydrogen absorbing alloy. Slot die coating Slot die coating is a coating technique for applying electrode slurry or extruded thin films to collector foils. The process was first developed for the industrial production of photographic paper. Sodium-ion-battery A sodium-ion battery works on the same principle as a lithium battery. However, the shuttle ion is sodium. Tape casting Tape casting (doctor blading, knife coating) is a casting process used to manufacture electrodes from a slurry that is cast in a thin layer onto a collector foil and then dried. The process is adapted from ceramic tape manufacturing. Thermal management The battery thermal management system is the device that is responsible for managing or dissipating the heat that is generated during the electrochemical processes that take place in the cells, allowing the battery to operate in a safe and efficient manner.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00247-4

585

586

Cell and Battery Design – Batteries | Bipolar Plates and Batteries

Key points

• • • •

1

Explain the idea of bipolar batteries Describe bipolar design principles Describe special issues for bipolar design in different battery chemistries Summary and outlook

Introduction

Electrochemical energy storage systems batteries or accumulators have evolved considerably in recent decades and are even key to the future transformation of energy systems. Due to the widespread use of batteries in various fields, a life without batteries seems impossible today. As a result, the demands on batteries are constantly increasing. This mainly concerns high energy and power densities, but also lifetime, safety and geometry. Electromobility and other mobile applications have been and still are the main drivers of this trend, as a high ratio of storage capacity to volume and mass volumetric and gravimetric energy density is essential. In conventional battery design, the volume or mass to capacity ratio is mainly determined by the system design, which means that the energy density is significantly reduced from the cell to the module to the system. To meet the growing demand for improved energy and power density of batteries at the cell, module and system levels, several optimization strategies have been introduced in recent years. The most important aspect for optimal storage capacity is to improve the cathode and anode material itself. In addition, the storage capacity of the active material should be maximized, e.g. by using the optimum potential range and avoiding side reactions. A second very important design principle for optimizing energy density is to reduce the so-called inactive components and materials that are not involved in energy storage. This improves the ratio of active to inactive components. In addition to the cell, the module and system design must be as compact as possible. To optimize power density, it is also essential to minimize ohmic losses in the cell and system configuration. To ensure this over the lifetime of the battery, the degradation processes and the formation of interfacial resistance must be understood and minimized. The bipolar battery architecture is a promising approach to address these challenges. In this design, multiple cells are connected in series, but not by connecting individual, separately packaged cells via external interconnects, as is the case in conventionally designed module configurations. A bipolar design eliminates packaging and external interconnects; instead, cells are connected directly through the substrate of each electrode - the bipolar substate - eliminating the need for external wiring. This method provides an efficient way to stack and interconnect cells, optimizing factors such as electrical resistance, volume, weight and cost. The concept of bipolar battery stacking has its roots in the Voltaic pile from the early days of electrochemical science and has been studied for over a century for various battery chemistries as it considerably simplifies the battery stack configuration. In essence, a bipolar battery is a highly integrated series connection of individual cells in a single package. Serial cell integration results in increased total voltage, which in an n-cell configuration is n times the voltage of individual cells with the same capacity as a single cell. As a result, the overall efficiency and performance of the battery system is increased. This is possible by reducing the amount of inactive components and mitigating ohmic losses and voltage drops, which is particularly beneficial in applications requiring high charge and discharge performance. The bipolar configuration principle has been implemented in several different ways for various battery chemistries. What is common to all bipolar batteries, however, is that the high degree of integration allows the elimination of components, resulting in an increase in volumetric and gravimetric energy and power density compared to the classic series connection. This makes the concept attractive and generates a high level of interest in suitable solutions. However, there are only a few examples of bipolar batteries in commercial use, which indicates that there are also challenges in realizing the specific bipolar configuration. In addition to rechargeable accumulators or batteries secondary galvanic elements in which the storage material is contained within the cell and must be reversibly charged and discharged, the bipolar design has been used with great success in fuel cells and flow batteries. These tertiary galvanic elements receive their fuel externally. They rely on electrochemical reactions to convert chemical energy provided by an external liquid or gaseous reactant into electrical energy. Therefore, the bipolar substrate has the additional function of guiding the reactant through integrated flow fields to the porous electrodes where the electrochemical conversion takes place. This is covered elsewhere. In this chapter, the focus will be on battery or accumulator chemistries that do not have an external supply of reactant.

2

Design principles of bipolar batteries

In monopolar cells, the electrodes are connected in parallel, which means that the capacity is added, but the voltage is the same as that of the single cell. In contrast, the bipolar battery is a highly integrated series connection of individual cells, see Fig. 1. This means that the total output voltage of the bipolar pack is equal to the sum of the individual cell voltages. The optimum number of cells per cell stack is based on the total stack voltage to be achieved and the increased production and quality

Cell and Battery Design – Batteries | Bipolar Plates and Batteries

587

Fig. 1 Design of bipolar battery stack. Adapted from Döring, H.; Clasen, H.; Zweynert, M.; Garche, J.; Jörissen, L. Materials for Bipolar Lead-Acid Batteries. In: New Promising Electrochemical Systems for Rechargeable Batteries; Barsukov, V., Beck, F. (eds.); NATO ASI Series. pp. 3–13. Vol. 6, 1996. Springer, https://doi.org/10. 1007/978-94-009-1643-2_1.

requirements resulting from a higher number of cells in the stack. Since the output current and thus the capacity of the bipolar battery is the same as that of a single cell, maximum cell area and cathode loading are desired, and the weakest cell in the pack is the determining factor. Alternatively, bipolar stacks can be connected in parallel to increase capacity, but at the expense of energy density. The extremely high integration density of the bipolar battery design is achieved by bringing the individual cells, consisting of cathode and anode and the electrolyte separator, very close together. To achieve this, the substrates of the cathode and adjacent anode are either placed in direct contact with each other, or both electrodes are placed on either side of a common bipolar substrate. Thus, adjacent cells share a bipolar substrate so that the anode of one cell is placed on one side and the cathode of the next cell is placed on the other side of the substrate the bipolar electrode is both the anode and cathode of two adjacent cells. This approach creates a series connection of planar cells that requires no additional external interconnects and no single cell housing. Planar stacking results in homogeneous current distribution and flow perpendicular to the cell surface over the entire electrode area. This eliminates localized high current densities, resulting in a significant reduction in ohmic resistance and internal Joule heat generation, which is particularly beneficial during high-power charging and discharging. As a result, the thermal management of bipolar batteries can be greatly simplified, leading to an additional reduction of inactive components at the system level. The bipolar battery design can be realized from any cell chemistry, helping to reduce the gap between theoretical and practical energy density of battery storage devices. In practice, however, the transfer of cell concepts previously developed for monopolar configuration to a bipolar structure leads to different challenges in cell design as well as in manufacturing and engineering for the individual battery chemistries. In particular, the bipolar substrate must have the required electrochemical and chemical stability compatible with both the anode and cathode. In addition, the bipolar substrate provides the intercell connection and isolates the electrolyte within the cells. As a result, it must be highly electron conductive and isolate ion transport. Since the substrate also serves as the cell housing, it must be stable to the electrolyte and mechanically robust to the cell seal. A second component that usually requires a complete rethinking in a bipolar battery is the sealing of the individual cells. In monopolar single cells, this is typically accomplished with massive cases that introduce a large amount of additional inactive components into a battery system. The bipolar design eliminates these heavy and massive components by packing only the bipolar stack in a separate housing. As a result, the individual cells are formed by compact seals between the individual layers of the bipolar substrates, which prevent electrolyte loss while also preventing shorts within the individual cells. The stability of the seals, e.g. in the presence of gaseous degradation products, is crucial for the durability of the bipolar cell stack. Therefore, liquid electrolyte fill ports or gaseous degradation product valves are mandatory for selected battery chemistries. With solid electrolytes, cell sealing is much simpler than with liquid electrolytes. This certainly explains why the bipolar concept is of great interest in the context of solid-state batteries. In order to minimize the difficulties caused by the bipolar configuration with regard to the corrosion stability of the bipolar substrate and the seals, so-called “pseudo” bipolar designs are described and applied,1 see Fig. 2. These configurations were developed primarily to avoid corrosion effects on the bipolar substrate in lead-acid batteries. The “pseudo” bipolar design can be realized by a side-by-side configuration where the connection area is limited to the edge of the electrodes. A similar effect can be achieved by a wrap-around-configuration, where side-by-side cells are wrapped around an inert cell case, or by a wire-throughwall-design with multiple connections. Due to the specific cell connection of “pseudo” bipolar cells, ohmic resistance effects are more relevant than in “classic” bipolar designs. Transitioning from a monopolar to a bipolar battery configuration requires significant adjustments to the production line. In the manufacturing of a bipolar battery, ensuring the accurate production of bipolar substrates, electrodes, and sealant materials is paramount. It should be noted, however, that a bipolar battery assembly line can result in an increase in capital costs due to process

588

Cell and Battery Design – Batteries | Bipolar Plates and Batteries

Fig. 2 ‘Pseudo’ bipolar battery configurations; left: side-by-side; right: wrap-around resp. wire-through- wall design. Adapted from Döring, H.; Clasen, H.; Zweynert, M.; Garche, J.; Jörissen, L. Materials for Bipolar Lead-Acid Batteries. In: New Promising Electrochemical Systems for Rechargeable Batteries; Barsukov, V., Beck, F. (eds.); NATO ASI Series. pp. 3–13. Vol. 6, 1996. Springer, https://doi.org/10.1007/978-94-009-1643-2_1.

adaptations. For example, in a bipolar design where the anode and cathode are coated on opposite sides of the same current collector, the sequencing of the coating process adds processing time and reduces throughput because the cathode or anode is applied after the counterpart has dried. In addition, large electrode stacks with multiple individual cells require very precise and reliable manufacturing processes, as the weakest cell in a series connection determines the overall performance of the system. This applies in particular to the manufacture of the bipolar electrode and the incorporation of the electrolyte and seal, which require reliable and economical process engineering solutions. High manufacturing accuracy and low variation are of utmost importance. In the event of a failed cell, the entire stack would have to be scrapped, significantly impacting production yield and operating costs. During operation of the bipolar battery, the homogeneous current distribution and the avoidance of local hotspots are expected to result in significantly less degradation than in monopolar cells, where inhomogeneities in current density cannot be avoided. However, as described above, if a cell fails or malfunctions, the entire cell stack must be replaced because it is not possible to replace individual cells. This results in high costs for repair, removal or recycling. To be able to respond appropriately to failures and degradation, much effort is often required to diagnose and manage the bipolar multi-cell module. In particular, avoiding overcharging or discharging of individual cells is a critical issue when the deviation of cell capacity within the stack becomes significant after a certain lifetime. The above issues result in a trade-off between optimized energy and power density on the one hand, and manufacturing and replacement costs on the other. To increase energy density, cell stacks with the largest possible capacity and a high number of individual cells are desirable. However, this makes the structure of the cell stack more complex, which increases manufacturing and repair costs. These considerations are critical to the practical applicability of bipolar batteries and must be solved specifically for each battery cell chemistry.

3

Application of bipolar concept in different cell chemistries

Despite its many advantages, the bipolar design presents particular challenges, especially in terms of implementation. These challenges vary considerably depending on the electrochemical couple chosen and require customized solutions. In the literature, bipolar studies have been reported on a variety of battery chemistries, including Nickel-Cadmium, Aluminum, Manganese Dioxide-Zinc, and more. In this section, examples will be given to illustrate the approaches that have been explored in this domain.

3.1

Bipolar lead-acid batteries

Lead-acid batteries2,3 have found their main application as starter batteries in automobiles, where their high current capacity is exploited. They are also used as traction batteries, particularly in industrial trucks, and as uninterruptible power supplies for stationary applications. The first reports of bipolar lead-acid batteries date back more than 100 years. A patent application by Paget in 18994 describes the idea of bipolar electrodes for lead-acid batteries, and Kapitza and Heath in 19235,6 published extensive work on developing bipolar lead-acid batteries. The goal was to optimize the rate capability of this established battery chemistry to provide energy storage for mobile applications or short, high discharge pulses. The introduction of the bipolar concept in lead-acid batteries has therefore led to a significant improvement in performance. LaFollette et al.7 confirmed this by comparing selected aqueous bipolar battery chemistries using a simple model. They found that over the range of current densities studied, the lead-acid battery with thin planar electrodes achieved optimum performance.

Cell and Battery Design – Batteries | Bipolar Plates and Batteries

589

However, this performance improvement was accompanied by the observation of increased corrosion in the battery electrodes, resulting in higher self-discharge rates. At the same time, challenges such as electrolyte leakage and mixing emerged, prompting the development of advanced electrolyte technologies. These innovations involved immobilizing the acid in gels or absorbent glass mats. However, electrolytes posed significant sealing problems. As a result, the effectiveness of seals and vents became critical to overall cell performance and stability. In recent decades, the focus has been on the development of stable bipolar substrates and seals and vents. The first breakthrough was achieved in the 1960s, when specially adapted seals and valves were developed so that a separation of the individual cells could be achieved that was stable in aqueous sulfuric acid.8 The main challenge for lead-acid bipolar substrates is that they must be stable in sulfuric acid as well as in the battery potential window to avoid corrosion and high oxygen and hydrogen overpotentials.9 Therefore, conductive materials that are less susceptible to corrosion have been under investigation for performance improvement. Theoretically, lead and lead alloys, metals with a dense lead protective layer, ceramic materials, and plastic matrix-filled materials can be used. In practice, however, the lead alloys have insufficient corrosion resistance, leading to holes in the bipolar substrate and short circuits. The other materials are corrosion resistant but do not provide good electrical contact between the active mass and the bipolar substrate. In order to combine the advantages of both groups of materials, various methods have been described in the literature. One such method is to use the active mass as the contact material and to use corrosion resistant components to prevent electrolytic connection and short-circuiting between the cells. Alternatively, various “pseudo” bipolar batteries have been developed that address the systematic problem of electrochemical and chemical stability of the bipolar substrate. The result is an increase in internal resistance, which is still much lower than that of the conventional connection. To date, there are several companies claiming to have commercial or pilot production of bipolar lead-acid batteries.

3.2

Bipolar nickel-metal hydride batteries

Nickel-metal hydride batteries10 are rechargeable batteries consisting of a nickel-oxide hydroxide cathode, a metal hydride anode and an alkaline electrolyte. These batteries have proven their versatility and are widely used in applications such as portable electronics, hybrid vehicles and power tools. Compared to other advanced battery systems, the nickel-metal hydride battery system offers high specific energy and power while operating with an aqueous electrolyte at ambient temperatures. The nickel-metal hydride couple is based on a relatively simple ion transfer reaction. In particular, since this reaction does not involve the electrolyte in the overall cell reaction, it provides an opportunity to realize a bipolar design for battery stacks with a limited amount of starved electrolyte. Sealed nickel-metal hydride batteries in a bipolar cell design have attracted particular interest for energy storage applications due to the expected increase in specific performance. Benefits expected from bipolar nickel-metal hydride batteries include a reduction in internal resistance, improved power density and the potential for more compact designs. Since the electrolyte in nickel-metal hydride chemistry does not directly participate in the electrochemical reaction and is therefore not subject to significant aging or decomposition, its amount can be reduced to fill the pores of the components and enable the ionic transition. Because excess electrolyte can be avoided, the sealing of the individual cells is much less complicated. However, the electrochemical reaction is associated with a phase equilibrium between gaseous hydrogen and hydrogen adsorbed on the bulk of the electrode.11 During charging, hydrogen is generated on the surface of the negative electrode and stored in the hydride alloy structure of the electrode.12 During discharge, the hydrogen stored in the hydride alloy is electrochemically reacted and the nickel electrode is reduced. In addition, at the end of the charge and during overcharge, oxygen is generated on the surface of the nickel electrode and recombined on the surface of the hydride electrode. To achieve stable operation, the overcharge rate is limited to the recombination capabilities of the design to ensure that there is no excessive pressure build-up in the cell that could lead to cell failure. Specific challenges in the design of bipolar nickel-metal hydride batteries arise from the need for special precautions to avoid electrolytic short circuits between cells across the bipolar substrates, and the sensitivity of the bipolar substrates on their positive side to corrosion reactions that increase the electrical resistance. In addition, the need to apply anode and cathode materials to a bipolar substrate in a single manufacturing step has been described as a challenge. As a result, “pseudo” bipolar configurations are preferred. The development of bipolar nickel-metal hydride batteries in the 1980s was driven primarily by mobile and aerospace applications. For these applications in particular, solutions with high energy and power density were sought. However, bipolar nickel-metal hydride batteries were not successful in this context and became less common and often research-oriented as lithium-ion batteries gained popularity in the 1990s. As a result of their significantly improved characteristics, lithium-ion batteries became the preferred choice, overtaking technologies such as nickel-metal hydride batteries in many applications.

3.3

Bipolar lithium-ion and sodium-ion batteries

Since the introduction of the first practical lithium-ion battery by Sony Corporation in 1991, extensive research has been conducted to improve both the material base and the production process. Despite these efforts, the production of conventional lithium-ion batteries remains a very complex and highly controlled process.

590

Cell and Battery Design – Batteries | Bipolar Plates and Batteries

The electrodes of lithium-ion batteries are currently designed as monopolar electrodes, where a thin metallic current collector foil serves as a support for the actual storage materials. Typically, aluminum is used for the cathode and copper for the anode because aluminum forms alloys with lithium at low potentials, making it unsuitable as a current collector for graphite electrodes. Electrode materials are typically applied as a slurry - a suspension of storage material particles, a polymer binder material, carbon components such as carbon black to enhance electrical conductivity, and an aqueous or organic solvent. For the monopolar electrode, the same electrode material is sequentially or simultaneously applied to both sides of the current collector using processes such as tape casting or slot die coating. The active materials of the cathode are typically lithium metal oxides with tailored powder qualities, while the anode is typically graphite. The coating process results in thin layers of electrode materials on the metal foils, which are then dried to remove solvents from the coating and compacted by calendering, leaving a dry electrode. The cathode, separator and anode are then cut into pieces and rolled or stacked together to form a cell structure commonly known as a “jellyroll.” To prevent direct contact and potential short circuits, the separator, typically made of a porous polymer material, is placed between the cathode and anode. This jellyroll is carefully inserted into the cell housing. The electrolyte, which contains organic solvents and lithium salt, is filled into the cell to facilitate the movement of ions between the cathode and anode. The battery cell undergoes a formation process that includes multiple charge and discharge cycles, a critical step in stabilizing the battery’s performance and capacity. It is important to recognize that the specific details of the manufacturing process can vary between different manufacturers and battery chemistries. This underscores the importance of the complexity of lithium-ion battery manufacturing, where the smallest variations in the processes can affect the overall performance and characteristics of the final product. Lithium-ion batteries have the potential for exceptionally high energy and power density due to the high capacity of the storage materials used. However, the storage density of the electrode/separator assembly within the cell is significantly reduced when transferred to the module and system level, limiting the exploitation of the intrinsic material properties. Commercial lithium cells offer more than 650 Wh/l. Due to the necessary integration into the battery system, the volumetric energy density is significantly lower as safety components, connectors, housing and thermal management are added. The actual battery cells make up only about 40–60% of the system volume, with the remaining space taken up by assembly and interconnect technology, battery management, and cooling. Mathematically, a system-level energy density of at least 450 Wh/l is a prerequisite for achieving relevant ranges in vehicle operation. The bipolar design can reduce energy density reduction by eliminating individually housed cells and the need for individual cell current collectors and electrical cell connections. The design of bipolar lithium-ion batteries faces significant engineering challenges, with a particular focus on the development of electrochemically stable bipolar electrodes, as lithium-ion batteries are characterized by a higher cell potential compared to other chemistries. This means that the half-cell potentials of the two electrodes are ideally far apart. Pairing individual cathode and anode materials results in cell voltages in the range of 1.5 to 4.2 V. In the bipolar electrode, the half-cell potentials of these electrode materials act on both sides, and the current collector material must be stable over this entire range. In addition, the assembly of these electrodes into a perfectly sealed battery stack, which improves the overall safety and performance of the lithium-ion battery, presents a different set of challenges. Efforts to apply the bipolar principle to lithium batteries have used conventional electrode materials, such as graphite on the anode side and metal oxides for the cathode. Since graphite requires copper as the current collector on the anode side, while aluminum can be used on the cathode side, the bipolar current collector is formed as a copper/aluminum composite. In the meantime, there are commercially available double-layer foils made of aluminum and copper, which can be processed as a metal composite foil. The graphite anode on the copper side and the cathode materials on the aluminum side can be successively applied to these metal composite foils. It is important to note that both the cathode and anode slurries have different properties: Graphite slurries are usually aqueous in nature and must be applied first. Cathode slurries may be based on organic solvents due to the moisture sensitivity of nickel-containing cathode powders. In this case, they are applied in the second step to eliminate potential interactions with the aqueous slurry. Since the combination of copper and aluminum as a metal composite foil still poses significant engineering and process challenges, various approaches have been explored to overcome this difficulty. The focus has been on exploring alternative current collector foil materials or anode active materials. The use of graphitic carbon compounds operating at 0.1 V (vs. Li/Li+) becomes impractical when reducing cell complexity by designing a bipolar cell with only aluminum current collectors. This is due to the effect that aluminum reacts with lithium at 0.4 V before lithium can intercalate into the graphitic carbon structures. Lithium titanate (Li4Ti5O12) with an operating voltage of 1.55 V vs. Li/Li+ has been used as an alternative to graphite active material. However, this reduces the achievable cell potential and consequently the energy density of the cell. As a result, the potential advantages of the bipolar concept would be significantly limited. An operating potential range of 0.5–1.0 V for the negative electrodes would ensure optimal performance and mitigate issues related to the reactivity of aluminum with lithium and the resulting reduction in energy density. Therefore, layered LiMS2 (M ¼ Ti, V; with Ti3+/Ti2+ and V3+/V2+ redox couples) and intercalated metal-organic frameworks (iMOF) (e.g. 2,6-naphthalene dicarboxylate dilithium) have been investigated.13 Alternatively, polymer-carbon composite current collector foils were also investigated.14 They demonstrated voltage stability up to 5 V vs. Li/Li+ and negligible Li intercalation losses in the carbon primer. The advantage of nearly 50% lower raw material density of the polymer-carbon composite current collector foil compared to metal aluminum foil, along with expected improvements in collector thickness reduction and cost savings due to a scaled industrial manufacturing approach, could provide an opportunity to

Cell and Battery Design – Batteries | Bipolar Plates and Batteries

591

significantly reduce the mass loading of the collector in the battery cell. The polymer-carbon composite current collector foil appears to be particularly advantageous for bipolar battery architectures, as it can form a very stable edge seal when combined with polymer sealing materials. For the production of bipolar electrodes, wet deposition processes are preferred, resulting in elaborated process sequences. In addition, drying conditions and electrode densification parameters must be matched to the requirements of the cathode and anode materials. New processes that increase the solids content of electrode slurries or enable dry coating and printing processes could bring significant advantages in the future. It should not be forgotten that the reproducibility and quality of the electrode coating is particularly important in bipolar design, since the weakest cell in the bipolar series determines the performance of the entire stack. Since the quality requirements are much higher than for monopolar cells due to the large electrode surfaces and the number of cells in the stack, significant challenges are expected. The bipolar stack in lithium-ion batteries is formed by stacking individual bipolar electrodes and separators. To form electrochemically separated cells, the electrodes are joined at the edges with electrically and ionically non-conductive sealing material to act as intercell separators. This manufacturing step presents several technical challenges that must be addressed in order for it to be successfully implemented. In particular, precise alignment during stack fabrication and application of seals or other insulating materials are essential to avoid potential sealing problems at the intercell connections. Therefore, the use of polymeric sealants and adhesives is described. The sealing material must be of high chemical and mechanical quality to ensure cell sealing over the entire lifetime. In particular, high stability against the electrolytes used, which are typically based on organic solvents, must be achieved to prevent electrolyte leakage and short circuits resulting in capacity degradation. It should be noted that dissolution and degradation of polymeric sealing materials due to reactions with the liquid electrolyte can result in organic contamination of the electrolytes. In addition, the application of sealants results in electrochemically inactive edge regions, reducing the electrochemically active cell area. As a result, the increased energy density of lithium-ion bipolar batteries is only effective in large-area stacks when the ratio of active area to edge geometry is significant.15 The system-level integration of the lithium-ion bipolar stack includes other components that ensure safe and reliable operation. For example, a cell balancing system must be implemented and integrated to compensate for potential variations in the capacity of individual cells in the bipolar stack due to factors such as manufacturing influences or differential aging of the cells connected in series. In addition, bipolar battery systems are expected to require less or no thermal management because the battery system generates relatively less heat due to a reduced number of heating points, such as metallic interconnects or junctions, which are the source of Joule heating. In addition, the special geometry allows a bipolar battery to withstand higher discharge currents and is expected to generate less heat than a monopolar configuration.16 In recent years, the lithium-ion bipolar battery has also become of interest to the industry, but the commercial availability of lithium-ion bipolar batteries is still limited due to technical challenges. While there have been research and development efforts to explore bipolar battery designs for potential advantages such as simplified construction and reduced weight, widespread commercial adoption has not yet been realized. Meanwhile, sodium batteries are considered promising as the next generation of future battery storage.17 Although the energy density is expected to be lower, the availability of sodium is significantly better than that of lithium. It is also expected that existing manufacturing and cell assembly technologies can be used and that lithium can be relatively easily replaced by sodium. Therefore, there are interesting investigations to transfer the bipolar concept to sodium, as some disadvantages of sodium chemistry, such as reduced energy density and cell voltage, could be compensated by bipolar concepts compared to lithium monopolar cells. A major advantage is that sodium, unlike lithium, does not alloy with aluminum at low potentials and aluminum remains stable over a wide potential range of 0–5 V vs. Na/Na+. Compared to materials such as titanium and stainless steel, which are also considered potential-conducting materials, aluminum has been well tested and proven to be highly processable in cell manufacturing. Therefore, aluminum can be used as a bipolar substrate for both the anode and cathode, greatly simplifying the fabrication of bipolar electrodes and the assembly of the cell stack.18 Regarding the use of liquid electrolytes, similar considerations apply to sodium batteries as described above for lithium. Sodium batteries also use electrolytes based on organic solvents, which requires the use of seals made of highly chemically resistant materials to prevent electrolyte leakage and mixing problems.

3.4

Bipolar solid-state batteries

The transition from liquid electrolytes, which typically rely on organic solvents, to solid electrolytes with minimal or reduced flammability has become a key strategy in battery development in recent years. This transition not only enhances the safety of batteries, even in challenging or abusive conditions, but also represents a significant improvement in the safety profile of lithium and sodium batteries in particular. The challenges associated with liquid electrolytes, their filling and the sealing of cell assemblies also make solid electrolytes very attractive for the construction of bipolar battery stacks. Therefore, in recent years, there has been an increasing focus on solid electrolytes, which are ion conductors of a polymer, sulfide, oxide or hybrid nature, particularly in the context of bipolar lithium batteries and, to some extent, sodium batteries and other battery chemistries. Another notable advantage of solid electrolytes is their potential to facilitate the use of high-capacity metal anodes, particularly in lithium batteries, as lithium metal is considered the ultimate anode material for solid-state lithium batteries due to its high energy

592

Cell and Battery Design – Batteries | Bipolar Plates and Batteries

density. However, critical technical issues remain, such as uncontrollable lithium dendrite growth during cycling and high resistance at the solid electrolyte-lithium metal interface. Recently, several approaches have been developed to engineer or stabilize the solid electrolyte/lithium metal interfaces. As such, solid electrolytes play a central role in preventing dendrite formation, thus contributing to the overall stability and safety of metal-anode batteries. Therefore, with the advent of solid electrolytes, the concept of bipolar design has received renewed attention, especially in the research community, as it can be more easily implemented with solid electrolytes because solid electrolytes are less mobile than liquid electrolytes. The risk of electrolyte loss and mixing is minimized, and the production of multilayer cells becomes much simpler. This is accompanied by the low ohmic losses of the bipolar cell structure, the reduced thermal stress and the adapted requirements for the thermal management, which are particularly attractive for future generations of batteries. The effect of optimized thermal management of a bipolar stack was demonstrated in a study by Pang et al.16 In this study, the thermal behavior of a stacked bipolar arrangement of lithium solid-state cells in series was compared to a stack of monopolar cells made from the same materials. The main results provide a representative insight into bipolar solid-state batteries. The authors describe that compared to the monopolar design, a more uniform current distribution over the entire surface area can be achieved because the bipolar stack efficiently utilizes the entire surface area to transfer current from one unit cell to the adjacent unit cell. On the other hand, the stack capacity is limited due to the serial connection of each unit cell in the bipolar stack, so a high-load positive electrode configuration with an optimized electrode design is essential to increase the stack capacity. By considering that the volumetric and gravimetric energy densities of the bipolar stack exceed those of the parallel stack. The negligible heat generation in the bipolar stack contrasts significantly with the higher heat generation due to Joule heating of the tabs observed in the parallel stack. This highlights the need for reduced thermal management in the case of bipolar stacks, which further improves the volumetric and gravimetric density of the bipolar battery at the system level. In addition, ensuring the reliable operation of each unit cell is critical, as the failure of one unit cell can lead to the failure of the entire bipolar stack. The use of solid electrolytes instead of liquid electrolytes provides the opportunity to safely manufacture large volume cells, but the solid-state nature also introduces new requirements for bipolar battery manufacturing. While liquid electrolytes are typically introduced in the final step of cell manufacturing, diffusing into the pores and forming solid/liquid interfaces, solid electrolytes must be incorporated into the component manufacturing process from the outset to form stable, reliable, and materially coherent solid/solid interfaces. This is critical because most solid electrolytes, with the exception of some sulfide electrolytes, have significantly lower ionic conductivity than liquid electrolytes, and the quality of the electrode/electrolyte interface is essential for charge transport at the interface and the generation of interfacial resistance. In addition, the electrochemical window of most solid electrolytes is more limited than that of established liquid electrolytes, requiring the introduction of protective layers at the interface between the electrode material and the electrolyte. This, combined with the different process requirements resulting from the different electrolyte materials, potentially leads to highly complex multilayer manufacturing processes. These manufacturing processes must be adapted to the requirements of fabricating bipolar cell architectures. For example, polymer electrolytes and interlayers can be polymerized, infiltrated or coated in situ.19,20 Thermoplastic polymers can be incorporated directly into the composite electrode. Sulfide electrolytes can be coated together with the electrode slurry by wet coating or dry coating by compaction. Oxide electrolytes require a sintering process at temperatures of several hundred degrees after fabrication of the individual components, which densifies the structure and creates ion-conducting paths. From the perspective of cell design and achieving optimal cell performance, it is essential to optimize the whole manufacturing sequence in that way to maximize the capacity of the electrodes, which is correlated with the coating thickness, while minimizing the thickness of the electrolyte separator, all while ensuring high mechanical stability. Therefore coating, like doctor blading and slot die coating, and printing processes21 involving slurries and inks are established and shown to be scalable methods. Bipolar batteries can be manufactured by applying electrolyte and/or active material slurries with specific rheological properties onto supporting components. Since especially printable slurries can easily adapt to the substrate’s shape, it is possible to realize battery architectures with various shapes, form factors and functionalities. The stacking of individual components and cells into a bipolar stack can be achieved either by sequential coating or by laminating free-standing films. Both require high precision. The sequential coating process is performed on a mechanical support, which can be the current collector, the thick composite cathode, or the electrolyte separator. The remaining functional coatings are then sequentially applied to this component. In the lamination process, the free-standing components are successively laminated together to form the individual cells and subsequently the cell stack. This requires components with adequate mechanical stability, which typically leads to over-dimensioning in the case of separators. For the future scale-up of the production of bipolar solid-state batteries, Jung et al.22 conclude that the necessary development steps should be investigated in the future: Advanced solid electrolyte materials with high Li + conductivity are critical, accompanied by composite electrodes with continuous networks for both Li + and electron conduction. In addition, conformal interfaces between the solid electrolyte and electrode active materials based on interfacial engineering are required. These advances must be complemented by current collector materials that remain stable under highly reductive and oxidative conditions. In connection with the stacking or lamination of a large number of cells within a bipolar battery, advanced stack design and a manufacturing process with precise stacking are essential for the production of solid-state bipolar batteries without internal short circuits. In order to improve both speed and accuracy, automation of the stacking or lamination process needs to be further developed.

Cell and Battery Design – Batteries | Bipolar Plates and Batteries

593

The primary focus should be on minimizing the performance variation of individual cells in a bipolar stack, as this is essential for achieving long cycle life. While individual cells in a stack may initially exhibit similar performance, over time their charge-discharge behavior may diverge, resulting in a reduction in overall capacity during continuous operation. In summary, the introduction of bipolar solid-state electrolyte batteries is a promising step forward in the evolution of battery systems. In fact, some companies are claiming to have mastered the bipolar design of the solid electrolyte cell.

4

Summary and outlook

Bipolar batteries have been considered for various electrochemical applications since the invention of Volta’s pile. The ability of the bipolar architecture to significantly increase energy and power density by connecting multiple cells in series, resulting in shorter electron paths and lower internal resistance, has continued to attract research and development interest. The sharing of the current collector, called the bipolar substrate, between the anode of one cell and the cathode of the next cell is the core of the bipolar design. This arrangement poses a number of challenges in terms of cell and system design, as well as the stability of the individual components, which must be solved on a case-by-case basis for each battery chemistry. The individual cells in the stack must be reliably separated electrochemically to prevent electrolyte leakage and mixing, which may lead to cell capacity degradation. The bipolar substrate requires materials negligible ionic conductivity compared to electronic conductivity, as well as stability against oxidative or reductive stress and corrosion from electrolytes. This allows for the use of both metallic bipolar substrates as well as carbon polymer composite materials. Effective manufacturing processes are required to realize bipolar multilayer structures with high precision and reliability. These processes differ somewhat from those for monopolar cells due to the overall higher complexity of the manufacturing sequences. In addition, both manufacturing and operation must avoid deviations in the performance of individual cells, since the weakest cell determines the performance of the bipolar stack. Despite the potentially significant benefits of bipolar batteries, there are few examples of successful commercialization to date. With the use of solid electrolytes, the prospects for producing stable cell stacks in a bipolar design are better than ever. Until then, bipolar rechargeable batteries remain the Holy Grail of battery development.

References 1. Döring, H.; Clasen, H.; Zweynert, M.; Garche, J.; Jörissen, L. Materials for Bipolar Lead-Acid Batteries. In New Promising Electrochemical Systems for Rechargeable Batteries; Barsukov, V., Beck, F., Eds.; NATO ASI Series, Vol. 6; Springer, 1996; pp. 3–13. https://doi.org/10.1007/978-94-009-1643-2_1. 2. Garche, J. Advanced Battery Systems - the End of the Lead-Acid Battery? Phys. Chem. Chem. Phys. 2001, 3, 356–E367. https://doi.org/10.1039/b005451h. 3. Bullock, K. R. Progress and Challenges in Bipolar Lead-Acid Battery Development. J. Electrochem. Soc. 1995, 142 (5), 1726. https://doi.org/10.1149/1.2048646. 4. Paget, L. Storage Battery and Method of Preparing Electrodes Therefor; US 6 27 009, 1899. 5. Kapitza, P. L.; Heath, H. F. Electric storage apparatus; US 1 656 203 A, 1923. 6. Kapitza, P. L. A Method of Producing Strong Magnetic Fields. Proc. Royal Soc. London Series A. 1924, 105, 691–710. https://doi.org/10.1098/rspa.1924.0048. 7. LaFollette, M.; Bennion, D. N. Design Fundamentals of High Power Density, Pulsed Discharge, Lead Acid Batteries. J. Electrochem. Soc. 1990, 137, 3693. 8. Biddick, R. E.; Nelson, R. D. Lead-Acid Bipolar Battery for Multisecond Pulse Discharge. IECEC 1968, 1, 47–51. 9. Kao, W.-H. Substrate Materials for Bipolar Lead/Acid Batteries. J. Power Sources 1998, 70 (1), 8–15. https://doi.org/10.1016/S0378-7753(97)02605-0. 10. Wiesener, K.; Ohms, D.; Benczúr-Ürmössy, G.; Berthold, M.; Haschka, F. High PowerMetal Hydride Bipolar Battery. J. Power Sources 1999, 84, 248–258. https://doi.org/ 10.1016/s0378-7753(99)00325-0. 11. Ohms, D.; Kohlhase, M.; Benczúr-Ürmössy, G.; Wiesener, K.; Harmel, J. High Performance Nickel-Metal Hydride Battery in Bipolar Stack Design. J. Power Sources 2002, 105 (2), 120–126. https://doi.org/10.1016/S0378-7753(01)00929-6. 12. Klein, M. G.; Eskra, M.; Plivelich, R.; Ralston, P. Bipolar Nickel Metal Hydride Battery. https://haszstudios.com/electro/products/technicalpapers/BipolarNickel.pdf. accessed 2023-12-13. 13. Ogihara, N.; Yasuda, T.; Kishida, Y.; Ohsuna, T.; Miyamoto, K.; Ohba, N. Organic Dicarboxylate Negative Electrode Materials with Remarkably Small Strain for High-Voltage Bipolar Batteries. Angew. Chem. Int. Ed. 2014, 53 (43), 11467–11472. https://doi.org/10.1002/anie.201405139. 14. Fritsch, M.; Coeler, M.; Kunz, K.; Krause, B.; Marcinkowski, P.; Pötschke, P.; Wolter, M.; Michaelis, A. Lightweight Polymer-Carbon Composite Current Collector for Lithium-Ion Batteries. Batteries 2020, 6, 60. https://doi.org/10.3390/batteries6040060. 15. Wolter, M.; Schilm, J.; Nikolowski, K.; Kusnezoff, M.; Partsch, U.; Freytag, C. Construction or Manufacture of Accumulators Having Only Flat Construction Elements, i.e. Flat Positive Electrodes, Flat Negative Electrodes and Flat Separators; EP 17734316.7A, 2017. 16. Pang, M.-C.; Wei, Y.; Wang, H.; Marinescu, M.; Yan, Y.; Offer, G. J. Large-Format Bipolar and Parallel Solid-State Lithium-Metal Cell Stacks: A Thermally Coupled Model-Based Comparative Study. J. Electrochem. Soc. 2021, 167 (16), 160555. https://doi.org/10.1149/1945-7111/abd493. 17. Sundaram, P. M.; Soni, C. B.; Vineeth, S. K.; Sanjaykumar, C.; Kumar, V. Reviving Bipolar Construction to Design and Develop High-Energy Sodium-Ion Batteries. J. Energy Storage 2023, 63, 107139. https://doi.org/10.1016/j.est.2023.107139. 18. Rudola, A.; Wright, C. J.; Barker, J. Explorations into the Viability of High Voltage Bipolar Na-Ion Cells Using Liquid Electrolytes. Front. Energy Res. 2022, 10, 852630. https://doi. org/10.3389/fenrg.2022.852630. 19. Röchow, E. T.; Coeler, M.; Pospiech, D.; Kobsch, O.; Mechtaeva, E.; Vogel, R.; Voit, B.; Nikolowski, K.; Wolter, M. In Situ Preparation of Crosslinked Polymer Electrolytes for Lithium Ion Batteries: A Comparison of Monomer Systems. Polymers 2020, 12 (8), 1707. https://doi.org/10.3390/polym12081707. 20. Coeler, M.; van Laack, V.; Langer, F.; Potthoff, A.; Höhn, S.; Reuber, S.; Koscheck, K.; Wolter, M. Infiltrated and Isostatic Laminated NCM and LTO Electrodes with Plastic Crystal Electrolyte Based on Succinonitrile for Lithium-Ion Solid State Batteries. Batteries 2021, 7 (1), 11. https://doi.org/10.3390/batteries7010011. 21. Zhou, S.; Li, M.; Wang, P.; Cheng, L.; Chen, L.; Huang, Y.; Yu, S.; Mo, F.; Wei, J. Printed Solid-State Batteries. Electrochem. Energy Rev. 2023, 6, 34. https://doi.org/10.1007/ s41918-023-00200-x. 22. Jung, K.-N.; Shin, H.-S.; Park, M.-S.; Lee, J.-W. Solid-State Lithium Batteries: Bipolar Design, Fabrication, and Electrochemistry. ChemElectroChem 2019, 6 (15), 3842–3859. https://doi.org/10.1002/celc.201900736.

Cell and Battery Design – Batteries | Cell Balancing Yevgen Barsukov, Battery Management Systems, Texas Instruments Inc., Dallas, TX, United States © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 2.1 2.2 2.3 3 3.1 3.2 3.3 3.4 4 4.1 4.2 4.3 4.3.1 4.3.2 4.4 4.5 5 5.1 5.2 5.3 5.4 5.5 6 References

Introduction Types of battery cell unbalance affecting charge/discharge voltage State of charge (SOC) unbalance Total capacity differences Impedance differences How does unbalance harm performance? Premature cells degradation through exposure to overvoltage Safety hazards from overcharged cells Early charge termination resulting in reduced capacity Early discharge termination Hardware implementation of balancing Resistive current bypass—Passive balancing Energy re-distribution—Active balancing Adjacent cell energy redistribution Charge shuttles Switching converter-based cell balancing of adjacent cells Passing energy from any cell to arbitrary cell in the pack Passing energy from any cell to the whole pack Balancing algorithms Cell voltage based SOC based SOC and total capacity based Effect of different algorithms of hardware current capability requirements depending on battery chemistry Overall reduction of pack energy decreases due to imbalance and delaying the end of life: Comparison of different balancing methods Summary

595 596 596 598 599 601 601 602 602 602 603 603 604 605 605 606 607 607 610 610 611 612 614 616 618 618

Abstract In battery packs where multiple cells are connected in series, the cell voltage is not always equal to pack voltage divided by the number of cells. What are the causes of this effect, what are the consequences for the pack energy degradation, and how it can be prevented? This paper will explore these questions in detail, identify the causes and solutions from both hardware and algorithmic standpoint, tied to specific battery chemistries, in addition with cost/benefit analysis of different solutions. Some examples of active and passive balancing systems based on TI ICs and demonstration designs are used in this paper.

Glossary C rate Rate of charge or discharge that will result in full charge or discharge in 1 h. It can be found by finding mAh capacity of the cell and divide it by 1 h time value, for example for 2000 mAh cell 1C rate is 2000 mA, ½ C rate is 1000 mA, etc. C-rate can be used for both charge and discharge. CC/CV Constant current/constant voltage charging protocol, which maintains constant CC charging current, until pre-defined CV voltage is reached, after which control changes to maintain constant voltage at CV level, while current gradually decreases. Charging terminates after I ¼ I_balance_min defined below. I_balance_min ¼ I_CC
Max_Imbalance_%/100, so for example if SOC imbalance for the cell with its largest value is 5%, balancing current will be I_CC
0.05. This can be quite a substantial current, because CC current can often be as high as 1C rate (1C means full charge in 1 h), but it is still much lower compared with balancing current needs of dV based balancing, where whole imbalance has to be removed only during last 10% of discharge, so balancing current will be 10 of this value. For variable current systems it makes sense to keep the balancing current to minimal value possible to maximize system efficiency, while peak worst-case current requirement remains the same as I_balance_min computed above. Continuous balancing current for each cell can be computed so that balancing will be finished right at the end of charging, and so it will depend on estimated remaining charge time, and dQbypass values for each cell. Based on this consideration, each cell balancing current would be: Ibalance ½i ¼ dQbypass½i=chargetime In case of active balancing, energy is not waisted and so it makes sense to balance not only on the top but also on the bottom, end of discharge point. However, if cell capacities are equal, balancing on the top automatically also balances on the bottom, so it can be done just once in charge direction. However, for different capacity cells algorithm will be discussed below.

5.3

SOC and total capacity based

Further improvement of above method takes into account also differences in total cell capacities (Qmax). This becomes possible if cell-balancing algorithm is an integral part of more complex gas-gauging algorithm that is monitoring state of each cell and capable of measuring changes in total cell capacity of each cell as it is the case for example with bq40z50R5 cell-balancing. Overall balancing method is similar to that described in Section 5.2, except for calculation of dQcharge and correspondingly dQbypass, which now considers each cell individual capacity. Fig. 17 shows progress of passive cell-balancing causing changes of open-circuit voltage at the end of charge in a 3-cells pack, where cell 1 was individually discharged and cell 2 charged by 2% prior to the test. By-pass resistance used in this test was 700 Ohm. For the case of different Qmax values between cells, the algorithm for active balancing in discharge direction is somewhat different. It requires following steps: a) Determine the initial SOC for each series cell bank separately. One of determination methods is using open circuit voltage correlation with the state of charge, but for better accuracy for flat voltage chemistries more complex algorithms that assign different weights depending on the flatness of the curve are needed. Such method can only be implemented in a microcontroller with flash memory and significant computational resources. b) Determine Qmax for each cell, for example using opportunistic algorithm that takes into account previous SOC and current integration history. c) Find chemical SOC_min[i] for each cell at which discharge termination Vload[i] ¼ Vterm_cell would occur if this cell would be discharged by itself. Here Vterm_cell ¼ Vterm/N_serial is minimal system voltage that system can operate at, at cell level, and Vload[i] ¼ OCV(SOC[i]) + I
R[i](SOC[i], T) has to be determined under consideration of up to date cell impedance and

Cell and Battery Design – Batteries | Cell Balancing

613

Cell Voltage, mV

4180

4170

4160

4150

4140

0

5

cell 0 cell 1 cell 2

10

15 20 Cycle Number

25

30

Fig. 17 Evolution of cell voltages during SOC/Qmax balancing, starting from initial 2% down (cell 1) and 2% up (cell 2) unbalance.

discharge current or system power and temperature. Detailed description of such determination is outside of the scope of this chapter, but can be found here.2 This is the target SOC all cells have to reach in order for total pack voltage to become Vterm at discharge termination, those assuring that no early termination of discharge occurs and discharge energy is maximized. d) Find how much discharge is needed for each cell to reach its own target SOC_min[i] fully charged state, dQdischarge[i]. This requires knowledge of total capacity of each cell, Qmax[i], and present SOC, given that: SOC_min[i] ¼ SOC[i] + dQdischage [i]/Qmax[i], so dQdischarge[i] ¼ (SOC_min[i] – SOC[i])
Qmax[i]. Note that here SOC is represented as a fraction of 1. e) To find all of such values, the cell with median amount of charge that needs to pass to fully discharge state, dQdischarge_median, is found. After that for all cells that have higher amount of discharge needed, dQsub values are computed as difference from the median value dQdischarge_median, and for cell with lower amount of discharge needed, dQadd are computed same way. Further it will be referred to as dQbypass, assuming that it can be either positive or negative. Subsequent action depends if active balancing system is only capable to provide fixed balancing current I_balance when switching is operating as in system shown in Fig. 15, or it is able to modulate the current in certain range, like PWM controller-based system shown in Fig. 14. For fixed current systems the next task is to find balancing time for all cells as Balancing time[i] ¼ dQbypass[i]/I_balance Where I_balance is the current of the active balancing system. Once balancing time is exceeded, balancing stops, as the required charge already passed. The peak current requirement is determined by the worst-case scenario where balancing time ¼ maximal discharge time. Since discharge time is a complex function of battery impedance, temperature, average system load and capacity, we can just state that typical gauging systems are capable to compute this value. For the purpose of simplified overestimation, we can assume that balancing rate needs to keep up with highest sustained discharge current, Imax_discharge, e.g., system needs to be designed so that I_balance > ¼ I_balance_min defined below: I_balance_min ¼ Imax_discharge
Max_Imbalance_%/100, so for example if SOC imbalance for the cell with largest imbalance is 5%, balancing current will be Imax_discharge
0.05. This can be quite a substantial current, because discharge current can often be as high as 1C rate, and so balancing current twould be 0.05
C rate (1C means full discharge in 1 h), but it is still much lower compared with balancing current needs of dV based balancing, where whole imbalance has to be removed only during last 10% of discharge where discharge enters its steep portion so dV/dSOC value becomes large enough to exceed balancing dV threshold, so dV algorithm balancing current will be roughly 10 of this value. For variable current systems it makes sense to keep the balancing current to minimal value possible to maximize system efficiency, while peak worst-case current requirement remains the same as I_balance_min computed above. Continuous balancing current for each cell can be computed so that balancing will be finished right at the end of discharging, and so it will depend on estimated remaining discharge time, and dQbypass values for each cell. Based on this consideration, each cell balancing current would be: I balance½i ¼ dQbypass½i=discharge time While it is possible to estimate: discharge time ¼ dQdischage½i=Imax discharge above estimation is assuming that all imbalances will be removed by the end of discharge, but more accurate methods of estimation that consider actual cell temperature, discharge current or system power and cell impedance exist.2

614 5.4

Cell and Battery Design – Batteries | Cell Balancing Effect of different algorithms of hardware current capability requirements depending on battery chemistry

In case of by-pass balancing on top methods, they can gradually converge to equal state of charge at the top over multiple cycles, as shown in Sections 5.1 and 5.3, and after that just need to provide “maintenance” balance to account to cell self-discharge differences and incremental Qmax changes, their peak current requirements are quite low and defined by above maintenance current level. This maintenance level depends on self-discharge rate of particular chemistry, but is generally very low for all Li-ion chemistries (218

material

Source of Business

Electronic fluorinated fluids

The Novec 7100 is the same as the HFE-7100 HFE-6512 mineral oil E5 TM 410 AmpCool AC-110

hydrocarbon

esters Silicone oils Water-based fluids category Electronic fluorinated fluids hydrocarbon esters Silicone oils Water-based fluids

MIVOLT-DF7 MIVOLT-DFK Silicone oils dimethyl silicon oil Ethyl silicone oil deionized water Aqueous glycol solution Alumina nanofluids 0.4% material The Novec 7100 is the same as the HFE7100 HFE-6512 mineral oil E5 TM 410 AmpCool AC-110 MIVOLT-DF7 MIVOLT-DFK Silicone oils dimethyl silicon oil Ethyl silicone oil deionized water Aqueous glycol solution Alumina nanofluids 0.4%

Density at 20
C

Thermal conductivity 0.075

category

USA 3 M

Kinematic viscosity at 20
C 0.32

Zhejiang Huikai Dingrui Dutch Shell USA ENGINEERED FLUIDS UK M&I Materials UK M&I Materials USA super lube

−57 −75 250 >300 >155 80 N

Non-flammable inflammable inflammable inflammable

dielectric constant

16 16 80.2 64.92 Eco-friendliness ODP ¼ 0 GWP ¼ 530 ODP ¼ 0 GWP ¼ 1 GWP ¼ 0 ODP ¼ 0 GWP < 1 ODP ¼ 0 GWP < 1

Cell and Battery Design – Batteries | Thermal Management

Table 1

Cell and Battery Design – Batteries | Thermal Management Table 2

643

Two-phase immersion cooling medium.30

Two-phase immersion cooling medium Density at 20
C

Thermal conductivity 0.059 0.075 0.057 0.077

specific heat capacity 1103 1300 1100 1200

dielectric constant 1.8 7.4 1.75 32

flashing point

security

Eco-friendliness

49

Latent heat of vaporization 88

−122

34

142

N

FC-72

−90

56

88

N

SF33

−107

33.4

166

N

category

material

Kinematic viscosity at 20
C 0.4 0.32 0.18 0.3

material

Source of Business USA 3 M USA 3 M USA 3 M USA chemours freezing point

Electronic fluorinated fluids

Novec649 Novec7000 FC-72 SF33

category Electronic fluorinated fluids

Novec649

−108

Novec7000

boiling point

Fig. 30 Schematic diagram of dual-phase static immersion cooling.33

Fig. 31 Schematic diagram of dual-phase dynamic immersion cooling.34

1600 1400 1680 1383.5

Non-flammable

ODP ¼ 0 GWP ¼ 1 ODP ¼ 0 GWP ¼ 530 ODP ¼ 0 high GWP ODP ¼ 0 GWP ¼ 2

644

Cell and Battery Design – Batteries | Thermal Management

Fig. 32 Variation curves of heat flux (q) and nucleation rate with the difference between wall temperature and saturation temperature.11

Fig. 33 Boiling image of the cooling working fluid at the beginning of boiling and the boiling at the end of discharge.35

Cell and Battery Design – Batteries | Thermal Management

645

Fig. 34 The function of BESSs in power transmission.36

phase change cooling and heat pipe cooling are still in the theoretical research stage. Fig. 35 shows a comparison of the four different thermal management techniques in terms of thermal conductivity index, heat dissipation rate and cost.

8.1

Air cooling energy storage system

8.1.1 Principles and characteristics Air cooling is an energy storage thermal management technology that uses air as the heat transfer medium, as shown in Fig. 36. When the temperature of the energy storage system is too high, the surface heat of the battery is absorbed through the airflow to achieve

Fig. 35 Comparison of four cooling methods in terms of thermal performance.

646

Cell and Battery Design – Batteries | Thermal Management

Fig. 36 Air cooling principle.37

cooling. Air cooling can be divided into natural cooling and forced cooling. Natural cooling mainly utilizes natural wind pressure to dissipate heat from the battery module, also known as passive cooling; Forced cooling, also known as active cooling, is widely used in energy storage systems by introducing air into the battery module through a fan to remove the heat generated by batteries and dissipate it into the air. Fig. 37 shows the thermal management control strategy for a forced-cooling energy storage system.

8.1.2 Market applications of air-cooled energy storage systems Air cooling, as a simple, feasible, and low-cost cooling method, is widely used in the early energy storage market. As shown in Fig. 38, the 1500 V series air-cooled outdoor cabinet energy storage system developed by East Group Co., Ltd. has been successfully applied in the Hebei Kangbao Intelligent Microgrid Demonstration Project. The energy storage system has a modular design for both the battery cluster and battery cooling system, and adopts a design of grouped batteries without parallel connection for intelligent air conditioning cooling. It has the characteristics of economic efficiency, convenience, flexibility, and safety. In the current energy storage market, the user side energy storage market mainly uses air cooling, while the grid side energy storage market has shifted towards liquid cooling.

8.2

Liquid cooling energy storage system

8.2.1 Principles and characteristics Liquid cooling technology can save more than 50% of space and it is more suitable for future large-scale energy storage plants with a capacity of more than 100 MWh. However, liquid cooling also has the disadvantages of high equipment cost, complex system, and maintenance difficulties. Liquid cooling can be divided into indirect liquid cooling and direct liquid cooling.

Fig. 37 Thermal management control strategy for air-cooled systems.38

Cell and Battery Design – Batteries | Thermal Management

647

Fig. 38 Air-cooled energy storage system. Source: East Group Co., Ltd.

Fig. 39 Liquid cooling container. Source: Serlattice.

8.2.1.1 Indirect liquid cooling energy storage system Indirect liquid cooling energy storage system is one of the most widely used thermal management systems for energy storage. Compared to air-cooled systems, liquid-cooled system can reduce power consumption by up to 30% and increase service life by up to 13%. Fig. 39 shows the Xinghan SMT-ESS series of indirect liquid cooling energy storage box from Shanghai Sermatec Energy Technology Co., Ltd. The Energy Storage System mainly consists of box, batteries, liquid coolers, liquid cooling pipelines, combiner cabinets, fire cabinets, distribution boxes, and monitoring systems. Indirect liquid cooling energy storage system has the advantages of high compatibility, easy installation, space saving and long service life. 8.2.1.2 Direct liquid cooling energy storage system Direct liquid cooling, also known as immersion cooling, is an important direction for future energy storage cooling systems. Immersion cooling technology achieves direct, fast and sufficient cooling by immersing the storage battery in a special insulating coolant, as shown in Fig. 40. Non-oil-based immersion coolants are usually used for immersion cooling because oil-based immersion coolants have the disadvantages of high viscosity, flammability, and high cost. However, immersion cooling also has the risk of coolant leakage and requires a high degree of sealing of the energy storage system. The application of immersion cooling technology will, to a certain extent, solve the safety hazards of battery cells, greatly improve the cooling efficiency of the battery thermal management system, effectively improve the consistency of the battery operating temperature, and realize that the temperature difference between different areas of the battery is less than 2
C. 8.2.1.3 Market application of direct liquid cooling energy storage system In March 2023, the world’s first submerged liquid-cooled energy storage station was put into operation at the Meizhou Baohu Energy Storage Station of the Southern Power Grid (SPG), and the energy storage system was provided by Kortron. The capacity of each immersed liquid cooled battery compartment at Meizhou Baohu Energy Storage Power Station is 5.2 MWh, equipped with 14 battery clusters. Meanwhile, they are all submerged in the coolant. It can manage the temperature rise during battery operation of no more than 5
C, and the temperature difference between different batteries is no more than 2
C. Compared with the traditional

648

Cell and Battery Design – Batteries | Thermal Management

Fig. 40 Immersion liquid cooling. Source: Esonline.

air-cooled energy storage system, the battery heat dissipation efficiency is also increased by 50%. With the improvement of related technologies, the immersion cooled energy storage system will become the mainstream of the future energy storage market.

8.3

Phase change cooling and heat pipe cooling energy storage system

The application of phase change cooling and heat pipe cooling in energy storage is still in the research state, cost and stability are the main factors affecting its application. With the further research, the application of phase change cooling and heat pipe cooling in energy storage will be gradually realized. Phase change cooling is a technology that utilizes phase change materials (PCM) for battery cooling. The phase change materials are in direct contact with the battery cells, the structure is shown in Fig. 41. PCM undergo a phase transition (between solid and liquid) within a specific temperature range to absorb heat and achieve battery cooling. However, the volume of the phase change material will change significantly after absorbing heat, which will easily affect the working performance of the energy storage system. Heat pipe cooling is a device that uses the medium at the heat absorbing end to evaporate and carry away heat, and then transfers the heat to the outside world through condensation at the heat releasing end. As an efficient heat exchange element, heat pipes are rarely used in high-capacity battery systems, and related research is still in the laboratory stage. However, the cost is high, and they are widely used in electronic devices such as mobile phones. When applied to energy storage systems, the cost-effectiveness is not high.

9

Sensors in the battery thermal management system

Sensors play a crucial role in thermal management systems. They are responsible for detecting specific measured signals within the system and converting them into usable output signals. These converted signals serve various purposes, including information transmission, processing, recording, display, and control. By adhering to specific rules, sensors transform these signals into electrical quantities such as voltage or current, which are then transmitted to the thermal management system. In battery thermal management systems, sensors enable automatic detection and control, making them a vital component in achieving efficient and reliable operation.

Fig. 41 Schematic of phase change material cooling.37

Cell and Battery Design – Batteries | Thermal Management 9.1

649

Temperature sensor in battery thermal management system

The temperature sensor in the battery thermal management system is mainly used to measure and monitor the surface temperature of the power battery, the temperature of the battery cooling system medium, etc. It plays an important role in ensuring the safety, performance and life of the power battery. The principle of power battery temperature sensor is mainly to utilize the relationship between different physical quantities and temperature, convert the temperature into an electrical signal to carry out temperature measurement. At present, the types of temperature sensors commonly used in the battery thermal management system are mainly thermistor temperature sensors and thermocouple temperature sensors.

9.2

Common temperature sensors and their working principles

9.2.1 NTC thermistor temperature sensor The principle of thermistor temperature sensors is to utilize the property of resistance to change with temperature. Thermistors are divided into three categories: positive temperature coefficient thermistors (PTC), negative temperature coefficient thermistors (NTC), and critical temperature resistors (CTR). Among them, the NTC negative temperature coefficient thermistor (Fig. 42) is the most widely used sensor., Negative temperature coefficient refers to the decrease of the resistance with the rise in temperature. This is an exponential relationship between temperature and resistance. Its principle is to utilize the properties of semiconductor materials, as the temperature rises, the number of carriers increases, so the resistance value decreases. NTC thermistors work on the principle of temperature sensitivity of semiconductor materials with a certain degree of accuracy and stability. As shown in Fig. 43, one of the advantages of NTC thermistors is their fast response time, which allows them to accurately measure temperature changes in a relatively short period of time. In addition, its small size and light weight make it easy to install and integrate into a variety of devices.

9.2.2 Type K thermocouple temperature sensor The principle of the thermocouple temperature sensor is to use the thermoelectric effect, as shown in Fig. 44, that is, when two different metals or alloy wires are connected into a circuit, if the temperatures of the two junctions are different, a thermoelectric potential will be generated in the circuit. The size of the thermoelectric potential is directly proportional to the temperature difference between the two junctions. Common types of thermocouple temperature sensors include S-type, R-type, B-type, K-type, E-type, J-type, T-type, N-type, etc., among which the K-type is the most commonly used. The K-type thermocouple has a wide temperature measurement range, which can reach from −200
C to 1300
C. Its features include fast response time, strong anti-interference ability, high linearity, etc. In addition, the K-type thermocouple is relatively low in price and easy to use, so it has been widely applied.

Fig. 42 NTC negative temperature coefficient thermistor sensor in different packaging modes. Source: Internet.

650

Cell and Battery Design – Batteries | Thermal Management

Fig. 43 (a) NTC negative temperature coefficient thermistor resistance and temperature curve38 (b) Equivalent circuit of NTC thermistor sensor. Source: Baidu Encyclopedia.

Fig. 44 How thermocouples work.39

9.3

Common temperature sensor probe types and their application scenarios

The temperature sensor probe is an important part of the temperature sensor. It directly contacts the object or medium being measured and converts the temperature into an electrical signal or digital signal. According to the principle of temperature measurement and shape, temperature sensor probes can be divided into insertion (immersion) type probes, patch type probes, air type probes, etc. Depending on the type of temperature sensor probe, temperature sensors are widely used in battery thermal management systems for temperature detection of the battery body, temperature detection of battery cooling media, and temperature detection of the control board of the battery thermal management system.

9.3.1 Temperature sensor probe for battery surface temperature detection The patch type probe is used for battery surface temperature detection. As shown in the figure, the probe is generally attached to the surface of the battery cell to collect the temperature of the battery cell. The advantages of the patch type probe are flexible structure, simple installation, and no influence of the object. The disadvantage is that it can only collect the surface temperature of the battery and cannot collect the internal temperature of the battery, and it is greatly affected by the environment. Fig. 45 shows the two different forms of battery surface temperature detection sensors.

9.3.2 Temperature sensor probe for temperature detection of cooling media The insertion (immersion) type probe can be used to detect the temperature of the cooling medium in the power battery cooling system, etc. Accurate temperature data is obtained by immersing it in the cooling medium of the battery thermal management

Fig. 45 Two different forms of battery surface temperature detection sensors. Source: iYHBW.

Cell and Battery Design – Batteries | Thermal Management

651

system. The probe is generally made of metal materials, such as stainless steel, copper, etc. The advantages of the insertion type probe are simple structure, easy installation, and fast response speed. Its structure is not suitable for measuring surface temperature. Fig. 46 shows the battery pack coolant temperature sensor used in BYD Qin/Song models.

9.4

Pressure sensor in battery thermal management system

In order for the battery thermal management system to effectively monitor and control the working status of the power battery, it is necessary to install other sensors in the system in addition to the temperature sensor. Among them, the pressure sensor is an important sensor. It can measure the air pressure inside the battery pack, the hydraulic pressure of the battery liquid cooling system, and the expansion force between the battery cells, thereby judging whether the battery has thermal runaway, whether the cooling effect is good, and whether the battery structure is intact.40

9.4.1 Commonly used pressure sensors and their working principles 9.4.1.1 Strain gauge pressure sensor The strain-type pressure sensor uses the strain effect of elastic elements to convert pressure changes into changes in strain resistance. The strain-type pressure sensor is a sensing device that uses elastic sensitive elements and strain gauges to convert the measured pressure into corresponding resistance value changes. As shown in Fig. 47 The advantages of the strain-type pressure sensor are simple structure, fast response speed, and good linearity. However, the disadvantages are low sensitivity, large temperature influence, and short life span. 9.4.1.2 Piezoresistive pressure sensor The piezoresistive pressure sensor uses the piezoresistive effect to convert pressure changes into changes in the resistance of the piezoresistive element. The piezoresistive effect refers to the phenomenon that when a semiconductor is subjected to stress, the change in the energy band caused by the stress, the movement of the energy valley, causes its resistivity to change. The advantages of the piezoresistive pressure sensor are high sensitivity, easy temperature compensation, and long life, but the disadvantages are

Fig. 46 Battery pack coolant temperature sensor used in BYD Qin/Song models. Source: BYD.

Fig. 47 Structure of pressure sensor using strain effect.39

652

Cell and Battery Design – Batteries | Thermal Management

Fig. 48 Piezoresistive pressure sensor structure schematic diagram. Source: Sensor expert network.

complex structure, slow response speed, and poor linearity. Fig. 48 shows Piezoresistive pressure sensor structure schematic diagram.

9.5

Common pressure sensor probe types and their application scenarios

Different forms of pressure sensor packaging probes can be applied in different measurement scenarios, such as detecting the air pressure inside the battery pack, the hydraulic pressure of the battery liquid cooling system, and the expansion force between the battery cells, etc.

9.5.1 Pressure sensor probe for oil pressure detection in cooling systems The pressure sensor used for detecting the hydraulic pressure of the cooling system is mainly used to monitor the hydraulic changes in the battery liquid cooling system, to judge the cooling effect of the battery and the working status of the hydraulic system. The hydraulic changes in the battery liquid cooling system reflect the flow, resistance, and leakage of the coolant, etc. Regulating the flow rate and pressure of the coolant, optimizing the cooling effect of the battery, ensuring the normal operation of the battery thermal management system, and ensuring the safety of the battery are of great significance. Fig. 49 shows the arrangement of pressure sensors in the battery cooling system and cooling system pressure sensor.

9.5.2 Pressure sensor probe for pressure detection inside battery pack The pressure sensor used for detecting the internal pressure of the battery pack is used to monitor the changes in the air pressure inside the battery pack, to determine whether the battery has thermal runaway or other abnormal conditions. Therefore, timely detection of changes in the air pressure inside the battery pack is an important means of ensuring battery safety by collecting battery status in battery thermal management. Fig. 50 shows the layout of the pressure sensor inside the battery pack and chip type gas pressure sensor.

Fig. 49 (a) Arrangement of pressure sensors in the battery cooling system (Source: Zhihu). (b) Cooling system pressure sensor. Source: NSTRUMENT.

Cell and Battery Design – Batteries | Thermal Management

653

Fig. 50 (a) Layout of the pressure sensor inside the battery pack (Source: HIScore Microsemiconductor). (b) Chip type gas pressure Sensor. Source: First Sensor.

9.5.3 Pressure sensor probe for battery cell expansion pressure detection The thin-film piezoresistive force sensor is used to measure the expansion force between the battery cell and the adjacent battery, in order to determine the degree of deformation and structural integrity of the battery. Therefore, measuring the expansion force between battery cells is an important indicator for the battery thermal management system to evaluate the condition of the battery. Currently, this type of pressure sensor has not been widely used in battery products, but its application prospects are obvious. Fig. 51 shows different structures of thin-film pressure sensors for cell expansion pressure detection.

Fig. 51 Thin-film pressure sensors for cell expansion pressure detection of different structures. Source: LEGACT.

654

10

Cell and Battery Design – Batteries | Thermal Management

Conclusion

The article introduces the battery thermal management system (BTMS), which is essential for ensuring the safety and performance of batteries. It reviews five main cooling methods for BTMS: air cooling, liquid cooling, phase change cooling, direct cooling, and heat pipe. It explains the principles, advantages, and challenges of each method, as well as the current market applications and future development trends. 1. Air cooling is simple but struggles with high-energy-density batteries. Liquid cooling is efficient but complex. Phase change cooling, a promising technique, uses latent heat but faces technical and economic hurdles. Direct cooling, a new technology, directly cools battery cells efficiently but faces temperature control and refrigerant containment challenges. Heat pipe, a highly efficient heat transfer device, reduces battery temperature and enhances safety and lifespan. 2. The article discusses battery thermal management in energy storage applications and sensors used in these systems. Energy storage, a key battery application, requires effective thermal management for system reliability and stability. It explores the impact of usage scenarios on thermal management technology selection, current technologies like air and liquid cooling, and future trends towards immersion and mixed cooling. Sensors, essential for automatic detection and control in these systems, are also discussed. 3. As battery energy density increases and high-power charging/discharging becomes more common, battery thermal management faces challenges. Uneven heat distribution can cause localized overheating, affecting battery performance and lifespan. Effective thermal management is crucial under dynamic conditions as batteries generate varying heat loads. Dynamic regulation in thermal management systems is key for adapting to these conditions. Modern electric vehicles’ need for lightweight design and high integration imposes new requirements on battery systems, necessitating significant improvements in traditional thermal management approaches. 4. Breakthroughs in material technology, artificial intelligence, and the theory of multi-physics field coupling have brought about significant innovation and development in battery thermal management. The introduction of new materials such as graphene and nanomaterials has opened new possibilities for enhancing battery thermal management techniques. Leveraging artificial intelligence algorithms to optimize control strategies for thermal management systems has elevated the intelligent capabilities of these systems. The utilization of multi-physics field coupling simulation methods allows for an in-depth exploration of the thermal-electric coupling effects within the battery, providing a more reliable theoretical foundation for precise thermal management.

References 1. Wang, Q.; Jiang, B.; Li, B.; et al. A Critical Review of Thermal Management Models and Solutions of lithium-Ion Batteries for the Development of Pure Electric Vehicles. Renew. Sustain. Energy Rev. 2017, 74, 107–128. 2. DJI Innovations. Mavic 2 Technical Specifications. (2018-08-23) [2023-03-23]. https://www.dji.com/cn/mavic-2. 3. Xpeng Motors: Xpeng's P5 Sky Voyager X2 successfully granted special flight permit. (2023−01−30) [2023-03-23]. https://www.xpeng.com/news/company_news/4775.html. 4. Sun, H.; Dixon, R. Development of Cooling Strategy for an Air Cooled Lithium-Ion Battery Pack. J. Power Sources 2014, 272, 404–414. 5. Lu, Z.; Yu, X.; Wei, L.; Qiu, Y.; Zhang, L.; Meng, X.; Jin, L. Parametric Study of Forced Air Cooling Strategy for Lithium-Ion Battery Pack with Staggered Arrangement. Appl. Therm. Eng. 2018, 137, 28–40. 6. Zhu, C.; Li, X.; Song, L.; Xiang, L. Development of a Theoretically Based Thermal Model for lithium Ion Battery Pack. J. Power Sources 2013, 223, 155–174. 7. Saw, L. H.; Poon, H. M.; Thiam, H. S.; Cai, Z.; Chong, W. T.; Pambudi, N. A.; King, Y. J. Novel Thermal Management System Using Mist Cooling for Lithium-Ion Battery Packs. Appl. Energy 2018, 223, 147–158. 8. Yan, W. Numerical Study on Flow and Heat Transfer Characteristics of Liquid Cold Plate [D]; Xidian University, 2013. 9. Jiaqiang, E.; Xu, S.; Deng, Y.; Zhu, H.; Zuo, W.; Wang, H.; Chen, J.; Peng, Q.; Zhang, Z. Investigation on Thermal Performance and Pressure Loss of the Fluid Cold-Plate Used in Thermal Management System of the Battery Pack. Appl. Therm. Eng. 2018, 145, 552–578. 10. Sheng, L.; Zhang, H.; Su, L.; Zhang, Z.; Zhang, H.; Li, K.; Fang, Y.; Ye, W. Effect Analysis on Thermal Profile Management of a Cylindrical Lithium-Ion Battery Utilizing a Cellular Liquid Cooling Jacket. Energy 2021, 220, 119725. 11. Roe, C.; Feng, X.; White, G.; Li, R.; Wang, H.; Rui, X.; Li, C.; Zhang, F.; Null, V.; Parkes, M.; et al. Immersion Cooling for Lithium-Ion Batteries—A Review. J. Power Sources 2022, 525, 231094. 12. Jarrett, A.; Kim, I. Y. Design Optimization of Electric Vehicle Battery Cooling Plates for Thermal Performance. J. Power Sources 2011, 197, 10359–10378. 13. Huo, Y.; Rao, Z.; Liu, X.; Zhao, J. Investigation of Power Battery Thermal Management by Using mini-Channel Cold Plate. Energ. Conver. Manage. 2015, 89, 387–395. 14. Huang, Y.; Mei, P.; Lu, Y.; Huang, R.; Yu, X.; Chen, Z.; Roskilly, A. P. A Novel Approach for Lithium-Ion Battery Thermal Management with Streamline Shape mini Channel Cooling Plates. Appl. Therm. Eng. 2019, 157, 113723. 15. Deng, T.; Zhang, G.; Ran, Y. Study on Thermal Management of Rectangular Li-Ion Battery with Serpentine-Channel Cold Plate. Int. J. Heat Mass Transf. 2018, 125, 143–152. 16. Wu, W.; Liu, J.; Liu, M.; Rao, Z.; Deng, H.; Wang, Q.; Qi, X.; Wang, S. An Innovative Battery Thermal Management with Thermally Induced Flexible Phase Change Material. Energ. Conver. Manage. 2020, 221, 113145. https://doi.org/10.1017/j.enconman.2020.113145. 17. The first successful flight of the RX4E, a four-seat electric aircraft developed by Liaoning General Aviation Research Institute under the Liaoning Provincial Department of Education. [EB/OL]. (2019-10-30) [2023-03-23]. https://mp.weixin.qq.com/s/8WtV5CURZbS3dPpaiIF5zQ. 18. Wu, W.; Yang, X.; Zhang, G.; Ke, X.; Wang, Z.; Situ, W.; Li, X.; Zhang, J. An Experimental Study of Thermal Management System Using Copper Mesh-Enhanced Composite Phase Change Materials for Power Battery Pack. Energy 2017, 113, 909–917. https://doi.org/10.1017/j.energy.2017.07.119. 19. Wang, F.; Cao, J.; Ling, Z.; Zhang, Z.; Fang, X. Experimental and Simulative Investigations on a Phase Change Material Nano-Emulsion-Based Liquid Cooling Thermal Management System for a Lithium-Ion Battery Pack. Energy 2020, 207, 118215. https://doi.org/10.1017/j.energy.2020.118215. 20. Lei, N.; Minchi, W.; Yao, Z.; et al. Experimental Study on Thermal Management System of Refrigerant Direct-Cooled Battery for Pure Electric Vehicles. J. Refrig. 2020, 41 (04), 52–58.

Cell and Battery Design – Batteries | Thermal Management

655

21. Weragoda, D. M.; Tian, G.; Burkitbayev, A.; Lo, K.-H.; Zhang, T. A Comprehensive Review on Heat Pipe Based Battery Thermal Management Systems. Appl. Therm. Eng. 2023, 224. 22. Lin, K.-T.; Wong, S.-C. Performance Degradation of Flattened Heat Pipes. Appl. Therm. Eng. 2013, 50 (1), 46–54. ISSN 1359-4311. 23. Hong, S.; Zhang, X.; Wang, S.; Zhang, Z. Experiment Study on Heat Transfer Capability of an Innovative Gravity Assisted Ultra-Thin Looped Heat Pipe. Int. J. Therm. Sci. 2015, 95, 106–114. ISSN 1290-0729. 24. Wang, Q.; Rao, Z.; Huo, Y.; Wang, S. Thermal Performance of Phase Change Material/Oscillating Heat Pipe-Based Battery Thermal Management System. Int. J. Therm. Sci. 2016, 102, 9–16. ISSN 1290-0729. 25. Bernagozzi, M.; Georgoulas, A.; Miché, N.; Marengo, M. Heat Pipes in Battery Thermal Management Systems for Electric Vehicles: A Critical Review. Appl. Therm. Eng. 2023, 219. Part A. 119495. ISSN 1359-4311. 26. Xie, Y.; Li, H.; Li, W.; Zhang, Y.; Fowler, M.; Tran, M. K.; Zhang, X.; Chen, B.; Deng, S. Improving Thermal Performance of Battery at High Current Rate by Using Embedded Heat Pipe System. J. Energy Storage 2022, 46, 103809. ISSN 2352-152X. 27. Zhang, W.; Qiu, J.; Yin, X.; Wang, D. A Novel Heat Pipe Assisted Separation Type Battery Thermal Management System Based on Phase Change Material. Appl. Therm. Eng. 2020, 165, 114571. ISSN 1359-431. 28. Smith, J.; Singh, R.; Hinterberger, M.; Mochizuki, M. Battery Thermal Management System for Electric Vehicle Using Heat Pipes. Int. J. Therm. Sci. 2018, 134, 517–529. 29. Zhe, L.; Hua, Z.; Lei, S. Progress of Cooling Technology for Submerged Lithium-Ion Batteries in New Energy Vehicles. J. Refrig. 2024, 1–12. http://kns.cnki.net/kcms/detail/11. 2182.tb.20231108.1613.004.html. 30. Zeng, S.; Wu, W.; Liu, J.; Wang, S.; Ye, S.; Feng, Z. A Review on Immersion Cooling Technology for Lithium-Ion Batteries. Energy Storage Sci. Technol. 2023, 12 (9), 2888–2903. https://doi.org/10.19799/j.cnki.2095-4239.2023.0269. 31. Liu, Q.; Sun, C.; Zhang, J.; Shi, Q.; Li, K.; Yu, B.; Xu, C.; Ju, X. The Electro-Thermal Equalization Behaviors of Battery Modules with Immersion Cooling. Appl. Energy 2023, 351, 121826. https://doi.org/10.1016/j.apenergy.2023.121826. 32. Hao-Wen, G. Simulation and Experimental Study of Submerged Liquid-Cooled Battery Packs for Pure Electric Vehicles; M.S, Zhejiang University, 2022. https://doi.org/10.27461/ d.cnki.gzjdx.2022.000191. 33. Li, Y.; Bai, M.; Zhou, Z.; Wu, W.-T.; Lv, J.; Gao, L.; Huang, H.; Li, Y.; Li, Y.; Song, Y. Experimental Investigations of Liquid Immersion Cooling for 18650 Lithium-Ion Battery Pack under Fast Charging Conditions. Appl. Therm. Eng. 2023, 227, 120287. https://doi.org/10.1016/j.applthermaleng.2023.120287. 34. Wang, Y.-F.; Wu, J.-T. Thermal Performance Predictions for an HFE-7000 Direct Flow Boiling Cooled Battery Thermal Management System for Electric Vehicles. Energ. Conver. Manage. 2020, 207, 112569. https://doi.org/10.1016/j.enconman.2020.112569. 35. Wu, C.-H. Heat Transfer Performance of Lithium-Ion Batteries for Electric Vehicles Based on Immersion Cooling; M.S, Guilin University of Electronic Science and Technology, 2022. https://doi.org/10.27049/d.cnki.ggldc.2022.000867. 36. Zhu, X.; Xu, X.; Kong, B.; Wang, J.; Shi, H.; Jiang, Y. Coupling Simulation of the Cooling Air Duct and the Battery Pack in Battery Energy Storage Systems. Phys. Scripta 2023, 98, 075907. 37. Xinlong, Z. H. U.; Junyi, W. A. N. G.; Jiashuang, P. A. N.; et al. Present Situation and Development of Thermal Management System for Battery Energy Storage System. Energy Storage Sci. Technol. 2022, 11 (01). 38. Tian, G.; Zjamg, L.; Niu, Z.; Li, Z.; Luo, J.; et al. Design of Thermal Management for Container-Type Energy Storage System. Chin. J. Power Sources 2021, 45 (03), 317–319. 329. 39. Fraden, J. Handbook of Modern Sensors: Physics, Designs, and Applications, 4th ed.; Springer Verlag, 2010. https://searchworks.stanford.edu/view/9115874. 40. Wolinski, T.; Konopka, W.; Bock, W.; et al. Progress in Liquid Crystalline Optical fiber Systems for Pressure Monitoring; SPIE, 1998.

Cell and Battery Design – Batteries | Electrical Management Yanan Wanga, Xuning Fengb, Xuebing Hanb, and Minggao Ouyangb, aDepartment of Mechanical and Energy Engineering, Beijing University of Technology, Beijing, China; bDepartment of Vehicle and Mobility, Tsinghua University, Beijing, China © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 1.1 1.2 2 2.1 2.2 2.3 2.4 3 3.1 3.1.1 3.1.2 3.2 3.3 3.3.1 3.3.2 3.4 3.4.1 3.4.2 4 References

Introduction Battery hardware Battery system hardware Batteries hardware Cell structure Connections Topology Module/pack shell Systems hardware BMS category Functional types System Topology Monitoring and communication Safety and Thermal management Active safety and fault diagnosis Thermal management Balance and control Battery balance Charging control Conclusion and outlook

657 657 657 657 657 658 659 659 660 660 660 662 662 662 662 664 665 665 665 666 666

Abstract This chapter presents an overall view of battery and its system hardware. From the cell level to module level then to system level, various devices are introduced, such as power supply module, charging module, measurement module, analog IO module, digital IO module, clock module, communication module, thermal management, balancing, and high-voltage protection module. This chapter analyzes the main functions of different modules and tries to give a overview of hardware for batteries and its system. It also provides a perspective about the trends of hardware development at the end of this chapter.

Key points

• • •

An overall introduction about the hardware of battery and its management system. Hardware presentation from cell level to system level. Topology and categories analysis by functional module.

Abbreviations BMS CAN CC-CV MCU PTC SEMS SOC SOH

656

Battery management system Controller area network Constant current and constant voltage Main control unit Positive temperature coefficient Screw and Washer Assemblies State of charge State of health

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00248-6

Cell and Battery Design – Batteries | Electrical Management

1

657

Introduction

Batteries and systems are interdisciplinary management objects that integrate electrochemistry, electrical, mechanical, and thermophysics. In the hardware aspect, the battery system generally includes battery cells to module to pack, then integrated with power electronic devices (such as connectors, fuses, contactors, sensors), to construct the final system. Moreover, by using power electronic devices including high-power switching inverters, DC converters, and chargers, the battery system is connected to the load for the final power output of the battery system.1 In this chapter, the hardware sections inside battery and its system are introduced. This chapter is mainly presented in two parts: one part is the hardware units before the batteries are formed into a system, such as cells, modules, clusters, and connecting devices among them; the other part is the various functional units in system level, such as charging unit, monitoring unit, thermal management unit, etc.

1.1

Battery hardware

Before cells are formed into a system, the design of battery hardware units concerns about the structure, connections, topology and mechanical shell for cell, module or cluster.2 Commonly used cells generally can be divided into three types, that is, pouch cell, prismatic cell, and cylindrical cell. Although the manufacturing processes of these three types cells are different, such as the winding and stacking process, the basic structural components are roughly the same, as shown in Fig. 1.

1.2

Battery system hardware

At the system level, battery hardware concerns about the design of various functional modules with Main Control Unit (MCU) as the core module to build a comprehensive battery management.3 The system hardware includes power supply module, charging module, measurement module (voltage, current, temperature, etc.), analog IO module, digital IO module, clock module, communication module, thermal management, balancing, and high-voltage protection module, as shown in Fig. 2. The system hardware contains components both with force electricity and control electricity.

Fig. 1 Battery hardware in cell level and grouped from cell to module.

658

Cell and Battery Design – Batteries | Electrical Management

Fig. 2 Battery system hardware.

2 2.1

Batteries hardware Cell structure

To construct a single cell, such as pouch cell, prismatic cell, or cylindrical cell, electrodes, electrolyte and separator are the basic components in electrochemical aspect. While in hardware aspect, more components are included to realize the functional operation of a cell. This section concludes the necessary structural parts to consist a single cell as follows.4 1. Positive terminal A conductive component that facilitates external circuit connection to the positive pole of the battery. 2. Negative terminal A conductive component that facilitates the connection of external circuits to the negative terminal of a battery 3. Tab The metal conductor that connects the internal electrode plates and terminals of a cell. 4. Cell can Package the internal components of the battery and provide protective components to prevent direct contact with the outside. 5. Cell lid The part used to cover the battery casing, and it usually includes electrolyte injection port, safety valve, and terminal outlet port. a. safety valve gas vent: a structural component designed to release gas inside the battery to avoid excessive internal pressure, and a pressure relief threshold is designed based on the characteristics of the battery. b. electrolyte injection port: The port of controlling the amount and injection time of liquid electrolyte to inject it into the battery. The main purpose is to form ion channels, achieving reversible charging/discharging. c. terminal outlet port: the output port of negative terminal and positive terminal. 6. Gasket insulator An insulating material placed between the electrode group and the bottom of the cell can to strengthen insulation. 7. Washer The material made of insulating material, used in conjunction with the cell lid, is fixed on the outside of the cell lid to provide insulation. 8. Protective tape To prevent short circuit between positive and negative electrodes and provide insulation protection, including tab protection tape, electrode protection tape, bottom tape, and side tape. 9. Positive Temperature Coefficient (PTC) element Semiconductor materials or components with high positive temperature coefficients of the resistance, protecting cell from overcharge or overheating.

2.2

Connections

Besides the basic structural parts of a single cell, the connections design for the current input (power supply) and the current output (load) of the cell is also a vital aspect. As mentioned previously, positive terminal, negative terminal, and tab are the three main parts attached to a cell. While the connections outside a cell are realized by contactors, power cable, fastening connectors (stud, nut, and washer).5

Cell and Battery Design – Batteries | Electrical Management

659

1. Contactor Contactor is defined as a high-power relay on the battery current power input or output path. For example, pre-charge contactor is a contactor designed for charging/discharging of battery cells to avoid surge current. The connectors on the current path can only use the components specified in the engineering design to avoid electrical arcing and further fire accidents, and need to be fixed to avoid electromagnetic compatibility problems. 2. Cable The selection of electrical connection wires is equally important, with copper cables and aluminum bus bars being better choices. The connection method of the cables will also affect the subsequent current sensor, current splitter configuration, and consideration of whether the current flows in both directions. 3. Fastening connector Generally, the flat washer or Screw and Washer Assemblies (SEMS) is used for fastening the battery terminal. The flat washer comes from the automotive industry, and the open washer comes from the battery manufacturing industry. At present, due to the different expansion rates of various metals, the open washer can form a continuous and effective pressure at the cell terminal. In addition, the nut on the fixed inner stud should not pass over the stud, otherwise, the electrical connection will pass through the nut instead of the stud, resulting in an increase in internal resistance, and the current will pass through the thread lock between the stud and nut.

2.3

Topology

Cells can form battery modules and clusters through a variety of topological structures, and battery manufacturers will assemble them in series, parallel or a combination of series-parallel to meet specific needs. The unit in series can meet higher voltage requirements, and the unit in parallel can meet larger capacity requirements. The common topology is first in parallel and then in series, or first in series and then in parallel. In terms of stability, the method of first parallel and then series is higher than the method of first series and then parallel, but in actual use, for example, in case of short circuit, the method of first parallel and then series will lead to short circuit of the cell in parallel with the short-circuit cell, and the remaining series cell will bear greater load.6 In addition, after the cells are grouped according to different topologies, the battery module will also include metal cover plate, harness, adhesive, module control unit, etc., which will also affect the number of external connected devices, such as current and voltage sensors, and electrical control boards, etc. as shown in Fig. 3.1 Battery group design should consider minimizing current and conduction losses. However, due to the inconsistency of the battery (capacity, resistance, self-discharge rate, etc.), the mismatch rate of the series cells increases, resulting in the uneven degradation of the whole module, which involves a major control problem of the Battery Management System (BMS) hardware, namely battery equalization.

2.4

Module/pack shell

The mechanical shell of batteries and battery packs need to meet the requirements of safety, durability, battery life, and thermal management.7

Fig. 3 Cell topology and its influence on the number of external connected devices, (a) parallel cells share a control board, (b) every cell has a control board.

660

Cell and Battery Design – Batteries | Electrical Management

1. A small amount of preload force on batteries (whether pouch cell or prismatic cell) can suppress electrode gradient and layering, which helps with battery life, and winding can also generate preload force. 2. Having a certain degree of mechanical rigidity, the battery can be fixed under the worst-case vibration load conditions that can be expected 3. The shell of modules and other components must withstand certain mechanical external forces (such as compression and collision). 4. Capable of withstanding high drops or being able to withstand collisions while the vehicle is in motion. 5. Capable of guiding battery exhaust to a safe exit, even in the event of battery failure.

3 3.1

Systems hardware BMS category

In addition to the hardware of the battery cell module, the hardware of the battery system is the main consideration of the whole BMS. A typical BMS construction is shown in Fig. 4. However, BMS varies to different forms and types in different manufacturers. In general, the system hardware should consider the battery module or battery pack as the management object, and form a complete monitoring and management system from the power supply, control unit, signal acquisition, cell monitoring, and communication modules.

3.1.1 Functional types The main functions of BMS include battery status monitoring, active safety and thermal management, battery balancing, charging control, data communication storage, electromagnetic compatibility, and so on,8,9 as shown in Fig. 5. According to the different main functional modules, it can be divided into chargers, monitors, equalizers, and protectors, which are introduced as follows.

12 V

Power management Protection and signal conditioning Voltage regulators

HV power interface High-side switches Interface Optional ring config SPI interfaces

Pre-charge contactor Contactor + Contactor -

Isolated transceivers

Battery management Control Unit

Cell management 1

MCUs

Battery pack Sensing V,I,T

Battery 1 48V-800V

CAN bus

Wired connectivity

Serial EEPROM

Cell management n

CAN transceivers

Sensing

Battery n

V,I,T

Interface

CAN dataline protection SPI interfaces

Isolated transceivers

BMS main controller

Fig. 4 A typical automotive battery management system (https://www.st.com/en/applications/electro-mobility/automotive-battery-management-system-bms.html).

Cell and Battery Design – Batteries | Electrical Management

661

Fig. 5 The main function of BMS.

1. Charger The simplest BMS should have the ability to control battery charging and discharging, even with a constant current and voltage charger. In addition, some signal acquisition functions such as battery voltage and voltage can be integrated to prevent overcharging of the battery,5 as shown in Fig. 6. 2. Monitor In addition to simple charge and discharge control, existing BMS has more functions for monitoring the status of battery modules and even individual cells, as well as monitoring devices. A monitor is a module that monitors parameters. A monitor generally includes the sensors required to measure battery data, and the main computing unit with software embedded to estimate parameters,5 as shown in Fig. 7. 3. Equalizer or balancing unit If the battery pack involves a certain scale of individual groups, then the equalizer is an essential functional module of BMS. Equalizer is essentially a power electronic converter,5 as shown in Fig. 8. The complexity, efficiency, and speed of balanced circuits are the three main factors to consider during design. Balance circuits include passive balancing and active balancing, which can be referred to Section 3.4.1 in this chapter. 4. Protector The main function of the protector is to turn off the current and is usually integrated into the BMS control unit, responsible for power control. The protector usually includes power switches, such as solid-state switches such as transistor and MOSFETs, which are used to cut off the current path if the protection condition is triggered during the charging and discharging process,5 as shown in Fig. 9.

Fig. 6 The structure of charger.

Fig. 7 The structure of monitor.

662

Cell and Battery Design – Batteries | Electrical Management

Fig. 8 The structure of equalizer.

Fig. 9 The structure of protector.

3.1.2 System Topology The battery system can be divided into distributed (Table 1a), centralized (Table 1b), and modular (Table 1c)one according to the topology type. The currently commonly used topology is centralized, as it requires fewer control units and has lower costs.10 The selection of the three topology types depends on the number of cells. For example, with the high-voltage requirement of batteries such as power battery in electric vehicles (EVs), modular topology (Table 1c) is gradually applied.

3.2

Monitoring and communication

The BMS monitors the status of battery modules, cell units, and configured components throughout their entire lifecycle. For batteries, key parameters such as SOC and SOH are necessary estimates for the system.11,12 Hence, voltage, current, temperature, SOC, SOH, and so on are measured or estimated through the corresponding hardware. For other components of the system, the main functions of system monitoring are whether the input and output signals of the module are abnormal, whether the power supply is abnormal, and whether the power exceeds a set limit.13,14 In addition to software estimation and data processing, the system hardware needs to meet certain conditions to support the implementation of various monitoring functions, mainly as shown in Table 2. In addition to monitoring the critical state of the system, collaborative operation with other external modules requires communication. Communications are primarily realized by Controller Area Network (CAN) bus. Under the communication protocol definition of the central control unit, various components within the system interact with each other through bus deployment, serving as inputs and outputs to achieve functions such as data reading, storage, and outsourcing.

3.3

Safety and Thermal management

3.3.1 Active safety and fault diagnosis Ensuring safety is the primary task for all systems during operation, and as a complex electrochemical component, battery safety monitoring and thermal management are important considerations for the system. The hardware related to system security functions such as fault diagnosis, safety warning, and risk assessment has been designed and developed.15,16 The general requirements for safety thermal management hardware are shown in Table 3.

Table 1 No.

Distributed, centralized, and modular topology of battery system. Topology

Characteristics • Every cell has a sub controller, for signal acquisition and equalization; • The sub controller communicates with the main controller via bus; • The main controller is responsible for monitoring the status of the battery pack.

(b)

• Only one main controller; • Cell and module signal acquisition and equalization are all completed by the main control; • Save the number of controllers and bus layout.

(c)

• Fixed cell units share one sub controller; • A main controller for overall control; • Strong scalability, balancing the advantages and disadvantages of the distributed and the centralized one.

Cell and Battery Design – Batteries | Electrical Management

(a)

663

664

Cell and Battery Design – Batteries | Electrical Management Table 2

An example of work conditions of monitoring hardware.

No.

Index

Values

No.

Index

Values

1 2 3 4 5 6

processor type clock frequency Internal RAM Flash EEPROM power supply

32bit/64bit chip 80 MHz 256 Kbyte 4 Mbyte 16 Kbyte 9–32 V DC

7 8 9 10 11 12

humidity altitude pressure temperature Rest mode low power mode

15%  90%RH −100–5000 m 56.9–106.3 kPa −40  C  +85  C 1 mA 5 mA

Table 3

The list of safety hardware functions.

No.

Index

Design requirements

1

MCU

2

Power function

3

Communication function

4 5 6

Driver function Input detection Charging function

7

current sampling

8 9

Storage function real-time clock

10

High voltage relay

• Low voltage diagnosis • EEPROM verification fault diagnosis • External FLASH verification fault diagnosis • Diagnosis of overvoltage/undervoltage • Diagnosis of voltage source category • Diagnosis of the source of fault/category • CAN communication failure (no timeout/BUS OFF) • Communication failure of sampling, control (no/timeout) • Abnormal communication faults in the main MCU Short-circuit/overload/overcurrent/over-temperature/overvoltage Input detection fault diagnosis • Charger temperature (over temperature/low temperature) • Sensing short circuit to ground fault • PE broken needle detection • Electromagnetic lock fault (short circuit/overload temperature) • Diagnosis of sensor harness breakage • Short circuit diagnosis of sensor harness (to ground) Block Damage Diagnosis • Battery level self-diagnosis • Diagnosis of data anomalies Diagnosis of sticking/breaking faults

3.3.2 Thermal management Batteries operate under wide temperature range environmental conditions and require temperature management to maintain maximum performance of each individual cell within a given reference temperature range. Thermal management mainly considers three aspects: low-temperature heating, high-temperature cooling, and temperature consistency of battery pack.17,18 1. High-temperature cooling Battery cooling is aimed at addressing the accelerated aging of batteries under high temperature conditions and the potential for overheating to cause malfunctions. It is necessary to cool the battery to eliminate the heat generated during operation. Research has shown that for every 10  C temperature rise in batteries, the heat increases exponentially, even leading to irreversible thermal runaway. The commonly used cooling methods currently include gas, liquid, refrigerant, or phase change materials. And the hardware design of cooling channel is the core of the cooling system development. 2. Low-temperature heating On the other hand, operating a battery below low temperature (0  C) weakens its electrochemical activity and it cannot release enough energy on discharge and leading to serious battery degradation phenomena such as lithium deposition during charging. Therefore, rapid low-temperature heating of batteries is also an important part of thermal management. The hardware method mainly involves adding external electric heaters, heat pumps, and other methods to quickly heat up the battery. 3. Temperature consistency of battery pack As for the consistency of battery pack temperature, it is the imbalance caused by the influence of temperature on current. The thermal distribution is always considered when the hardware structure of battery pack is designed.

Cell and Battery Design – Batteries | Electrical Management 3.4

665

Balance and control

3.4.1 Battery balance It was mentioned previously that due to cell inconsistency, there is differential aging of individual cells within the module, and the capacity of the module or battery pack is limited by the most severely aged cells. If cell balancing is not carried out, the battery pack will quickly deteriorate to an unusable level.19 Battery balance can be divided into two main types: active equalization and passive equalization.20 Passive equilibrium dissipates excess heat energy through resistance. Active balancing transfers heat energy from high-voltage batteries to low-voltage batteries through more complex devices such as capacitors, inductors, and transformers, achieving voltage or energy balance of battery modules. The balanced cell can recycle during the charging and discharging process, maximizing the synchronous aging rate, thereby improving the cycle life of the entire battery module. The design of battery balance should concern about the complexity, efficiency, and speed of the balance circuit.

3.4.2 Charging control Charging control is mainly responsible for controlling the charging and discharging behavior of the battery. With the development of battery system functions, charging control is no longer just a simple constant current and constant voltage (CC-CV) charging strategy.21 The existing intelligent BMS will be strategically optimized and upgraded to guide the operation of this functional module, such as the emergence of multi-stage charging strategy or pulse charging strategies. For the hardware upgrade and improvement of the charging module, the existing BMS charging interface meet certain conditions. For example, Table 4 is a typical design requirement of charging interface, which includes operating temperature range and accuracy, signal detection function of electromagnetic control, charging inspection function, and so on. Chargers have several connection ways according to the power level of chargers, as shown in Fig. 10.

Table 4

An example of operating conditions of charging hardware.

No.

Index

Design requirements

1

Charger temperature

2 3

Electromagnetic lock control Charging interface connection

Number of detection channels: 6 channels range: −40–125  C; accuracy: −30  65C  2  C Drive capability>3A; feedback signal detection AC/ DC charging inspection

Fig. 10 Several battery connection ways with different power chargers.

666

4

Cell and Battery Design – Batteries | Electrical Management

Conclusion and outlook

This chapter presents the hardware aspect of battery and its system. Battery hardware and BMS hardware are introduced with details, separately. Battery hardware includes cell structure, connections, series-parallel topology, and module shell. BMS hardware includes four functional categories and system topology, then three main aspects of BMS hardware are introduced furtherly, that is, monitoring and communication part, safety and thermal management part, and balance and control part. Basic knowledge of these hardware parts is provided to construct an overview on the constituent of battery and its system. With the trends of solid-state batteries, high-energy battery, and high-density battery, it will require cell structures to withstand higher levels of current, then higher safety requirements are raised, such as the impact of mechanical stress on the lifespan of batteries. Moreover, battery cover components are developing toward flexibility and multifunctionality. Due to the increasing demand for battery perception, embedded multi-sensor technology will promote the increase of sensing hardware in various parts of the battery cover, interior, etc. In addition, the trend of system intelligence and digitization will stimulate the intelligent upgrade of system hardware, such as the improvement of data communication, accuracy of module acquisition, expansion of hardware interfaces. For processing the massive data collected by the system, expansion of hardware computing power would also be thriving in future.

References 1. Rahn, C. D.; Wang, C. Y. Battery Systems Engineering; John Wiley & Sons, 2013. 2. Izadi, Y.; Beiranvand, R. A Comprehensive Review of Battery and Super-Capacitor Cells Voltage-Equalizer Circuits. IEEE Trans. Power Electron. 2023. 3. Kim, T.; Ochoa, J.; Faika, T.; et al. An Overview of Cyber-Physical Security of Battery Management Systems and Adoption of Blockchain Technology. IEEE J. Emerging Sel. Top. Power Electron. 2020, 10 (1), 1270–1281. 4. Li-Ion Battery Manufacturing Terminology, T/CIAPS0011—2021; 2021. China. 5. Andrea, D. Battery Management Systems for Large Lithium-Ion Battery Packs; Artech House, 2010. 6. Wang, Y. X.; Zhong, H.; Li, J.; et al. Adaptive estimation-based hierarchical model predictive control methodology for battery active equalization topologies: Part I–Balancing strategy. J. Energy Storage 2022, 45, 103235. 7. Santhanagopalan, S.; Smith, K.; Neubauer, J.; et al. Design and Analysis of Large Lithium-Ion Battery Systems; Artech House, 2014. 8. Naseri, F.; Karimi, S.; Farjah, E.; et al. Supercapacitor Management System: A Comprehensive Review of Modeling, Estimation, Balancing, and Protection Techniques. Renew. Sustain. Energy Rev. 2022, 155, 111913. 9. Zhao, C.; Andersen, P. B.; Træholt, C.; et al. Grid-Connected Battery Energy Storage System: A Review on Application and Integration. Renew. Sustain. Energy Rev. 2023, 182, 113400. 10. Sugumaran, G.; Amutha, P. N. A Comprehensive Review of Various Topologies and Control Techniques for DC-DC Converter-Based Lithium-Ion Battery Charge Equalization. Int. Trans. Electr. Energy Syst. 2023, 2023. 11. Badawy, M. O.; Sharma, M.; Hernandez, C.; et al. Model Predictive Control for Multi-Port Modular Multilevel Converters in Electric Vehicles Enabling HESDs. IEEE Trans. Energy Convers. 2021, 37 (1), 10–23. 12. Han, X.; Lu, L.; Zheng, Y.; et al. A Review on the Key Issues of the Lithium Ion Battery Degradation among the Whole Life Cycle. ETransportation 2019, 1, 100005. 13. Goetz, S. M.; Peterchev, A. V.; Weyh, T. Modular Multilevel Converter with Series and Parallel Module Connectivity: Topology and Control. IEEE Trans. Power Electron. 2014, 30 (1), 203–215. 14. Han, X.; Feng, X.; Ouyang, M.; et al. A Comparative Study of Charging Voltage Curve Analysis and State of Health Estimation of Lithium-Ion Batteries in Electric Vehicle. Automot. Innov. 2019, 2, 263–275. 15. Feng, X.; Ouyang, M.; Liu, X.; et al. Thermal Runaway Mechanism of Lithium Ion Battery for Electric Vehicles: A Review. Energy Storage Mater. 2018, 10, 246–267. 16. Feng, X.; Ren, D.; He, X.; et al. Mitigating Thermal Runaway of Lithium-Ion Batteries. Joule 2020, 4 (4), 743–770. 17. Zhang, F.; Feng, X.; Xu, C.; et al. Thermal Runaway Front in Failure Propagation of Long-Shape Lithium-Ion Battery. Int. J. Heat Mass Transfer 2022, 182, 121928. 18. Feng, X.; Zheng, S.; Ren, D.; et al. Investigating the Thermal Runaway Mechanisms of Lithium-Ion Batteries Based on Thermal Analysis Database. Appl. Energy 2019, 246, 53–64. 19. Liu, F.; Zou, R.; Liu, Y.; et al. A Modularized Voltage Equalizer Based on Phase-Shift Modulation for Series-Connected Battery Strings. IEEE Trans Ind Electron 2023. 20. Wei, Z.; Peng, F.; Wang, H. An LCC-Based String-to-Cell Battery Equalizer with Simplified Constant Current Control. IEEE Trans. Power Electron. 2021, 37 (2), 1816–1827. 21. Tomaszewska, A.; Chu, Z.; Feng, X.; et al. Lithium-Ion Battery Fast Charging: A Review. ETransportation 2019, 1, 100011.

Cell and Battery Design – Batteries | Hardware Waleri Milde and Stephan Lux, Fraunhofer ISE, Freiburg im Breisgau, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 1.1 1.2 1.3 2 2.1 2.1.1 2.1.2 2.1.3 2.2 2.2.1 2.2.2 2.3 3 3.1 3.1.1 3.2 3.2.1 4 4.1 4.1.1 4.1.2 4.1.3 4.1.4 4.2 4.2.1 4.3 5 References

Introduction Structure and integration of the BMS BMS and interfaces to the power electronics Integrated interfaces and communication protocols Functions of a battery management system Measurement Measurement of the cell voltage Measurement of the cell current Measurement of the cell temperature Safety Communication and shutdown Cell balancing Extra functionalities System concepts Modular approach Application Centralized approach Application State estimation State of charge estimation Coulomb counting Kalman filter H-Infinity filter Electrochemical impedance spectroscopy (EIS) State of health estimation Methods for SOH estimation Prediction remaining useful life Outlook

668 669 670 670 670 671 671 671 672 672 673 673 673 674 674 675 675 676 676 677 677 678 679 679 681 681 682 682 683

Abstract Lithium-ion batteries are becoming increasingly popular, and it is impossible to imagine today’s world without them. Every lithium battery needs a battery management system for monitoring and control. Care is required in large systems such as electric cars and home storage systems, as failures and damage can have catastrophic consequences. The battery management system measures all critical parameters of a battery at fixed intervals. These are voltage, current and temperature. As soon as one of these parameters leaves its valid value range, the battery management system must intervene and disconnect the battery from the system in an emergency. Accurate state of charge determination is also essential at the same time. The state of charge of a battery cannot be measured directly at the battery but must be calculated. There are various methods for this. The best known are Coulomb Counting and the Kalman Filter. Each of these methods has its advantages and disadvantages and it must be decided for each system individually which method is used. Stability, reliability, and real-time capability are of particular interest in electromobility. Research is far from finished here and better measurement methods are being developed all the time. More sophisticated battery models supported by artificial intelligence are becoming more widespread and more important as the battery market grows stronger and faster.

Glossary AI Artificial Intelligence refers to the development of algorithms and computer technologies capable of performing tasks that typically require human intelligence. This encompasses abilities such as problem-solving, learning, language understanding, visual perception, speech recognition, and decision-making. Algorithm An algorithm is a step-by-step procedure or set of rules designed to perform a specific task or solve a particular problem. In computer science and mathematics, algorithms serve as a systematic approach to solving problems and achieving objectives.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00253-X

667

668

Cell and Battery Design – Batteries | Hardware

Battery management system A Battery Management System is an electronic control unit utilized in battery packs to monitor and control their performance, lifespan, and safety. It is a critical component in battery storage systems, electric vehicles, and other applications employing rechargeable batteries. Coulomb Counting Coulomb Counting is a method used in battery management systems (BMS) to estimate the state of charge (SOC) of a battery by keeping track of the total electric charge (in coulombs) that flows into and out of the battery during charge and discharge cycles. Electrochemical impedance spectroscopy Electrochemical Impedance Spectroscopy (EIS) is a technique used in electrochemistry to study the electrical properties of electrochemical systems. It involves applying a small amplitude alternating current (AC) signal to a system and measuring the resulting voltage response over a range of frequencies. Kalman filter The Kalman filter is a mathematical algorithm widely used in signal processing and control systems. Its primary purpose is to estimate the state of a system based on a series of measured values that are subject to noise and inaccuracies. The filter optimizes the estimation by considering both the current measurements and the previous estimates of the state. MCU A Microcontroller Unit (MCU), often simply referred to as a microcontroller, is a compact integrated circuit that contains a processor core, memory, and programmable input/output peripherals. MCUs are designed for embedded systems and are widely used in various applications, ranging from consumer electronics to industrial automation. Warburg resistance The Warburg impedance or Warburg resistance is a term used in electrochemical impedance spectroscopy (EIS) to describe a diffusion-related impedance element in electrochemical systems. It is named after German chemist Otto Warburg, who made significant contributions to the understanding of electrochemical processes.

Key points

• • •

1

Design guidelines for battery management systems Description of features for battery management systems in large systems Identify potential risks in battery systems

Introduction

In the field of advanced battery storage technologies, battery management systems (BMS) play a crucial role. A BMS is responsible for monitoring, controlling, and protecting batteries and plays an important role in maximizing their performance and lifetime. There are several market segments in the stationary sector, including home storage, commercial and industrial storage, and neighborhood storage as well as utility-scale storage. Each segment has specific requirements for BMS design, integration, interfaces to power electronics, and performance parameters. In this chapter, we will look at BMS in the stationary sector and take a closer look at the market segments and their differences and requirements.1 Home storage systems are batteries, typically combined with PV systems and used in private households. They allow homeowners to store and use excess energy from renewable sources to increase self-sufficiency and therefore reduce electricity purchases from the grid. Home storage systems typically have a much smaller capacity compared to commercial or industrial storage systems and are designed to meet the energy needs of a household. The BMS for home storage must be cost-effective, easy to integrate and to operate. It should have interfaces that allow it to communicate with power electronics and the energy management system.2 Commercial and industrial storage systems are also behind-the-meter applications but typically provide much bigger capacities and power rates. They are used to store and harness energy to provide peak load balancing, grid relief, and energy recovery besides PV self-sufficiency. The BMS for these applications must be able to monitor and control large amounts of energy and depend on the operating control strategy, also high-power rates. It should include advanced features to monitor and predict battery performance, diagnose faults, and optimize operations. Integration of the BMS with the enterprise energy management system is critical to enable seamless control and use of the storage. Neighborhood storage systems, also called district storage systems, are energy storage systems designed for the shared use of energy in a specific residential neighborhood or building complex. They enable the exchange of surplus energy between individual residential units, supporting efficient energy use in the district. Neighborhood storage systems can have different technologies and capacities, depending on the requirements of the district and the number of participating households. The BMS for neighborhood storage should have communication interfaces to coordinate energy exchange between storage units. It should also provide a function to monitor the overall system and optimize energy use.3 Utility-scale storage systems are so called front-of-the-meter applications, typically with large capacities and power rates from a couple of MWh/MW up to the range of GWh/GW nowadays. They can be installed as stand-alone applications to provide control power (e.g., Frequency Containment Response FCR). As energy markets are getting more and more volatile this kind of battery storage is used for participating in the intraday and day ahead markets (arbitrage). Nowadays, PV and wind power plants are also

Cell and Battery Design – Batteries | Hardware

669

combined with battery storage. In such hybrid solutions various technical grid integration topics can be addressed and new business models can be enabled.4 For utility-scale applications highly sophisticated and reliable battery management systems are needed to secure safe, reliable, and high performance for the various operating control strategies. Especially, they must provide precisely inner state value and the current performance capabilities as well as prediction of the remaining useful lifetime in dependence of the several use-cases. Furthermore, they must consider effects of operating modes on the aging and communicate this information to the energy management system for economical evaluation and optimization to increase the return of investment. Battery management systems for electric vehicles (EVs), have a unique role and specific requirements compared to stationary storage. The BMS in automotive storage systems is critical to the performance, safety, and lifetime of the battery in electric vehicles. It plays an important role in ensuring vehicle range, monitoring the state of charge, and ensuring an efficient and reliable power supply while the vehicle is in operation. Automotive storage systems must have high power density to provide sufficient power for acceleration and vehicle operation. The BMS must be able to handle high currents and control the flow of energy in real time to ensure optimal performance. The range of an electric vehicle is directly dependent on the capacity and efficiency of the battery. The BMS must closely monitor the battery’s state of charge and provide accurate range predictions so drivers can reliably estimate the remaining range. Safety is a top priority in automotive storage systems. The BMS must protect the battery from overcharging, deep discharging, overheating and short circuits to avoid possible fires or other hazards.5

1.1

Structure and integration of the BMS

The structure of the BMS usually includes the following components: Sensors and Meters: the BMS uses sensors and meters to collect data such as voltage, current, temperature, and other parameters from the battery. This data serves as input for the BMS’s calculations and decisions. Control and monitoring unit: The BMS has a control and monitoring unit that analyzes the collected data, monitors the state of charge, controls the operation of the battery, and takes protective measures to ensure safe and efficient use of the battery. Communication interfaces: The BMS has communication interfaces to communicate with other systems, such as the building’s energy management system or other storage units in the neighborhood. This allows the BMS to be integrated into the overall system and facilitates monitoring and control of battery operation. The BMS in the stationary sector requires interfaces to the power electronics to control the energy flow between the battery and the power grid or consumers. This includes interfaces such as direct current to alternating current (DC-AC) converters or alternating current to direct current (AC-DC) converters that adjust the current flow and optimize energy efficiency. The BMS in automotive storage systems is usually closely connected to the vehicle’s power electronics (Fig. 1). It monitors the battery’s state of charge, voltage, current and temperature, and communicates with the engine control unit and the power electronics system to regulate the flow of energy. The BMS also controls the vehicle’s charging to protect the battery from damage and ensure optimal charging efficiency.

Fig. 1 The illustration shows a schematic diagram of a battery management system consisting of several modules, a current sensor and communication via CAN bus. The BMS is designed for an automotive application, with inverter, vehicle controller, and charger.

670 1.2

Cell and Battery Design – Batteries | Hardware BMS and interfaces to the power electronics

Interfaces to power electronics are crucial to control and optimize the flow of energy between the battery and other systems, such as the power grid or consumers. There are different types of interfaces: The most important interface is the bi-directional inverter. It converts the direct current generated by the battery into alternating current to suit the connected AC loads. The alternating current to direct current (AC-DC) inverter mode converts the alternating current from the power grid or other power sources to direct current to charge the battery. This inverter enables the control of the charging process and the adjustment of the charging current to the battery parameters provided by the battery management. Another important type is the bidirectional DC-DC converter. It allows the conversion of the direct current between different voltage levels. This interface is used to adjust the battery voltage to the requirements of the connected loads or to connect several batteries with different voltages. On the generation side, a bidirectional DC-DC converter allows the DC coupling of a PV system. This enables efficient use of battery energy. In addition to pure power interfaces, communication interfaces are also important to connect the battery management system (BMS) to other relevant components of energy systems. This enables the transmission of data, status information and control commands for efficient, safe, and reliable control of battery operation and integration into the overall system. When selecting the right power electronics interfaces for a particular stationary battery storage system, factors such as capacity, capability, operating voltage, battery type and power flow requirements must be considered. The right interfaces enable efficient use of battery energy, optimized charging and discharging, and seamless integration into the power system.

1.3

Integrated interfaces and communication protocols

The battery management system is a complex device that requires various connections to monitor, control and communicate with the battery and other components of the storage system. Some of the common connections and communication interfaces are explained below. The BMS is directly connected to the battery cells or battery modules to monitor and control the battery current, voltage, and temperature. Temperature sensors are usually directly connected to the BMS, placed near the battery cells or battery modules to monitor battery temperature and control temperature management. Current sensors can be installed directly in the module and connected to the BMS. Usually, due to cost reasons just one current sensor is used in a serial string of battery modules. For larger systems with high currents, it is also a good idea to use an external sensor and connect it directly to the main computing unit. In small systems, like smartphones or other consumer products, the current can be measured directly on the board, since the currents are small. The BMS is equipped with various communication interfaces to communicate with other components of the storage system and the power grid. These include CAN bus (Controller Area Network), RS485, Ethernet, Modbus, or wireless protocols such as Bluetooth or WLAN. CAN bus is a widely used bus system in automotive applications, including BMS for electric vehicles. It provides reliable and robust data communication between various components of the vehicle and enables the rapid transfer of data between the BMS and other vehicle systems.6 RS485 is a serial communication standard commonly used in industrial applications. It allows communication over long distances and is well suited for use in stationary storage systems. Ethernet is a standardized communication protocol that offers a high data transfer rate and is used in many industrial applications. It enables fast and reliable communication between different components of the BMS and other networks. Modbus is another widely used communication protocol in industrial applications. It enables communication between different devices and systems via serial interfaces.7 Wireless communication capabilities, such as Bluetooth or WLAN, are increasingly being integrated into BMS to enable flexible data transmission and remote monitoring. Power Line Communication (PLC) enables communication over the existing power supply infrastructure and is used in some applications for data transmission between BMS components.8,9

2

Functions of a battery management system

A battery management system (BMS) for advanced battery storage technologies, such as lithium-ion, must ensure a wide range of tasks (Fig. 2). The main task of the BMS is to ensure safe operation at any time. For this purpose, the operating limits of the cells must be considered, and the BMS is responsible for safely shutting down the battery system in the event of a fault. Cell limits are usually based on the cell manufacturer’s information. In advanced approaches the BMS will also be fed with data from precise cell tests to detect and, if possible, reduce further hazard to the cells, such as possible lithium plating at high currents and low temperatures. The BMS can detect further faults in the battery system. These can be differences in temperatures or voltages of single cells occurring especially in large systems with a huge number of cells. The measured data can be stored in the BMS or transmitted to other components of the energy system; besides onboard data processing a more sophisticated evaluation of these

Cell and Battery Design – Batteries | Hardware

671

Fig. 2 The illustration shows the basic functions of a BMS, divided into four categories. It is divided into measurements, safety, control, and additional features.

measured values can be made via the so-called cloud computing remotely. The options for recording data and storing it locally or forwarding it to higher-level computing units vary greatly depending on the BMS type and battery storage application. As mentioned above, a communication interface is required to connect the BMS with external units such as the inverter or an external energy management system. Furthermore, a BMS may also contain functionalities for direct control of external units like pumps, relay, or fans. Beyond that, there are high level functionalities to calculate values such as inner state values and information on the application specific remaining useful lifetime. This information must be provided for the user. The BMS in automotive storage systems is usually closely connected to the vehicle’s power electronics. It monitors the battery’s state of charge, voltage, current and temperature, and communicates with the engine control unit and the power electronics system to regulate the flow of energy. The BMS also controls the vehicle’s charging to protect the battery from damage and ensure optimal charging efficiency. Interfaces to the vehicle’s power electronics mainly include communication with the engine control unit and the inverter, which converts the battery’s direct current into alternating current for the electric motor. The BMS gives commands to the inverter to regulate the flow of energy according to driving conditions and battery condition. It ensures optimal vehicle performance and range while maintaining battery safety.

2.1

Measurement

The recording of current, voltage and temperature of each cell or cell block in a battery system represents the core task of a battery management system. On the one hand, this is always the basis for safe operation; on the other hand, this data enables further evaluation to derive higher-level values such as inner states, e.g., state of charge (SOC), or to validate the predicted properties of the system in operation.

2.1.1 Measurement of the cell voltage Measuring the cell voltage in the battery system is a demanding task. Voltages must be measured accurately in the range of mV; the environment is EMC loaded and the occurring currents can be superimposed with harmonics caused by the connected power electronics. The cells are connected in series in the string, which makes galvanic decoupling necessary to keep high voltages away from the measurement electronics. There are many integrated circuits on the market that are designed for this task, and they bring along some more functionalities that are described in this chapter. For example, the LTC 6813–1 from Analog Devices or the MC33774 from NXP can measure the voltage of 18 cells with a maximum measurement error of 2.2 mV and a typical error below 1 mV for most temperature ranges.10,11 The chip is transmitting the signal in digital form via a galvanic isolated isoSPI Bus (two-wire bus). It is described as a multicell battery stack monitor and in a large stationary storage system these circuits may be coupled via iso-SPI in a daisy chain with one host processor connection. They gain their supply voltage directly from the cells measured in the module with a supply current in the sleep mode of 6 mA. Thus, the values of a module can be perfectly recorded and transmitted.

2.1.2 Measurement of the cell current Cell current can be sensed by a variety of sensors, e.g., the LTC 6813-1 allows a Hall-effect sensor to be connected to its GPIO inputs to measure current in sync with cell voltage. However, the current is usually measured using so-called shunts as precise measurement resistors that allow the measurement of a voltage drop. Both, simple resistors, and integrated solutions are available. These integrated solutions (e.g., IVT-S series from Isabellenhütte12) usually bring along an AD converter and provide the information

672

Cell and Battery Design – Batteries | Hardware

about the current, galvanically isolated (1000 VDC) via CAN bus. Measuring values are transmitted every 10 ms. Additionally, up to three system voltages and the system temperature can be measured independently and the values are transferred to the BMS via CAN Bus. However, it should be noted that in battery systems generally the total current is measured, the distribution of the currents (string currents) in parallel connected strings is not recorded. In the case of the division of the currents in the cell block (a parallel connection of individual cells to increase the capacity), no commercial solution is known that allows the individual current to be recorded.

2.1.3 Measurement of the cell temperature Cell temperatures provide crucial information about the status of battery systems and have an influence on specific battery parameters such as the currently available capacity and the current inner resistance. It must be noted that the surface temperature that is measured is different from the core temperature, especially for larger cells. It is recommended to measure each single cell temperature for safety reasons, in real life this is done usually when large cells are used. In battery systems with small cells this is not the case, due to cost reduction pressure. E.g., in systems using parallelized cells (mostly 18,650 type) up to one sensor per cell block is used. The sensor might be NTC, PT 100 or comparable. The LTC6813 mentioned before provides 9 general purpose I/O (GPIO) that can be configured as analog input for temperature measurement. Additionally, by using three of these GPIO, a multiplexer may be added that expands the measurement of analog signals on 16 channels.10

2.2

Safety

To operate the system in a safe condition at any time, it is mandatory to keep the lithium-ion cells in the so-called operating window. This is a specification that is usually delivered by the cell manufacturer to avoid overcharge, unallowed deep discharge levels, too high charge and discharge currents, severe effects such as lithium-plating as well as overheating. It must be noted that all values given are usually dependent on other parameters, so charge current is usually limited at lower temperatures and charge and -discharge current might also be derated with higher temperatures. The safe operating window of a lithium-ion cell is depicted (Fig. 3). There are several mechanisms in a lithium-ion cell that may lead to hazardous effects if the cell is operated outside the safe operating window 1. Given values for temperature and voltage are arbitrary chosen as there are cells on the market with other specifications. Overcurrent is not depicted in this figure. 1. 2. 3. 4. 5. 6.

Safe operation window Undervoltage (deep discharge), dissolution of the copper anode Lithium plating caused by overcharge (3a) or charging at deep temperatures (3b) Cell failure with overtemperature disintegration of solid electrolyte layer (SEI), thermal runaway possible High temperature and voltage, gassing, fire, smoke, thermal runaway possible Melting of the separator, thermal runaway, oxygen release possible

Fig. 3 The illustration shows the different areas in which a battery operates. The block with the number 1 is marked in green, as this is the safe operating range of the battery. Areas in which the battery can no longer be operated safely and damage may occur are marked in yellow.

Cell and Battery Design – Batteries | Hardware

673

All these effects may overlap, and BMS’ main function is to keep the cell in the safe operating window.

2.2.1 Communication and shutdown Communication plays a crucial role in a Battery Management System (BMS). It is essential to distinguish between external and internal communication. Internal communication involves collecting data from sensors and facilitating communication among these sensors. For instance, frontend ICs such as the previously mentioned LTC1813–1 communicate via an isolated SPI bus, which allows them to be interconnected in a daisy-chain fashion. Each frontend IC is responsible for monitoring one battery module, and all the gathered information is transmitted to the module BMS or directly to the master BMS. Additionally, information of other sensors might be transferred like current, voltage and temperature via CAN bus. The master BMS, in turn, communicates with external units:

• • • •

Power electronics – BMS usually delivers maximum operating values to control power electronics. On the other hand, often BMS system need release signal to come in operation and switch relay on. Energy Management System – BMS internal information is crucial for external energy management system, BMS provides information on internal states such as state of charge, state of health and state of power. Contactor, fans, pumps, pyro fuse – BMS might control contactors that are usually off and fans and pumps for cooling purposes. Usually BMS communicates default values (maximum and minimum values) to keep the battery in a safe state. In case of the malfunction of power electronics or energy management system, this default values can be exceeded and the BMS might derate the power rate of the battery system or switch the battery off by use of pyro fuse or contactor. Data transfer for monitoring, control, and safety purposes. – BMS data is transferred to external server systems (Cloud computing). This brings the benefit of working on a large battery database and allows to work with high computing power. KI and machine learning algorithms are used and might detect severe battery error in an early state.

Communication is not standardized and is e. g. possible via CAN, Modbus over ethernet, RS 485, Bluetooth.

2.2.2 Cell balancing A battery system contains many individual lithium-ion cells. These have slight differences due to production, e.g., internal resistance, self-discharge rate and available capacity can be different. Due to the installation position in the battery system and the possibly uneven current flow, the temperature distributions in the system are also not completely homogeneous. Due to the differences in production and the influences of operation, the charge levels of the batteries and thus the cell voltages tend to diverge. A battery system in which, for example, the state of charge of the individual and in serial switched cells differs by 8% will suffer a loss of capacity at the system level of 8% and will only provide 92% of the capacity, even if each individual cell still has an SOH of 100%. This is since the cell with the highest state of charge limits the charging process even if other cells are not yet fully charged. To avoid this, so-called balancing systems are installed in battery systems which equalize the voltages or the states of charge of the individual cells. A distinction can be made between passive and active balancing systems. Active systems take charge from the cell(s) with the highest state of charge and supply it via appropriate converter stages to the cells with the correspondingly low states of charge, until the system is balanced again. When using passive balancing systems, shunt resistors are usually connected in parallel via FET to the cell or cells with the highest state of charge to discharge the cell or to charge it more slowly than the remaining cells during a charging process. The LTC 6813–1 mentioned above provides up to 200 mA balancing current without using external FET. An active balancing system generally generates less waste heat than a passive system, but its structure is considerably more complex since energy flows must be controlled accordingly. The higher number of components also increases the susceptibility to errors. Furthermore, active balancing systems also result in higher cost. With an improvement in cell quality through mass production as well as with improved module and systems designs allowing more homogeneous operating conditions (e.g., in terms of temperature distribution), the use of balancing systems does not become superfluous, but the balancing currents required, and the waste heat generated in passive systems as a result are smaller. As an outcome, passive balancing appears to be very advantageous and dominates the market in many areas. Modern front-end components already contain the logic and control for passive balancing, so that only a few external components are required for a functional. System.

2.3

Extra functionalities

In addition to the aspects described before, it is possible to implement higher level functionality on board of the battery management system. This might include state of charge algorithm (SOC) and algorithm to estimate the state of health (SOH) and the state of power (SOP) of the single battery module or battery system. However, it is a decision of the system designer if those algorithms are implemented on the BMS itself or on an external processing unit. The operating mode is influencing the aging of a battery system, so long rest times at a high state of charge limit lifetime, especially at elevated temperatures. Also, higher currents at low temperatures, leading to lithium-plating, are detrimental. So,

674

Cell and Battery Design – Batteries | Hardware

optimization of battery operation mode along with energy management functionality might also be implemented on the BMS to set appropriate default values to the power electronics. For several reason data storage might be provided on BMS to capture operational data. On the one hand this gives information of system behavior in case of an error even when data of external data storage or server is not available; on the other hand, this information is available for the qualification of used batteries in 2nd life applications. This is also a requirement of the upcoming EU battery passport as part of the EU battery regulation.13 It is crucial to display information to the user. A BMS might integrate a display, e.g., controlling a hardware display integrated in the battery housing or provide an interface to an external display, this might be an app running on a smartphone connected by Wi-Fi or Bluetooth. The functionality provided might be useful to set up the system and provide operating data to the user or service staff.

3

System concepts

Battery Management Systems (BMS) are essential components in any battery-powered application, providing critical functions such as monitoring, protection, and control. A BMS ensures that the battery pack operates within safe limits, extends the life of the battery pack, and enhances the overall performance and reliability of the system. The concept of a BMS is simple: to monitor and control the individual cells within a battery pack to ensure that the pack operates within safe limits and provides optimal performance. This includes monitoring critical parameters such as voltage, current, temperature, and calculation of inner states, such as state of charge, as well as providing protection functions to prevent overcharging, over-discharging, and critical-temperature conditions. There are two main approaches for BMS design: the centralized approach, where a single central controller is responsible for monitoring and controlling all cells in the battery pack, and the modular approach, where multiple microcontroller units (MCUs) are used to monitor and control specific sections/modules of the battery pack. Both approaches have their own benefits and limitations, and the choice of which approach to use depends on the specific requirements of the application. In this chapter, we will provide an overview of BMS systems and concepts, the two main approaches for BMS design, and the key factors to be considered when choosing a BMS for a particular application. Whether you are a battery engineer, system designer, or end-user, this chapter provides a comprehensive introduction to the world of Battery Management Systems.

3.1

Modular approach

A modular BMS is designed with a distributed architecture, where multiple microcontroller units (MCUs) are used to monitor and control each cell in the battery pack. The MCUs are connected to a main controller, which acts as a coordinator for the system. This approach provides several benefits over traditional monolithic BMS designs, including improved scalability, reduced cost, and enhanced reliability. A decentralized BMS needs a fuse, a current sensor, and a switch (Fig. 4). The BMS controls all peripherals and is connected to only one module controller. The modules are interconnected and send data to each other in daisy chain mode. The BMS receives the data and evaluates it. One of the main advantages of modular BMS is scalability. As the demand for larger and more complex battery packs continues to grow, it is becoming increasingly difficult to manage these systems with a single central controller. With a modular approach, the number of MCUs can be easily increased or decreased to accommodate the size and complexity of the battery pack. This allows for a flexible and scalable solution that can be adapted to meet the changing needs of the system.

Fig. 4 The illustration shows a BMS that was set up with the “distributed BMS” topology. Several battery modules can be seen, all of which are connected to a BMS front-end. The current sensor measures the current of the entire system. Additional safety is ensured by the fuse and the switch.

Cell and Battery Design – Batteries | Hardware

675

Another advantage of modular BMS is cost reduction. By using multiple smaller MCUs instead of a single large controller, the overall cost of the BMS is reduced. This is due to the lower cost per unit of smaller MCUs and the reduced complexity of the design, which reduces the costs associated with development, testing, and certification. Furthermore, the modular approach allows for the use of standard components, which are widely available and less expensive than custom-designed components. Reliability is also improved with the modular approach. By dividing the battery management system into multiple components, the risk of a single point of failure is reduced. This is because each MCU operates independently, and if one unit fails, the others can continue to operate, ensuring that the battery pack remains functional. Furthermore, the use of multiple MCUs can improve the accuracy of the system, as each cell in the battery pack can be monitored and controlled more precisely. Furthermore, this approach also allows the estimation of the inner states of each single cell. One of the challenges associated with modular BMS is the coordination of the MCUs. The main controller must be able to effectively communicate with each MCU and coordinate the monitoring and control of the battery pack. This requires a robust communication protocol that can handle the large amounts of data generated by the system. In addition, the MCUs must be designed to work together in a consistent and reliable manner, which can be a complex task. Despite these challenges, the modular approach to battery management systems is becoming increasingly popular in the industry. Many battery manufacturers are now offering modular BMS as a standard option, and the technology is being used in a wide range of applications, from small portable electronics to large-scale electric vehicle charging stations. In conclusion, the modular approach to battery management systems offers several advantages over traditional monolithic designs, including improved scalability, reduced cost, and enhanced reliability. While there are some challenges associated with this approach, such as the coordination of multiple MCUs, these can be overcome with careful design and implementation. As battery technology continues to evolve and grow in importance, the modular approach to BMS will become increasingly relevant, providing a flexible and scalable solution that meets the needs of a wide range of applications. At least if battery systems are designed in a modular way and cell2pack concepts will still be the exception.

3.1.1 Application As described above, a modular BMS approach is preferred in applications where a high degree of flexibility and scalability is desired. Modular BMS systems are designed with multiple microcontroller units (MCUs), each of which is responsible for monitoring and controlling a specific section of the battery pack. This allows the system to be easily expanded or scaled up to accommodate changes in the battery pack size or configuration. One example of an application where a modular BMS approach is preferred is in electric vehicles (EVs) in case the battery system is designed with several battery modules, where the battery pack may need to be expanded or reconfigured as the vehicle evolves over time. With a modular BMS, it is possible to add or remove individual modules as needed, without affecting the overall performance of the system. Another example is the market segment of large-scale energy storage systems (C&I as well as utility-scale), where the battery pack may consist of hundreds or thousands of individual cells. A modular BMS allows for the monitoring and control of each cell or sub-section of the battery pack, providing ka high degree of flexibility and scalability. Modular BMS systems are also preferred in applications where a high degree of fault tolerance is desired. With a modular BMS, if one module fails, only a specific section of the battery pack is affected, rather than the entire system. This allows for continued operation of the battery pack, with reduced risk of catastrophic failure.

3.2

Centralized approach

The centralized approach to BMS uses a single central controller to monitor and control all cells in the battery pack. The central controller communicates with each cell through a network of voltage and temperature sensors and current monitors. This approach provides several benefits, including improved accuracy and ease of use, as well as reduced cost in smaller systems and simplified design. In a centralized BMS with a fuse, a current sensor, and a switch the BMS controls all peripherals and monitors the battery cells at the same time (Fig. 5). One of the key advantages of the centralized approach is improved accuracy. By using a single central controller, the BMS can monitor and control the entire battery pack with high precision, ensuring that all cells are operating within safe limits. This is particularly important in applications where safety is a primary concern, such as electric vehicles and energy storage systems. Another advantage of the centralized approach is ease of use. With a single central controller, the BMS is easier to design, test, and implement than a modular system, where multiple microcontroller units (MCUs) are used. This is because there is only one central controller to design and test, reducing the complexity of the system and allowing for a more streamlined design process. One of the challenges associated with the centralized approach is the risk of a single point of failure. If the central controller fails, the entire BMS will fail, potentially causing serious problems with the battery pack. To mitigate this risk, the central controller must be designed with high reliability and must be tested thoroughly to ensure that it can operate reliably under a wide range of conditions. Another option is to use redundancies. E.g., put two MCUs on the board and if one fails the other can take over and guarantees a safe operation until maintenance can take place.

676

Cell and Battery Design – Batteries | Hardware

Fig. 5 The illustration shows a central BMS. Here, all cells are connected directly to the BMS master. This takes over the measurements, communication and data processing. A fuse and a switch are also installed here.

Despite these challenges, the centralized approach to battery management systems continues to be a popular choice for many applications. Many battery manufacturers offer centralized BMS as a standard option, and the technology is widely used in a range of applications. In conclusion, the centralized approach for battery management systems provides several benefits, including improved accuracy, ease of use, and reduced cost. While there are some challenges associated with this approach, such as the risk of a single point of failure, these can be mitigated with careful design and testing. The centralized approach continues to be a popular choice for many small-scale applications, and will likely remain so in the future, providing a simple and reliable solution for monitoring and controlling battery packs.

3.2.1 Application A centralized Battery Management System (BMS) is ideal for cost-effective, simple, and precise applications. It employs a single central controller to monitor and control all battery pack cells, resulting in lower costs, streamlined design, and enhanced accuracy. One example of an application where a centralized BMS approach is preferred is in small portable electronics, such as laptops and smartphones, where the battery pack is relatively small, and cost is a major consideration. With a centralized BMS, the overall cost of the system is lower than a modular system, as there is only one central controller required, rather than multiple microcontroller units (MCUs). In addition, the centralized approach is also preferred in applications where ease of use and ease of integration are important. With a single central controller, the BMS is easier to design, test, and integrate into the overall system, reducing the complexity of the system and allowing for a more streamlined design process. In summary, a centralized BMS approach is preferred in applications where cost, simplicity, and accuracy are prioritized. The centralized approach provides a simple and reliable solution for monitoring and controlling battery packs and is a popular choice for applications such as small portable electronics and applications where safety is a primary concern.14

4

State estimation

State estimation in battery systems is a critical process aimed at precisely assessing and managing various key parameters that define a battery’s performance and health. These parameters include State of Charge (SOC), State of Health (SOH), State of Power (SOP), and State of Energy (SOE). SOC represents the current level of charge in a battery, essentially indicating how much energy is available for use. SOH reflects the overall health and degradation of the battery, providing insights into its remaining lifespan and capacity. SOP signifies the battery’s ability to deliver power instantaneously, which is crucial in applications requiring rapid energy release. SOE encompasses the comprehensive state of a battery, encompassing SOC, SOH, and SOP to provide a holistic view of its capabilities. Accurate state estimation is essential for optimizing battery performance, enhancing safety, and prolonging its operational life. This process involves advanced algorithms and sensors to continuously monitor and update these states in real-time, enabling precise control and management of battery systems in applications ranging from electric vehicles to renewable energy storage. In essence, state estimation is the foundation upon which efficient and reliable battery management and utilization are built. In this chapter we will provide an overview of the most common techniques for state of charge estimation of modern batteries.

Cell and Battery Design – Batteries | Hardware 4.1

677

State of charge estimation

The state of charge (SOC), provides information about the current charge content of a battery compared to its actual capacity: SOC ¼

Q Cactual

Accurate and reliable SOC determination allows users to plan energy consumption, optimize battery life, and avoid unexpected failures. In this section, we will look at different methods and techniques for determining the state of charge of a battery and explore their importance for a wide range of applications. The state of charge of a battery can be expressed as a percentage, which indicates how much charge is still available compared to the maximum capacity. It ranges from 0%, which corresponds to a fully discharged state, to 100% for a fully charged state. Accurate knowledge of SOC is important for users of portable electronics as well as for applications in the automotive, industrial and energy sectors. However, determining the state of charge of a battery is not a simple task. It is influenced by several factors, including battery technology, depth of discharge, temperature, aging effects, and the battery management system (BMS) itself. Therefore, accurate SOC determination requires the use of advanced measurement and analysis techniques and sophisticated algorithms. There are several methods for state-of-charge determination based on different principles. These include voltage measurement, current integration, Kalman filtering, impedance spectroscopy, and many others. Each method has its own advantages and disadvantages in terms of accuracy, cost, complexity, and applicability in different battery systems.15–18 Accurate state of charge determination is of great importance for many applications. In automotive applications, accurate SOC determination enables calculation of remaining range and helps drivers better plan their trips and avoid unexpected battery discharges. In portable devices such as cell phones or laptops, SOC is a key indicator of remaining operating time and helps users recharge their devices in a timely manner. In addition, accurate SOC determination is also critical for the efficient use of battery storage systems in the renewable energy industry, e.g., to optimize PV self-sufficiency in behind-the-meter PV battery applications. In this section, we will provide a comprehensive look at the different methods for determining the state of charge of a battery, discuss their advantages and disadvantages, and highlight the importance of accurate SOC determination for a variety of applications. By better understanding the state of charge determination, we can optimize battery performance and improve their use in numerous applications.

4.1.1 Coulomb counting Coulomb Counting, also known as current integration, is a widely used method for determining the State of Charge (SOC) of a battery. It is based on the principle of conservation of charge and calculates the charge using current measurement while charging and discharging. In this chapter, we will take a closer look at Coulomb Counting by examining its advantages, disadvantages, complexity, and required computational power. The formula for calculating the SOC using the Coulomb Counting method is: Z t iðt Þ dt SOCðt Þ ¼ SOCðt 0 Þ + t 0 Qnom Where: SOC(t) is the actual state of charge at time t, SOC(t0) is the state of charge at time t0, i(t) is the current and Qnom is the nominal capacity of the battery. Aging effects can be included in the value of Qnom which means that the estimated value of the SOC of a new cell can be different than the SOC of an aged cell. The integration of the current flow on the MCU can be performed using an integrator or using a mathematical model. Example Assuming the battery has a rated capacity of 100 Ah and the current flow is 5 A (discharge currents have a negative sign). When the battery is fully charged, the SOC is 1 or 100%. After 1 h, the SOC is calculated to: SOC ¼ 1 +

− 5A1h ¼ 0:95 100Ah

SOC ¼ 1 +

− 5A2h ¼ 0:90 100Ah

After 2 h, the SOC is calculated:

4.1.1.1 Advantages of coulomb counting Coulomb Counting is comparatively simple compared to some other state of charge determination methods. It does not require complex measurement instruments or elaborate calculations. The basic idea is to measure the current and integrate the charge introduced into or extracted from the battery to calculate the SOC. In Coulomb Counting, the measurement is made directly on the battery via the current sensor. This avoids measurement errors due to resistance losses or voltage fluctuations that can occur with indirect measurement methods.

678

Cell and Battery Design – Batteries | Hardware

Implementation of Coulomb Counting generally does not require expensive measurement equipment or special sensors. However, the use of an accurate current sensor is important to ensure accurate measurements. 4.1.1.2 Disadvantages of Coulomb Counting Coulomb Counting assumes that the calculation of the charge introduced or removed based on an accurate current measurement. However, cumulative errors can occur when small measurement errors accumulate over time. This can result in low accuracy for long-term state-of-charge tracking. Temperature influences the accuracy of Coulomb Counting. Temperature dependent effects such as the change of the measurement resistance of the sensor can affect the measurements and lead to inaccuracies. 4.1.1.3 Complexity and computing power required The Coulomb Counting approach does not require complex calculations. The main task is to integrate the measured current and sum the charge over time. Modern battery management systems (BMS) can perform these calculations in real time and continuously update the SOC based on the measured currents. The computing power required depends on the accuracy and desired update rate of the SOC. The more frequently the SOC is to be updated, the more computing power is required to perform real-time charge integration. In summary, Coulomb Counting provides a simple and cost-effective method for determining the state of charge of a battery. Although there are some drawbacks and challenges, it has become a useful tool for monitoring and controlling batteries in stationary storage and other renewable energy applications.

4.1.2 Kalman filter The Kalman filter is an advanced method for determining the State of Charge of a battery. It is a mathematical algorithm based on statistical estimation that provides an optimal estimate of the current SOC, considering measurements and system models. In this chapter, we will take a closer look at the Kalman filter by examining its advantages and disadvantages, complexity, and required computational power.19 The Kalman filter works with two estimates: The updated estimate is the best estimate of the SOC based on the available information. The future estimate is the estimate of the SOC based on the updated estimate and the expected future development of the SOC. The Kalman filter updates the updated estimate using the following formula:   Updated estimate ¼ current measurement ∗ measurement uncertainty + updated estimate ∗ updated estimate uncertainty = ðcurrent measurement uncertainty + updated estimate uncertainty Þ The future estimate is created using the following formula: future estimate ¼ updated estimate + expected future development The current measurement is the measured current flow. Measurement uncertainty is the uncertainty of the measured current flow. The updated estimate uncertainty is the uncertainty of the updated estimate. The expected future development is the expected future development of the SOC.20 4.1.2.1 Advantages of the Kalman filter High accuracy: The Kalman filter provides high accuracy in estimating the state of charge of a battery. It considers both measurements and system model and optimizes the estimate by weighing the different pieces of information. This allows the Kalman filter to effectively compensate for inaccuracies and noise in the measurements. The Kalman filter is robust to disturbances and measurement errors. By continuously updating the estimate based on actual measurements and predicting system behavior, the filter can compensate for unwanted effects such as noise or sudden changes in current or voltage. Also, the filter can be adapted to different battery models and system configurations. It allows the integration of additional information such as temperature measurements, aging effects, or dynamic load profiles to further improve the accuracy of the stateof-charge estimation.21 4.1.2.2 Disadvantages of the Kalman filter The Kalman filter is more complex compared to some other state-of-charge estimation methods. It requires detailed knowledge of the battery model and accurate modeling of measurement errors and noise sources. Therefore, the implementation and calibration of the filter requires expertise and experience. The Kalman filter requires some computational power, especially when using extensive battery models or when processing large amounts of data in real time. The efficiency and speed of the computations depend on the implementation and available hardware.

Cell and Battery Design – Batteries | Hardware

679

4.1.2.3 Complexity and required computing power Implementing a Kalman filter requires a detailed model of the battery that matches the physical properties and behavior of the battery under various operating conditions. Calibration of the filter requires measurements and tests to adjust the model parameters and improve the accuracy of the estimates.22 The computational cost of the Kalman filter depends on the size of the battery model, the number of measurements, and the desired update rate of the SOC. Modern battery management systems (BMS) can implement the Kalman filter in real time and continuously update the SOC estimate. However, the required computational power may vary depending on the complexity of the filter and the available hardware resources.

4.1.3 H-Infinity filter The H-Infinity filter is another advanced method for determining the State of Charge (SOC) of a battery. It is based on the concept of H-Infinity control, which aims to maximize the filter’s robustness to uncertainties and disturbances. In this chapter, we will take a closer look at the H-Infinity filter by examining its advantages, disadvantages, complexity, and required computational power. In addition, we will compare the H-Infinity filter with the Kalman filter and discuss their respective strengths and applications.23,24 4.1.3.1 Advantages of the H-Infinity filter The H-Infinity filter is particularly robust to uncertainties, disturbances, and model errors. It can account for these factors and adjust charge state estimates accordingly. As a result, the H-Infinity filter can provide accurate estimates even under difficult operating conditions and in the presence of disturbances.25 The filter strives for optimal accuracy in state-of-charge estimates. It uses mathematical methods to obtain the best estimates based on available measurements and model information. By accounting for uncertainty, the H-Infinity filter can provide more accurate estimates than traditional methods such as Coulomb Counting. The H-Infinity filter can be adapted to different battery models and system configurations. It allows the integration of additional information such as temperature measurements, aging effects, or dynamic load profiles to further improve the accuracy of stateof-charge estimation. 4.1.3.2 Disadvantages of the H-Infinity filter The H-Infinity filter is more complex compared to conventional state-of-charge estimation methods. It requires a detailed battery model and mathematical knowledge to model and implement the filter. The complexity of the filter can make it difficult to implement and calibrate. The filter requires some computational power, especially when using extensive battery models or when processing large amounts of data in real time. Efficient implementation of the filter requires specialized computational methods and consideration of available hardware capacity.26 4.1.3.3 Complexity and required computing power The implementation of an H-Infinity filter requires a detailed model of the battery that considers the physical properties and behavior of the battery under different operating conditions. The mathematical implementation of the filter requires advanced knowledge of control engineering and signal processing. The computational cost of the H-Infinity filter depends on the size of the battery model, the number of measurements, and the desired update rate of the SOC. The implementation usually requires powerful hardware and an efficient computational methodology to perform the SOC estimation in real time. 4.1.3.4 Comparison to the Kalman filter Both the H-Infinity filter and the Kalman filter are advanced methods for determining the state of charge of a battery. Both offer high accuracy and robustness but have different approaches and characteristics. The Kalman filter is based on statistical estimation and considers both measurements and system models. It is well established and successfully used in many applications. The Kalman filter is known for its accuracy and flexibility but requires detailed modeling and calibration as well as some computational power. The H-Infinity filter, on the other hand, is based on the H-Infinity control concept and optimizes the estimation of the SOC considering uncertainties and disturbances. It is characterized by its robustness and its ability to perform accurately even in systems with high uncertainty. However, the H-Infinity filter requires even more detailed modeling and higher computational power compared to the Kalman filter. The choice between the Kalman filter and the H-Infinity filter depends on the specific requirements of the application and the available resources. Both filters offer advanced methods for determining the state of charge of a battery and can be used in various application areas, including renewable energy and stationary storage, to ensure efficient and reliable use of battery capacity.

4.1.4 Electrochemical impedance spectroscopy (EIS) Electrochemical impedance spectroscopy (EIS) is an advanced method for determining the State of Charge (SOC) of State of Health (SOH) of a battery. It is based on electrochemical impedance analysis, which examines the behavior of the battery at different

680

Cell and Battery Design – Batteries | Hardware

frequencies. In this chapter, we will take a closer look at EIS by examining its advantages and disadvantages, complexity, and the computational power required. In addition, we will highlight the differences between the EIS and filtering techniques such as the Kalman filter and the H-infinity filter.27,28 The State of Charge of a battery can be determined using electrochemical impedance spectroscopy (EIS). With EIS, a small, sinusoidal AC signal is applied to the battery and the resulting voltage response is measured. The impedance of the battery is then calculated from the voltage response. Battery impedance is a measure of the battery’s resistance to current flow. The impedance of the battery changes with the SOC of the battery. This is because the properties of the battery’s electrolytes and electrodes change with SOC. The impedance of the battery can be analyzed using various methods to determine the SOC of the battery. One commonly used method is the impedance shift method. In this method, the impedance of the battery is measured at different SOC values. The impedance shift is then the difference between the impedance of the battery at a given SOC value and the impedance of the battery at a reference SOC value. The impedance shift can then be used with a model or algorithm to determine the SOC of the battery. The model or algorithm is trained on a data set of EIS measurements of batteries with known SOC values.29 When creating a model for a battery, there are four common parameters that represent battery chemistry. These are30:

•

• • •

Electrolytic (Ohmic) Resistance—RS o This corresponds to the resistance of the electrolyte in the battery. o It is affected by the electrodes and wire length used when performing the test. o It increases as the battery ages. o It is dominant together with inductive effects when the frequency is greater than 1 kHz. Double Layer Capacitance—CDL o It is located between the electrode and the electrolyte. o It is composed of two parallel layers of opposite charges encompassing the electrode. o It is dominant in the 1 Hz to 1 kHz frequency range. Charge Transfer Resistance—RCT o This resistance occurs from transferring electrons from one phase to another, that is, a solid (electrode) to a liquid (electrolyte). o It changes with temperature and state-of-charge of the battery. o It is dominant in the 1 Hz to 1 kHz frequency range. Warburg (Diffusion) Resistance—W o This represents a resistance to mass transfer, that is, diffusion control. o It typically exhibits a 45 phase shift. o It is dominant when the frequency is less than 1 Hz.

4.1.4.1 Advantages of electrochemical impedance spectroscopy (EIS) Direct characterization: EIS allows direct characterization of the battery’s electrochemical properties, including internal resistance, capacitances, and reaction kinetics. The state of the battery can be inferred by analyzing the impedance spectra. The state of charge (SOC) is not captured directly by EIS, but in comparison to previous EIS measurements made at different SOC. Non-invasive: EIS is a non-invasive method that requires no modifications to the battery. It can therefore be applied both in the laboratory and in real time during battery operation. Broad frequency spectrum: EIS examines the behavior of the battery over a broad frequency spectrum. This allows detection of various physical and electrochemical processes that can lead to battery aging or degradation.31 4.1.4.2 Disadvantages of electrochemical impedance spectroscopy (EIS) Complexity: EIS requires a certain level of expertise and experience in interpreting impedance spectra. Characterization and analysis of the complex impedance data requires advanced mathematical and electrochemical knowledge. Measurement equipment: performing EIS requires specialized measurement equipment that can acquire impedance spectra over a wide range of frequencies. This equipment can be expensive and requires specialized training for proper operation. Also, power electronics might influence the measuring results. 4.1.4.3 Differences between EIS and filter techniques (Kalman filter, H-Infinity filter) EIS is based on comparative measurement of electrochemical impedance to derive the state of charge of a battery. The filtering techniques such as the Kalman filter and the H-Infinity filter, on the other hand, rely on statistical estimates and models to determine the state of charge. The EIS provides information about the electrochemical properties of the battery related to its aging and degradation. Filtering techniques, on the other hand, use measurements such as the battery’s current and voltage to estimate the state of charge. Specialized knowledge and equipment are needed to acquire and analyze impedance spectra. Filtering techniques such as the Kalman filter and the H-infinity filter require detailed modeling and calculations to estimate the state of charge, but they are generally simpler to implement and require less computing power than EIS.32

Cell and Battery Design – Batteries | Hardware

681

EIS is commonly used in research laboratories and battery development to obtain detailed information about electrochemical behavior. Filtering techniques such as the Kalman filter and the H-Infinity filter are typically applied in battery management systems (BMS) for real-time state-of-charge monitoring and efficient battery control.33

4.2

State of health estimation

The condition of a battery is a critical factor in its performance and lifetime. To evaluate the condition of a battery, the term “State of Health” (SOH) is often used. The SOH provides information about the current health of a battery and helps to estimate its remaining capacity compared to its original capacity. SOH ¼ C actual=C initial Or in terms of power the SOH of a battery is assessed by its capacity to reliably supply the required power. A battery with good SOH should be capable of consistently delivering the necessary electrical energy without significant degradation or voltage drops, ensuring it meets the power demands of the intended application. In this chapter, we will look at how the SOH of a battery is determined or calculated, its applications and limitations, and possible differences between the stationary and automotive sectors.34 The State of Health (SOH) of a battery is a measure of its current condition. It is expressed as a percentage of the original capacity of a battery. A SOH of 100% means that the battery still has its full capacity, while a SOH of 80% means that the battery only has 80% of its original capacity. The SOH of a battery decreases over time as the battery ages. There are several factors that can affect the SOH of a battery, including temperature, charge cycles, charge and discharge rates, depth of discharge, and storage conditions. In the stationary sector, such as when batteries are used for energy storage of solar or wind power systems, the focus is often on long-term stability and optimization of battery lifetime. Here, the SOH is an important parameter for determining the optimal time for maintenance or replacement work. In the automotive sector, on the other hand, the focus is on evaluating the remaining capacity to determine the range of the vehicle and to take timely measures to ensure sufficient performance. Determining the SOH of a battery is not an exact science and is subject to some limitations. One challenge is that SOH is affected by many factors, as already mentioned. It can be difficult to account for all these factors in a single measurement. In addition, different battery technologies may exhibit different behaviors, making it difficult to compare SOH values. There are several methods to determine the SOH of a battery. One commonly used method is to perform load profile tests where the battery is subjected to certain loads while its voltage and capacity are monitored. By analyzing the changes in voltage and capacity compared to reference values, the SOH can be derived. Another method is impedance spectroscopy, which examines the electrochemical properties of the battery to draw conclusions about its condition.

4.2.1 Methods for SOH estimation Determining the State of Health (SOH) of a battery is a complex process that involves various methods and techniques. There are several approaches to determining SOH based on different parameters and measurements. It is important to note that determining the SOH of a battery brings its own challenges. The SOH is affected by many factors, including operating conditions, temperature, depth of discharge, rate of charge and discharge, aging effects, etc. Therefore, it is difficult to determine the SOH precisely and accurately. In addition, different battery technologies may exhibit different behaviors, making it difficult to compare SOH values. In the following, some common methods are explained in more detail.35 4.2.1.1 Load profile and capacity tests Load profile tests are a commonly used method for determining the SOH of a battery. In this method, the battery is subjected to various loads while its voltage and capacity are monitored. By analyzing the changes in voltage and capacity compared to reference values, the SOH can be derived. This approach typically uses different discharge profiles to test the battery under different operating conditions and determine the SOH. A capacity test is a simple and effective method for determining the SOH of a battery. In a capacity test, the battery is fully charged and then discharged at a constant current until it reaches a specified voltage. The amount of charge over time that the battery can discharge is its capacity. Capacity tests and load profile tests are both effective methods for determining the SOH of a battery. However, there are some key differences between the two methods. Capacity tests are simpler and less expensive than load profile tests. However, they are also less accurate, as they do not consider the effects of different discharge rates or temperatures. Load profile tests are more accurate than capacity tests, but they are also more complex and expensive. They can also be more time-consuming, as they require the battery to be cycled through a series of different discharge rates and temperatures. 4.2.1.2 Impedance spectroscopy Impedance spectroscopy is an advanced technique for determining the SOH of a battery. It is based on analyzing the battery’s electrochemical impedance over a range of frequencies. By measuring impedance, conclusions can be drawn about the battery’s

682

Cell and Battery Design – Batteries | Hardware

condition, including its internal resistances, capacitances, and other electrochemical properties. Impedance spectroscopy enables detailed characterization of the battery and provides information about the aging and wear of the battery components. 4.2.1.3 Models and algorithms Another approach to determining SOH is to use mathematical models and algorithms based on measurements of voltage, current, and other parameters. These models often use empirical relationships to estimate SOH based on the behavior of the battery over time. Such models may be based on statistical analysis or machine learning algorithms and require continuous monitoring of the battery to determine its condition.36 Physical models describe electrochemical processes within the battery, while statistical models are trained on data sets of batteries with known SOHs. With machine learning algorithms features from measured parameters of a battery can be extracted. Models and algorithms can be a valuable tool for monitoring the SOH of batteries in real time. They can also be used to predict the remaining useful life of a battery.

4.3

Prediction remaining useful life

Batteries are an integral part of modern life, powering everything from smartphones and laptops to electric vehicles and renewable energy systems. The ability to accurately predict the remaining useful life of a battery is critical for ensuring the reliability and performance of these systems. The remaining useful life of a battery is defined as the amount of time that a battery can continue to provide useful power, considering its age, usage, and other factors. The estimation of the remaining useful life of a battery is a complex process, requiring the analysis of several parameters, such as battery capacity, voltage, temperature, and state of charge.37 The primary methods for estimating the remaining useful life of a battery are based on empirical models and physics-based models. Empirical models use data from previous battery tests to estimate the remaining useful life, while physics-based models use mathematical models to simulate the performance of the battery over time.38 One common approach to estimate the remaining useful life of a battery is using capacity fade. Capacity fade refers to the reduction in battery capacity over time and is often used as an indicator of the battery’s remaining useful life. This method involves monitoring the battery’s capacity over time and using that data to estimate its remaining useful life. Another approach to estimate the remaining useful life of a battery is using state of health (SOH) estimation. SOH estimation involves monitoring several parameters, such as voltage, temperature, and state of charge, and using that data to estimate the overall health of the battery. This method is particularly useful for predicting the remaining useful life of batteries in real-world applications, where the operating conditions can vary greatly. In addition to these methods, the use of artificial intelligence and machine learning algorithms is becoming increasingly popular for estimating the remaining useful life of batteries. These algorithms use data from previous battery tests, as well as real-world battery data, to train models that can accurately predict the remaining useful life of a battery. It is important to note that the remaining useful life of a battery is not a fixed value and can change over time based on the battery’s usage, operating conditions, and other factors. This means that accurate and up-to-date estimates of the remaining useful life are critical for ensuring the reliability and performance of battery-powered systems. In conclusion, the estimation of the remaining useful life of a battery is a critical component of battery management and plays a key role in ensuring the reliability and performance of battery-powered systems. With the use of empirical models, physics-based models, and artificial intelligence and machine learning algorithms, the prediction of the remaining useful life of batteries is becoming increasingly accurate and reliable.

5

Outlook

The use of batteries in stationary applications has seen rapid development in recent years. Battery management systems (BMS) play a crucial role in monitoring, controlling, and protecting batteries. In this chapter, we considered the various approaches of BMS in stationary applications, including requirements, system concepts and state estimation. Battery technology has made significant progress in recent years, and this trend is expected to continue. New materials, such as solid-state batteries, could increase energy density and improve safety. Furthermore, alternative technologies such as sodium-ion and zinc-ion are meanwhile on the market or close to market entry. In addition, intensive work is underway to develop batteries with longer lifetimes. These advances place increased demands on BMSs, as they must be able to accommodate the specific characteristics and requirements of these new battery technologies. The BMS of the future will need improved precision and adaptability to make the most of the performance of these advanced batteries. Its role terms of the EU battery passport approach as part of the EU battery regulation, recently introduced, and the corresponding requirements highlight these challenges.13 With the increasing use of renewable energy worldwide, the integration of BMS into advanced battery systems as part of energy systems is becoming more and more important. BMS plays a critical role in optimizing the storing and delivering of renewable energy. Future BMSs in combination with advanced energy management systems will be able to optimize the flow of energy between batteries, solar arrays, wind turbines and the power grid. They will also include intelligent control mechanisms to maximize the availability and use of renewable energy while ensuring grid stability.

Cell and Battery Design – Batteries | Hardware

683

Detailed monitoring and diagnostics of battery states including battery health are critical to maximizing battery performance and lifetime. Future BMSs will use advanced sensors and monitoring techniques to detect temperature, cell condition, and other relevant parameters as well as to calculate state of charge and state of health in real time conditions. By using advanced diagnostic algorithms and machine learning, deviations from normal conditions can be detected and analyzed in an early stage. This enables predictive maintenance and optimization of battery operation to ensure maximum performance and lifetime. In the future, different battery technologies could be combined to take advantage of different systems. For example, lithium-ion batteries could be used in conjunction with redox flow batteries to combine the benefits of high energy density and flexible capacity. However, the integration of different battery technologies poses challenges for the BMS, as it must be able to ensure efficient and reliable operation and control of such hybrid storage systems. The BMS of the future will consider both the individual requirements of the different battery technologies and the optimal use of the combined systems. With increasing connectivity and huge data growth, advances in data analytics and artificial intelligence (AI) will play an increasingly important role. Future BMSs will be able to process and analyze large amounts of data to provide valuable insights for optimizing battery operations. By using AI techniques such as machine learning and pattern recognition, BMSs can identify battery behavior patterns, make more accurate predictions, and develop effective control strategies. This enables optimized use of battery capacity, reduced energy loss, and extended battery life. The above points illustrate that the development of battery management systems in the stationary sector will undergo far-reaching advances. Through the integration of new battery technologies, the optimization of renewable energy integration, improved monitoring and diagnostic functions, the combination of different battery technologies and the use of advanced data analysis and artificial intelligence, future BMS will be able to make batteries even more efficient, reliable, and durable. This will enable sustainable and effective use of energy storage in the stationary sector and play a key role in the energy transition and the development of a sustainable energy infrastructure.

References 1. Hesse, H. C.; Schimpe, M.; Kucevic, D.; Jossen, A. Lithium-Ion Battery Storage for the Grid—A Review of Stationary Battery Storage System Design Tailored for Applications in Modern Power Grids. Energies 2017, 10, 2107. https://doi.org/10.3390/en10122107. 2. Vetter, M.; Lux, S. Chapter 11 – Rechargeable Batteries with Special Reference to Lithium-Ion Batteries. In Storing Energy; Letcher, T. M., Ed.; Elsevier, 2016; pp. 205–225. ISBN:9780128034408. https://doi.org/10.1016/B978-0-12-803440-8.00011-7. 3. Berga, K.; et al. Economic Evaluation of Operation Strategies for Battery Systems in Football Stadiums: A Norwegian Case Study. J. Energy Storage 2022, 34, 102190. 4. Vetter, M.; et al. PV Battery Power Plants in Europe – Status, Trends and Potentials; EES Europe, 2023. www.ees-europe.com. 5. Gamisch, S. Simulative Investigation of Measures to Prevent Thermal Runaway Propagation in Li-Ion-Battery Modules. In 16th International Renewable Energy Storage Conference, Düsseldorf; 2022. 6. International Norm: ISO 11898-1:2015, 2015. 7. The Modbus Organization. Modbus; Modbus, 2023. www.modbus.org. 8. Landinger, T. F. Power Line Communications for Automotive High Voltage Battery Systems: Channel Modeling and Coexistence Study with Battery Monitoring. Energies 2021, 14 (7), 1851. 9. Talei, A. P. Considerations for a Power Line; Elektrotechnik & Informationstechnik, 2021. 10. Analog Devices. LTC6813; 2023. www.analog.com. 11. NXP. MC33774; 2023. www.nxp.com. 12. Isabellenhuette. IV-S Series; www.isabellenhuette.de, 2023. 13. BMWK. Battery Pass; www.thebatterypass.eu, 2023. 14. Jäger, M. Building Energy Systems; 2023. www.learn.libre.solar. 15. Ning, Z.; et al. Co-Estimation of State of Charge and State of Health for 48 V Battery System Based on Cubature Kalman Filter and H-Infinity. J. Energy Storage 2022, 56, 106052. 16. Chen, Y.; et al. SOC Estimation of Retired Lithium-Ion Batteries for Electric Vehicle with Improved Particle Filter by H-Infinity Filter. Energy Rep. 2023, 9, 1937–1947. 17. Vedhanayaki, S.; et al. Certain Investigation and Implementation of Coulomb Counting Based Unscented Kalman Filter for State of Charge Estimation of Lithium-Ion Batteries Used in Electric Vehicle Application. Int. J. Thermofluids 2023, 18, 100335. 18. Wang, C.; et al. A Novel Hybrid Machine Learning Coulomb Counting Technique for State of Charge Estimation of Lithium-Ion Batteries. J. Energy Storage 2023, 63, 107081. 19. Kalman, R. E. A New Approach to Linear Filtering and Prediction Problems. J. Basiy Eng. 1960, 35–45. 20. Plett, G. L. Extended Kalman Filtering for Battery Management Systems of LiPB-Based HEV Battery Packs: Part 2. Modeling and Identification. J. Power Sources 2004, 134 (2), 262–276. 21. Campestrini, C. Practical Feasibility of Kalman Filters for the State Estimation of Lithium-Ion Batteries; Doctoral Dissertation, Technische Universität München, 2018. 22. Zhang, L. Intelligent Computing for Extended Kalman Filtering SOC Algorithm of Lithium-ion Battery. Wireless Personal Commun. Int. J. 2018, 102, 2063–2076. 23. Zhou, K.; et al. Robust and Optimal Control; Prentice Hall, 1996. 24. Doyle, C.; Glover, K. State-Space Formulae for all Stabilizing Controllers that Satisfy an H-Infinity-Norm Bound and Relations to Relations to Risk Sensitivity. Systems Control Lett. 1988, 11 (3), 167–172. 25. Li, L. State of Charge Estimation for Lithium-Ion Power Battery Based on H-Infinity Filter Algorithm. Appl. Sci. 2020, 10 (18), 6371. 26. Chen, Z. Synthetic State of Charge Estimation for Lithium-Ion Batteries Based on Long Short-Term Memory Network Modeling and Adaptive H-Infinity Filter. Energy 2021, 228, 120630. 27. Fischer, O.; et al. Comparative Study of Excitation Signals for Microcontroller-Based EIS Measurement on Li-Ion Batteries. In 2021 International Workshop on Impedance Spectroscopy (IWIS); 2021. 28. Yadav, J.; Arya, S.; Kushwaha, A.; Ansar, S. A.; Khan, R. A. Reinforcing Li-Ion Batteries with Electrochemical Impedance Spectroscopy. Mater. Today Proc. 2023, https://doi.org/ 10.1016/j.matpr.2023.02.105. 29. Westerhoff, U. Electrochemical Impedance Spectroscopy Based Estimation of the State of Charge of lithium-Ion Batteries. J. Energy Storage 2016, 8, 244–256. 30. Analog Devices. Electrochemical Impedance Spectroscopy (EIS) for Batteries; 2023. www.analog.com. 31. Simatupang, D.; Benshatti, A.; Park, S. Y. Embedded Electrochemical Impedance Spectroscopy into Battery Management System. In 47th Annual Conference of the IEEE Industrial Electronics Society; 2021.

684

Cell and Battery Design – Batteries | Hardware

32. Wang, L.; Song, Z.; Zhu, L.; Jiang, J. Fast Electrochemical Impedance Spectroscopy of Lithium-Ion Batteries Based on the Large Square Wave Excitation Signal. iScience 2023, 26 (4). 33. Cui, B.; et al. Internal Short Circuit Early Detection of Lithium-Ion Batteries from Impedance Spectroscopy Using Deep Learning. J. Power Sources 2023, 563, 232824. 34. Birkl, C. R.; Roberts, M. R.; McTurk, E.; Bruce, P. G.; Howey, D. A. Degradation Diagnostics for lithium Ion Cells. J. Power Sources 2017, 341, 373–386. 35. Zou, Y. Combined State of Charge and State of Health Estimation over Lithium-Ion Battery Cell Cycle Lifespan for Electric Vehicles. J. Power Sources 2015, 273, 793–803. 36. Salucci, C. B.; Bakdi, A.; Glad, I. K.; Vanem, E.; De Bin, R. A Novel Semi-Supervised Learning Approach for State of Health Monitoring of Maritime Lithium-Ion Batteries. J. Power Sources 2023, 556, 232429. 37. Jia, J. SOH and RUL Prediction of Lithium-Ion Batteries Based on Gaussian Process Regression with Indirect Health Indicators. Energies 2020, 13 (2), 375. 38. Wu, Y.; Li, W.; Wang, Y.; Zhang, K. Remaining Useful Life Prediction of Lithium-Ion Batteries Using Neural Network and Bat-Based Particle Filter. IEEE Access 2019, 7, 54843–54854.

Batteries – Batteries General – Introduction | Overview Eduardo Cattaneoa, Juergen Garcheb, and Bernhard Riegela, aHOPPECKE Batterien GmbH & Co. KG, Brilon, Germany; b Ulm University, Ulm, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 2.1 2.2 3 3.1 3.2 3.3 3.4 4 5 5.1 5.2 5.3 5.4 5.5 6 6.1 6.2 7 7.1 7.2 7.3 7.3.1 7.3.2 7.4 8 References Further reading

Introduction History Primary cells Secondary cells Working principles Cell voltage Thermodynamic equilibrium Electrochemical equilibrium Electrochemical reaction kinetics Cell design principles Cell components Active mass Current collectors Electrolyte Separator Case Cell design Redox-flow cell design Battery and system Battery market and applications Global market Portable applications Automotive and transportation application Automotive application Transportation application Stationary applications Outlook

687 687 687 688 689 689 689 691 691 692 692 692 692 692 693 693 693 695 696 696 696 697 698 698 699 699 700 700 701

Abstract The present chapter is the introduction chapter for the battery section in the encyclopedia and provides a general overview on primary and secondary battery systems including a chronological account of the main battery storage developments, a short description of the main cell components and cell design criteria. To attain a brief, yet comprehensive introduction, electrochemical reactions at the electrodes and several technical details were omitted as they are going to be treated in other chapters. This chapter also summarizes the typical battery system applications like automotive, transportation, portable, stationary, motive power, and large-scale storage with the corresponding market share.

Glossary Equilibrium voltage (reversible voltage) The difference in the reversible potentials of the two electrodes that make up the cell under equilibrium conditions. Open-circuit voltage The voltage of a battery (or cell) when there is no net current flow; it can be measured directly. This value can differ from the equilibrium voltage because OCV is measured under non-thermodynamic equilibrium conditions, non-equilibrium processes like corrosion can contribute to the value of the OCV. The open-circuit voltage (OCV) is generally defined as the voltage between the terminals of an electrochemical cell when no current flows through the cell. Primary battery (or cell) A battery (or cell) that contains a fixed amount of stored energy when manufactured, and that cannot be recharged after that energy is withdrawn. Secondary battery (or cell) A battery (or cell) that is capable of repeated charging and discharging. Also known as a “rechargeable battery (or cell).”

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00170-5

685

686

Batteries – Batteries General – Introduction | Overview

Key points

• • • •

Chronological overview of primary and secondary electrochemical storage systems Working principles, cell components and battery design criteria Battery global market and market shares for different applications Outlook and prospects

Abbreviations DG DH DS ABS AGM BESS BMS CAGR CD DMC EC GPS HEV ICD ICE LAB LFP LIB NMC OCV PC PHEV PP PPO SAN SLI UPS VRLA

Gibbs energy change enthalpy change entropy change acrylonitrile butadiene styrene absorptive glass mat+ battery energy storage systems battery management system compound annual grow rate compact disk dimethyl carbonate ethylene carbonate global positioning system hybrid electric vehicle implantable cardioverter defibrillator internal combustion engine lead acid battery lithium iron phosphate lithium battery lithium nickel manganese cobalt oxides open circuit voltage polycarbonate plug-in hybrid electric vehicle polypropylene polyphenylene oxide styrene acrylonitrile start light ignition uninterrupted power supply valve-regulated lead–acid

Symbols A Ag Ah
C C Cl Cu Al Cd Co EV F Fe GWh

ampere silver ampere-hour Celsius carbon chlorine copper aluminum cadmium cobalt electric vehicle fluorine iron gigawatt-hour

Batteries – Batteries General – Introduction | Overview

H La Li Mn Mo Na Ni O R¼ S T TWh V V Zn

1

687

hydrogen lanthanum (rare earth) lithium manganese molybdenum sodium nickel oxygen universal gas constant sulfur temperature terawatt-hour vanadium volt zinc

Introduction

This section Batteries in the Encyclopedia, on one hand, gives a general overview about battery related topics like charging/ discharging, safety, and environmental issues. On the other hand, this section covers also diverse battery systems based on the elements Al, Fe, Pb, Li, Mn, Ni, Co, Na, and Zn. Battery systems are usually named by the anode. Ni systems are an exception to this rule, as they are historically named by the cathode.a This section describes also the primary and secondary systems including the special case of redox-batteries and reserve cells. The title of the present chapter is Batteries. In colloquial language this term is used to designate an assembly of primary and secondary cells. Accurately speaking, a battery is a storage system with at least two cells, regardless of whether they are primary or secondary cells. The term battery goes back to the military field, where “battery” referred to a cluster of cannons. In the electrical area Benjamin Franklin first used the term “battery” in 1749 while doing experiments with electricity using a set of linked capacitors. The present chapter also provides a chronological account of battery developments. This includes a section about the working principles of the different storage types with their specific cell design, and the blocks to build battery modules. Furthermore, this chapter gives a brief survey of battery applications in transportation, stationary and portable devices including the development of the battery market.

2

History

An electrochemical cell that is a device that converts chemical energy, stored in the reactants, directly into electrical energy. These electrochemical storage devices are classified in primary and secondary. The state of charge of the discharged primary cells cannot be restored to the original condition, because in general the electrochemical reaction occurring in the cell is not reversible, i.e., after depleting the capacity they have to be disposed. The state of charge of discharged secondary systems, on the contrary, can be returned to the original condition, i.e., they are rechargeable.

2.1

Primary cells

Primary cells were the first electrochemical storage devices that were able to provide a steady current with constant voltage. The first successful electrochemical primary battery was invented by Alessandro Volta (1800) which consisted of a pile of alternating discs of copper and zinc that were separated by pasteboard discs soaked in brine. Volta’s pile became the only source to generate a constant current in those days. For long time operation the electrical current flow of Volta’s battery was, however, hampered by the buildup of an insulating film of hydrogen bubbles on the copper electrode.1 This shortcoming was eliminated with the introduction of cathodes of metallic oxides which reacted with the hydrogen. The most efficient was manganese dioxide (MnO2) used for cell cathodes first by G. Zamboni (1812) and later by G. Leclanche (1866). The “zinc-carbon wet cell” by Leclanche achieved the breakthrough for the practical use. Carbon was here misleading, because the positive active material was really MnO2 packed around a carbon current collector. A solution of ammonium chloride served as the electrolyte and a zinc rod or plate, sometimes amalgamated with mercury (H2 gassing suppressor), formed the anode.2 a The term anode in electrochemical storage, is commonly used for the negative and cathode for the positive electrode. This applies for both, discharging and the charging, although it is the correct term only for the discharge process.

688

Batteries – Batteries General – Introduction | Overview

Fig. 1 Cylindrical bobbin type zinc/alkaline/manganese dioxide primary cell. Drawing based on Marsal, P.; Kordesch, K.; Urry, L. F. Dry Cell, US2960558 patent filed 1957 and granted in 1960 to Union Carbide Corporation.

A further disruptive invention by Leclanché and Gassner (1888), was the “dry” cell with an immobilized electrolyte, consisting of a paste made with starch or other absorbent materials and a zinc anode in the form of a can which served as negative and container simultaneously. The dry cells became the ideal electrochemical storage for portable devices as they could be used in any orientation due to the almost closed container and their spill-proof electrolyte. In the decade 1960–1970 the primary cells made a great step forward, with the introduction of the more powerful and high-capacity zinc/alkaline/manganese dioxide cell with a KOH electrolyte and a high-purity zinc powder anode that provided greater surface area for the anode.3 The shape of the cells was also cylindrical, and the active materials were set like in Leclanche’s cell in a bobbin construction but with a central anode (Fig. 1). A rechargeable zinc/alkaline/manganese version was also commercialized in the 90s of the last century.4 Other relevant alkaline primary battery systems are:

• • •

Zinc/alkaline/silver oxide battery. A high-power primary battery with an alkaline electrolyte, a negative electrode of zinc and a positive electrode containing silver oxide. The Zinc/alkaline/air battery with a carbon positive electrode covered by a catalyst to permit the oxygen reduction reaction, i.e., oxygen serves as the active material at the positive electrode. Aluminum/alkaline/air belongs like Zn-air to the metal-air battery type. This system has a high gravimetric energy density, but with a lower gravimetric power density than Zn-air.

In the late 1960s, different non-aqueous lithium primary batteries were introduced into the market. These primary cells showed high capacities with high voltages: Li-(CFx)n (US3536532), Li-MnO2 (US4216247), Li-CuO with OCVs between 2.8 and 3 V. In this type of Li primary cells, the electrolytes were organic aprotic solvents like ethylene carbonate (EC) and dimethyl carbonate (DMC). The solutes that provided the ionic conductivity were lithium salts like LiClO4, LiAsF6 and the nowadays preferred LiPF6. There is a second class of lithium primary cells: i. Li/SO2 where the positive reactant is a gas: sulfur dioxide (SO2) dissolved in a solvent with a salt as solute (US3423242) and 2.85 V OCV ii. Li/SOCl2 where the positive reactant is a liquid: thionyl-chloride (SOCl2) and 3.50 V higher OCV Due to safety reasons these inorganic electrolyte lithium cells were mainly deployed in military, special industrial and space applications.

2.2

Secondary cells

The rapid development and widespread of the nowadays well-known practical rechargeable cells, took place more than 50 years after Volta’s primary battery and was boosted by the invention of the first charger, i.e., the invention of the dynamo (commutator) in 1871. In early stationary applications like powering railway signaling and Morse electric telegraphs the primary and secondary systems, were competitors. The primary cells however, due to their lower weight and size became rapidly the standard storage for powering portable devices.

Batteries – Batteries General – Introduction | Overview

689

The first rechargeable cell was developed 1802 by J.W. Ritter with a cell consisting of layered discs of copper and cardboard soaked in a brine. The rechargeable cell for practical use was developed later in 1859 by G. Planté. His lead-acid cell consisted of a lead anode and a lead dioxide cathode immersed in sulfuric acid. The lead-acid battery has since then undergone many improvements. The most outstanding innovation in the last quarter of the 20th century was the development of the valve-regulated (VRLA) system. Lead-acid in both, in the standard vented and VRLA versions celebrated an unparalleled success in the electrochemical storage, becoming worldwide the most-produced battery system until around 2020. In 1899, E.W. Jungner developed the nickel–cadmium accumulator with rechargeable negative cadmium and positive Ni-hydroxide electrodes with a potassium hydroxide electrolyte. The nickel–cadmium system prevailed on the market for long time behind the lead-acid favorite, especially for applications requiring high discharge rates, low temperatures, and long cycle life. T. Edison introduced in 1900 a second alkaline rechargeable storage, the nickel oxide-iron (NiFe) accumulator, with a negative made from fine powder iron and a positive nickel-hydroxide with 20–30% graphite flakes electrode. Due to the high self-discharge and poor energy efficiency this system almost disappeared from the market excepting niche applications. At the beginning of the 90s of the last century nickel-metal hydride (Ni-MH) cells were launched into the market, consisting of nickel-hydroxide cathode and a H2 storage metal hydrides anode based on lanthanum, and other rare earth elements. Thanks to its high power and cycle life performance NiMH remained long time the choice to power HEVs. Basic research for Li rechargeable batteries started early in the 1970s and in 1973 M.S. Whittingham first reported the charge-discharge cycling with a lithium metal anode and a di-chalcogenide Li intercalation cathode (GB1468244).5 The first commercial Li metal rechargeable battery with a MoS2 cathode (US4224390) was the MoliCel introduced into the market between 1986 and 1988 by Moli Energy Ltd. in Canada. The cell distribution, however, was halted in October 1989 due to a cellphone explosion that caused minor burns to the user. Later insights on this incident concluded that the lithium-metal anodes under low rate discharge with fast charge operation reacted strongly with the liquid electrolytes to build dendrites leading to shorts. The later developed, safer, rechargeable lithium-ion batteries utilize intercalation anodes made usually of a graphitic carbon6 and metal oxide layered cathodes like the LiCoO2 studied by the research group of J. B. Goodenough in 1980.7 A lithium-ion secondary battery based on the lithium intercalation in layered compounds for both, the positive and negative electrodes was filed for a patent in 1986 by A. Yoshino and co-authors (US4668595). The breakthrough in the rechargeable battery market was achieved by SONY in 1991 with a lithium-ion battery based on graphite anodes and lithium cobalt dioxide cathodes used to power portable devices. Li-ion became since then the dominating battery technology. Primary systems with high charge retention, however, are still the choice for devices with low power consumption or cases where there is no infrastructure or time to recharge the cells.

3 3.1

Working principles Cell voltage

The first cell to be used in practice was the Zn-Cu cell called Daniel Element named after its inventor J.F. Daniell in 1836. Each of the two Cu and Zn electrodes in his cell were immersed into different electrolytes separated by a porous earthenware partition—the Cu cathode in a copper sulfate solution and the Zn anode in dilute sulfuric acid. The schematic drawing in Fig. 2 shows the two electrolytes separated by a salt-bridge.8 The overall equation of the cell is: ZnðsÞ + Cu2+ ðaqÞ > Zn2+ ðaqÞ + CuðsÞ

(1)

The equilibrium voltage at 25
C can be derived from the corresponding standard electrode half-cell potentials: Cu2+ ðaqÞ + 2e− ¼ Cu ðsÞ

jo ¼ +0:34 V

(2)

Zn2+ ðaqÞ + 2e− ¼ Zn ðsÞ

jo ¼ −0:76 V

(3)

Subtracting (2) minus (3) we obtain the open circuit cell voltage (OCV): Uo ¼ jo ðCu=Cu2+Þ − jo ðZn=Zn2+Þ ¼ 1, 10 V

(4)




The equilibrium potential of Daniell’s cell for 1 M ion concentrations at 25 C was used to define the Volt, i.e., the unit for the electromotive force in the international system of units.

3.2

Thermodynamic equilibrium

The standard half-cell potentials (2) and (3) correspond to the maximum electrical energy obtainable from the reaction of that element or compound. All standard potentials are referred to the standard reaction of H2/H+ (aq) which is conventionally taken as zero.8 The electrode reactions follow the laws of the classical thermodynamics:

690

Batteries – Batteries General – Introduction | Overview

Fig. 2 Daniell’s type cell with two different electrolytes separated with a salt bridge to assure charge neutrality. From Panero, S. Electrochemical Theory: Thermodynamics. In Encyclopedia of Electrochemical Power Sources; Garche, J., Dyer, C. K., Moseley, P. T., Ogumi, Z., Rand, D. A. J., Scrosati, B., Eds.; Elsevier: Amsterdam, 2009; vol. 2, pp. 1–7.

i. The Gibbs free energy, also called the free enthalpy of the reaction: DG ¼ DH − T DS

(5)

DG is the energy liberated or absorbed in a reversible process at constant pressure and constant temperature. Its value results from the difference between the total energy in the system, i.e., the change of the enthalpy DH (heat of reaction at constant pressure) minus the entropy change DS times the temperature T in Kelvin degrees, where T DS is the amount of energy in the system that is not available to do useful work. If DG < 0 the reaction is called exergonic which means, work producing. In this case the reaction proceeds spontaneously. A reaction for which DG > 0 is called endergonic which means work-consuming, i.e., the reaction can be made to occur only by doing work on it.9 ii. Gibbs free energy of a chemical reaction is given by: DG ¼ DG0 + RT ln Q

(6)

Q ¼ activity of products=activity of reactants

(7)

DG ¼ Standard Gibbs energy Q ¼ reaction quotient R ¼ universal gas constant T ¼ absolute temperature in kelvins 0

The activity is a measure of the “effective concentration” of a species in a reacting system. By convention, it is a dimensionless quantity and the activity of pure substances in condensed phases (liquids or solids) is taken as unity. For example, for the reaction written formally: aA+bB > cC+dD

(8)

Q ¼ fCgc :fDgd =fA ga :fBgb

(9)

DG ¼ DG0 + RT ln fCgc :fDgd =fA ga :fBgb

(10)

is

Replacing (9) in (5):

If DG ¼ 0 then Q ¼ K the thermodynamic equilibrium constant for the reaction.

Batteries – Batteries General – Introduction | Overview 3.3

691

Electrochemical equilibrium

The cell voltage Ucell ¼ jcathode − janode > 0 is connected to the Gibbs free energy by the equation. DG ¼ −z F Ucell F ¼ 26,802 Ah mol−1 (Faraday constant) and z ¼ number of exchanged electrons.

(11)

In equilibrium the standard Gibbs energy DG0 equals the corresponding electrical energy driven by the standard cell potential U0: DG0 ¼ −z F U0 At equilibrium DG ¼ 0. Replacing in (10) the DG in (12): h

i U0 ¼ RT=zF ln K ¼ RT=zF ln fCgc :fDgd − ln fA ga :fBgb

(12)

0

(13)

fequilibrium activitiesg

Eq. (13) allows the calculation of equilibrium constants. However, when the conditions differ from the standard equilibrium state (12), the cell potential can be calculated with the Nernst equation (14). Ucell ¼ U0 − RT=zF ln Q In an approximate form of the Nernst equation (14), molal activities in (7) are replaced by molar concentrations.

3.4

(14) 8

Electrochemical reaction kinetics

For systems in thermodynamic equilibrium no detailed knowledge is required about the reactions occurring at the anode and cathode, i.e., the previous results are valid at defined temperature and electrolyte concentrations. However, for cells outside thermodynamic equilibrium, it is not easy to find laws valid for all electrochemical systems. An exception would be the empirical law published in 1897 by W. Peukert. His equation valid for many electrochemical storage systems, relates the discharge current with the discharge time (15), in the form: In  t ¼ IN n−1 CN

(15)

I ¼ actual discharge current in Ampere t ¼ actual time to discharge the battery expressed in hour (h) n ¼ Peukert coefficient (e.g., lead-acid with thick plates n ¼ 1,4 and thin plates n ¼ 1,2–1,25) CN ¼ Nominal capacity in ampere.hour (Ah) IN ¼ Discharge current of the nominal capacity The understanding of the electrode’s reaction kinetics, however, require detail knowledge of the electrochemical and physic-chemical processes occurring at the electrodes. The early way to describe departures from thermodynamic equilibrium for the electrodes was using the term “polarization”1 mentioned in the previous section. Although still used in the praxis, the term polarization has been over time increasingly replaced by the more precise concept, “overpotential.” The difference of the cell voltage with current flow minus the voltage at open circuit is the cell overvoltage.10 Consequently, the cell overvoltage is the sum of the overpotentials at each of the two electrodes. Due to the electrode overpotentials the cell voltage of an electrochemical cell during discharge is always lower than the open-circuit value, reversely, it is always higher during charging, i.e., the overpotential or polarization of an electrode is a measure of the reaction inertia of an electrode to the flow of current. In the electrolyte the charge is carried mostly by the positive and negative ions; solid electrolytes have mostly an ionic conductivity which is carried only by the cation, sometimes also only by the anion. Electrons are transported in the electrode via its electronic conducting parts (active masses, additives, collectors) and the terminals. The transition from electronic to ionic conductivity and vice versa is the cause of the electrode polarization, which is governed by a whole series of processes, like11: i. Charge transfer between the electrolyte and electrode exponentially dependent on overpotential and temperature (Butler-Volmer equation) ii. Resistance to the flow of the ionic current in the electrolyte (electric field driven ion migration) iii. Ionic diffusion driven by concentration gradients in the electrolyte (Fick’s law) iv. Temperature dependent ion adsorptions at the electrode surface v. Solubility of the electrode reaction products vi. Buildup of passive layers at the electrode surface vii. Changes of the crystal structure of the electrodes The electrode potentials normally determined by the main reaction are often shifted by side reactions like the generation hydrogen and oxygen, relevant for aqueous electrolytes.

692

Batteries – Batteries General – Introduction | Overview

The electrodes behavior is usually described quantitatively with the help of mathematical models. In most cases a complex task consisting in solving numerically, a set of coupled partial differential equations describing the processes listed above.

4

Cell design principles

A cell consists of an assembly of electrodes with active masses and collectors, separators, electrolyte, container, and terminals. Early electrochemical primary and secondary cell assemblies in the 19th century were the result of empirical trials. The cells were manually assembled using a limited available range of suitable materials. On the contrary, the prototype cells developed in the 20th century could benefit from the higher variety of materials and a deeper insight in electrochemical processes. For the choice of electrode material and electrolytes apply following criteria:

• • • • • • • • • • • • • • 5 5.1

High open circuit cell potential High versatility, the cell is designed to perform in high and low current applications as well Low self-discharge rate. Not always reachable for aqueous electrolyte cells Limited potential variation during cell discharge High ionic conductivity of the electrolyte Ability to sustain high (discharge and charge) current densities. High gravimetric and volumetric energy density Chemical stability of the electrodes with respect to other cell components (e.g., the electrolyte) Electrochemical stability of cell components in the operating potential window and temperature range Low solubility of the active mass discharge species in the electrolyte Low diversity for metals and compounds in the cell (relevant for manufacturing and recycling) Low cost of cell components and the world-wide availability of the raw materials Low toxicity and flammability Recyclability of the components

Cell components Active mass

The positive and negative active mass are the electrochemical active components of the cell, i.e., are the dischargeable and rechargeable materials. The composition of the positive active mass defines the distinguishing properties of the cell. The dominant element in the active mass usually gives the name to the storage, for example Na-S, Li-I2, Ni-Cd, Ni-Zn, Ni-H2 or the cathode compound like Li-MnO2, Zn-Ag2O.

5.2

Current collectors

The positive and negative current collectors are generally electronic conducting materials mostly metals or metallic alloys in the form of thin foils or grids. The current collectors have a double function: 1. carry the electrical current out of the cells. 2. support mechanically the active masse. Typical requirements for the current collectors are good electronic conductivity, low weight, and compatibility with the electrolyte to avoid corrosion, dissolution, or passivating layers.

5.3

Electrolyte

The electrolyte allows the motion of cations and anions towards the electrodes. It is an ionic conducting solution consisting of a solute and a solvent. In aqueous electrolytes the solvent is water and the solute can be an acid, a hydroxide or a soluble salt. In aprotic electrolytes the solvent is a mixture of organic solvents, with Li or a Na salt as solute. There are also solid-state electrolytes (e.g., b-alumina), molten-salt and polymeric electrolytes. An important requirement for the electrolytes apart from the high conductivity is the electrochemical stability of its components in the cell potential and temperature working range. In spill-proof design cells the electrolyte is immobilized. For example, in spill-proof VRLA-gel cells the H2SO4 is immobilized in a SiO2 gel dispersion. A main difference between lead-acid compared to almost all other electrochemical storage technologies is that the electrolyte also takes part in the main electrochemical reaction. See overall cell equation (16): 2PbSO4 + 2H2 O > Pb + PbO2 + 2H2 SO4

(16)

Consequently, electrolyte properties like ion conductivity or diffusion coefficient vary as a function of the state of charge of the cell.

Batteries – Batteries General – Introduction | Overview 5.4

693

Separator

The separator materials are basically electronically non-conductive, microporous structures with good wettability, small and uniform pore-size, good strength and flexibility, and low ionic resistance in order to maintain the ionic conductivity. In the cell the separator is used to separate mechanically the anode and cathode cell electrodes. The separators, depending on the system, have supplementary functions like improve the retention of the active mass to prevent shedding, to hinder or at least minimize dendritic growth. The materials used to make separators are specific to the cell chemistry characteristics, i.e., there are separators for alkaline, lead-acid, redox and lithium cells. A general requirement is the chemical stability at high temperature and in contact with the electrolyte.

5.5

Case

The case or jar of a cell or a block is an essential component. The material of the case depends on the chemistry and its shape depends on the form factor of the cell, i.e., prismatic, cylindrical or coin. Cylindrical cells have usually metallic cases like stainless steel for Li cylindrical cells. A prominent example is the rechargeable Li-cell 18,650 (18 mm diameter and 65 mm length) introduced in 1991 used in EVs (see cell Li-ion rechargeable types in Fig. 3). Lithium-ion prismatic rechargeable cells also have aluminum cases to reduce weight. High temperature cells of sodium-sulfur (Na-S) and sodium-nickel-chloride (Na-NiCl2) have also cylindrical metal cases. On the other hand, vented and VRLA lead-acid, NiCd and other alkaline cells and blocks with aqueous electrolytes have plastic containers. For example, polypropylene, polystyrene, polyphenylene oxide (PPO), ABS or SAM.

6

Cell design

Large capacity cells like Ni-Cd and some lead-acid used in industrial applications have prismatic designs with flat plates. The prismatic, vented and VRLA gel lead-acid industrial cells in Fig. 4 are built with plates groups connected in parallel. The left drawing in Fig. 5 shows schematically a standard battery of interconnected monopolar cells. The lugs of the flat plates from the positive and negative stacks are connected to a busbar. In the case of lead-acid batteries like the starter batteries the busbars are in most cases welded through the jar partition. The usual way to achieve high discharge power with monopolar cells is by assembling large numbers of thin plates stacked close together. The prismatic cell design with flat plates is used for Li-ion batteries as well. For these batteries the electrode stacks are not only packed into stable metallic or plastic housings like in the case of prismatic cells but also into Al-plastic foils (Fig. 6). This so-called pouch or coffee bag cells are lighter than prismatic cells. More modern battery designs have bipolar plates like schematically shown in the drawing right of Fig. 5. In the bipolar plates the current flows always horizontally, this means that there is no need for the top-lead busbar, i.e., the bipolar cell design allows higher

Fig. 3 Cutaway view of a cylindrical spiral wound Li-MnO2 primary cell. Source: Hoppecke.

694

Batteries – Batteries General – Introduction | Overview

Fig. 4 Lead-acid cells for stationary industrial applications. Left: OPzS tubular plates flooded cell Right: OPzV type VRLA-gel cell with tubular plates. Source: Hoppecke Batterien.

Fig. 5 Prismatic cell with monopolar and bipolar plate configurations (arrows show the current flow).

Fig. 6 Rechargeable Li-ion cells: (a) cylindrical, (b) pouch, (c) prismatic.

Batteries – Batteries General – Introduction | Overview

695

gravimetric and volumetric energy densities compared to the conventional monopolar equivalents. The capacity of the bipolar plate battery is however limited by the plate size because the active mass of each cell must be pasted on a single plate. Primary and rechargeable batteries for portable devices can either be built in the prismatic or cylindrical cell designs: There are two configurations for cylindrical cells: a. The bobbin is the usual construction for zinc-carbon and alkaline-manganese dioxide cells and for Li-SOCl2 primary cells. The active mass is molded into two concentric cylinders as shown in the patent drawing for the zinc/alkaline/manganese dioxide cell in Fig. 1. In this configuration it is possible to maximize the amount of the active material, but at the expense of the electrodes surface area. This increases capacity but reduces the high-rate performance of the cell. b. The spirally wound is the usual construction for high power cells like the Li-MnO2 primary in the cutaway view of Fig. 3. The anode was a Li metal foil and the cathode consisted of a mixture of g-MnO2 and carbon pasted on an Al expanded metal current collector. The advantage of the spirally wound construction compared to the bobbin is the shorter average distance of the positive active mass to the current collector which reduces the ohmic IR-drop. The spirally wound design is often used for rechargeable Li-ion and high power VRLA-AGM lead-acid starter batteries. Coin cells are primary cells used to power small electronic devices. The coin cell design12 is common for zinc-air and Li-MnO2 (Fig. 7).

6.1

Redox-flow cell design

In this electrochemical storage the conversion of chemical energy in electricity is generated in a reactor cell by ion exchange through a membrane between an anode and a cathode. The voltage of a redox flow cell (RFC) results from the reaction between two electrolyte solutions that are separately pumped into the reactor cell—the anolyte and the catholyte solutions are stored in separate external tanks (Fig. 8). For example, the electrochemical reactions for a redox-flow vanadium cell are: V5+ + V2+ > V 3+ + V 4+

ðOCV ¼ 1, 7V Þ

(17)

Fig. 7 Cutaway view configuration of a Li-MnO2 coin cell. From Brodd, R. J. Notebooks: Batteries. In Encyclopedia of Electrochemical Power Sources; Garche, J., Dyer, C. K., Moseley, P. T., Ogumi, Z., Rand, D. A. J., Scrosati, B., Eds.; Elsevier: Amsterdam, 2009; vol. 2, pp. 22–29.

Fig. 8 Block diagram for a Redox-Flow storage system.

696

Batteries – Batteries General – Introduction | Overview

OFF-GRID SOLUTION Fields of application: independent off-grid solutions

Generator e.g. wind turbine, diesel generator

DC/DC converter PV plant

DC DC

Lead-acid battery

+ –

Li0n battery

+ –

AC/DC inverter

DC

AC

DC

DC

Island grid DC DC

5

Fig. 9 Stationary BESS and connection layout for off-grid applications. From Riegel, B.; Cattaneo, E.; Büngeler, J.; Herrmann, M.; Wu, E. New Requirements for Li-Ion and PbA Batteries in the Standard Industrial Applications. 9th International AABC Advanced Automotive Battery Conference, Strasbourg, France, 27–31 January 2019.

The main difference between a redox-flow cell with a conventional electrochemical cell is the spatial separation of the storage containers (electrolyte tanks) and the reaction site (membrane flow-cell with carbon felt bipolar electrodes as current collectors). This means that the stored electrical energy (capacity) and power are independently scalable and can be combined as needed for the application.

6.2

Battery and system

In the last years the batteries have been increasingly integrated into larger structures which include the battery modules, battery containers with normed dimensions and ventilated or air-conditioned battery rooms. These storage units are usually equipped with bi-directional converters and, depending on storage technology, with complex battery management systems (BMS). Fig. 9 shows an example of a stationary Battery Energy Storage System (BESS) with interconnected lead acid recombinant VRLA gel cells.13 The choice of the battery technology for battery electrical storage systems (BESS) for on-grid and off-grid energy applications depends mainly on the application.

7 7.1

Battery market and applications Global market

Fig. 10 shows the worldwide battery market related to value summarized for all market segments und all battery technologies for different years. Fig. 10 shows also that the secondary batteries are responsible for the largest part (c.90%) of the value of the global battery production with a compound Annual Growth Rate (CAGR) of 14% (2016–2022). The primary battery part amounted to about 10% of the global battery production and the CAGR in the same time amounts only c.4%. These numbers reflect the applications, because primary batteries are usually only used in portable applications, they are

Batteries – Batteries General – Introduction | Overview

697

Fig. 10 Global battery market in value ($bn) 2016–2022. From Pillot, Ch. The Rechargeable Battery Market, and Main Trends 2022–2030. Conference: Pb 2023, Athens, 21–23 June 2023.

Fig. 11 Global secondary battery market (2010–2030), (a) value ($Bn) on pack level, (b) volume (GWh). From Pillot, Ch. The Rechargeable Battery Market, and Main Trends 2022–2030. Conference: Pb 2023, Athens, 21–23 June 2023.

increasingly facing competition from secondary batteries. Primary batteries most in use are alkaline cells (60%) followed by Li cells (25%),14 as well as special cells as Li coin, Zn-air, Zn-Ag2O, and special alkaline. An analysis of the global value and the production capacity of important secondary batteries (Li-ion, Lead-acid, Ni-Cd, and Ni-MH) shows Fig. 11. It can be seen in Fig. 11 that up to the beginning of the 20th of this century the secondary battery market was dominated by the lead-acid battery. But then the lead-acid battery was overtaken relatively quickly by the Li-ion battery, to which the EV applications contributed. The LIB share in 2030, is estimated to be of 82%, with the total battery volume being expected to reach 4,1 TWh. Other sources assume 6,7 TWh in 2030 and 20 TWh in 2050. Elon Musk forecasted 300 TWh global battery capacity to reach net zero emissions by 2050.15 Related to applications the battery market is divided as shown in Fig. 12.

7.2

Portable applications

Batteries for portable devices including power tools have since years a relative fixed market segment with around 10% market share for 2022 and an annual growth of 4%—see Fig. 13.

698

Batteries – Batteries General – Introduction | Overview

Fig. 12 Value ($Bn) of the global battery market (1990–2022) on pack level related to applications. From Pillot, Ch. The Rechargeable Battery Market, and Main Trends 2022–2030. Conference: Pb 2023, Athens, 21–23 June 2023.

70000

Volume/MWh

60000 50000

Other Portable Electronics

40000

Tablets

30000

Portable PCs

20000 Cellular Phones

10000 0 00 05 10 15 20 25 30 20 20 20 20 20 20 20

Fig. 13 LIB market for consumer applications in energy (MWh) 2000–2030. From Pillot, Ch.; Renard, F. The Rechargeable Battery Market and Main Trends 2020-2030. AABC Europe, Mainz, Germany, June 2023.

These devices are used mainly for portable consumer applications as cellular phones, portable computers, cameras, camcorders, and cordless tools but also watches, calculators, personal digital assistants (PDAs), radios, electronic games, compact disk (CD) and digital video disk (DVD) players should be mentioned. Portable non-consumer applications are in the medical area (Pacemaker, implantable cardioverter defibrillator (ICD), drug delivery system, hearing aids etc.), in the military electronic (soldier equipment as laser lights, fire control systems, radio equipment, etc.) or special measurement systems, global positioning system (GPS). Originally, especially primary batteries were used for portable applications, which had a significantly higher energy density compared to the secondary batteries available at the time and were also cheaper. As primary batteries are used mostly alkaline cells (60%) followed by Li cells (25%),15 and special cells as Li coin, Zn-air, Zn-Ag2O, and special alkaline. With the development of LiBs, which have a specific energy up to 300 Wh/kg, a large part of the portable applications, particularly the high-quality ones, which are mostly used on a daily basis, are covered by secondary Li-ion batteries. The distribution of the LIB on the various portable consumer applications today and in the future is shown in Fig. 13.

7.3

Automotive and transportation application

7.3.1 Automotive application The main automotive battery application is related to Starting, Ignition and Lighting of the car. These three functions are performed by the SLI battery based on the Lead-acid technology. Additional functions have recently been added, namely the start-stop function as well as the regenerative braking and motor assist function, which is used in micro-hybrid cars. The start-stop function and the motor assist function could be fulfilled by the classic SLI battery, but not the regenerative braking function with extremely high charging currents since the charging rate of the classic SLI battery is limited. Further developments of the LAB led to a significant improvement in the charging rate. Currently more than 50% of the newly produced passenger cars are micro-hybrid vehicles and are equipped with the improved type of LAB. This automotive application market amounted to about 15% (2022) of the total battery market by value.

Batteries – Batteries General – Introduction | Overview

699

7.3.2 Transportation application In 2010 the market share of battery vehicles (EV, PHEVs) was with 0.01% very low16 (100%: 70 Mill. cars).17 The enormous development of electric battery vehicles sales up to 2023 with about 17.7% market share (100%: 70.8 Mill. cars) can be seen in Fig. 14. In the future even stronger growth is expected with approx. 30% CAGR (2020–2030). There are two reasons for the strong growth: On the one hand, the performance of Li batteries has improved significantly (e.g., spec. energy 300 Wh/kg) and the price has reduced strongly (approx. 150 EUR/kWh pack) and on the other hand, the pressure to reduce CO2 emissions has increased enormous. The transport sector contributes to about 16.2% of the global emissions.18 The CO2 emission reduction will be mainly done by the replacement of the ICE vehicle technology by electric vehicle technology with batteries. So, the EU officially banned the sale of petrol and diesel cars with internal combustion engine (ICE) vehicles, from 2035 on.19 For an effective CO2 reduction by EVs, however, the CO2 footprint especially in the production of batteries must be reduced. It depends strongly on materials used in the battery and the CO2 emission/kWhelectricity on the location of the production. It is estimated that a lithium-ion battery with NMC622 chemistry has a CO2 footprint of 78 kgCO2/kWhbatt produced with EU grid electricity in 2022.20 In the new EU battery law,21 the focus is on the CO2 footprint which must not exceed a maximum value specified by the EU in 2026.

7.4

Stationary applications

The main stationary applications are telecom, UPS, and energy storage, where a strong market growth is expected especially in the energy storage market—see Fig. 15. Telecom batteries provide backup power in situations where the main power source is insufficient or unavailable. Typically, the frequency of battery use is for good networks about 15 times and for poor networks up to 300 times per year. Uninterrupted power supply (UPS) batteries act when utility power fails to ensure that critical equipment can safely shut down to protect the operation of single computers or big data centers, buildings, and power plants. Batteries for energy storage systems (BESS or only ESS) are used for residential and commercial storage, grid-connected utility grid-scale energy storage, and off-grid energy storage. Their share in the stationary application area is growing disproportionately. The reason for this is the constantly increasing share of renewable energy in the electrical network (approx. 40%22) and the need for grid balancing to use the energy more economically.

EV & PHEV market worldwide, 2015-2023 18,0

18,0%

17,7%

16,0

16,0% 14,3 14,0% 13,0%

12,0

12,0% 10,5

10,0

10,0%

8,0

8,0%

6,0

77% 6,0% 4,2% 3,3

4,0

73%

2,0

EV & PHEV Market share

Million EV & PHEV

14,0

4,0% 2,0%

70% 0,6% 0,0 62% 15 016 017 018 019 020 021 022 023 2 2 2 2 2 20 2 2 t2 as c re Fo EV

PHEV

0,0%

EV & PHEV %

Fig. 14 Global EV and PHEV market (2015–2023). From Pillot, Ch.; Renard, F. The Rechargeable Battery Market, and Main Trends 2020-2030. AABC Europe, Mainz, Germany, 20–22 June 2023.

700

Batteries – Batteries General – Introduction | Overview

700 000 600 000

MWh

500 000 400 000 300 000 200 000 100 000 2015

2020

ESS Residential (LlB + LAB) UPS

2022

2025

2030

ESS Grid (LlB + LAB) TELECOM

GENERATION

Fig. 15 Global battery storage market (2022–2030) for different applications. From Pillot, Ch.; Renard, F. The Rechargeable Battery Market, and Main Trends 2020-2030. AABC Europe, Mainz, Germany, 20–22 June 2023.

8

Outlook

The battery development from the first available power source with sufficient capacity ever (VOLTA pile 1800) to the multitude of applications that have become an integral part of our lives. It is clear to see that new applications became possible whenever the battery performance was significantly improved. The improvements were made continuously through increase of parameters of the various battery systems, but also “revolutionary” through the introduction of new battery systems. A milestone in this development was the commercialization of the Li-ion battery with its high specific energy and, in the meantime, low costs, which made many applications particularly in the transportation and portable sectors only possible. This development is supported by the energy transformation that replaces fossil energies with regenerative energies. The batteries become a vital storage tool which ensures a stable power supply. Demand for Lithium-Ion batteries to power electric vehicles and energy storage has seen exponential growth, increasing from just 0.5 GWh in 2010 to around 526 GWh a decade later. Demand is projected to increase on about 3360 TWh (see Fig. 11) or even higher. This tremendous increase in the expected production rate leads to enormous challenges in the resources sector. To be able to realize these numbers, a strong expansion of lithium, cobalt, and nickel supply chains is necessary, whereby ecological and socio-economic boundary conditions have to be respected. Furthermore, materials with a reduced or zero Ni and Co content should be used (e.g., LFP), or systems with different resource requirement (e.g., Na-ion), as well as systems with a high specific energy where more electrical energy can be extracted from a given mass (all solid-state batteries). In addition, strict recycling and circular economy must be standard.23 For an effective CO2 reduction using batteries, however, the CO2 footprint especially in the production of batteries must be reduced. It depends strongly on materials used in the battery and the CO2 emission/kWhelectricity on the location of the production. In the new EU battery law,21 the focus is on the CO2 footprint which must not exceed a maximum value specified by the EU in 2026.

See also: Batteries – Battery Types – Lead-Acid Battery: Automotive Batteries: Applications and Market Trends; Batteries – Battery Types – LeadAcid Battery: Automotive Batteries: Products; Cell Components – Electrodes: Active Materials - Microstructures and Interphases; Cell Components – Separators: Overview; Electrochemical Fundamentals: Thermodynamics of Electrode Reactions

References 1. Rand, D. A. A Journey on the Electrochemical Road to Sustainability. J. Solid State Electrochem. 2011, 15, 1579–1622. 2. Barak, M. Primary Batteries - Introduction. In Comprehensive Treatise of Electrochemistry, Volume 3: Electrochemical Energy Conversion and Storage; Bockris, J. O.’M., Conway, B. E., Yeager, E., White, R. E., Eds.; Springer, 1981. Chapter 4. 3. Marsal, P.; Kordesch, K.; Urry, L. F.. Dry Cell, US2960558 patent filed 1957 and granted in 1960 to Union Carbide Corporation, n.d.

Batteries – Batteries General – Introduction | Overview

701

4. Kordesch, K.; Daniel-Ivad, J. Rechargeable Zinc/Alkaline/Manganese Dioxide Batteries. In Handbook of Batteries; Linden, D., Reddy, T. B., Eds, 3rd ed.; McGraw Hill, 2003. Chapter 26. 5. Whittingham, M. S. Lithium Batteries and Cathode Materials. Chem. Rev. 2004, 104, 4271–4301. 6. Yasami, R.; Touzain, P. A Reversible Graphite-Lithium Negative Electrode for Electrochemical Generators. J. Power Sources 1983, 9, 365–371. 7. Mizushima, K.; Jones, P.; Wiseman, P.; Goodenough, J. B. LixCoO2 (01000 cycles at 100% depth of discharge (DoD) has been demonstrated. Cycling at C/3 (charge for 65 min and C/1.6 discharge for 35 min) at 40% DoD with an overcharge factor of 1.02 at 22  C has been demonstrated for over 8000 cycles. Under similar conditions, a Ni–H2 cell requires 1.1 overcharge factors. Nickel–metal hydride cells have a higher rate of self-discharge in relation to Ni–Cd cells. Highly facile oxygen recombination reaction at MH alloy electrode can accelerate the decomposition of positive active material and gaseous hydrogen in equilibrium with the metal hydride alloy can reduce the charged positive electrode as shown below: 2NiOOH + H2 ! 2NiðOHÞ2

(IX)

It is noteworthy that hydrogen oxidation is inhibited by anodically formed Ni(OH)2/NiOOH. But cathodically deposited nickel hydroxide does not inhibit the reaction to the same degree. Shuttle reactions owing to the presence of impurities, such as nitrate present in the electrodes and the electrolyte, contribute to self-discharge to a significant extent in a fresh cell. Oxidation of organic impurities that are extracted or leached out from the separator at the positive electrode also contributes to self-discharge process. Metal ions leached out from the MH alloy may have adverse reaction at the positive electrode. It is reported that the main contributing factor to the decomposition of conventional polyamide separator is the production of ammonia and amine, which participate in shuttle reactions similar to NO3− ions in Ni–Cd battery. The charge retention can be drastically improved by using chemically stable separators such as sulfonated polypropylene. Fig. 7 depicts the comparison of charge retention characteristics of Ni–MH cells while using different separator materials. Addition of aluminum and zirconium improves the charge retention in AB5-type alloys.

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Overview 1.6

3.0

National Panasonic 25 Ah sealed NiMH cell

1.5

2C

1C

2.5 0.3C

0.1C

Cell voltage (V)

1.4 2C 1.3

1C

Cell voltage

0.3C

1.2

0.1C

2.0 1.5 1.0

Internal pressure 0.5

1.1 1.0 0.9

Ambient temperature 20 qC 0

5

10

15

20

25

30

35

Internal pressure (kg cm2)

678

0 0.5 40

Capacity (Ah)

Fig. 4 Pressure profiles for a Ni–MH cell during its charging at various rates. Reproduced from Shukla, A.K.; Venugopalan, S.; Hariprakash, B. Nickel-based Rechargeable Batteries. J. Power Sources 2001, 100, 125–148.

1.6

Gold peak NiMH cell

Charge: 0 1 C, 14 h Temperature: 20 qC

Cell voltage (V)

1.4 0.2 C 0.5 C 1C 2C

1.2

1.0

3C Nominal capacity 3000 mAh Specific energy 80 Wh kg1 Energy density 260 Wh l1

0.8

0.6

0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Capacity (Ah)

Fig. 5 Effect of discharge rates on the capacity of a Ni–MH cell. Reproduced from Shukla, A.K.; Venugopalan, S.; Hariprakash, B. Nickel-based Rechargeable Batteries. J. Power Sources 2001, 100, 125–148.

4

Current status, opportunities, and challenges

A very common problem with hydrogen storage materials is the severe volume expansion during charge-discharge process, leading to cracking and pulverization of the alloy and making it amenable to oxidation. In addition, dissolution of the alloy in the electrolyte contributes to capacity decay. Present materials science strategies are concentrated in combining different phases and microstructures to overcome shortcomings associated with the bulk metal hydrides. To increase rate capability, materials with high surface area are being produced by powder metallurgy, mechanical alloying, and chemical/electrochemical etching. All the materials with high surface area invariably suffer from lower cycle life owing to the increased oxidation of their surfaces. Metal hydride electrodes are susceptible to widespread cracking and irreversible oxidation, which affect their cycle life, stability, and rate capability. Metallurgical processes such as encapsulation (electroless plating of copper and nickel), doping with palladium and cerium, rapid solidification, and macroalloying have shown only a limited success in reducing the rate of degradation. Pure LaNi5 electrodes in contact with potassium hydroxide vapor undergo brittle fracture during hydriding, resulting in a rapid decay of capacity with cycling. Cobalt addition with aluminum or silicon has been found to significantly improve the cycling behavior of Ni–MH cells. Laboratories around the world have exhausted nearly all the elements in the periodic table to synthesize various AB2 and AB5 alloys in their efforts to improve the cycle life and capacity of metal hydride cells. The major problem with metal hydride electrodes is that the function of the alloying elements either acting alone or in combination with other alloying elements cannot be predicted with certainty. Conventional AB5 and AB2 alloys based on

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Overview

679

120

Nominal capacity (%)

110 100 90 Aerospace cell Commercial cylindrical cell

80 70 60 50 20

10

0

10

20

30

40

50

Temperature (qC)

Fig. 6 Effect of temperatures on the capacity of Ni–MH cells. Reproduced from Shukla, A.K.; Venugopalan, S.; Hariprakash, B. Nickel-based Rechargeable Batteries. J. Power Sources 2001, 100, 125–148.

100

Test temperature 20 qC

Charge retention (%)

90

80

70

Polyolefin separator

60

Sulfonated hydrophilic polypropylene separator state-of-the-art cell Hydrophilic grafted polypropylene separator Conventional NiCd with polyamide separator

50

40

1st generation NiMH cell with polyamide separator 0

20

10

30

Time (days)

Fig. 7 A comparison of charge retention characteristics of Ni–MH cells with different separators. Reproduced from Shukla, A.K.; Venugopalan, S.; Hariprakash, B. Nickel-based Rechargeable Batteries. J. Power Sources 2001, 100, 125–148.

LaNi5 and (Ti, Zr)Ni2 have relatively low coulombic capacity values between 300 and 450 Ah kg−1 and, hence, research focused on alloys such as TiZrNi2 and Mg2Ni as low-cost, lightweight, and safer alternatives. In order to increase the energy density of the Ni–MH battery, it is mandatory to improve the performance of metal hydride electrodes. With a combination of modifications in the alloy composition and new methods of electrode preparation, discharge capacities between 630 and 780 Ah kg−1 have been achieved at a discharge current density of 50 A kg−1 for the magnesium-based Mg1.9Al0.1Ni0.8Co0.1Mn0.1 alloy electrodes. An amorphous structure appears to be central in achieving high discharge capacities. These results indicate that the kinetics of hydriding/dehydriding reactions of magnesium-based alloy electrodes can be greatly improved by ball milling and chemical coating. Efforts are therefore being directed toward maintaining particle-to-particle electrical contact by the use of polymer binders, compaction of porous nickel foam, surface plating, and doping and compaction with conductive powders. The specific energy of Ni–MH batteries can vary from 40 to 110 Wh kg−1 depending on the requirement. For example, where device runtime is paramount, Ni–MH batteries need not have high-power capability. On the other hand, for extremely high power charge and discharge requirements, factors that affect the specific energy of Ni–MH batteries are: (a) extra current collection, (b) high N/P ratios (proportion of excess negative electrode capacity to positive electrode capacity), and (c) specific cell design and construction.

680

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Overview Table 1

Evolution in specific energy and energy density values of Ni–MH batteries.

Year

Specific energy (Wh kg−1)

Energy density(Wh L−1)

Ni–MH AA cell capacity (mAh)

1991 1993 1996 2000 2002 2003 2005

54 70 80 92 95 102 107

190 235 255 300 345 385 428

1100 1400 1600 1900 2100 2300 2600

Reproduced from Fetcenko, M.A.; Ovshinsky, S.R.; Reichman, B. et al. Recent Advances in NiMH Battery Technology. J. Power Sources 2007, 165, 544–551.

Table 2

Evolution in specific power values of Ni–MH batteries. Specific power (W kg−1)

State of charge

100 80 50 20

1991

1993

1996

1999

2000

2002

2003

170 150 140 120

340 330 320 260

770 710 630 500

1060 1000 950 800

1280 1290 1280 1090

1650 1900 1880 1580

1780 1990 1960 1630

Reproduced from Fetcenko, M.A.; Ovshinsky, S.R.; Reichman, B. et al. Recent Advances in NiMH Battery Technology. J. Power Sources 2007, 165, 544–551.

Table 1 presents evolution in specific energy values of Ni–MH batteries during the during the period 1990–2005 in consumer cylindrical Ni–MH cells for portable power. The specific energy of Ni–MH batteries is kept between 90 and 110 Wh kg−1 for portable power applications, between 65 and 80 Wh kg−1 for EV batteries, and between 45 and 60 Wh kg−1 for hybrid electric vehicle (HEV) and other high-power applications. Although gravimetric energy density is imperative for advanced battery technologies, in many cases volumetric energy density (Wh L−1) happens to be more important. It has been possible to achieve energy density values as high as 420 Wh L−1. Cost reduction is at the center stage of Ni–MH battery development. In recent years, high-volume battery production has pushed the cost of Ni–MH batteries below that of Ni–Cd batteries. Table 2 shows the evolution in the specific-power values of Ni–MH batteries for the period 1990–2003. Persistent efforts to engineer the surface oxide have vastly improved the low-temperature performance. The oxide forms a support matrix on ultra-fine metallic catalysts but contains insufficient pore structure to allow rapid transfer of reactants at low temperatures. By realizing increased porosity and specifically sized pore channels, low-temperature power density of Ni–MH batteries has increased from practically zero at −30  C to over 300 W kg−1. Excellent high-temperature performance was obtained for a cobalt-rich zinc-poor powder containing coprecipitated calcium and magnesium. Nickel-metal hydride battery charging efficiency at 65  C was increased from 36% to 85% by formulating the nickel hydroxide active material so as to suppress oxygen evolution. By introducing extremely small metallic nickel alloy inclusions throughout the surface oxide of the metal hydride alloy by the method of preferential corrosion, the catalytic activity was significantly increased through reduced charge transfer resistance and a specific power of 1900 W kg−1 was attained.

5

Conclusion

The performance of Ni–MH batteries has seen continuous improvements over the years since their inception in 1991 through a variety of novel approaches such as high-density negative electrodes, thinner separators, upgraded positive electrodes, and improved packaging efficiencies. At present, Ni–MH batteries are being produced in high volumes for portable power applications. Nickel–metal hydride batteries have now become proven power source for HEVs. In addition to the essential performance targets of energy, power, cycle life, and operating temperatures, flexible vehicle packaging, easy application to series and series/parallel strings, choice of cylindrical/prismatic cells, safety, tolerance to electrical and mechanical abuse, maintenance-free operation, ability to utilize regenerative braking energy, simple and less expensive charging and electronic control circuits, and environmentally acceptable and recyclable materials have established the eminence of Ni–MH batteries. However, the declining costs and mounting performances of secondary lithium batteries are threatening to replace Ni–MH batteries.

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Overview

681

Further readings 1. Cui, N.; Luo, J. L.; Chuang, K. T. Nickel–Metal Hydride (Ni–MH) Battery Using Mg2Ni-Type Hydrogen Storage Alloy. J. Alloys Compd. 2000, 302, 218–226. 2. Davolio, G.; Sorangi, E. The ‘Memory Effect’ on Nickel Oxide Electrodes: Electrochemical and Mechanical Aspects. J. Appl. Electrochem. 1998, 28, 1313–1319. 3. Dhar, S. K.; Ovshinsky, S. R.; Gifford, P. R.; Corrigan, D. A.; Fetcenko, M. A.; Venkatesan, S. Nickel/Metal Hydride Technology for Consumer and Electric Vehicle Batteries – A Review and Up-Date. J. Power Sources 1997, 65, 1–7. 4. Fetcenko, M. A.; Ovshinsky, S. R.; Reichman, B. Recent Advances in NiMH Battery Technology. J. Power Sources 2007, 165, 544–551. 5. Fukunaga, H.; Kishimi, M.; Matsumoto, N.; Ozaki, T.; Sakai, T.; Tanaka, T.; Kishimoto, N. A Nickel Electrode with Ni-Coated 3D Steel Sheet for Hybrid Electric Vehicle Applications. J. Electrochem. Soc. 2005, 152, A905–A912. 6. Furukawa, N. Development and Commercialization of Nickel–Metal Hydride Secondary Batteries. J. Power Sources 1994, 51, 45–59. 7. Gifford, P.; Adams, J.; Corrigan, D.; Venkatesan, S. Development of Advanced Nickel/Metal Hydride Batteries for Electric and Hybrid Vehicles. J. Power Sources 1999, 80, 157–163. 8. Ikoma, M.; Hoshina, Y.; Masumoto, I.; Iwakura, C. Self-Discharge Mechanism of Sealed-Type Nickel/Metal–Hydride Battery. J. Electrochem. Soc. 1996, 143, 1904–1907. 9. Kohler, U.; Antonius, C.; Bauerlein, P. Advances in Alkaline Batteries. J. Power Sources 2004, 127, 45–52. 10. Kohno, T.; Yamamoto, M.; Kanda, M. Electrochemical Properties of Mechanically Ground Mg2Ni Alloy. J. Alloys Compd. 1999, 293–295, 643–647. 11. Kritzer, P. Separators for Nickel Metal Hydride and Nickel Cadmium Batteries Designed to Reduce Self-Discharge Rate. J. Power Sources 2004, 137, 317–321. 12. Kuriyama, N.; Sakai, T.; Miyamura, H.; Ishikawa, H. Solid State Metal Hydride Batteries Using Tetramethylammonium Hydroxide Pentahydrate. Solid State Ion. 1992, 53–56, 688–693. 13. Leblanc, P.; Blanchard, P.; Senyarich, S. Self-Discharge of Sealed Nickel–Metal Hydride Batteries, Mechanisms and Improvements. J. Electrochem. Soc. 1998, 145, 844–847. 14. Linden, D.; Reddy, T. B. Handbook of Batteries, 3rd Edn.; McGraw-Hill: New York, 2002. 15. Morioka, Y.; Narukawa, S.; Itou, T. State-of-the-Art Alkaline Rechargeable Batteries. J. Power Sources 2001, 100, 107–116. 16. Ohms, D.; Kohlhase, M.; Benczur-Urmossy, G.; Schadlich, G. New Developments on High Power Alkaline Batteries for Industrial Applications. J. Power Sources 2002, 105, 127–133. 17. Ovshinsky, S. R.; Fetcenko, M. A.; Ross, J. A Nickel Metal Hydride Battery for Electric Vehicle. Science 1993, 260, 176–181. 18. Pistoia, G. Batteries for Portable Devices; Elsevier: New York, 2005. 19. Potter, B. G.; Duong, T. Q.; Bloom, I. Performance and Cycle Life Test Results of a PEVE First-Generation Prismatic Nickel/Metal-Hydride Battery Pack. J. Power Sources 2006, 158, 760–764. 20. Rand, D. A. J.; Woods, R.; Dell, R. M. Batteries for Electric Vehicles; Research Studies Press, 1998. 21. Shinyama, K.; Magari, Y.; Akita, H. Investigation Into the Deterioration in Storage Characteristcs of Nickel–Metal Hydride Batteries During Cycling. J. Power Sources 2005, 143, 265–269. 22. Shukla, A. K.; Venugopalan, S.; Hariprakash, B. Nickel-Based Rechargeable Batteries. J. Power Sources 2001, 100, 125–148. 23. Sun, L.; Liu, H. K.; Bradhurst, D.; Dou, S. X. The Electrode Properties of Mg1.9Al0.1Ni0.8Co0.1Mn0.1 Alloy by Mechanical Grinding With Ni Powders. Electrochem. Solid St. 1999, 2, 164–166. 24. Taniguchi, A.; Fujioka, N.; Ikoma, M.; Ohta, A. Development of Nickel/Metal-Hydride Batteries for EVs and HEVs. J. Power Sources 2001, 100, 117–124. 25. Vincent, C. A.; Scrosati, B. Modern Batteries, 2nd Edn.; Arnold: London, 1997. 26. Weinstock, I. B. Recent Advances in the US Department of Energy’s Energy Storage Technology Research and Development Programs for Hybrid Electric and Electric Vehicles. J. Power Sources 2002, 110, 471–474. 27. Yang, C.-C. Polymer Ni–MH Battery Based on PEO-PVA-KOH Polymer Electrolyte. J. Power Sources 2002, 109, 22–31. 28. Yang, X.-G.; Liaw, B. Y. Numerical Simulation on Fast Charging Nickel Metal Hydride Traction Batteries. J. Electrochem. Soc. 2004, 151, A265–A272.

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Metal Hydrides PHL Nottena,b and M Latrochec, aPhilips Research Laboratories, Eindhoven, The Netherlands; bEindhoven University of Technology, Eindhoven, The Netherlands; cCMTR-ICMPE, UMR 7182, CNRs, Thiais, France © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of P.H.L. Notten, M. Latroche, SECONDARY BATTERIES – NICKEL SYSTEMS | Nickel–Metal Hydride: Metal Hydrides, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 502–521, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00164-7, with revisions made by the Editor.

1 Introduction 2 Basic principles of Ni–MH batteries 2.1 Energy storage reactions 2.2 Side reactions 2.2.1 Overcharging 2.2.2 Overdischarging 2.2.3 Self-discharge 3 Hydride-forming electrode materials 3.1 AB5-type compounds 3.1.1 Structural properties 3.1.2 Structural properties of metal hydrides 3.2 Nonstoichiometric AB5+y compounds 3.3 Other metallic compounds suitable for electrochemical application 3.3.1 AB2 compounds 3.3.2 AB compounds 3.3.3 AB3–4 compounds 3.3.4 Magnesium-based compounds 3.3.5 Comparison between the various compounds See also Further reading

683 683 683 685 685 687 688 690 690 690 692 693 695 695 696 696 698 699 700 700

Abstract The properties and characteristics of rechargeable Nickel-Metal Hydride batteries are essential dictated by the chemical and structural characteristics of the metal hydride. This chapter depicts the relationships between the chemical composition, structural characteristics and properties of the intermetallic compounds employed in these batteries. A general overview is provided on the mechanisms taking place during insertion/deinsertion of the hydrogen, together with the intrinsic limitations of the materials.

Key points

• •

Brief overview of the basic principles of Mi-MH batteries and of the side reactions Overview of the hydride-forming materials, their structural characteristics and their properties

Nomenclature

Symbols and units

C E Eeq EMH ENi ENI–MH Eo F I
Ia Ic

682

Specific electrochemical storage capacity (mAh g−1) Potential (V) Equilibrium potential (V) Electrode potential MH electrode (V) Electrode potential Ni electrode (V) Battery voltage (V) Redox potential (V) Faraday constant (As eq−1) Exchange current (A) Anodic current (A) Cathodic current (A)

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00323-6

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Metal Hydrides

n PH2 Qd R RA RB Re T W WNi–MH DS h

683

Number of electrons in electrochemical reactions Partial hydrogen pressure (Pa) Discharge capacity mAh g−1 Gas constant (J Mol−1 K−1) Radius atom A Radius atom B Electrolyte resistance (O) Temperature (K) Formation of heat (W) Heat evolution Ni–MH batteries (W) Entropy change (J Mol−1 K−1) Overpotential (V)

Abbreviations FCC HEV NMR P(H)EV PCT SoC

1

Face-centered cubic Hybrid electrical vehicle Nuclear magnetic resonance Plug-in (hybrid) electrical vehicle Pressure composition temperature State-of-charge

Introduction

Many potential applications have been suggested since the discovery of hydride-forming metals and related intermetallic compounds, ranging from hydrogen purification membranes, and hydrogen sensors to hydrogen-driven heat pumps. To date, however, only one application has proven its widespread commercial feasibility. The application of hydride-forming compounds to the electrochemical field of rechargeable batteries has become a major success over the past decade. This is, among other things, due to their high storage capacity, good rate capability, reliability, environmental friendliness, and low cost. The discovery in the late 1960s at the Dutch Philips Research Laboratories that the intermetallic compound LaNi5 was able to absorb reversibly large amounts of hydrogen gas has initiated this tremendous technical and commercial success. Soon after this discovery it was realized that electrodes made of this type of hydride-forming materials would be highly favorable as electrochemical energy storage medium and could become a serious alternative for the frequently used nickel–cadmium (Ni–Cd) batteries. It took, however, until the late 1980s before nickel-metal hydride (Ni-MH) batteries had turned into a mature and reliable battery system and, subsequently, had become widely accepted in our present-day portable society. The reason for this long introduction time was related to the fact that the thermodynamic properties and the cycle-life performance of the parent compound LaNi5 in the required alkaline electrochemical environment were rather poor. The development at the Philips Research Laboratories of electrochemically stable multicomponent, AB5-type hydride-forming, compounds eventually solved this serious problem and accomplished the required breakthrough, needed for the realization of Ni–MH batteries. For many years now, commercial Ni–MH batteries have employed mischmetal-based, AB5-type hydride-forming, compounds as negative electrode material but, nowadays, new compounds are developed, leading to a promising future for these alkaline-type batteries. In this chapter the materials research making the Ni–MH battery feasible is reviewed. First, the basic electrochemical principles underlying the energy storage reactions and the various side reactions, occurring during (over)charging and (over)discharging, and open-circuit conditions are highlighted. Subsequently, the properties of hydride-forming materials are outlined in detail, including the crystallographic and electrochemical characteristics of both stoichiometric and nonstoichiometric, intermetallic, AB5+x compounds. Taking into consideration the near future some points regarding the potentially new materials, offering, for example, high potentials as far as energy density is concerned are discussed. For the crystallographic and electrochemical properties of the second (nickel) electrode inside Ni–MH batteries, the reader is referred to the excellent review written by J. McBreen.

2 2.1

Basic principles of Ni–MH batteries Energy storage reactions

A schematic representation of a Ni–MH battery, containing a hydride-forming (MH) electrode and a nickel electrode, is shown in Fig. 1. A separator electrically insulates the electrodes. Both separator and electrodes are impregnated with a strong alkaline solution

684

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Metal Hydrides

Nickel electrode

Hydride electrode

Overcharge 4OH Ni 2H2O+O2+4e 

O2

MH

e H2O NiOOH Charge capacity

Charge

MHx

ch –

OH Ni(OH)2

Overdischarge

xOH

d

M

OH

xH2O

H2

xe

2e+2H2O Ni 2OH+H2 MH

Discharge

Separator impregnated with KOH solution

Fig. 1 The concept of a sealed rechargeable Ni–MH battery seen from a schematic point of view. Reproduced with permission from Notten, P. H. L. In: Grandjean. F.; Long, G. J.; Buschow, K. H. J. (eds.) Rechargeable Nickel–Metal Hydride Batteries: A Successful New Concept; 1994; Vol. 281; Chapter 7; p. 151. NATO ASI Series E. London, ISBN 0-7923-3299-7.

(usually of the order of 7 mol L−1 KOH) that provides for the ionic conductivity between the two electrodes. The overall electrochemical reactions, occurring at both electrodes during charging (ch) and discharging (d) can, in their most simplified form, be represented by ch

NiðOHÞ2 + OH− $ NiOOH + H2 O + e− d

ch

M + H2 O + e− $ MH + OH− d

[I] [II]

During charging divalent Ni(II) is oxidized into the trivalent Ni(III) state and water is reduced to hydrogen atoms at the metal (M) electrode, which are, subsequently, absorbed by the hydride-forming compound. The reverse reactions take place during discharging. The net effect of this reaction sequence is that hydrogen is transported from one electrode to the other. The reaction takes place without any water consumption preventing any drying of the battery contrary to Ni–Cd batteries. It should be noted that the hydrogen stored in the MH electrode is in equilibrium with that in the gas phase. In general, exponential relationships between the partial anodic/cathodic currents and the applied electrode potential are observed under kinetically controlled conditions, as is schematically depicted in Fig. 2 (dashed curves). The potential scale is given with respect to an Hg/HgO reference electrode. The equilibrium potential of the nickel electrode under standard conditions is eq more positive (Eeq Ni ¼ +439 mV) than that of the MH electrode (EMH) – depends on the partial hydrogen pressure of the hydride-forming materials, according to MH + MH ! 2M + H2 "

[III]

Because the preferred partial hydrogen pressure of MH electrode materials is of the order of up to a few 0.01 bars, Eeq MH ranges generally between −930 and −860 mV. This implies that the theoretical open-circuit potential of a Ni–MH battery is approximately 1.3 V (ENi–MH ¼ ENi–EMH), indeed similar to that of Ni–Cd batteries, making these two aqueous battery systems very compatible. During galvanostatic charging with a constant current an overpotential (
) will be established at both electrodes. The magnitude of each overpotential component (
Ni and
MH in Fig. 2) is determined by the kinetics of the charge transfer reactions. An electrochemical measure for the kinetics of a charge transfer reaction is generally considered to be the exchange current Io, which is defined at the equilibrium potential, Eeq, at which the partial anodic current equals the partial cathodic current (see Fig. 2). In the case of the Ni electrode, IoNi is reported to be relatively low, which implies that at a given constant anodic current, IaNi, the established overpotential at the Ni electrode is relatively high (Fig. 2). In contrast, the kinetics of the MH electrode is reported to be strongly dependent on the materials composition. Assuming a highly electrocatalytic hydride-forming compound, this implies that the current–potential curves, characteristic for the MH electrode, are very steep in comparison to those for the nickel electrode, resulting in a much smaller value for
MH at the same cathodic current IcMH, as is schematically shown in Fig. 2. It is evident that the battery voltage under current flow is a summation of the open-circuit potential and the various overpotential contributions, including the ohmic potential drop (
IR) caused by the electrical resistance of the electrolyte (Re). The reverse processes occur during discharging, resulting in cell voltage lower than 1.3 V.

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Metal Hydrides

Ni(OH)2 + OH

ch

685

NiOOH + H2O + e

Ia I qNi

I qMH

Ia

E eq

0

E eq

Ic

Open-circuit potential

MH

Ic

M+H2O+e

ch

E qH

E qH

2O/H2

–1.0

Ni

MH+OH 2O/MH

E qNO

3/NO2

–0.5

E qOH/O

2

0 E (V vs Hg/HgO)

E qNi II/III 0.5

1.0

1.5

Fig. 2 Schematic representation of the current–potential curves for a Ni and MH electrode (solid lines), assuming kinetically controlled charge transfer reactions. The partial anodic and cathodic reactions are indicated as dashed lines. The exchange currents (Io) are defined at the equilibrium potentials (Eeq). Potentials are given with respect to an Hg/HgO reference electrode. Besides the redox potentials (Eo) of the main electrode reactions, those of some side reactions are also indicated. Reproduced with permission from Notten, P. H. L. In: Grandjean. F.; Long, G. J.; Buschow, K. H. J. (eds.) Rechargeable Nickel–Metal Hydride Batteries: A Successful New Concept; 1994; Vol. 281; Chapter 7; p. 151. NATO ASI Series E. London, ISBN 0-7923-3299-7.

2.2

Side reactions

To ensure the well functioning of sealed rechargeable Ni–MH batteries under a wide variety of operating conditions, the nickel electrode is designed to be the capacity-determining electrode, as is schematically depicted in Fig. 1. Such a configuration forces side reactions to occur at the nickel electrode both during overcharging and overdischarging.

2.2.1 Overcharging During overcharging OH− ions are oxidized at potentials more positive with respect to the standard redox potential of the OH−/O2 redox couple (about 0.3 V with respect to Hg/HgO reference in Fig. 2) and oxygen evolution is induced at the nickel electrode, according to Ni

4OH− ! O2 + 2H2 O + 4e−

[IV]

As a result, the partial oxygen pressure inside the sealed cell starts to rise. Advantageously, oxygen can be transported to the MH electrode, where it can be reduced at the MH/electrolyte interface at the expense of the hydride-formation reaction [II] according to MH

O2 + 2H2 O + 4e− ! 4OH−

[V]

Both the oxygen evolution and the so-called oxygen recombination reaction are schematically represented in Fig. 3. Because the overpotential for the recombination reaction at the MH electrode is relatively high, it has been argued that its rate is most probably transport-controlled by the oxygen supply through the electrolyte. The oxygen recombination mechanism ensures that the partial oxygen pressure inside the Ni–MH battery will be maintained low. It should be noted that both oxygen and hydrogen gas are present during overcharging as has recently been analyzed and simulated. Although thermodynamically more favorable (EoNi > EoO2), the parasitic oxygen evolution reaction (Eq. [IV]) only takes place at significant rates at more positive potentials than the voltage range at which the basic nickel reaction generally occurs (Eq. [I]). This is fortunately due to the much poorer kinetics of the oxygen evolution reaction (IoO2 < IoNi) compared to those of the nickel reaction. This generally results in a rather sharp increase of the battery voltage at the end of the charging process, at the point where the overcharging process takes over. This is indeed confirmed experimentally, as Fig. 4 reveals. It is also clear from this figure that the pressure inside the battery sharply rises at the end of the charging process, around 100% state-of-charge (SoC), and tends to level off at higher SoC. This pressure rise is, in fact, dictated by the competition of the oxygen evolution reaction and the nickel reaction and is found to be strongly dependent on the rate at which the Ni–MH battery is charged. Besides the gas pressure build-up inside Ni–MH batteries, the development of the battery temperature is also of considerable importance and influences the

686

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Metal Hydrides

Ni

MH

Separator Gas phase

O2(g)

O2(g)

O2(I)

O2(I) do

2

(c)

(b) (a)

O2(g) formation

O2

Electrolyte

Fig. 3 Schematic representation of the oxygen recombination cycle. Oxygen formation is initiated at the nickel electrode/electrolyte interface. Small gas bubbles are formed and the gas will be transported to the gas phase. Recombination starts by redissolution of oxygen in the electrolyte and will subsequently be reduced at the MH electrode/electrolyte interface. The recombination rate strongly depends on the diffusion layer thickness, through which oxygen has to be transported and is most favorable at the three-phase boundaries, as is schematically indicated by arrows (a)–(c).

thermodynamics and kinetics of the various processes. In addition, the temperature may also induce secondary effects, such as a reduced cycle life. The formation of heat (W) inside a Ni–MH battery can be represented by X   X − Ii TDSi (1) W Ni−MH ¼ + Ii j
i j + I2Ni−MH Re F where Ii are the partial currents flowing through the battery with INi–MH ¼ SIi, T is the temperature, and F is the Faraday constant. The factors, which contribute to the heat evolution during current flow, can be easily recognized in Eq. (1): the entropy changes (DSi) brought about by the various electrochemical reactions (I), the various overpotential components (
i), and the internal battery resistance (Re). As long as the basic electrochemical reactions (Eqs. [I] and [II]) proceed inside the battery, the overpotentials established at both the nickel and MH electrode are relatively small (see Fig. 2). This implies that the heat contribution, resulting from the electrode reactions, is limited, resulting, in turn, in a rather moderate temperature rise (Fig. 4). However, this situation changes drastically as soon as the oxygen recombination cycle at the MH electrode starts to occur. Because the MH electrode

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Metal Hydrides

7

687

70

6 60

1.6

P (bar)

1.4

E (V)

T P

1.2

1

0

25

50

75

100

5

50

4

40

3

30

2

20

1

10

125

T (qC)

E

0

State-of-charge (%)

Fig. 4 Development of the cell voltage (E), the internal gas pressure (P), and the cell temperature (T ) as a function of state-of-charge (SoC) for a Ni–MH battery during charging and overcharging with a high (3 A) current, that is, the battery is fully charged within 20 min. Note that during overcharging the temperature increases sharply and, consequently, the battery voltage decreases. From Notten, P. H. L. In: Grandjean. F.; Long, G. J.; Buschow, K. H. J. (eds.) Rechargeable Nickel–Metal Hydride Batteries: A Successful New Concept; 1994; Vol. 281; Chapter 7; p. 151. NATO ASI Series E. London, ISBN 0-7923-3299-7.

potential is at least 1 V more negative with respect to the standard redox potential of the OH−/O2 couple, this implies that the established overpotential for the oxygen recombination reaction is extremely high (>1.2 V). Considering Eq. (1), it is expected that the heat evolution inside a battery will sharply increase as soon as the oxygen recombination cycle starts. This is indeed in agreement with the pronounced temperature increase found during overcharging in the experiments (see Fig. 4) and has been confirmed by simulations. As a result of this severe temperature increase the voltage tends to decrease during overcharging. This so-called –DV/ dt effect is often used as signal to terminate the charging process.

2.2.2 Overdischarging Protection against overdischarging is another factor of importance, especially when Ni–MH batteries, which inevitably reveal small differences in storage capacities, are used in series. This implies that some batteries are already completely discharged while others contain small amounts of electrical energy. Continuation of the discharge process induces overdischarging to occur of the already fully discharged batteries. Under these circumstances, water is forced to be reduced at the nickel electrode, according to Ni

2H2 O + 2e− ! 2OH− + H2

[VI]

which also results in a pressure build-up inside the battery when no precautions are taken (see also Fig. 1). As the (electro)chemical affinity of the metal hydride (MH) electrode toward hydrogen gas is, in principle, excellent, it is evident that this gas can be again converted into water at the MH electrode during overdischarging, according to MH

H2 + 2OH− ! 2H2 O + 2e−

[VII]

or through a dissociative absorption reaction. Considering the similar nature of reactions [VI] and [VII], a cell voltage of close to 0 V is to be expected during overdischarging. The experimental result of such a process is shown in Fig. 5 and is in agreement with these expectations. During normal discharging, the battery voltage is located around 1.2 V and drops indeed toward an inverted voltage of −0.2 V when the overdischarging reactions take over. Fig. 5 also reveals that the pressure rise due to hydrogen evolution may be considerable, indicating that hydrogen recombination is not very favorable in this case. In the example given, the pressure was quickly built up to the critical level of approximately 20 bar, (1 bar ¼ 105 Pa). At that level, the safety vent was forced to open, which can be recognized on the small pressure decrease. Conclusively, monitoring the voltage(s) of individual cells or groups of cells is essential in order to protect these against severe (over)discharging, which is especially important for specific applications as (H)EV. A detailed mechanistic and kinetic study of the complex overdischarging reactions, occurring inside Ni–MH batteries, has recently been reported by A. Belfadhel-Ayeb and coworkers. It was concluded that the hydrogen evolution reaction at the nickel electrode occurs through the so-called Volmer–Heyrovsky mechanism and that the rate is characterized by two different Tafel slopes. The reaction order of hydrogen recombination reaction was found to be 0.5 and temperature independent in the whole temperature range investigated, indicating that dissociation of hydrogen molecules at the MH electrode is the rate-determining step.

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Metal Hydrides 2

25

1.5

20 E

E (V)

1

P 15

0.5

10

0

P (bar)

688

5

–0.5

0 0

100

50

150

State-of-discharge (%)

Fig. 5 Experimental result of the development of the cell voltage (E) and internal gas pressure (P) as a function of state-of-discharge for a Ni–MH battery during discharging and overdischarging. Reproduced with permission from Notten, P. H. L. In: Grandjean. F.; Long, G. J.; Buschow, K. H. J. (eds.) Rechargeable Nickel–Metal Hydride Batteries: A Successful New Concept; 1994; Vol. 281; Chapter 7; p. 151. NATO ASI Series E. London, ISBN 0-7923-3299-7.

2.2.3 Self-discharge It is well known that charged Ni–MH batteries, similar to Ni–Cd and lithium-ion batteries, lose their stored energy under open-circuit conditions to a certain extent. Typical self-discharge rates are of the order of 1% of the nominal storage capacity per day. These rates, however, strongly depend on the external conditions, such as state-of-charge and temperature. Fig. 6 shows that the self-discharge rate increases significantly at higher temperatures. Various mechanisms contribute to the overall self-discharge rate. These mechanisms are all electrochemical in nature. The most important mechanisms, contributing to the overall self-discharge rate, are as follows: (i) Owing to the more positive redox potential of the nickel electrode (+439 mV) compared to that of the competing oxygen evolution reaction (+300 mV; see Fig. 2), trivalent NiIII is thermodynamically unstable in an aqueous environment. Consequently, nickel oxyhydroxide (NiOOH) will, under open-circuit conditions, be reduced by hydroxyl ions according to NiOOH + H2 O + e− ! NiðOHÞ2 + OH− Ni

4OH− ! O2 + 2H2 O + 4e−

[VIII] [IX]

The electrons released by the OH− ions are transferred to the nickel electrode at the nickel electrode/electrolyte interface. Although the Ni(III) species are principally unstable, electrical charge can, however, relatively long be stored in the nickel electrode due to the fact that the kinetics of the oxygen evolution reaction are fortunately rather poor. Subsequently, the produced oxygen gas will be transported to the MH electrode, where it can be reconverted into OH− ions at the expense of charge stored in the MH electrode, that is, MH

O2 + 2H2 O + 4e− ! 4OH− −

MH + OH ! M + H2 O + e

−

[X] [XI]

These reactions also occur at the open-circuit potential at the MH electrode. The ultimate result is that the chemical energy stored in both the nickel and MH electrode is slowly released through an oxygen gas-phase shunt. (ii) A different type of gas-phase shunt is initiated by the MH electrode and is caused by the presence of hydrogen gas inside the battery. As the storage capacity of the MH electrode is considerably larger than that of the nickel electrode (Fig. 1) and the MH electrode contains a certain amount of ‘precharge’ in the form of a metal hydride, a minimum partial hydrogen pressure is established inside the Ni–MH battery, according to the chemical equilibrium MH $ M +

1 H 2 2

[XII]

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Metal Hydrides

100

689

0 qC

Residual capacity (%)

80 20 qC 60

40

20

0

45 qC

0

5

10

15

20

25

30

Time (days)

Fig. 6 Self-discharge characteristics of a Ni–MH battery at various temperatures. Gray regions indicate the spread between different batteries. Reproduced with permission from Notten, P. H. L. In: Grandjean. F.; Long, G. J.; Buschow, K. H. J. (eds.) Rechargeable Nickel–Metal Hydride Batteries: A Successful New Concept; 1994; Vol. 281; Chapter 7; p. 151. NATO ASI Series E. London, ISBN 0-7923-3299-7.

As a result, hydrogen is inevitably in contact with the nickel electrode. Because the standard redox potential of the H2/H2O redox couple is much more negative than that of the Ni(II)/Ni(III) couple (Fig. 2), hydrogen can in principle be oxidized at the nickel electrode, whereas the nickel electrode is simultaneously reduced, according to Ni

H2 + 2OH− ! 2H2 O + 2e−

[XIII]

NiOOH + H2 O + e− ! NiðOHÞ2 + OH−

[XIV]

The overall electrochemical process occurs again under open-circuit, electroless, conditions at the nickel electrode and will be strongly influenced by the partial hydrogen pressure inside the battery. It has indeed been reported that the self-discharge rate at the nickel electrode is proportional to the partial hydrogen pressure. According to Eqs. [XII] and [XIV], the chemical energy stored in both the MH and nickel electrode is ‘wasted’ by a hydrogen gas-phase shunt and can no longer be employed for useful energy supply. (iii) A third possible self-discharge mechanism is related to the fabrication process of the nickel electrode. This electrode is often prepared by electrolytic reduction of an acidic salt electrolyte, for example, nickel nitrate (Ni(NO3)2). During this process NO3− ions are reduced to NH+4 ions. This results in a significant increase in pH near the electrode/electrolyte interface. Because the solubility product of nickel hydroxide (Ni(OH)2) exceeds, nickel hydroxide will subsequently precipitate on the substrate. A consequence of this process is that a certain amount of nitrate ions are inevitably incorporated in the nickel electrodes, which can be leached out during the battery cycle-life. These NO3− ions, dissolved in the electrolyte, form the basis of this third self-discharge mechanism. These ionic species can be reduced to lower oxidation states. It is generally assumed that a so-called nitrate/nitrite shuttle is operative in alkaline rechargeable batteries. Because the standard redox potential of the nitrate/nitrite redox couple is much more positive than that of the MH electrode (ENO2−/NO3−o ¼ − 91mV vs. Hg/HgO; see also Fig. 2), NO3− ions can be reduced at the MH electrode under open-circuit conditions, according to MH

NO3− + H2 O + 2e− ! NO2− + 2OH−

[XV]

MH + OH− ! M + H2 O + e−

[XVI]

The produced nitrite ions can diffuse to the nickel electrode and can be reconverted into nitrate while nickel oxyhydroxide is simultaneously reduced, according to Ni

NO2− + 2OH− ! NO3− + H2 O + 2e− −

NiOOH + H2 O + e ! NiðOHÞ2 + OH

[XVII] −

[XVIII]

This reaction sequence can proceed continuously because the electroactive nitrate and nitrite species are continuously produced at both electrodes and are not effectively consumed. The final result is again that charge stored in both the MH and nickel electrode is consumed and is no longer available for useful energy supply.

690

3 3.1

Batteries – Battery Types – Nickel Batteries | Nickel-Metal Hydride: Metal Hydrides

Hydride-forming electrode materials AB5-type compounds

3.1.1 Structural properties The first intermetallic compound tested as negative electrode for Ni–MH batteries was LaNi5 in the early 1970s. Since this very first work, several improvements have been achieved relating to hydride-forming materials. For about two decades, all the negative electrode materials used in Ni–MH batteries are based on LaNi5-type compounds. The crystallographic properties of LaNi5 were first reported by H. Nowotny. This phase crystallizes in the CaCu5-type hexagonal structure (P6/mmm space group) with La(1a) in (0,0,0), Ni(2c) in (1/3,2/3,0), and Ni(3 g) in (1/2,0,1/2) (Fig. 7). The structure can be seen as a stacking of planes containing both lanthanum and nickel hexagonal rings for z ¼ 0 and only nickel hexagonal ring for z ¼ 1/2. LaNi5 alloy exhibited very high equilibrium pressure (0.17 MPa) and very poor cycle life to be used in practical batteries. Such a poor behavior is illustrated in Fig. 8 where the storage capacity of LaNi5 is shown as a function of the number of electrochemical charge/discharge cycles. Despite a rather high initial specific capacity (372 mAh g−1), only 12% are recovered after 400 cycles. The insets in Fig. 8 show that the origin of this poor cycle life is related to a combination of particle size reduction, induced by repeated hydrogen absorption/desorption, and surface oxidation at the electrode/electrolyte interface. Such difficulties can be overcome by preparing the so-called pseudobinary alloys. Several research groups have investigated the evolution of the plateau pressure as a function of the substituting elements. It was demonstrated that a linear relation exists between the intermetallic cell volume and the logarithm of the plateau pressure. This so-called geometrical model has been widely applied for adapting the plateau pressure of the compound to the application needs. Such behavior is illustrated in Fig. 9 for various substituting atoms either on the nickel or on the lanthanum sublattice. A practical plateau pressure around 10−3 MPa suitable for electrochemical applications can be easily achieved by substituting nickel with other elements such as aluminum, manganese, or cobalt. Fortunately, aging problems in electrochemical medium may also be solved by making appropriate substitutions. Capacity loss upon cycling is generally attributed to the decomposition of the alloys into lanthanum hydroxide (La(OH)3) and nickel particles in concentrated potassium hydroxide. J. J. G. Willems at the Philips Research Laboratories obtained real improvements in terms of cycle life in the 1980s by adding cobalt to the nickel sublattice. Indeed, as can be seen in Fig. 10, the cycle life of a cobalt-substituted compound (La0.8Nd0.2Ni2.4Co2.5Si0.1) exhibits a stable reversible electrochemical capacity over 400 cycles, keeping about 86% of its initial value. The insets show that the electrodes hardly reveal any particle cracking and surface oxidation upon electrochemical cycling. Since then, numerous substitutions have been tempted, using various elements in order to obtain more efficient negative electrode materials. Both lanthanum and nickel atoms can be substituted by other elements preserving the hexagonal LaNi5 parent structure. Lanthanum can be easily replaced by 4 f elements in the whole range of concentration, leading to a complete solid solution for the rare earth site. Using this property, lanthanum can be easily replaced by cheaper natural mixture of rare earth, generally denoted as Mischmetal (Mm). On the nickel side, the substitution can be total for some neighboring elements but it is usually limited to a given range depending on the nature and the atomic radius of the substituting element. For example, the solid solution LaNi5–xMx is complete for M ¼ Co, Pt, or Cu but partial for Sn (x ZnMn2 O4

(7)

Since the concept was published, an assortment of zinc-insertion cathode materials and electrolyte combinations have been investigated with promising results. However, improved mechanistic understanding and standardized reporting guidelines are still needed to guide the development.44

3.2

Applications

Consumer electronics are a main application for primary zinc-based batteries, like alkaline and zinc-air. Alkaline batteries for these applications will be familiar to most readers in the cylindrical AAA, AA, C, and D formats that are available in a variety of shops around the world. Alkaline batteries are cheap and reliable, and they can be designed for both high-energy and high-power applications. This makes them suitable for a variety of household products like toys, flashlights, smoke detectors, remote controls, etc. The global market for alkaline batteries was worth $7.73 billion in 2021. Primary zinc-air batteries are more suited toward devices that require high energy density. For more than 40 years, hearing aids have made up the largest mainstream market for primary zinc-air button batteries. This market is dominated by many of the same companies that have been manufacturing primary zinc-based batteries for more than 100 years, including Rayovac, Duracell, Energizer, VARTA, Panasonic, and others. Demonstrators of zinc-based battery packs for electric mobility were shown to be feasible from the silver-zinc batteries in the 1960s, nickel-zinc batteries in the 1970s,45 zinc-bromine batteries in the 1980s,38 and zinc-air batteries in the 1990s.46 However, the advent of high-performance and affordable Li-ion batteries greatly slowed development on this front. Today, the performance metrics of zinc-based batteries struggle to compete in the electric mobility sector, which has stringent requirements for fast charging and long cycle life. Even though the inherent safety of aqueous zinc-based batteries does offer an advantage for some safety critical applications, current zinc-based battery technologies are mostly investigated as auxiliary or range-extender concepts rather than the main power source.47 Stationary battery energy storage systems (BESS) are one of the most promising applications for rechargeable zinc-based batteries. The growing need for stationary BESS is driven by a variety of forces including increased renewable energy sources in the electricity grid, demand for residential energy storage, the electrification of rural areas, and the uninterruptable operation of critical electronic infrastructure. Zinc-based batteries are often candidates for stationary utility electricity and micro-grid storage due to their modular versatility, low cost, and good safety. Zinc-bromine, zinc-air, nickel-zinc, and zinc-shuttle battery solutions are all commercially available and primarily targeting stationary applications.48–60 In stark contrast to the primary battery market, which is dominated by large and well-established businesses, most suppliers for zinc-based stationary energy storage solutions have been founded within the last 20 years. Emerging applications like Internet of Things (IoT) sensors and wearable electronics that demand flexible or fiber batteries are one new area for zinc-based batteries. The International Union of Physical and Applied Chemistry (IUPAC) highlighted fiber batteries as one of its Top 10 Emerging Technologies in Chemistry Initiative for 2022.61 Zinc-based batteries offer some natural advantages over Li-ion batteries for flexible application such as safety and environmental friendliness. Although the development of flexible Zn-based batteries is still at an early stage, there is a wealth of dedicated research on the topic and first steps toward commercialization.62

4

Summary

The history of zinc-based batteries can be traced all the way back to the birth of modern electrochemistry with the invention of the voltaic pile and Daniell cell. Over the last 223 years, zinc-based batteries have been the subject of national pride, helped win wars, revolutionized medical treatments, and taken human beings to the surface of the Moon and back. Today, they remain an important tool in our global effort to establish clean and sustainable energy infrastructure. Zinc-based batteries offer a reliable, affordable, and safe alternative to lithium-ion batteries for some applications. Recent advances in cell engineering and materials discovery are making rechargeable zinc-based batteries viable solutions for stationary energy storage ranging from grid-scale utilities to residential PV. Furthermore, novel application areas like flexible sensors and wearable electronics continue to open new opportunities for our oldest batteries.

758

Batteries – Battery Types – Zinc Batteries | Overview

References 1. Sangine, E. Mineral Commodity Summaries 2020, no. 703; 2020. https://doi.org/10.3133/mcs2020. 2. Piccolino, M. Animal Electricity and the Birth of Electrophysiology: The Legacy of Luigi Galvani. Brain Res. Bull. 1998, 46 (5), 381–407. https://doi.org/10.1016/S0361-9230(98) 00026-4. 3. Galvani, L. De viribus electricitatis in motu musculari commentarius, Vol. 7; De Bonoiensi Scientiarum et Artium Intituo atque Academie Commentarii, 1791; pp. 363–418. 4. Volta, A. On the Electricity Excited by the Mere Contact of Conducting Substances of Different Kinds. Phil. Trans. R. Soc. London 1800, 90, 403–431. 5. Desmond, K. Innovators in Battery Technology, Profiles of 95 Influential Electrochemists; McFarland & Company, Inc: Jefferson, NC USA, 2016. 6. Davy, H. XXIX. Farther Researches on the Magnetic Phenomena Produced by Electricity; with some New Experiments on the Properties of Electrified Bodies in their Relations to Conducting Powers and Temperature. Phil. Trans. R. Soc. 1821, 111, 425–439. https://doi.org/10.1098/rstl.1821.0030. 7. Mills, A. A. Early Voltaic Batteries: An Evaluation in Modern Units and Application to the Work of Davy and Faraday. Ann. Sci. 2003, 60 (4), 373–398. https://doi.org/ 10.1080/00033790110117566. 8. Daniell, J. F. X. On Voltaic Combinations. In a letter addressed to Michael Faraday, D. C. L., F. R. S., Fullerian Prof. Chem. Royal Institution, Corr. Memb. Royal & Imp. Acadd. of Science, Paris, Petersburgh, &c. By J. Frederic Daniell, F. R. S., Prof. Chem. in King’s College, London. Phil. Trans. R. Soc. 1836, 126, 107–124. 9. Daniell, J. F. An Introduction to the Study of Chemical Philosophy: Being a Preparatory View of the Forces which Concur to the Production of Chemical Phenomena; JW Parker, 1839. 10. Miller, K.; Patterson, G.; Thom, C.; Millikan, R.; McMeen, S. Cyclopedia of Telephony & Telegraphy. In A General Reference Work on Telephony, etc; Vol. 1; American School of Correspondence: Chicago, 1919. 11. Gassner, C. Galvanic Battery; US373064A, 1887. 12. Clarke, C. L. Galvanic Battery; 298175, 1884. 13. Laszcynski, S. Storage Battery; 714,201, 1902. https://doi.org/10.1145/178951.178972. 14. Bradley, C. S. Secondary Battery; 312,802, 1885. https://doi.org/10.1145/178951.178972. 15. Bowden, M. E.; Philadelphia, E.; C. H. Foundation. The Columbia Dry Cell Battery. Am. Chem. Soc. 2005, 1–4. 16. Haput, J. Vintage Eveready Columbia Gray Label Telephone Batteries, 6 Inches High, Made In USA. https://www.flickr.com/photos/51764518@N02/36670011780 (Accessed 29 November 2022). 17. Ruble, K. D. The RAYOVAC Story—The First 75 Years; RAY-O-VAC, 1981. 18. Bullock, K. R. Samuel Ruben: Inventor, Scholar, and Benefactor. Interface 2006, 15, 16–17. 19. Hintz, E. Portable Power : Inventor Samuel Ruben and the Birth of Duracell. Technol. Cult. 2009, 50 (1), 24–57. https://doi.org/10.1353/tech.0.0203. 20. Ruben, S. Alkaline Dry Cell; 2422045, 1947. 21. Friedman, M.; McCauley, C. E. The Ruben Cell—A New Alkaline Primary Dry Cell Battery: Its Design and Manufacture, Electrochemical Principles, Performance Characteristics, and Applications. Trans. Electrochem. Soc. 1947, 92, 195. 22. Greatbatch, W.; Holmes, C. F. The Past and Future of Implatable Batteries. In Proceedings on the Symposium on History of Battery Technology; The Electrochemical Society, Inc., 1987; pp. 232–254. 23. Saxe, C. G.; Brodd, R. J. History of Alkaline Zinc Manganese Dioxide Cells. In Proceedings on the Symposium on History of Battery Technology; The Electrochemical Society, Inc, 1987; pp. 47–66. 24. Kordesch, K. V.; Stark, R. E. Cathodes for Primary and Rechargeable Cells; 3113050, 1963. 25. Marsal, P. A.; Kordesch, K.; Urry, L. F. Dry Cell; US2960558A, 1960. 26. Andre, H. G. ACCUMULATOR; 2,317,711, 1943. 27. White, J. C.; Pierce, R. T.; Dirkse, T. P. Characteristics of the Silver Oxide—Zinc—Alkali Primary Cell. Trans. Electrochem. Soc. 1946, 90 (1), 467. 28. Howard, P. L. History of the Silver Oxide-Zinc Battery. In Proceedings on the Symposium on History of Battery Technology; The Electrochemical Society, Inc, 1987; pp. 175–179. 29. Bugga, R.; Brandon, E. Energy Storage for the Next Generation of Robotic Space Exploration. Electrochem. Soc. Interface 2020, 29 (1), 59–63. https://doi.org/10.1149/2. F08201IF. 30. McDivitt, J. A. Apollo 13 Mission Report; Houston, Texas, 1970. 31. Maiche, L. French Patent 127,069; 1878. 32. Heise, G. W. Air Depolarized Primary Battery; 1899615, 1933. 33. Heise, G. W.; Schumacher, E. A. An Air-Depolarized Primary Cell with Caustic Alkali Electrolyte. Trans. Electrochem. Soc. 1932, 62 (1894), 383. https://doi.org/ 10.1149/1.3493794. 34. Heise, G. W.; Cahoon, N. C. Fiftieth Anniversary: The Anniversary Issue on Primary Cell Systems: Dry Cells of the Leclanché Type, 1902–1952—A Review. J. Electrochem. Soc. 1952, 99 (8), 179C. 35. Drumm, J.J.; Burnell, A.G. Electric Storage Battery. Patent No. 2,004,552, United States Patent Office, 1933. 36. Putt, R. A. Assessment of Technical and Economic Feasibility of Zinc/Bromine Batteries for Utility Load-Leveling; Rolling Meadows: IL, 1979. 37. Bailleux, C. When Electrochemistry Paved the Way to the Sky. In Proceedings on the Symposium on History of Battery Technology; The Electrochemical Society, Inc, 1987; pp. 268–278. 38. Khor, A.; et al. Review of Zinc-Based Hybrid Flow Batteries: From Fundamentals to Applications. Mater. Today Energy 2018, 8, 80–108. https://doi.org/10.1016/j. mtener.2017.12.012. 39. Lex, P. J.; Mathews, J. F. Recent Developments in Zinc/Bromine Battery Technology at Johnson Controls. In IEEE 35th International Power Sources Symposium; 1992. https://doi. org/10.1109/IPSS.1992.282047. 40. Eisenberg, M. Alkaline Galvanic Cells; WO 93/26056, 1993. 41. Xu, C.; Li, B.; Du, H.; Kang, F. Energetic Zinc Ion Chemistry: The Rechargeable Zinc Ion Battery. Angew. Chem. Int. Ed. 2012, 51 (4), 933–935. https://doi.org/10.1002/ anie.201106307. 42. Parker, J. F.; Ko, J. S.; Rolison, D. R.; Long, J. W. Translating Materials-Level Performance into Device-Relevant Metrics for Zinc-Based Batteries. Joule 2018, 2 (12), 2519–2527. https://doi.org/10.1016/j.joule.2018.11.007. 43. Le, S.; et al. Review—Status of Zinc-Silver Battery. J. Electrochem. Soc. 2019, 166 (13), A2980–A2989. https://doi.org/10.1149/2.1001913jes. 44. Borchers, N.; Clark, S.; Horstmann, B.; Jayasayee, K.; Juel, M.; Stevens, P. Innovative Zinc-Based Batteries. J. Power Sources 2021, 484, 229309. https://doi.org/10.1016/j. jpowsour.2020.229309. no. December 2020. 45. Rajashekara, K. History of Electric Vehicles in General Motors. IEEE Trans. Ind. Appl. 1994, 30 (4), 897–904. https://doi.org/10.1109/28.297905. 46. Cook, R. Electric Car Showdown in Phoenix: Zinc-Air Battery Wins. Pop. Sci. 1991, 64–65. 47. Cano, Z. P.; Ye, S.; Banham, D. Batteries and Fuel Cells for Emerging Electric Vehicle Markets. Nat. Energy 2018, 3 (April), 279–289. https://doi.org/10.1038/s41560-0180108-1. 48. “Redflow.” https://redflow.com/ (accessed Nov. 29, 2022). 49. “Primus Power.” https://primuspower.com/en/ (accessed Nov. 29, 2022). 50. “Eos Energy Enterprises.” https://eosenergystorage.com/ (accessed Nov. 29, 2022).

Batteries – Battery Types – Zinc Batteries | Overview 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62.

“Æsir Technologies.” https://www.aesirtec.com/ (accessed Nov. 29, 2022). “ZAFSYS.” https://zafsys.com/ (accessed Nov. 29, 2022). “ZincFive.” https://zincfive.com/ (accessed Nov. 29, 2022). “Enzinc.” https://enzinc.com/ (accessed Nov. 29, 2022). “Zelos.” https://www.zelos.energy/ (accessed Nov. 29, 2022). “Urban Electric Power.” https://urbanelectricpower.com/ (accessed Nov. 29, 2022). “Zinc8.” https://www.zinc8energy.com/ (accessed Nov. 29, 2022). “e-Zinc.” https://e-zinc.ca/ (accessed Nov. 29, 2022). “Salient Energy.” https://salientenergy.ca/ (accessed Nov. 29, 2022). “Enerpoly.” https://enerpoly.com/ (accessed Nov. 29, 2022). Gomollón-Bel, F. IUPAC Top Ten Emerging Technologies in Chemistry 2021. Chem. Int. 2021, 43 (4), 13–20. https://doi.org/10.1515/ci-2021-0404. “Imprint Energy.” https://www.imprintenergy.com/ (accessed Nov. 29, 2022).

759

Batteries – Battery Types – Zinc Batteries | Zinc Electrode LS Caoa and D Lib, aFuel Cell System and Engineering Laboratory, Key Laboratory of Fuel Cells & Hybrid Power Sources, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian, China; bCollege of Light Industry and Chemical Engineering, Dalian Polytechnic University, Dalian, China © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of X.G. Zhang, SECONDARY BATTERIES – ZINC SYSTEMS | Zinc Electrodes: Overview, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 454–468, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00166-0.

1 2 3 4 5 6 7 8 8.1 8.2 8.3 9 10 11 12 13 References

Introduction Historic development Further development potential Physical characteristics of zinc electrode Electrochemical properties Porous electrode Morphology of zinc deposits Potential improvements for zinc battery electrode Morphological stability of rechargeable electrode Shape change Methods for controlling shape change Control of dendrite formation High rate performance Low and high temperature performance Reversibility Outlook

761 762 763 764 766 769 770 772 772 772 773 774 774 776 776 776 777

Abstract This article provides an overview of zinc electrodes used in different types of batteries from the perspective of the materials, the structural and morphological characteristics of the electrodes, and their relationship to electrochemical activity and reversibility. Specifically, the use of various forms of zinc materials, such as powder, fibers, dendrites, and pellets, for primary alkaline, rechargeable, zinc–air batteries and fuel cells is discussed. The structure and morphology of the porous electrode are analyzed with respect to electrode reactivity. The transformation of the morphology of zinc material during charging and discharging of the electrodes is discussed in terms of its impact on shape change, solid electrolyte interphase, and life of secondary zinc electrodes. Finally, the scope for the improvement of zinc electrodes for application in high-power systems and wide temperature window is discussed.

Glossary Autocatalytic dissolution The enhanced dissolution that is catalyzed by the dissolution reaction itself. Energy density The energy output from a battery per unit volume, expressed in Wh L−1 or Wh dm−3. Equilibrium potential See “Reversible potential.” Exchange current density The current per unit area that flows equally in the forward and backward directions when an electrode reaction is in equilibrium. Form factor Conditions related to dimension, shape, and volume of a cell or battery. Mechanically rechargeable Refers to a zinc–air battery in which the zinc negative electrode (anode) is changed manually or by mechanical means. Nonregenerative fuel cell Fuel cell for which the spent metal fuel is not regenerated but is shipped for recycling. Overpotential The shift in the potential of an electrode from its equilibrium value as a result of current flow. Power density The power output of a battery per unit volume, usually expressed in W L−1 or W dm−3 and quoted at 80% depth of discharge. Regenerative fuel cell Fuel cell for which the spent metal fuel is regenerated within the application system. Reversible potential The potential of an electrode when there is no net current flowing through the cell. Shape-change Change in the geometric area of the active material in a zinc electrode during charge–discharge operation.

760

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00038-4

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

761

Specific energy The energy output of a battery per unit weight, usually expressed in Wh kg−1. Specific power The power output of a battery per unit weight, usually expressed in W kg−1. Standard electrode potential The reversible potential of an electrode with all the active materials in their standard states. Note that usual standard states specify unit activity for elements, solids, 1 mol L−1 solutions, and gases at a pressure of 101.325 kPa.

Key points

• • • • • •

As power sources, zinc anodes can be paired with Ag, Cu, MnO2, air, Cl2, NiOOH, AgO, vanadium oxides, Prussian blue or organics cathodes. Zn electrodes feature porous structure and solid electrolyte interface. Electrochemistry of Zn electrodes in both alkaline and neutral electrolytes are described. Morphologies of Zn deposits are affected by electrolytes, current density, and temperature. Methods for controlling dendrite growth include novel Zn electrodes, electrolytes optimization, solid-electrolyte-interface, special separator materials, and specific charging methods. High rate, low and high temperature performance of reversible Zn electrodes in aqueous electrolytes are worthy of further investigation.

Symbols and Units

Eo i iactual io ip k Re Rp t1 tp

Standard electrode potential (V) Current density (A cm−2) Current density of the active surface (A cm−2) Exchange current density (A cm−2) Passivation current (A cm−2) A constant Electrolyte resistance (O) Polarization resistance, which results in a cell voltage loss when there is a flow of current through the surface of the electrode Time at which the current density is equal to passivation current Time required for passivation (s)

Abbreviations and Acronyms AC CPH DC i–V curve SHE

1

Alternating current Close-packed hexagonal Direct current Current density vs voltage plot Standard hydrogen electrode

Introduction

Zinc possesses a unique set of attributes that include a low equilibrium potential, electrochemical reversibility, stability in aqueous electrolytes, good tolerance to O2, good conductivity, low equivalent weight, high specific energy, high energy density, abundance, low cost, low toxicity, and ease of handling. The large overpotential for hydrogen evolution on zinc is particularly important, making zinc the metal of highest energy density among the common metals that can be efficiently electroplated in aqueous solutions. These attributes have made zinc a favorable negative-electrode (anode) material for electrochemical power sources since the invention of the battery more than 200 years ago, as illustrated in Fig. 1. Today, zinc-based battery systems still retain their importance as power sources for a wide range of applications and command about one-third of the world battery market.

762

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

1800 Volta piles Zn-H2O

1836 Daniell cell Zn-CuSO4

1941 Zn-AgO rechargeable

1872 Zn-HgSO4

1838 Zn-HNO3, Pt 1842 Zn-Na2Cr2O7, C Zn-FeCl3, C

1882 Alkaline Zn-MnO2 1883 Zn-CuO rechargeable 1884 Zn-HgO Commercial use 1947

1843 Zn-PbO2 1868 Leclanché cell Zn-MnO2

1884 Zn-Cl2

1885 Zn-Br2

1899 Zn-NiOOH rechargeable

1869 Zn-air Commercial use 1932 1800

1900

2012 Rechargeable zinc-ion battery

1950 Zn-air Mechanically rechargeable

2010s Zn-Prussian blue analogues Zn-Vanadium oxides Zn-Organics

1950s Alkaline Zn-MnO2 commercialization 1960s Zn-air Button cell 1971 Zn-air fuel cell Hydraulically rechargeable 1970s Alkaline Zn-MnO2 rechargeable

2000

Time

Fig. 1 Historic evolution of zinc-based power source systems.

2

Historic development

In 1800, Volta discovered the first means of generating a continuous flow of electricity. His device, subsequently known as the Volta pile, was a cell constructed of zinc disks as negative electrodes and silver or copper disks as positive electrodes. The electrodes were separated by pieces of cloth saturated with brine. Such cells were used as the only electrical power source until the invention of the Daniell cell, a copper/copper sulfate–zinc/sulfuric acid cell, in 1836. In the 1860s, the highly important Leclanché cell, which used zinc and a mixture of manganese oxide (MnO2) and carbon as the active materials with aqueous ammonium chloride as the electrolyte solution, came on the scene and has since dominated the primary battery market. The large market for Leclanché cells was developed due to the advent of the radio in the 1920s.1 The use of an alkaline electrolyte (potassium hydroxide) instead of chloride resulted in ZndMnO2 alkaline cell, which was first described as early as 1882 in a German patent. The desire for high-rate discharge capacity for primary cells led to the commercialization of the alkaline cell in the late 1950s; it became a commercial success in the 1970s. The principle of the zinc–air cell was discovered in 1868 when Leclanché noted that the performance of his cell was improved if the electrolyte was only half-filled so that the cell and the upper portion of the manganese dioxide and carbon mixture were only moist rather than immersed in the electrolyte. A year later, the first true zinc–air battery was demonstrated. Commercialization of zinc–air cells started only in 1932 when a porous carbon block impregnated with paraffin wax to prevent flooding was used as the positive electrode (cathode). Low-drain, large-size batteries were then built for radio and railway signaling applications. The first instance of mechanically rechargeable metal–air batteries occurred when a zinc–air battery designed in the 1950s allowed the spent zinc anode to be replaced with a fresh anode. By the 1960s, it was well recognized that such batteries could have significant application if a reliable and low-cost zinc anode became available. In the late 1960s, the use of hydrophobic Teflon film allowed the construction of high-energy zinc–air button cell, which has since been a dominant power source for hearing aids. Rechargeable zinc batteries have long been considered promising systems for high energy and higher power duty. The rechargeable nickel–zinc (NidZn) battery was invented toward the end of the nineteenth century and has since undergone extensive development. In the 2010s, zinc-nickel batteries are used for data center backup power. In the 1940s, the zinc–silver (ZndAg) battery was developed for commercial applications. Because of its high specific energy, safety, and reliability, the battery has been used in many military and space applications such as lunar spacecraft, deep-sea rescue vessels, satellites, torpedoes, submarines, high-value electronics, and medical equipment. In the 1990s, the zinc–bromine flow battery, which was invented more than 100 years ago, was developed for stationary energy storage. More recently, zinc–cerium flow cell battery has been developed. Since 2010s, rechargeable zinc-ion batteries, with manganese-based oxides, prussian blue analogs, vanadium-based oxides and organic compounds as cathode materials in neutral/mildly acidic aqueous electrolytes, accelerate application due to their low cost, material abundance, high safety, acceptable energy density and environmental friendliness. Manganese-based oxides are highlighted by low cost and rich redox chemistry of Mn. Vanadium-based oxides feature multi-electron redox reaction accompanied with accessible V oxidation states and large ion transfer channel. Prussian blue analogs are characterized by 3D open-framework structures with abundant redox-active sites and considerable structural stability. Organic compounds are marked by flexible structure designability, high element abundance, and sustainability.2

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

3

763

Further development potential

Developments in new designs and improvements of zinc electrodes continue as various zinc-based systems evolve. For ZndMnO2 alkaline batteries, despite their widespread applications, better performance is needed for a range of newly arrived, power-demanding, portable electronic devices. The energy utilization is low for high-rate applications, such as digital cameras, for which only less than one-third of the energy contained in conventional alkaline cells is utilized. Improving the rate of performance of alkaline batteries is an active development direction for battery manufacturers. The applications of rechargeable zinc batteries, ZndAg, ZndNi, Zn-vanadium-based oxides, Zn-prussian blue, ZndMnO2, and Zn-organics, are still limited by their relative short cycle-lives. For the ZndAg battery, for example, the short cycle-life limits the application beyond military/space applications. A major cause for capacity fading is related to the morphological stability of the zinc electrode with repetitive charging and discharging; it tends to develop dendrites, causing cell shorting and changes in shape, which reduce the reactive interface area. There has been significant development and progress in recent years to improve the cycle-life of zinc rechargeable systems to make them suitable for a range of specific applications.3 ZndMnO2 alkaline batteries feature long shelf storage life. After storage at 21
C for 4 years, ZndMnO2 batteries can provide 85% of initial capacity. However, storage at high temperatures and high humidity accelerate degradation of the batteries. At low temperature storage, the chemical activity is retarded and capacity is not greatly affected. Recommended storage conditions are 10–25
C with no more than 65% relative humidity. Mild aqueous zinc batteries faced the following two challenges: the increase of the specific energy of the full Zn-ion battery, and the prevention of the parasitic H2 evolution reaction occurring during the Zn electrodeposition step. To increase the energy density, the total mass of active materials should be increased and the cathode materials resulting in an average discharge voltage of the battery at least 1.0–1.2 V need to be developed. Moreover, research efforts should also address the optimization of the electrolyte composition (e.g., additives), toward the development of future aqueous electrolytes with the aim to hinder the parasitic hydrogen evolution reaction at the anode. Zinc–air has been a promising chemistry for high energy density systems since its invention one century ago. However, the electrically rechargeable zinc–air battery is not yet technically feasible owing to the fast degradation of the air cathode and the change of the morphology of zinc anode during cyclic discharging and charging. As a result, various types of nonelectrical rechargeable zinc–air battery systems have been explored as illustrated in Fig. 2. One method is to mechanically change the anode after each discharge; the spent anode is replaced physically with a fresh anode in the cell. Mechanically rechargeable systems have been developed for large power sources such as those used for motive power.4

Fig. 2 Ways for recharging zinc anodes in various zinc–air systems: (1) electrically rechargeable; (2a) and (2b) mechanically rechargeable with spent anode shipped to recyclers, oxide users, or zinc refiners; (3) mechanically rechargeable with anode recharged outside the cell in a different device; (4) mechanically rechargeable with new anode regenerated in a specialized plant; (5) in-unit regenerative hydraulically feeding fuel cell; (6a) and (6b) off-site regenerative fuel cell; (7a) and (7b) nonregenerative fuel cell in which the spent zinc is sent to recyclers, oxide users, or refiners.

764

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

Fig. 3 Schematic illustration of the flow of material in a zinc–air fuel cell system.

Alternatively, zinc–air fuel cell systems, in which the zinc active material is fed into the electrochemical cells hydraulically, have also been explored. Early designs used dendritic zinc as fuel because dendrites can easily be regenerated by electrodeposition. Later designs employed zinc pellets, which have a better fluid dynamic properties and a lower gassing rate. The operation of a regenerative zinc–air fuel cell is shown schematically in Fig. 3. Such a system allows independent scaling of either the runtime by changing the size of the fuel tank, or the power by changing the number of cell stacks.4 The cycle life and rate performance of zinc batteries can be affected by the temperature. At low temperature, the electrochemical performance of zinc batteries is determined by the resistance polarization caused by the ionic conductivity of electrolytes, the concentration polarization caused by the slow ion transport at the interface and the electrochemical polarization caused by the electrode reaction kinetics. To reduce polarization, the H-bond design of electrolytes promises the anti-freezing property, the solvation structure modification reduces the activation energy barrier of the interface, and the electrode structure regulation promotes the solid-state diffusion, which improve the electrochemical performance of zinc batteries. At high temperatures, aqueous electrolytes in zinc batteries is limited by the highly volatile character of water, the instability of the zinc-electrolyte interface, and the accelerated side reactions between water and zinc electrode. Hence, strategies for wide temperature applications should be proposed based on the underlying key factors. Zinc is a versatile energy carrier because a wide range of forms of zinc material can be engineered and utilized for the anode in different chemistries, as well as in different physical embodiments from conventional batteries, to flow batteries, to fuel cells. Some designs, such as flow batteries and fuel cells, have good prospects for electrochemical energy conversion and storage on a large scale. It is interesting to compare zinc with hydrogen, which has been widely regarded as a universal energy carrier. Hydrogen has a high specific energy, 33 kWh kg−1, vs zinc, 1.4 kWh kg−1. However, the energy density of hydrogen is 2.3 kWh L−1 for liquefied hydrogen, which is only about one-quarter of the energy density of zinc, 9.6 kWh L−1. The difference is much larger when the volume of the container is also considered. Zinc thus has a big advantage over hydrogen in storage and transportation when space, convenience, safety, and cost are considered. In fact, the problem of hydrogen storage and transportation is one of the major obstacles preventing the commercialization of hydrogen fuel cells.

4

Physical characteristics of zinc electrode

Except for the zinc–carbon cell, in which zinc anode is in solid form, the zinc electrode in all battery and fuel cell systems is of porous nature. The structure has a large active material–electrolyte interface and good availability of electrolyte near the surface, has space for storing dissolution products, and allows a high reaction rate on both weight and volume basis. A well-designed porous electrode will allow the electrode to discharge at a large current with minimal voltage loss and will sustain the longest period of discharge before the occurrence of passivation. Porosity of 60–80% is most commonly found in the zinc electrodes of different battery systems and is equivalent to a capacity of 1.2–2.2 Ah cm−3 for the zinc active material (zinc has a theoretical capacity of 5.86 Ah cm−3 in solid form). The porosity of the zinc anode is controlled by various means depending on the type of battery and the form of material used. Different forms of zinc materials that could be applied in various systems are shown in Fig. 4. Each has its own special set of characteristics with respect to physical, mechanical, and electrical properties, which provide specific possibilities for design, processing, and performance of a given battery system. At present, atomized zinc powder is used in alkaline ZndMnO2 and primary zinc–air cells. The large specific surface area of the powder (0.02 m2 g−1) allows the anode to deliver high currents, i.e., many times larger than that from sheet electrodes. Before being mixed with electrolyte, the zinc powder for alkaline battery applications typically has a density of 3–3.5 g cm−3, which is about

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

765

Fig. 4 Different forms of zinc materials produced by different methods.

42–50% zinc volume, or 50–58% porosity. To reach the required porosity, which is typically about 70%, manufacturers use a gelling agent to make a mix of zinc powder and electrolyte such that the zinc particles are not densely packed but suspended in the electrolyte gel. If the porosity is too small, the anode may not have good reactivity, whereas if it is too large, the anode may have poor conductivity. Dendritic powder is produced by electrodeposition from alkaline electrolyte and typically has a very large specific surface area, namely, as large as 1 m2 g−1. It is suitable for high-power applications such as primary ZndAg cells or large mechanically rechargeable zinc–air batteries. Because of the branching nature of dendrites, the powder particles can easily be bonded together mechanically through pressing and without the need for a bonding agent. Owing to the large specific surface area, dendritic powder is highly reactive—it produces a significant amount of hydrogen gas via corrosion during storage in the electrolyte and this limits its use in certain applications. Fibrous zinc materials, which are produced through machining or casting, are also suitable for high-power applications. An electrode made of fibrous material has a better electrical conductivity and can have a discharge performance comparable with an electrode made of dendrite powder, despite the fact that the specific surface area is much smaller. Another advantage of fibrous material is the ease of porosity control without the use of a gelling agent, which would increase both electrolyte resistance and material cost. Zinc pellets about 0.5 mm in size have been used as fuel for zinc–air fuel cells. They can be made by mechanically cutting zinc wires, by electrodeposition, and by direct casting. In operation, the pellets, which are stored in a fuel tank containing an abundance of potassium hydroxide electrolyte, are fed into the anode beds of the fuel cell system. Both size distribution and shape of the particles are important in terms of feeding hydrodynamics, packing density, and anode kinetics. Zinc oxide (ZnO) powder with a size typically in the range of about 0.1–3 mm is most commonly used as a starting material for making rechargeable electrodes. The structure and fabrication method for a rechargeable electrode are more complex than for the electrode in primary batteries. Bonding agents and zinc dust are usually mixed with the zinc oxide to provide certain mechanical stability and electrochemical properties. The formation of solid electrolyte interface (SEI) layer is triggered by the interfacial reactions between electrolyte and Zn electrode. This intrinsic SEI layer is electron insulating and Zn-ion conductive, which may prevent the continuous reaction between

766

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

Fig. 5 The cartoon (left) and deposit morphology (right) of proposed Zn2+-conducting solid electrolyte interphase, obtained from zinc trifluoromethanesulfonate electrolyte with trimethylethyl ammonium trifluoromethanesulfonate as an additive, characterized by small nodular particles embedded in a polymeric framework.

Zn electrode and electrolyte. In general, the SEI layer structure is composed of inorganic and organic components, although its composition and morphology largely depend on the selection of organic solvents and anions of zinc salts. As shown in Fig. 5, the inorganic inner layer (ZnF2, ZnSO4 and ZnCO3) forms near the electrode/SEI interface, allowing Zn-ion transport, while the heterogeneous organic porous outer layer is near the SEI/electrolyte interface and permeable to both Zn-ion and electrolyte molecules. The thickness of the formed solid electrolyte interface film is from a few nanometers to dozens of nanometers. The insolubility of organic, inorganic or organic/inorganic hybrid compounds within the solid electrolyte interface can prevent the solvent molecules in bulk electrolyte to be co-embedded in the deposited Zn, thus reducing the charge/discharge effect of the Zn anode. As a result, the solid electrolyte interface film can stabilize the Zn anode in aqueous and/or non-aqueous electrolytes and improve the service life of Zn batteries.5,6 A material can be characterized by its composition and by its size and geometric shape. The composition, including components and chemical bonds of the solid electrolyte interface, can be determined by X-ray photoelectron spectroscopy, synchrotron X-radiation, neutron scattering, X-ray diffraction, Raman and Fourier-transform infrared spectroscopy, etc. Size can be determined by means of definitive methods, like optical observation, scanning electron microscopy and transmission electron microscopy. The changes of electrode geometric shape in electrochemical cells during operation can be monitored by synchrotron-based X-ray techniques.5,7 Greater ability in characterization and engineering for forms of materials could make further improvement in battery systems. The benefits arising from the physical form of zinc powder materials, such as aspect ratio and shape, are currently utilized in the alkaline battery industry.

5

Electrochemical properties

Zinc is divalent in all its compounds. Compounds of zinc(I) do not exist naturally. The standard potential of the zinc electrode is Zn Ð Zn2+ + 2e − Eo ¼ −0:763 V

(I)

The oxidation of zinc electrode during discharge may involve several basic steps, which include oxidation of zinc atoms on the surface (i.e., the breaking up of the metallic bonding), solvation in the solution, diffusion in the electrolyte, and precipitation of the solid phase when the solubility limit is reached. The reverse is involved in the deposition process during recharging of the electrode. Thus, all the chemical species in the possible phases—electrolyte, solid active material, and precipitates—as well as the properties of inactive elements may play a role in the electrode process. The properties of zinc, zinc oxide, and potassium hydroxide solutions that are relevant to battery applications are listed in Table 1. The zinc electrode has a good electrochemical reversibility; it dissolves readily near its equilibrium potential with the formation of zinc divalent ions. A number of zinc complexes may form but the predominant species has been identified as the tetrahedral Zn(OH)42−. The electrode also has a rather high exchange current density, io, i.e., on the order of 0.1 A cm−2. By comparison, the io for hydrogen evolution on a zinc surface is on the order of 10−9 A cm−2. The high hydrogen overpotential means that zinc is stable in aqueous solutions and a high current efficiency is achieved during a deposition process with only minimal hydrogen evolution. The dissolution process at a zinc electrode in alkaline solutions can be expressed as. Zn + 4OH − ¼ Zn ðOHÞ4 2 − + 2e −

(II)

When the solution near the surface becomes saturated with the dissolved zinc species, precipitation of zinc oxide may occur according to the following reaction: Zn ðOHÞ4 2 − ¼ ZnO + 2OH − + H2 O

(III)

Zn + 2OH − ¼ ZnO + H2 O + 2e −

(IV)

Thus the overall reaction from zinc metal is.

Batteries – Battery Types – Zinc Batteries | Zinc Electrode Table 1

767

Properties of zinc relevant to battery applications (at room temperature).

Zn Atomic weight Crystalline structure Density Melting temperature Resistivity Ionic valence Ionic radius Stable dissolved form in KOH Standard potential Reversible potential in 35% KOH, 8.2 mol L Specific gravimetric capacity Specific volumetric capacity ZnO Density Molar volume ratio of ZnO–Zn Semiconductor Band gap Typical undoped resistivity KOH Density of 35% KOH, 8.2 mol L−1 Resistivity of 8 mol L−1 KOH Solubility of ZnO in 35% KOH, 8.2 mol L−1 Supersaturation of zincate in KOH Conductivity of 8 mol L−1 KOH + 1 mol L−1 Zn(OH)42− Ratio of diffusion coefficient of Zn(OH)42−/OH− Ratio of diffusion coefficient of Zn(OH)42−/K+ Ratio of i0 between reactions Zn $ Zn2+ and H2 $ H+ in 8 mol L−1 KOH

65.38 CPH 7.14 g cm−3 419.5
C 5.96 mO cm 2 0.74–0.83 A˚ Zn(OH)42− −0.763 VSHE −1.35 VSHE 0.82 Ah g−1 5.85 Ah cm−3 5.78 g cm−3 1.54 n type 3.2 eV 1 O cm 1.34 g cm−3 2.3 O cm 1M 2–4 times of solubility 3 O cm 0.1 0.25 1.4  108

i0, exchange current density; CPH, close-packed hexagonal; SHE, standard hydrogen electrode.

If the solution away from the surface is below saturation level, the oxide near the surface may further be dissolved following the reverse direction of reaction (III). There may also be intermediate processes involved in each of the main reactions. A typical i − V curve for a zinc electrode in an alkaline solution is shown in Fig. 6. Toward more negative potentials, zinc deposition dominates the reduction reaction until it reaches large negative potentials at which point hydrogen evolution becomes significant. These reactions are not encountered with primary batteries, but they are an integral part of the overall reaction scheme involved in every charging process of secondary batteries. At the open-circuit potential (i.e., when the external current is zero), zinc corrodes very slowly with simultaneous hydrogen evolution as described by reactions (II) and (V). These two processes are very detrimental to the performance of all types of zinc systems, whether primary or secondary batteries, or fuel cells. The buildup of gas pressure inside a battery may cause electrolyte leakage: 2H2 O + 2e − ! H2 + 2OH −

(V)

Toward more positive potentials, the current that results from the dissolution of zinc increases exponentially with increasing overpotential (Fig. 6). The current reaches a peak at a certain potential and then drastically drops, which signifies the occurrence of

Fig. 6 Typical shape of an i − V curve and dominant reactions of zinc electrode in alkaline electrolyte.

768

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

passivation. After the peak, the current stops decreasing, rises slightly, and reaches a plateau. With continuous dissolution, precipitation of zinc oxide occurs even at currents that are only a small fraction of the peak value. The precipitate is normally loose and flocculent but tends to thicken and may gradually turn into a compact layer that gives rise to passivation of the electrode. Two different films, both zinc oxide, have been found to form during anodic dissolution in concentrated potassium hydroxide. Type I is white and porous and forms by precipitation from a supersaturated layer of electrolyte. Type II forms directly on the electrode surface and ranges from light-gray to black. The dark color of type II film is due to an excess of zinc in the film. The formation of type II film has been considered to be responsible for the transition from the active to the passive state of zinc in alkaline solutions. The detailed mechanisms of zinc passivation in alkaline solutions under specific conditions are complex. Generally, the following basic steps are involved. Zinc is oxidized to form zinc oxide or zinc hydroxide (Zn(OH)2), which, in turn, dissolves in the electrolyte to form Zn(OH)42− or Zn(OH)−3. When the electrolyte can no longer dissolve the zinc oxide or zinc hydroxide that is produced by the dissolution process, a solid film forms and passivates the surface. In practice, passivation may occur at potentials much lower than the peak value with an extended time of dissolution, as illustrated in Fig. 6. Extended dissolution allows the oxide to precipitate and accumulate in front of the electrode at much lower currents. It is generally found that, for a wide range of conditions, equations of the type. ði − io Þtp 1=2 ¼ k

(1)

can be used to describe the relationship between i, the applied current density, and tp, the time required for passivation; io and k are constants that vary with electrolyte concentration, temperature, and convective conditions. The solubility of zinc oxide in potassium hydroxide is an important issue for the functionality and performance of zinc batteries and fuel cells. Either low or high zinc solubility is desirable depending on the system. Low solubility may be beneficial for a secondary battery in which shape change of the zinc electrode is a major issue, whereas high solubility is critical for smooth operation of hydraulic zinc–air fuel cells in which precipitation of oxide may cause clogging of the system. The solubility of zinc is a strong function of potassium hydroxide concentration, as shown in Fig. 7. The electrode potential decreases with increasing potassium hydroxide concentration. By contrast, the reaction kinetics in terms of exchange current and conductivity increase with potassium hydroxide concentration to about 30% and then decrease with further increases in concentration. For good conductivity and kinetic behavior, electrolytes with concentrations of 20–40% potassium hydroxide are commonly used. In this concentration range, the solubility is on the order of 1 mol L−1. Between 19
C and 145
C, there is very little temperature dependence of zinc oxide solubility in 35–46% potassium hydroxide. The solubility of electrochemically dissolved zinc in potassium hydroxide solutions is much higher than that from the dissolution of zinc oxide; as much as 3 times more zinc can be dissolved electrochemically resulting in a supersaturated solution. The supersaturated solution is very stable; the excess dissolved zinc very slowly precipitates as zinc oxide, and may take months or years to reach the equilibrium concentration. Neither seeding nor shock hastens the precipitation of zinc oxide from supersaturated solutions. Such behavior is beneficial for the operation of hydraulically refuelable fuel cells, as higher zinc solubility means higher energy density by using less electrolyte to dissolve the zinc in the system. The solubility of zinc in alkaline solutions can be modified via the addition of other salts. For example, calcium hydroxide, with which zincate tends to form less soluble compounds, reduces the solubility of zinc. On the contrary, some inorganic additives, such as titanate, aluminate, stagnate, sorbitol, and silicate, are found to increase the degree of supersaturation of zinc oxide in alkaline solutions. Unlike alkaline systems with zincates as charge carriers, zinc electrode involves the reversible plating/stripping of Zn2+ ions in the mild aqueous solutions. The reaction mechanism of Zn electrode can be summarized as: Discharge process : Zn ! Zn2+ + 2e −

Fig. 7 Different properties of the zinc–KOH system as a function of KOH concentration.

(VI)

Batteries – Battery Types – Zinc Batteries | Zinc Electrode Charge process : Zn2+ + 2e − ! Zn

769 (VII)

Simultaneously, given that Zn has a high electrochemical activity and thermodynamical instability in mild aqueous electrolytes, there are also anode-related disadvantages, such as dendrite growth, hydrogen evolution and corrosion.6,8 The Zn/ZnO standard reduction potential is −1.22 V vs standard hydrogen electrode (SHE) in alkaline electrolyte. Whereas, the standard reduction potential of Zn/Zn2+ is −0.76 V vs SHE in mild electrolyte, which is higher (less negative) than that of the Zn electrode in alkaline electrolyte. Therefore, the different electrode reactions of the Zn electrode in both electrolytes display different working potentials, which further affects the output voltage as well as the energy density of the battery.

6

Porous electrode

Direct use of Zn foil or Zn plates in alkaline media is still a challenge, because they have a low surface area, which easily leads to poor performance due to Zn passivation and dendrite growth.9 A porous electrode greatly improves the charge-transfer process because of the large effective surface area, good availability of electrolyte near the surface, and space for storing dissolution products. The high specific surface area allows high current without a large overpotential and sustains high current by virtue of its effect on reducing susceptibility to passivation.10 A characteristic property of porous electrodes is the current distribution from the outer surface to the interior of the electrode. There is a large difference between the current density at areas near the outer surface and that in the interior of the electrode because of the difference in the lengths of the electrolyte path. The relative amount of current between the surface and the interior of an electrode depends on the relative magnitude of the polarization resistance, Rp, and the electrolyte resistance Re. In the extreme case when the polarization resistance is negligible compared with electrolyte resistance such that Re/Rp ¼ 1, the current is essentially concentrated on the surface, as shown schematically in Fig. 8. In such a case, the benefit of the porous electrode is lost. On the contrary, when electrolyte resistance is negligibly small compared with that of the polarization resistance, i.e., Re/ Rp ¼ 0, the current is evenly distributed across the electrode. In practical situations, 0 < Re/Rp < 1, and there is a current distribution that is highest on the surface and decreases toward the interior. The larger the ratio of Re/Rp, the steeper the current distribution. Current distribution from surface to interior becomes steeper with increasing current. When the electrode kinetics are under activation control (current increases exponentially with potential), the increase in current is associated with decreasing polarization resistance. The benefit of a porous electrode becomes marginal at very high currents, as most current comes from the surface region with very little contribution from the interior of the electrode. When the reaction is limited by mass transport between the anode and cathode, the benefit of the large surface area is lost as the rate-limiting process is no longer in the interior of the electrode. In this case, the presence of pores in the electrode does not improve reaction kinetics as it does not participate in the mass transport between the anode and cathode. The character of the current—ionic or electronic—in a partially discharged electrode depends on the region in the electrode. The discharged part of a zinc electrode is a mixture of electrolyte and dissolution product, potassium hydroxide, and zinc oxide in a pasty form. In this state, the current is fully conducted by ions. In the active zone, the current is carried by ions in the electrolyte within pores and by electrons in the zinc metal; the relative amount of electronic current increases toward the inactive region where it is almost 100% electronic in nature. Discharge transforms metal atoms to ionic form and causes changes in the physical, chemical, and electrical properties of the anode–electrolyte assembly. There are several physical effects associated with discharging time. First, the reaction front proceeds

Fig. 8 Schematic illustration of current distribution in a porous structure for different ratios of electrolyte resistance to overpotential losses.

770

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

Fig. 9 Overpotential of gelled zinc powder electrode during discharge. Reproduced from Huot, J. Resistivity and Anodic Discharge of Mercury-Free Gelled Zinc Anodes. J. Power Sources 1993, 14, 177.

gradually from the surface toward the interior. Second, dissolution products (mainly zinc oxide) precipitate and cause volume expansion, reduction in porosity, and less access of electrolyte. Third, the amount of active material and its effective surface area, as well as the quantity of electrolyte, are gradually reduced with dissolution time. Furthermore, depending on the method by which the electrode is made, some particles of the active material may lose electronic connection with the electrode and thereby cause a loss of capacity. A typical discharge curve for a gelled zinc powder electrode at a current density of 20 mA cm−2 is given in Fig. 9. Four characteristic regions are displayed on the curve. An instant rise of voltage of 6 mV occurs upon imposing the current and is attributed to the activation overpotential and the electrical resistance of the anode. This is followed by a region of nonlinear overpotential rise (region A) for a total of about 60 mV that is caused by the transition to a steady state, which is largely controlled by mass transport processes. This region can vary from 7% to 12% of the total discharge capacity. After establishment of a steady-state condition, the overpotential varies almost linearly with time (region B) until the last stage when the voltage rises sharply (region C) because of the occurrence of passivation, which marks the end of life for the anode. The linear part of the discharging curve is largely due to the increased resistance of the longer ionic path associated with the receding of the reaction front on the electrode. The slope of the discharging curve can be used to estimate the resistivity of the discharged layer, which is a mixture of zinc oxide and electrolyte. The resistivity of the linear region for the gelled powder anode is found to be about 3.7 O cm. Discharged zinc anode is a mixture of zinc oxide and electrolyte and is typically in a muddy-pasty form, regardless of the initial form of the material (sheet, powder, or fiber) and type of battery. The discharged product appears light-gray at a low discharge current density whereas that discharged at a high current density tends to be blue, which has been attributed to the presence of undischarged interstitial zinc.

7

Morphology of zinc deposits

The morphology of zinc deposits obtained in alkaline, neutral/mildly acidic electrolytes plays a very important role in rechargeable batteries as zinc repeatedly undergoes dissolution and deposition during charge and discharge cycles. The subject has been investigated extensively. The major general findings from these studies are as follows:

• • • • • •

There are five principal types of deposits, namely mossy or spongy, layered, granular, dendrite, and cluster; these occur in the order of increasing current. A mossy type of deposit may also occur on top of a smooth deposit at a low current density at longer deposition times. Growth of dendrites is under diffusion control, whereas that of mossy is under activation control. Kinetically, there are preferred crystal directions and planes for dendrite growth. The condition of the substrate surface exerts an influence on the early stage of dendrite growth, but has little effect on the growth of dendrites. It is difficult to obtain compact deposits at a significant rate in alkaline solutions, except for the first fraction of a millimeter, because of the great tendency for the formation of dendritic deposits.

The five types of deposits, as characterized in a recent systematic study, are displayed in Fig. 10. Mossy deposits have the appearance of entangled whiskers with a diameter in the range of 50–200 nm and lengths that can exceed 5 mm. Layer-like deposits have the appearance of steps as a result of epitaxial growth, which is typically found at the beginning of deposition. Boulder deposits are

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

771

Fig. 10 Different types of deposit morphology obtained from KOH electrolyte; clockwise from the top left corner: cluster, dendrite, boulder, filament, mossy, and layer types.

granular shaped with clear planes and sharp angles characteristic of hexagonal crystalline structure. Dendrites are tree-like or leaf-like single crystallites generally oriented perpendicular to the substrate surface. Dendrites can be hexagonal columns if deposition is slow. Cluster deposits are agglomerates of chunky irregular boulders and highly branched dendrites developed at a very high current density. The layer and boulder deposits are compact, whereas mossy, dendrite, and cluster deposits are porous or dispersed. The layer and boulder deposits adhere to the substrate well, whereas the other three types are nonadherent. A summary of the characteristics of the five different types of deposits is given in Table 2. Layer and boulder types of zinc deposit are transient morphological patterns, developed only in the early stage of deposition. They tend to be replaced by other morphological patterns. As deposition proceeds beyond a certain time, a mossy deposit typically starts to form on top of layer or boulder deposits at a low current density; dendrites start to form at a high current density; and a Table 2

Characteristics of different types of zinc deposits.

Formation of current density Appearance Size of smallest dimension Aspect ratio Adherence Compactness Dimension of deposit growth Crystalline character Preferred growth plane Preferred growth direction Nucleation site selectivity Change of effective surface area with deposition time Dominating deposition mode Controlling process

Mossy

Layer-like

Boulder

Dendrite

Cluster

Filament whisker 50–200 nm Length/diameter >10 Poor Highly porous 1 Single crystallites

Layer, step 1 mm Step width/height 2–10 Strong Compact 2 Epitaxial crystallites

Granular 10 mm Length/width 1–2 Strong Compact 3 Equal axial crystallites

Bulky boulder >50 mm Length/width 1–2

(0001) [0001] or [1120], [1010] Highly selective Increase

(0001) [1120], [1010] Selective Little

None None Non-selective Little

Fern, leaf column 5–7 mm Length/thickness >10 Poor Dispersed 2 Discrete crystallites (1010) [1120], [1010] Non-selective Increase

None None Non-selective Little

Growth of crystallites Deposition site

Nucleation and growth Activation

Nucleation and growth Activation

Growth Diffusion

Nucleation and growth Electrolyte turbulence

Poor Dispersed 3 Poly crystallites

772

Batteries – Battery Types – Zinc Batteries | Zinc Electrode Table 3

Conditions that favor the formation of mossy and dendrite deposits.

Condition

Mossy

Dendrite

Current density Capacity Zincate concentration Stirring electrolyte Temperature

Low Low High Little effect High

High High Low Low Low

cluster deposit starts to form at a very high density. The conditions that favor the formation of a mossy deposit appear to be the opposite of that for dendrite deposits, as shown in Table 3.11 There is a clear crystal character for each type of deposit. Except for the cluster type, in which the individual grains are polycrystalline, the respective deposit of all other types is a single crystal. Also, except for boulder and cluster deposits, the preferred growth planes are {0001} and {1010} and the preferred growth directions are [1120] and [1010]. The exact orientation of individual deposits for each type with respect to the surface substrate depends on the orientation of the surface lattice on which the deposits are formed. The presence of metal impurities in the electrolyte can greatly affect deposit morphology, and different elements may impose very different effects. For example, the presence of indium results in larger and smoother grains, bismuth in finer grains with unclear crystalline character, and lead in amorphous-like grains. How different elements exactly affect the morphology of zinc deposits is still not understood. The effect of alloy additions on deposit morphology is perhaps one of the most important reasons for the large disparities found in the numerous investigations on the shape change of zinc electrode in the past. It is also a commonly employed strategy for controlling the shape change of zinc electrode.

8 8.1

Potential improvements for zinc battery electrode Morphological stability of rechargeable electrode

The main problem associated with zinc-based rechargeable batteries is the limited cycle-life principally due to the zinc electrode, the morphology of which tends to change on cycling. The change of morphology that causes an electrode to decline in performance or fail can be in several modes: shape change, densification, and dendrite formation. Shape change is a phenomenon involved with alteration of the zinc electrode geometric area, where zinc active material leaves one location and agglomerates in other locations. Most commonly, it is the edge and top that lose zinc, whereas the center and bottom parts of the electrode gain mass although the reverse can sometimes occur. Densification is the phenomenon whereby the electrode loses porosity and active surface area and thereby suffers a reduction in its kinetic capability and passivates easily. Densification is often observed together with shape change. Dendrite formation tends to occur during the charging period and may result in penetration through the separator causing shorting and instant cell failure.

8.2

Shape change

Shape change can be affected by numerous parameters in the system. The complexity of the factors and mechanisms involved in shape change have been reviewed in several papers. The parameters that have been investigated can be grouped into three major categories with respect to their effect on shape change, as shown in Table 4. The parameters that have the greatest effect are the stoichiometric ratio of cathode and anode, the type of separator, the depth of discharge, and the alignment of the cathode and the anode. There may also be other factors such as the ratio of solution volume to zinc, the unevenness in mass distribution in the electrode, and the difference in charging and discharging current. A number of mechanisms have been proposed for shape change, namely

• • • • • • • •

Osmotic pressure gradients associated with separators; Concentration gradients caused by a difference in current distribution during charge and discharge; Natural convection caused by density gradients; Autocatalytic dissolution from the periphery of the electrode; Lower charging efficiency for zinc deposition at electrode periphery owing to potassium hydroxide concentration gradients; Electrolyte flow during cycling induced by a density gradient across the electrode surface; Nonuniform current efficiency; Oxidation of zinc by oxygen that migrates from the cathode.

Uneven current distribution across the electrode surface is particularly important. This tends to be more pronounced during charge than during discharge. Uneven current distribution could result from a highly resistive separator, a wrinkling separator, misalignment of electrodes, or a misbalanced stoichiometric ratio of anode and cathode materials.

Batteries – Battery Types – Zinc Batteries | Zinc Electrode Table 4

773

Factors that have been identified to affect the shape change of rechargeable zinc electrode.

Enhancing shape change Increasing depth of discharge Loose packing High resistivity membranes Poor alignment of anode and cathode plates Increasing KOH concentration Amalgamation Electrolyte convection Nonuniformity in electrode (wrinkles in separator, bare spots and cracks in electrode, uneven mass distribution) Reducing shape change Increasing stoichiometric ratio of zinc to cathode active material Increasing the thickness of the separator at the edge Addition of ZnO to KOH Special charging methods Increasing concentration of zinc salt in neutral electrolytes No significant effects Excess electrolyte Electrolyte level Gravity Cell orientation Form of zinc powder or ZnO Rate of discharge and charge

The phenomenon of shape change and its manifestation with various factors are very complex. Generally, shape-change can be described as follows. A difference in the current density between the edge (presumed higher) and the center results in different zinc deposit morphologies, which, in turn, enhance the difference in the kinetics between the edge and the center. Force gradients (e.g., concentration, temperature, gravity, electric field) are created owing to uneven current distribution, particularly near the end of charging and discharging, and thus induce a net mass transport from the edge to the center. Shape change occurs when a sufficient amount of mass is transported to the center from the edge over a period of continuous cycling. Conversely, movement of the material in the reverse direction may occur if the initial current is lower at the edge than at the center. Fundamentally, the zinc electrode during discharge–charge cycling is associated with solid–liquid phase changes in the reaction, i.e.,

ðVIIIÞ

Because an individual atom goes through two liquid and two solid phases during each reaction cycle, it is highly unlikely that upon recharge it returns to exactly where it was in the electrode structure. There is a change of location for individual atoms after each cycle. A net change in microstructure, however minute, can occur in the electrode structure with the change of location of the atoms. The accumulation of minute lattice structural changes after many cycles could then lead to a noticeable morphological change at the scale of the electrode—the phenomenon of shape change.

8.3

Methods for controlling shape change

A wide range of methods have been explored for controlling the shape change of zinc electrode, namely the use of excess zinc, an oversized electrode, a low zinc solubility, special charging methods, inorganic additives, tight packing, a low-resistance membrane, a method for controlling electrode uniformity, an organic binder and gel, different mass or density from electrode center to edge, and variable separator thickness from center to edge. Various organic and inorganic additives may be mixed in the paste to improve the performance of the zinc electrode from different aspects, namely increased wetting of the zinc surface in the electrode, reduction of zinc solubility, dendrite suppression, formation of a continuous metal network to provide a base for more uniform deposition, complexation of dissolved zinc species to immobilize zinc, optimization of cross-sectional current distribution during the course of charging and discharging, and control and retention of the porous structure. Of the numerous metal additives that have been explored, the oxides or hydroxides of cadmium, lead, tin, indium, gallium, thallium, and bismuth have received the most attention. These elements have two common features. First, they all have a more noble standard potential than zinc such that, if present in ionic form in solution, they could deposit on the zinc surface during

774

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

charging, rest, or discharging so as to be always present. Second, they all have high hydrogen overpotentials and therefore their presence would possibly inhibit hydrogen evolution. Reducing the solubility of zinc in potassium hydroxide has been a common approach to controlling shape change. Methods include adding calcium oxide (CaO) or calcium hydroxide (Ca(OH)2) to the zinc electrode or using electrolytes of highly soluble salts, such as phosphate, borate, or fluoride, to have a minimal hydroxyl content. Electrolytes of low solubility can, however, lead to passivation during discharge at low currents. Some methods for shape control have been adopted in practice. These include the use of.

• • • • • •

Excess zinc over the stoichiometric amount of the cathode material; Oversized negatives; Special additives and binders in the mass of the electrode or as a surface coating; Lower concentration of potassium hydroxide and/or electrolyte additives that decrease the concentration of OH− ions; Special charging methods. Artificial or in-situ formed solid-electrolyte-interface.

More recent attempts tend to use a combination of multiple methods. Cycle-life associated with shape change is only one of the many performance and manufacturing criteria for a battery. Others may include specific energy, energy density, specific power, power density, charging time, form factor, shelf life, safety, convenience of use, manufacturing feasibility, and production cost. Any method that improves the cycle-life must also meet these other criteria. For instance, it must be compatible with the controlling strategies for issues such as dendrite formation, passivation, and gassing. Thus, depending on the system and circumstances, some methods may be feasible and some not. In the case of a NidAg battery, for example, the use of a low potassium hydroxide concentration to reduce zinc solubility and mass movement is not practical because a high potassium hydroxide concentration is required to limit the solubility of silver oxide (Ag2O). Also, such a tactic may cause zinc to passivate more easily and thus limit the current output.

9

Control of dendrite formation

Dendrites, as discussed above, are a particular type of morphology that form only under certain conditions. There are remarkable differences among Zn dendrites in different electrolytes. The Zn dendrites in alkaline system show a 1D ramified cone-like topology, which is likely to easily pierce the separators. And they can form even under moderate conditions of low capacity and current density, implying that their origin lies in thermodynamic instability. In contrast, Zn dendrites in neutral/mildly acidic electrolytes seem to be negligible in moderate conditions, while they are significant in extreme conditions (large capacity and current density). The Zn dendrites in neutral/mildly acidic electrolytes showed a 2D hexagonal morphology without ramification, remarkably weakening their piercing capability and enabling relatively facile dendrite protection along the physical shielding approach.12,6,8 Dendrite formation may not occur if the conditions during charge are not favorable. For example, it has been reported that dendrite formation normally does not occur if there is an excess of reducible zinc species in the zinc electrode. Methods for controlling dendrite growth include additives in the electrode, the use of excess active material, novel Zn electrodes, electrolytes optimization, artificial or in-situ formed solid-electrolyte-interface, special separator materials, and specific charging methods such as multisteps, superimposed alternating current and direct current, and pulse charging. Many additives, e.g., dimethyl sulfoxide, alkylammonium salt, polyethylene glycol 2000, polyvinyl alcohol, polyacrylamide, poly(sodium 4-styrenesulfonate), sodium dodecyl sulfate, cetyltrimethylammonium bromide, trimethyl phosphate and bismuth oxide, can have a drastic effect on the morphology of a zinc deposit in general and on dendrite formation in particular, as discussed in detail in a previous section.13 In the case of artificial protective layers, the large volumetric change in Zn deposition (1.7 mm under 1 mA h) leads to cracking or other damage to the protective layer. In contrast, the 3D copper skeleton current collector displayed dendrite-free Zn electrodeposition, which is ascribed to the uniform surface current distribution and electrode stability. High electrical conductivity and internal pores of the 3D copper skeleton can effectively accommodate massive volume changes and enhance Zn reversibility.6,14 Use of a special separator has been a common approach to suppress dendrite growth and prevent dendrites from penetrating through the separator and causing short circuits. It has been suggested that separators prevent zinc penetration to the positive electrode because the overpotential for zinc deposition in the separator is higher than that in the zinc electrode itself. Commonly used separators include cellulose and glass fiber membranes. Composite separators with impregnation or the coating of metals, metal oxide, and hydroxides, as well as multilayer separators, have also been used to provide better resistance to dendrite growth and penetration.

10

High rate performance

The capacity of a battery is influenced by the discharge rate; it typically decreases with increasing discharging rate. For a given cell and electrode design, the performance of the electrode depends strongly on the porous structure. For a porous zinc electrode, four processes may occur and lower the discharge capacity, namely depletion of electrolyte, pore plugging by dissolution products, passivation, and loss of matrix conductivity. Electrolyte depletion can be serious in the case of electrodes with low porosity or in a tight packing. The limited electrolyte reservoir and access for electrolyte transportation can cause early electrode failure. An electrode

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

775

Fig. 11 Qualitative illustration of passivation current density and actual current density as a function of discharging time for a zinc anode.

with small porosity can be easily blocked by the precipitation of zinc oxide, which has a 50% larger volume than zinc. Passivation, i.e., the electrode loses the ability to sustain current, occurs when a dense oxide film completely covers the electrode surface. Formation of a porous oxide layer on the surface does not generally cause passivation.15,16 The various effects on discharging capacity can be described based on the relationship between passivation and reaction kinetics as demonstrated in Fig. 11, which schematically illustrates the change in passivation current, ip, and current density of the active surface, iactual, as a result of the changes occurring inside the battery during discharge. Discharging causes a decrease in potassium hydroxide concentration, an increase in zincate concentration, and formation of zinc oxide. This means that, at a later stage of discharging, passivation will occur at a lower current density than at the beginning of discharge. On the contrary, at a constant discharging current, the actual current density on the active surface tends to increase as the effective surface area becomes smaller as a result of a reduced number of particles and a decreased particle size. As discharging continues, at a certain time, such as depicted by t1 in Fig. 11, when the current density is equal to the passivation current, passivation will occur and cause failure of the anode.16 It is also seen that a larger discharging current will have a higher actual current density and thus a smaller capacity. Thus any condition in anode formulation and fabrication that helps to sustain a high passivation current or that helps to slow down the increase of actual current density, i.e., by moving the solid lines to dashed lines in Fig. 11, will increase the discharge capacity from t1 to t2. Improving the material chemistry and the electrode porosity can help to move the passivation current curve from the solid line toward the dashed line, whereas increasing the specific area and the anode–cathode interface area will shift the actual current density from the solid line to the dashed line. For zinc–air systems, the capacity output of individual cells is independent of the active cathode material. In order to realize the advantage of its high energy density, the zinc anode needs to be relatively thick to reduce the number of cells in a battery. On the contrary, the power of the battery is directly proportional to the surface area of the cathodes and this obviously increases with the number of cells. An example of the relationships between energy density, power density, and zinc anode thickness is presented in Fig. 12. The data show that a battery with a 5-mm-thick anode has an energy density of more than 650 Wh L−1 with a power density close to 60 W L−1. The energy density decreases sharply with thinner anodes, but the power density significantly improves. The kinetics of the zinc anode is thus of particular importance for the design of zinc–air batteries. If the anode (also the cathode) is

Fig. 12 Energy density and power density for a zinc–air cell as a function of zinc anode thickness.

776

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

capable of discharging at a higher current density, the battery can use a thicker anode and have a higher energy density. Dendritic powder, which has a very high specific surface, is typically used for making high-power battery anodes. Zinc fibers, owing to their good connectivity, and thus electric conductivity, are also suitable for high-power anodes. For hydraulic zinc–air fuel cells, high-power performance can be improved by circulating electrolyte through the cell bed, which keeps renewing the electrolyte on the zinc surface and thus enhancing the reaction kinetics.16

11

Low and high temperature performance

Special practical applications, such as polar, aerospace, deep sea, and high-altitude region exploration, require the zinc-based energy storage device to operate at low or high temperatures. However, zinc batteries suffer from rate capacities, cycling life performance degradation and safety problems at extreme temperatures. If the temperature reaches the freezing point of the electrolyte, then crystallization will occur, leading to volume expansion of the electrolyte and interfacial separation between electrolyte and electrode. The H-bond design of the electrolyte ensures its anti-freezing property. H-bonding acceptor or donor, like dimethyl sulfoxide and polyethylene glycol, are introduced to suppress the formation of hydrogen bonds in water. Above room temperature, the inferior thermal and interfacial stabilities also pose a challenge to most of the zinc batteries regarding safe and long-term operation. The extreme heat inside the battery will accelerate the side reactions of water with zinc electrode, leading to depletion of the zinc metal anode. To ensure the high temperature performance, the thermal stability of the aqueous electrolyte should be increased by minimizing its volatility and controlling decomposition. Highly concentrated, nonaqueous/aqueous hybrid solvents and hydrogels, have been established based on this principle.17–20

12

Reversibility

The practical application of aqueous zinc batteries is challenged by poor reversibility and stability of the Zn electrodes. The irreversibility of Zn electrodes during repeated plating/stripping is due to the strong Coulombic interactions between solvated Zn2+ ions and their surrounding water shell. Water decomposition leads to the increase of local pH and the formation of Zn(OH)42− (zincate), which subsequently converts to electrochemically inert ZnO deposits. Moreover, the inhomogeneous morphologies of ZnO-based depositions on Zn induces the dendrite growth that compromises cycle life. Therefore, large excesses of Zn and electrolytes are used to mitigate Zn anode and water consumption, which decrease the overall energy density of batteries. To suppress water consumption and Zn dendrites, the bonding strength between Zn2+ ions and solvated water should be weakened, and a waterproof Zn-ion conductive solid electrolyte interphase should be constructed.13,17,14

13

Outlook

The unique set of physical and electrochemical attributes of zinc has made it a long-lasting and widely used electrode material. The diversity and versatility, by which zinc material can be processed and utilized, will allow zinc, in addition to its use in the many existing conventional battery systems, to be further explored and developed for nonconventional electrochemical power source systems. It could be expected that new embodiments of power source systems would continue to emerge and some may reach commercial success in the future with the advances in zinc materials science and engineering along with progresses in product design, system integration, operation control, and manufacturing capability. There is still much scope for the further advancement of existing and potential zinc-based power source systems. The energy utilization in primary batteries for many high-drain applications could be less than 30%. High-power zinc anode would allow the development of high-drain batteries with much larger capacity than presently available in the marketplace. For secondary zinc batteries, better control of zinc electrode morphology, areal capacity, current density, electrolyte amount and depth of discharge should improve the cycle-life and power density of MnO2-Zn, NidZn battery and ZndAg battery and allow them to succeed in mainstream commercial applications. The prospects of mechanically and hydraulically refuelable and/or regenerative systems would be realized with the advancement in operation and service model, equipment, and infrastructure. Many aspects of the electrochemistry of the zinc electrode are worthy of further investigation and development. They include, for example, time-based relationship between electrode kinetics and transformation of zinc electrode during dissolution and deposition; salvation process of dissolved zinc dissolution products in relation to electrolyte pH, salt types, solubility and supersaturating; the overpotential of zinc deposition/dissolution in relation to desolvation energy and interface kinetics of zinc ions; form and role of zinc dissolution products with respect to anode structure and operation conditions; effect of impurity and alloying elements in both dissolution and deposition processes; conduction mechanism between zinc particulates in the electrolyte during discharging using in-operando methods; and electrochemical kinetics and physical stability of electrode in relation to forms of material; energy density in relation to the pair of Zn anode with the cathode, electrolyte, separator, and current collector.

Batteries – Battery Types – Zinc Batteries | Zinc Electrode

777

References 1. Zhang, N.; Chen, X.; Yu, M.; Niu, Z.; Cheng, F.; Chen, J. Materials Chemistry for Rechargeable Zinc-Ion Batteries. Chem. Soc. Rev. 2020, 49, 4203–4219. 2. Blanc, L. E.; Kundu, D.; Nazar, L. F. Scientific Challenges for the Implementation of Zn-Ion Batteries. Joule 2020, 4, 771–799. 3. Wang, X.; Zhang, Z.; Xi, B.; Chen, W.; Jia, Y.; Feng, J.; Xiong, S. Advances and Perspectives of Cathode Storage Chemistry in Aqueous Zinc-Ion Batteries. ACS Nano 2021, 15, 9244–9272. 4. Smedley, S. I.; Zhang, X. G. A Regenerative Zinc–Air Fuel Cell. J. Power Sources 2007, 165, 897–904. 5. Cao, L.; Li, D.; Pollard, T.; Deng, T.; Zhang, B.; Yang, C.; Chen, L.; Vatamanu, J.; Hu, E.; Hourwitz, M. J.; et al. Fluorinated Interphase Enables Reversible Aqueous Zinc Battery Chemistries. Nat. Nanotechnol. 2021, 16, 902–910. 6. Du, W.; Ang, E. H.; Yang, Y.; Zhang, Y.; Ye, M.; Li, C. C. Challenges in the Material and Structural Design of Zinc Anode towards High-Performance Aqueous Zinc-Ion Batteries. Energy Environ. Sci. 2020, 13, 3330–3360. 7. Ma, L.; Zhi, C. Zn Electrode/Electrolyte Interfaces of Zn Batteries: A Mini Review. Electrochem. Commun. 2021, 122, 106898. 8. Li, D.; Cao, L.; Deng, T.; Liu, S.; Wang, C. Design of a Solid Electrolyte Interphase for Aqueous Zn Batteries. Angew. Chem. Int. Ed. 2021, 60, 13035–13041. 9. Coates, G.; Hampson, N. A.; Marshall, A.; Porter, D. F. The Anodic Behaviour of Porous Zinc Electrodes. II. The Effects of Specific Surface Area of the Zinc Compact Material. J. Appl. Electrochem. 1974, 4, 75–80. 10. Parker, J. F.; Chervin, C. N.; Pala, I. R.; Machler, M.; Burz, M. F.; Long, J. W.; Rolison, D. R. Rechargeable Nickel–3D Zinc Batteries: An Energy-Dense, Safer Alternative to Lithium-Ion. Science 2017, 356, 415–418. 11. Wang, R. Y.; Kirk, D. W.; Zhang, G. X. Effects of Deposition Conditions on the Morphology of Zinc Deposits from Alkaline Zincate Solutions. J. Electrochem. Soc. 2006, 153, C357. 12. Cao, L.; Li, D.; Deng, T.; Li, Q.; Wang, C. Hydrophobic Organic-Electrolyte-Protected Zinc Anodes for Aqueous Zinc Batteries. Angew. Chem. Int. Ed. 2020, 59, 19292–19296. 13. Cao, L.; Li, D.; Hu, E.; Xu, J.; Deng, T.; Ma, L.; Wang, Y.; Yang, X.-Q.; Wang, C. Solvation Structure Design for Aqueous Zn Metal Batteries. J. Am. Chem. Soc. 2020, 142, 21404–21409. 14. Ma, L.; Schroeder, M. A.; Borodin, O.; Pollard, T. P.; Ding, M. S.; Wang, C.; Xu, K. Realizing High Zinc Reversibility in Rechargeable Batteries. Nat. Energy 2020, 5, 743–749. 15. Huot, J.-Y.; Malservisi, M. High-Rate Capability of Zinc Anodes in Alkaline Primary Cells. J. Power Sources 2001, 96, 133–139. 16. Zhang, X. G. Fibrous Zinc Anodes for High Power Batteries. J. Power Sources 2006, 163, 591–597. 17. Cao, L.; Li, D.; Soto, F. A.; Ponce, V.; Zhang, B.; Ma, L.; Deng, T.; Seminario, J. M.; Hu, E.; Yang, X.-Q.; et al. Highly Reversible Aqueous Zinc Batteries enabled by Zincophilic–Zincophobic Interfacial Layers and Interrupted Hydrogen-Bond Electrolytes. Angew. Chem. Int. Ed. 2021, 60, 18845–18851. 18. Chen, M.; Xie, S.; Zhao, X.; Zhou, W.; Li, Y.; Zhang, J.; Chen, Z.; Chao, D. Aqueous Zinc-Ion Batteries at Extreme Temperature: Mechanisms, Challenges, and Strategies. Energy Storage Mater. 2022, 51, 683–718. 19. Liu, S.; Zhang, R.; Mao, J.; Zhao, Y.; Cai, Q.; Guo, Z. From Room Temperature to Harsh Temperature Applications: Fundamentals and Perspectives on Electrolytes in Zinc Metal Batteries. Sci. Adv. 2022, 8, eabn5097. 5091–5024. 20. Zhang, Q.; Ma, Y.; Lu, Y.; Li, L.; Wan, F.; Zhang, K.; Chen, J. Modulating Electrolyte Structure for Ultralow Temperature Aqueous Zinc Batteries. Nat. Commun. 2020, 11, 4463.

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon K Kordesch and W Taucher-Mautner, Graz University of Technology, Graz, Austria © 2009 Elsevier B.V. All rights reserved. This is a reproduction of K. Kordesch, W. Taucher-Mautner, PRIMARY BATTERIES – AQUEOUS SYSTEMS | Leclanché and Zinc–Carbon, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 43–54, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00097-6.

Introduction History of Zinc–Carbon Batteries Basic Constructions of Zinc–Carbon Batteries Cylindrical Cells Flat Cells Battery Materials Manganese Dioxide Zinc Corrosion Inhibitors Carbon Black Electrolytes Separators Chemistry of Zinc–Carbon Systems Types of Zinc–Carbon Cells Leclanché Cells Zinc Chloride Cells Cell Performances Discharge Characteristics Temperature Sensitivity Shelf Life Rechargeability Conclusions Further Reading

779 779 781 781 782 784 784 784 784 784 784 784 785 785 785 785 786 786 789 789 789 790 790

Abstract This article describes the history, basic constructions, materials, chemistry, types, and cell performances of zinc–carbon batteries that have been well known for 140 years. The oldest cell type, the Leclanché battery, contains aqueous ammonium chloride as electrolyte, a mixture of manganese dioxide and carbon black contacted with a carbon rod as cathode, and solid zinc as anode. The second type of zinc–carbon batteries, the zinc chloride cell, utilizing paper separator and zinc chloride electrolyte providing improved performance on heavy-drain applications, was developed considerably later. Owing to their relatively low price and reliability, both cell types remain among the most widely used of all primary systems worldwide, especially in emerging third world countries. Zinc–carbon cells are produced either in cylindrical geometry or as flat multicell battery increasing volumetric density nearly twice. The discharge characteristics of Leclanché and zinc chloride systems of different battery grades (general purpose, heavy duty, and extra heavy duty) are compared to each other. In general, continuous discharge and higher current increases performance difference between battery grades of the same size. Intermittent discharge reduces the difference between grades and systems. Zinc chloride cells most evidently outperform Leclanché cells when used continuously and also exhibit better low-temperature performance.

Glossary Anode The electrode in an electrochemical cell where oxidation takes place. During discharge, the negative electrode of the cell is the anode. During charge, the situation reverses and the positive electrode of the cell is the anode. Capacity The total number of ampere-hours (Ah) that can be withdrawn from a fully charged cell or battery under specified conditions of discharge. Capacity retention The fraction of the full capacity available from a battery under specified conditions of discharge after it has been stored for a period of time. Cathode The electrode in an electrochemical cell where reduction takes place. During discharge, the positive electrode of the cell is the cathode. During charge, the situation reverses and the negative electrode of the cell is the cathode. Continuous discharge A discharge during which a battery is discharged to a prescribed end-point voltage without interruption.

778

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00350-9

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

779

Corrosion (From Latin ‘corrodere’ meaning ‘gnaw to pieces’) Corrosion of metals is the spontaneous chemical (oxidative) destruction of metals under the effect of their environment. Most often it follows an electrochemical mechanism, where anodic dissolution (oxidation) of the metal and cathodic reduction of an oxidizing agent occur as coupled reactions. Cutoff voltage The battery voltage at which the discharge is terminated. It is also called end voltage. Depth of discharge The ratio of the quantity of electricity (usually in ampere-hour) removed from a cell or battery on discharge to its rated capacity. Energy density The ratio of the energy from a battery to its volume (Wh l−1). Galvanic cell An electrochemical cell that converts chemical energy into electrical energy by electrochemical action. Intermittent discharge A discharge during which a battery is subjected to alternate periods of discharge and rest according to specified discharge regime. Polarization The change of the potential of a cell or an electrode from its equilibrium potential caused by the passage of an electric current. Separator An ion-permeable, electronically nonconductive spacer or material, which prevents electronic contact between electrodes of opposite polarity in the same cell. Shelf life The duration of storage under specified conditions at the end of which a cell or battery still retains the ability to give a specified performance.

Nomenclature

Abbreviations and Acronyms AMD CMD COV EHD EMD GP HD LIF NMD

activated manganese dioxide chemically synthesized manganese dioxide cutoff voltage extra/super heavy duty electrolytic manganese dioxide general purpose heavy duty light intermittent flashlight natural manganese dioxide

Introduction Zinc–carbon batteries or ‘dry’ cells are galvanic cells that have been well known for 140 years. There are two types of zinc–carbon batteries in use today, the zinc chloride and the Leclanché systems, providing an economical power source. From the earliest inception in the 1860s, the Leclanché cell was commercially successful because the zinc of the anode, naturally occurring manganese dioxide for the cathode, and ammonium chloride salt for the electrolyte were sufficiently available and inexpensive. The zinc chloride battery technology was developed later on, in the 1960s, comprehending stable thin separators, improved seals, and zinc chloride as electrolyte, yielding substantially improved performance on heavy-drain applications and less cell leakage. Both types remain among the most widely used of all primary batteries worldwide, although their use in the United States and Europe is declining. In general, the zinc–carbon battery can be characterized as having low cost, ready availability, and acceptable performance for many applications, which are especially important attributes for power supply in emerging third world countries.

History of Zinc–Carbon Batteries The Leclanché cell was invented by the French telegraphic engineer G. L. Leclanché (1838–82) in the 1860s. In Figure 1, the first Leclanché wet cell is shown. It originally consisted of a square glass jar containing an aqueous solution of ammonium chloride. The positive electrode or cathode was made of a one-to-one mixture of manganese dioxide and powdered coke that was pressed into a porous earthenware pot containing a carbon rod current collector. An amalgamated zinc rod served as negative electrode or anode. Both electrodes were immersed in the electrolyte solution. The manufacturing of Leclanché cells was inexpensive. The cell was simple and safe in operation, it provided excellent shelf storage, and the electrical energy was adequate at that time. In 1868, already 20 000 cells were produced per year attributing to rapid extension of telegraph and telephone networks throughout Europe. However, a long stage of development was needed to create the dry cell type of today. Replacing the zinc rod by a zinc sheet formed into a cup and finally by a zinc can that also served as cell container, made the cell unbreakable. Powdered coke was replaced

780

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

Figure 1 First Leclanché wet cell. Reproduced with permission from Huber R (1974) Leclanché batteries. In: Kordesch K (ed.) Batteries, pp. 1–239. New York: Marcel Dekker, Inc.

by graphite that was mixed with natural manganese dioxide (NMD), pressed into cylindrical form, and wrapped in paper or gauze to prevent disintegration. A carbon rod in the center served as positive terminal. This construction is well known as ‘bobbin’ electrode. Figure 2 illustrates the oldest form of the dry cell, the pasted cell, with its bobbin. The immobilization of the electrolyte was an essential step. The companies Hellesens in Denmark and also Leclanché in France produced immobilized cells around 1880–90. C. Gassner is generally considered to be the inventor of the first dry cell containing ammonium chloride solution thickened with plaster of paris and zinc oxide to form an electrolyte paste, in 1887. Moreover, for industrial production naturally occurring carbohydrates, such as flour and starch, were used as gelatinizing agents. When graphite was completely replaced by highly absorbent and conductive acetylene black in the 1930s, the wrapping of the bobbin became redundant and the achievable service life doubled. Other significant developments were related to cell design. The invention of portable radio receivers after World War I led to the flat cell design. For further improvement of the cylindrical construction, the paper-lined cell was built with a separator carrying a thin layer of paste. Another design variation, the sectored cell, was made of four sectors each containing an individual paper-lined cell that was reformed by pressing into sector shapes. This assembly was inserted into a common cylindrical zinc–carbon cell and the 7 8 6 2 1

2

5

1

3 4 (a)

(b)

Figure 2 Pasted cell. (a) Sectional drawing: (1) cathode, (2) carbon rod, (3) zinc can, (4) cardboard or plastic disk, (5) immobilized electrolyte, (6) washer, (7) brass cap, and (8) bitumen seal. (b) In tissue-wrapped bobbin: (1) paper or gauze and (2) linen thread. Reproduced with permission from Huber R (1974) Leclanché batteries. In: Kordesch K (ed.) Batteries pp. 1–239. New York: Marcel Dekker, Inc.

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

781

sectors were connected in parallel. In the inside-out cell type, the zinc can was replaced by a leak-proof graphite container and the zinc anode, in the shape of vanes, was centrally located in the cell. Inside-out cells and sector cells gave improved performance on heavy discharge applications, thereby showing minor tendency of leaking. Further construction details of a few designs of zinc–carbon cells are described later. Since 1960, much effort has been directed toward developing the zinc chloride cell system by adding zinc chloride to the ammonium chloride electrolyte. Thus, improved performance on heavy-drain applications was provided. From the 1980s onward, the development has been focused on environmental issues, such as the elimination of mercury, cadmium, and other heavy metals from the zinc–carbon dry cell. Toward the end of the twentieth century, the zinc–carbon dry cell was produced in many billions per year, but now it has reached a plateau due to the competition with the more powerful alkaline primary batteries, which are used at a level of 25 billions, 80% of it in the United States, Japan, and Western Europe. In Asia, emerging third world, and Eastern European countries, the relatively inexpensive zinc–carbon batteries are still predominant. Alkaline manganese zinc batteries are described in Primary Batteries – Aqueous Systems: Alkaline Manganese–Zinc. The history of primary cells and the principles of primary cells are given in History: Primary Batteries and Primary Batteries: Overview.

Basic Constructions of Zinc–Carbon Batteries Zinc–carbon batteries are produced in two general configurations: cylindrical and flat. Cylindrical cells are either available as unit cells or are assembled in multicell batteries. The flat cell construction is built only as a multicell battery. In both constructions, the chemical ingredients used are similar.

Cylindrical Cells In the cylindrical Leclanché cell, as shown in Figure 3, the zinc can is simultaneously used as anode and cell container. The manganese dioxide is mixed with acetylene black, soaked with electrolyte, and built as a bobbin. The carbon rod of the bobbin serves as current collector of the manganese dioxide mix and provides structural strength. Owing to its porosity, the carbon rod allows accumulated gas to escape but circumvents electrolyte leakage. The cathode and the anode are separated by a paste separator or by a coated Kraft paper (with starch or polymer) in the paper-lined cell. The separator permits ions to pass through and prevents electrical short circuit between the electrodes. The advantage of the paper-lined type is a thinner separator spacing leading to lower internal resistance and supplemental volume (about 10%) for active cell components. Single cells are covered with metal, cardboard, plastic, or paper jackets for visual appearance and to minimize electrolyte leakage through the cell container. The basic construction of the cylindrical zinc chloride cell, illustrated in Figure 4, differs from that of the typical Leclanché cell by means of a resealable venting seal. In this cell type, the carbon rod is sealed with wax to restrict venting only to the seal path. The One-piece metal cover (+)

Top washer Wax ring seal Support washer Asphalt seal Anode (zinc can) Carbon electrode Paste separator flour, starch, electrolyte

Cathode mix – manganese dioxide, carbon, electrolyte Air space Zinc can Kraft Label

Jacket-labeled polyethylene bonded tube Metal bottom cover (−)

Plastic film Paste separator Cup and star bottom

Figure 3 Construction details of cylindrical Leclanché cell with paste separator and asphalt seal. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill.

782

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

Plastic closure

One-piece metal cover (+)

Sealant Cathode mix – manganese dioxide, carbon, electrolyte

Carbon electrode Separator coated paper

Zinc can

Anode (zinc can)

Kraft Label

Jacket-labeled polyethylene bonded tube

Plastic film Separator

Metal bottom cover (−)

Cup and star bottom

Figure 4 Construction details of cylindrical zinc chloride cell with coated paper separator and plastic seal. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill.

One-piece metal cover (+) Conductive lining

Paper jacket Cathode mix Separator Zinc vane Wax inner seal Washer Metal bottom cover (−)

Figure 5 Construction details of cylindrical Leclanché inside-out cell. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill.

sealing system allows gaseous corrosion products to escape (e.g., hydrogen from zinc corrosion) but largely excludes oxygen from outside during shelf storage. Furthermore, it protects the cell against drying out. In the inside-out cylindrical cell type of Figure 5, the cell container is formed from graphite instead of zinc. The cathode mix is pressed directly into an impact-molded graphite container. The negative electrode, consisting of two zinc vanes forming a cross, is enclosed in a laminate separator and located at the center of the cylindrical cell. Although this construction yields more efficient zinc utilization and less leakage, it has not been manufactured since the late 1960s.

Flat Cells The flat cell construction, shown in Figure 6, increases the available volume in comparison to the cylindrical version, as the cathode mix contains no voids (expansion chamber) or carbon rod. Thus, the energy density is remarkably enlarged. In this flat cell geometry, as illustrated in the sectional view of the unit cell in Figure 6, the zinc plate is coated with conductive carbon (paint or plastic film) providing electrical contact. In addition, the coating isolates the zinc from the cathode of the adjacent cell unit functioning as cathode collector as well. A plastic envelope confines the active cell materials. In multicell battery assemblies (e.g., 9 V battery, ‘Eveready’ 216), depicted in Figure 6, the rectangular form of the flat cell also reduces wasted space (no voids between the cells) and increases the volumetric density nearly twice compared to cylindrical cell

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

783

Positive contact

Negative contact Lithographed steel jacket Sectional view of unit cell

Wax coating Connector strip Wax coating

Plastic envelope Cathode mix – manganese dioxide, carbon, electrolyte Separator Anode (zinc) Carbon conductive coating

Figure 6 Construction details of Leclanché flat cell and 9 V battery (e.g., Eveready 216). Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill.

assemblies. The assembled battery is attached to battery terminals via metal contact strips. The final assembly is usually encapsulated in wax, shrink film, or plastic to avoid leakage. In the early 1970s, Polaroid introduced a new instant camera-film system, the SX-70, which included a battery in the film pack rather than in the camera. The innovation was that the battery was changed with each new film and thus, the photographer would not have to be concerned with the freshness of the power supply. This flat-pack zinc–manganese dioxide battery (Polaroid P-80 battery, 6 V) can be regarded as special design of the zinc–carbon system. For the construction of a single cell (1.5 V), as illustrated in Figure 7, the anode zinc is coated on a conductive vinyl web. The manganese dioxide is mixed in a slurry containing electrolyte salts. Zinc chloride with some ammonium chloride is the main component of the electrolyte. The anode and the cathode are separated by a thin film of cellophane. In the P-80 battery, the four cells are connected by vinyl frames to each other and aluminum collector plates. The thin layers have to stay in close contact to maintain the low resistance of the flat geometry and gassing has to be minimized. Other key parameters of the flat battery are quite similar to those of cylindrical cells.

Conductive cathode collector Cathode (MnO2, carbon, electrolyte) Separator (cellophane) Anode (zinc)

Conductive anode collector (+) Separator (−)

Figure 7 Exploded view of a single cell from Polaroid P-80 battery pack. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill.

784

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

Battery Materials Manganese Dioxide The type of manganese dioxide (MnO2) and its quantity used in dry cells are mainly responsible for cell capacity. Performance characteristics depend on individual crystal structure, varying degrees of hydration, and the activity of the manganese dioxide. Manganese dioxide potentials are additionally affected by the pH of the electrolyte. Most zinc–carbon batteries are cathode limited. Four different types of manganese dioxide are applied in dry cells: (NMD alpha- and beta-structure), activated manganese dioxide (AMD), chemically synthesized manganese dioxide (CMD, delta-structure), and electrolytic manganese dioxide (EMD, gamma-structure). Activated manganese dioxide is prepared by chemical treatment of natural ore (roasting, sulfuric acid treatment). The CMD type may be obtained either as a by-product of an oxidation process with potassium permanganate or by thermal decomposition of other manganese compounds followed by oxidation. Electrolytic manganese dioxide is obtained by anodic oxidation of manganese sulfate in hot sulfuric acid. Although more expensive, EMD has the advantage of yielding higher cell capacity with improved rate capability (for heavy or industrial applications) and its polarization is significantly lower compared to other types.

Zinc The anodic material used in Leclanché or zinc–carbon dry cells is an alloy based on zinc that usually contains some lead and cadmium. Modern forming techniques have diminished the amount of the alloying compounds in zinc cans (0.03–0.06% cadmium, 0.2–0.4% lead). Although lead improves the forming quality, cadmium stiffens the alloy. Moreover, both materials act as corrosion inhibitor as well. For environmental reasons, manganese is a suitable stiffening substitute for cadmium, however, without corrosion resistance. Of the three different ways of producing the zinc can – soldered zinc, deep drawn, and impact extruded – the latter is now the method of choice. Metallic impurities such as iron, copper, nickel, and cobalt have to be avoided so as not to provoke corrosive reactions particularly in ‘zero’ mercury cells. The chemistry and electrochemistry of zinc is covered in Chemistry, Electrochemistry, and Electrochemical Applications: Zinc.

Corrosion Inhibitors Corrosion inhibitors are usually added directly to the electrolyte or during the coating procedure of the paper separator to protect the zinc against corrosion. In the past, corrosion of the zinc alloy was reduced by amalgamation, chromate treatment of the surface, or by addition of organic inhibitors. Environmental concerns have generally eliminated mercury and cadmium as alloying compound. However, these restrictions are crucial for zinc chloride cells in the lower pH region due to excessive hydrogen gas formation. Alternative substitutes are gallium, indium, lead, tin, and bismuth, either as zinc alloy or as soluble salt, added to the electrolyte. Organic inhibitors, such as glycols and silicates, can also be used.

Carbon Black Manganese dioxide has only light semiconductor properties. Therefore, chemically inert graphite or acetylene black is added to the cathode to improve its electronic conductivity. Another important property of carbon is its absorptive behavior toward electrolyte and it provides compressibility and elasticity during processing of the cathode mix. Nowadays, acetylene black is mainly used as a conductive additive as it can keep more electrolyte, an important property for zinc chloride cells. Graphite is used only to some extent. Cells containing acetylene black usually deliver excellent intermittent service, whereas cells with graphite operate well for high flash currents or continuous drains.

Electrolytes The Leclanché cell contains an aqueous solution of ammonium chloride mixed with zinc chloride as typical electrolyte formulation (26 wt% ammonium chloride, 8.8 wt% zinc chloride). The addition of zinc chloride prevents formation of ammonium chloride crusts and influences the gelling rate of the paste. Zinc chloride cells typically comprise only zinc chloride electrolyte (15–40 wt%) but may also contain some ammonium chloride for high rate performance. The amount of zinc corrosion inhibitor added to the electrolyte is generally low (Leclanché cell: 0.25–1 wt%, zinc chloride cell: 0.02–1 wt%).

Separators In Leclanché and zinc chloride cells, two different separator types are used, either the gelled paste or the paper-lined type. For the first kind, the paste containing starch and/or flour is directly dispensed into the zinc can. During insertion of the bobbin the paste is pushed up the can walls. Gelatinization of the paste is carried out at room temperature or at elevated temperature (60–96
C) depending on paste formulation. The paper-lined type comprises a special paper (Kraft paper), which is single- or double-sided, coated with cereal or methylcellulose as gelling agent, formed into a cylinder, and inserted into the zinc can. Subsequently, the

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

785

manganese dioxide mixture or a complete bobbin is placed inside the zinc can. Inserting of the carbon rod compresses the bobbin, releases some electrolyte from the cathode mixture, and thus wets the coated paper. The paper-lined version is also applied for flat cells. In relation to capacity, the paper-lined type is superior to the paste type (thick layer of paste), as it contains up to 10% or even more manganese dioxide.

Chemistry of Zinc–Carbon Systems Zinc–carbon batteries contain manganese dioxide as cathode material that is mixed with carbon (acetylene black) to improve conductivity and to absorb the electrolyte. The anode consists of a metallic zinc can and an aqueous solution of ammonium chloride and/or zinc chloride as electrolyte. The simplified cell reaction can be written as: Zn + 2MnO2 ! ZnO + Mn2 O3

[I]

Based on this reaction, the theoretical specific capacity is 224 Ah kg−1. The highest specific capacity under ideal conditions can be 96 Ah kg−1 including all components in the calculation. The actual specific capacity in practice ranges from 75 Ah kg−1 (very light discharge) to 35 Ah kg−1 (heavy-duty, intermittent discharge). Despite the long history of the Leclanché cell, the chemistry of zinc–carbon systems is actually more complex than that of many other batteries. The efficiency of the chemical reaction depends on electrolyte concentration, cell geometry, discharge rate and temperature, depth of discharge, diffusion rate, and on the manganese dioxide type. A more comprehensive description of the chemistry of zinc–carbon systems can be summarized as follows: Leclanché batteries (ammonium chloride electrolyte): Zn + 2MnO2 + 2NH4 Cl ! 2MnOOH + ZnðNH3 Þ2 Cl2 ðlight dischargeÞ

[II]

Zn + 2MnO2 + NH4 Cl + H2 O ! 2MnOOH + NH3 + ZnðOHÞCl ðheavy dischargeÞ

[III]

Zn + 6MnOOH ! 2Mn3 O4 + ZnO + 3H2 O ðprolonged dischargeÞ

[IV]

Zinc chloride batteries (zinc chloride electrolyte): Zn + 2MnO2 + 2H2 O + ZnCl2 ! 2MnOOH + 2ZnðOHÞCl ðlight=heavy dischargeÞ

[V]

4Zn + 8MnO2 + 9H2 O + ZnCl2 ! 8MnOOH + ZnCl2  4ZnO  5H2 O

[VI]

Zn + 6MnOOH + 2ZnðOHÞCl ! 2Mn3 O4 + ZnCl2  2ZnO  4H2 O ðprolonged dischargeÞ

[VII]

or

Moreover, it has to be mentioned that prolonged discharge of manganite (MnOOH) does not contribute further to useful operating voltage for practical applications.

Types of Zinc–Carbon Cells Zinc–carbon batteries are produced in a variety of different shapes, sizes, and capacities for numerous applications. Single-cell and multicell batteries are classified by cell type, either Leclanché or zinc chloride, and by battery grade (general purpose (GP), industrial heavy duty (HD), and extra/super heavy duty (EHD)) as described in the following sections.

Leclanché Cells The typical Leclanché cell with starch paste separator is nearly completely displaced by the more powerful zinc chloride cell exhibiting a low leakage characteristic. Nevertheless, in developing countries there is still need for this least expensive aqueous battery system. The GP type, containing ammonium chloride electrolyte with some zinc chloride and NMD as cathode material, is mainly applied for intermittent low-rate discharge. Another kind of Leclanché cell is the industrial HD type applied for intermittent medium- to heavy-rate discharge. This battery grade comprises chemically or electrochemically synthesized manganese dioxide alone or is being mixed with NMD. The separator can be made of starch paste or it can be of the paper-lined version.

Zinc Chloride Cells The low-priced zinc chloride cell of the GP type is used for intermittent and continuous low-rate discharge performance. The electrolyte consists of zinc chloride optionally including small amounts of ammonium chloride. Natural manganese dioxide is used as cathode. The zinc chloride cell of the industrial HD type is applied for low to intermediate continuous and intermittent heavy-rate

786

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

discharge performance. Natural manganese dioxide together with electrolytic manganese dioxide builds the cathode and a coated paper separator with cross-linked or modified starch is employed. The electrolyte composition is the same as for the GP grade. The EHD type is the premium grade of the zinc chloride cell recommended for medium and heavy continuous and for heavy intermittent discharge but with higher price compared to other zinc chloride types. The electrolyte mainly consists of zinc chloride with perhaps a small amount of ammonium chloride. This premium zinc chloride cell type containing electrolytic manganese dioxide, paper-coated separators, and special electrolyte compositions attains enhanced low-temperature characteristics and diminished electrolyte leakage.

Cell Performances Discharge Characteristics The discharge characteristics of zinc–carbon cells show a sloping profile (single- or double-S curve). This behavior can be attributed to increasing cell resistance and polarization of the active materials during discharge procedure. Depending on application requirements, typical end or cutoff voltages (COV) of the 1.5 V zinc–carbon cell are 1.2 V (for electronic devices), 0.9 V (for flashlights), and 0.75 V or below (for radios). The performance of zinc–carbon cells most notably depends on discharge mode, such as continuous or intermittent, and on current drain. Cell performance is also sensitive to the operating temperature and storage conditions. During continuous discharge, the cell is discharged without interruption until a chosen COV is reached. In accordance with the application test, an intermittent performance stops discharging and repeats the discharge procedure after a specified rest period. Intermittent testing can also simulate practical applications (e.g., use of a torch) and is performed until the voltage drops below COV. Table 1 lists typical performance of the more popular battery sizes at various loads under a 2 h day−1 intermittent discharge. Data for multicell batteries are not included as they are beyond the scope of this contribution and available in various manufacturersʼ catalogs or via Internet. The AAA batteries, the smallest size of the zinc–carbon system, are typically used in remote control and other electronic devices. The AA-size batteries are applied in penlights, photoflash, and electronic applications. The classical C- and D-size batteries are mainly used in flashlights. Figure 8 shows a three-dimensional view of the general effects of intermittent schedule and discharge current on the capacity of a GP D-size zinc–carbon cell to 0.9 V COV. On extremely low current drains the benefit of intermittent discharge is minimal for both systems as probably the reaction rate proceeds more slowly than the diffusion rate. However, it must be pointed out that under extremely low-rate discharge conditions, other factors, such as aging, reduce capacity. Most applications work in the moderate (radio) to high rate range (flashlight) and for those, intermittent usage can even triplicate delivered capacity. The performance of cells of different sizes (AAA, AA, C, D), continuously discharged at a relatively high resistance of 150 O, is illustrated in Figure 9. With increasing cell sizes, the amount of active material and the electrode surface area increases, thus maintaining the voltage of larger cell sizes at a higher level. For the larger cell sizes C and D, the performance of intermittent and continuous discharge is similar at low current (about 10 mA). Some benefit could be gained for smaller sizes of Leclanché cells (AAA, AA) by using intermittent discharge. The smaller the zinc chloride cells, the lesser the difference relative to continuous discharge. Characteristics of zinc–carbon cells

Table 1

Weight (g)

Diameterb (mm)

Heightb (mm)

Leclanché current (mA)

Leclanchécapacity (mAh)

Zinc chloridecurrent (mA)

Zinc chloridecapacity (mAh)

R03

8.5

10.5

44.5

AA

R6

15

14.5

50.5

C

R14

41

26.2

50

D

R20

90

34.2

61.5

– – – 1 10 100 300 5 25 100 300 10 50 100 500

– – – 950 800 400 180 1900 1875 600 510 4000 3500 2500 1500

1 10 20 1 10 100 300 5 20 100 300 10 50 100 500

520 550 520 1200 1100 800 300 4000 3000 2000 1650 7000 6750 5500 3000

Size

IEC

AAA

a

Typical values for 2 h day−1 service to 0.9 V cutoff. a IEC, International Electrotechnical Commission; nomenclature for typical primary round cells. b Maximum dimensions.

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

787

6 5

Capacity (Ah)

4 4 3 2 1 10 20

Ty

pic

al

24

50

dis

100

ch

arg

ec

300

urr

2

4

8

en

t (m

A)

ty Du

−1 )

(h ule

day

ed

sch

Figure 8 Battery performance (capacity to 0.9 V) as a function of discharge current and operating schedule for a general-purpose D-size zinc–carbon cell at 20
C. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill.

1.8

1.6

Voltage (V)

1.4

1.2

AAA size

1 AA size

C size

D size

0.8

0.6 0

100

200

300

400

500

600

700

800

900

1000

Hours of service

Figure 9 Continuous discharge of zinc–carbon cells through 150 O at 20
C. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill.

In Figure 10, the discharge curves of Leclanché and zinc chloride D-size cells of different battery grades, such as GP, HD, and EHD, as described in the previous section, are displayed at 2.2 O continuous discharge. The GP zinc chloride cell exceeds the GP Leclanché cell about 30% to 0.9 V COV. Under these conditions, HD zinc chloride cells outperform Leclanché cells of the same grade by even about 50%. The best result for this application can be obtained with the EHD zinc chloride cell that provides 4.4 times more duration of service in comparison to the GP Leclanché cell. Furthermore, continuous discharge of various battery grades at the lighter 3.9 O results in prolonged service time and minor difference between the grades. The light intermittent flashlight LIF test of various battery grades at 2.2 O, illustrated in Figure 11, yields increased performance of all batteries and less difference between individual grades compared to continuous discharge at the same load. A moderate discharge at 24 O for 4 h day−1, simulating a transistor radio and an electronic equipment battery test, is shown in Figure 12. This application mode implicates tighter grouping of all grades and increased service time. In general, continuous discharge and higher current increases the performance difference between various grades of the same size. Intermittent discharge tends to reduce the differences between grades and systems. This is based on chemical recuperation during the rest period and on redistribution of reaction products. Moreover, a zinc chloride cell can support heavier drains much better due to its improved transport mechanism.

788

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

1.8

1.6

Voltage (V)

1.4 GP Leclanché 1.2

GP zinc chloride HD Leclanché

1

HD zinc chloride EHD zinc chloride

0.8 0.6 100

0

200

300

400

600

500

700

800

Minutes of service

Figure 10 Comparison of Leclanché and zinc chloride D-size batteries of various grades. Cells were continuously discharged through 2.2 O at 20
C. GP, general purpose; HD, heavy duty; EHD, extra heavy duty. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill. 1.8

1.6

Voltage (V)

1.4

1.2 HD Leclanché HD zinc chloride EHD zinc chloride

1

0.8

GP Leclanché GP zinc chloride

0.6 0

100

200

300

400

500

600

700

800

Minutes of service

Figure 11 Comparison of Leclanché and zinc chloride D-size batteries of various grades. Cells were discharged on the light intermittent flashlight (LIF) test (4 min h−1, 8 h day−1) through 2.2 O at 20
C. GP, general purpose; HD, heavy duty; EHD, extra heavy duty. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill. 1.8

Voltage (V)

1.6 HD Leclanché

1.4

HD zinc chloride EHD zinc chloride

1.2 1

GP zinc chloride

0.8

GP Leclanché

0.6 0

20

40

60

80

100

120

140

160

Hours of service

Figure 12 Comparison of Leclanché and zinc chloride D-size batteries of various grades. Cells were discharged through 24 O for 4 h day−1 at 20
C. GP, general purpose; HD, heavy duty; EHD, extra heavy duty. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill.

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

789

Temperature Sensitivity Zinc–carbon cells are normally designed to operate best in a temperature range of 20–30
C. The higher the battery temperature during discharge, the greater the energy output. However, prolonged exposure to high temperatures of 50
C and more will cause very rapid deterioration of the battery. Otherwise, the service time at low temperatures is reduced due to diminished chemical activity in the cell. Furthermore, these effects are more pronounced for heavy current drains. The capacity of the Leclanché cell drops down very rapidly with decreasing temperatures, yielding only about 65% of its original capacity at 0
C, as illustrated in Figure 13. Below –20
C, this battery is usually inoperative unless special low-temperature electrolytes are being used. Proper insulation of the Leclanché battery may be helpful at low ambient temperatures. Zinc chloride batteries exhibit a better low-temperature performance still providing 80% of the room temperature capacity at 0
C. However, at –20
C typical zinc chloride electrolytes become slushy, and below –25
C ice formation occurs making the cell inoperable.

Shelf Life The storage quality of zinc–carbon batteries mainly depends on charge condition. Hence, the shelf life of unused cells is better than that of partially discharged ones. Capacity degradation results from parasitic reactions like zinc corrosion, chemical side reactions, and moisture loss. In addition, shelf life is strongly affected by storage temperature. High temperatures accelerate capacity loss and low storage temperatures diminish it. The capacity retention of zinc chloride batteries comprising coated paper separator and improved sealing with molded polypropylene or polyethylene is far better than that of Leclanché cells. The capacity loss of a zinc chloride battery after storage at 0, 20, and 40
C is illustrated in Figure 14. By means of refrigerated storage the shelf life can be further increased. Cooled storage at –20
C retains 80–90% of the initial capacity after 10 years. Nevertheless, there are some aspects to bear in mind for storing zinc–carbon batteries at low temperatures. Repeated cycling from high to low temperatures has to be avoided. Refrigerated cells need some time for room temperature equalization before usage, and moisture condensation has to be eliminated. Furthermore, case materials and seals must have similar expansion coefficients to avoid the problem of cracking.

Rechargeability There were also some efforts of recharging zinc–carbon dry cells in the past. Several experiments were carried out to reactivate partly discharged cells. One of the most important aspects was to avoid excessive charging voltages (beyond 1.7 V) leading to decomposition of the electrolyte. The charging procedure was carried out very slowly, preferably with direct current comprising an alternating current component. Completely or heavily discharged cells could not be recharged. Successful charging techniques were attained for the rechargeable alkaline manganese dioxide–zinc system that was developed at Graz University of Technology and at Battery Technology Inc., Canada. This alkaline rechargeable battery system that is commercially produced since 1993 and licensed in the United States and other countries worldwide has the capability to replace 100 throw-away batteries.

120

Available capacity (%)

100

Zinc chloride

80 Leclanché 60 40 20 0 −20

−10

0

10

20

30

40

Temperature (°C)

Figure 13 Effect of temperature on available capacity of zinc–carbon D-size batteries (general-purpose Leclanché and heavy-duty zinc chloride systems) for moderate-drain radio-type discharge. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill.

790

Batteries – Battery Types – Zinc Batteries | Zinc–Carbon

110

Capacity retained (%)

100 0 °C

90

80 20 °C 70 40 °C

60 50 0

1

2

3

4

5

Shelf age (years)

Figure 14 Capacity retention of paper-lined zinc chloride batteries with plastic seal at 40, 20, and 0
C. Reproduced with permission from McComsey DW (2002) Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In: Linden D and Reddy TB (eds.) Handbook of Batteries, 3rd edn., pp. 8.1–8.45. New York: McGraw-Hill.

Conclusions The Leclanché and the zinc chloride systems comprise inexpensive manganese dioxide and zinc as battery materials. These batteries are produced in different cell geometries and types that can be used for a large variety of applications. On the basis of the long historical tradition and reliability of zinc–carbon batteries going back to the 1860s, they are still among the most widely used primary batteries worldwide being economically very important for emerging third world countries.

Further Reading 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23.

Bregazzi, M. Acetylene black. Electrochemical Technology 1967, 5, 507–513. Brodd, R.J.; Kozawa, A.; Kordesch, K.V. 75th anniversary review series, primary batteries 1951–1976. Journal of the Electrochemical Society 1978, 125, 271C–283C. Cahoon NC and Heise GW (eds.) (1976) The Primary Battery, ch. 1, vol. 2, pp. 8–106. New York: John Wiley & Sons. Crompton, T.R. In Small Batteries, Primary Cells John Wiley & Sons: New York, 1983; vol. 2, pp 17–41. Crompton, T.R. Battery Reference Book. Butterworths: London, 1990; pp 7.1–7.4. Dentch, M.; Hillier, A. Polaroid’s manganese dioxide–zinc battery without mercury. Progress in Batteries & Solar Cells 1990, 9, 120–130. Ding, Y.; Wu, D.; Zhu, J. A study of high temperature shelf life tests of dry batteries. Progress in Batteries & Solar Cells 1991, 10, 161–164. Everatt, P.R.; Tye, F.L. An analysis of the behavior of Leclanché batteries on simulated application tests. Part I, R20 cells containing electrodeposited manganese dioxide. Progress in Batteries & Solar Cells 1988, 9, 128–132. Heise, G.W.; Cahoon, N.C. Dry cells of the Leclanché type, 1902–1952 – A review. Electrochemical Society, 1902-Fiftieth Anniversary-1952, Primary Battery Issue Journal of the Electrochemical Society 1952, 99, 179C–187C. Huber R (1972) Trockenbatterien (3.Auflage). Varta Fachbuchreihe. Düsseldorf: VDI Verlag GmbH. Huber, R. Leclanché batteries. In Batteries; Kordesch, K., Ed.; Marcel Dekker, Inc: New York, 1974; pp 1–239. Kordesch, K.; Daniel-Ivad, J. Rechargeable zinc/alkaline/manganese dioxide batteries. In Handbook of Batteries, 3rd edn.; Linden, D., Reddy, T.B., Eds.; McGraw-Hill: New York, 2002; pp 36.1–36.18. Kozawa, A.; Powers, R.A. Electrochemical reactions in batteries – Emphasizing the MnO2 cathode of dry cells. Journal of Chemical Education 1972, 49, 587–591. McComsey, D.W. Zinc–carbon batteries (Leclanché and zinc chloride cell systems). In Handbook of Batteries, 3rd edn.; Linden, D., Reddy, T.B., Eds.; McGraw-Hill: New York, 2002; pp 8.1–8.45. Miyazaki, K.; Kagawa, K. New alloy composition for zinc can for carbon–zinc dry cells. Progress in Batteries & Solar Cells 1987, 6, 110–112. Schumm, B. Models for diffusion studies in carbon–zinc cell cathodes. Progress in Batteries & Solar Cells 1990, 9, 150–157. Schumm, B. Behavior of carbon–zinc cells, paste or paper-lined in tropical tests. Progress in Batteries & Solar Cells 1991, 10, 76–80. Schumm, B. Commercial non-rechargeable battery systems. In Modern Battery Technology; Tuck, C.D.S., Ed.; Ellis Horwood Limited: New York, 1991; pp 87–111. Schumm B and Dereska JS (1988) Long shelf life in Leclanché R32 cells. Proceedings of 33rd International Power Sources Symposium, 13–16 June 1988, pp. 728–734. New Jersey: The Electrochemical Society. Uetani, Y. Recently developed ZnCl2 cells in Japan. Progress in Batteries & Solar Cells 1978, 1, 54–58. Uetani, Y.; Sasama, H.; Iwamaru, T. Reactions in zinc chloride cells compared to Leclanché cells. Proceedings of the Symposium on Advances in Battery Materials and Processes 1985, (4), 475–483. New Jersey: The Electrochemical Society. Vinal, G. Primary Batteries. John Wiley & Sons: New York, 1950. Warburton DL (1963) Leclanché battery shelf life study. 17th Annual Power Sources Conference, 21–23 May 1963, pp. 138–142.

Batteries – Battery Types – Zinc Batteries | Zinc–Manganese Gautam G Yadava and Josef Daniel Ivadb, aUrban Electric Power, Pearl River, NY, United States; bjdi energy consulting, Newmarket, ON, Canada © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is a update of J. Daniel-Ivad, SECONDARY BATTERIES – ZINC SYSTEMS | Zinc–Manganese, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 497–512, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00171-4.

1 Introduction 2 Brief history 3 Electrochemistry 3.1 Electrolyte 4 The MnO2 electrode 5 First electron or proton insertion battery 6 The second electron battery or the conversion battery 7 Separator 8 The Zn electrode 9 Future of Zn|MnO2 batteries 10 The advent of high voltage Zn|MnO2 batteries 11 Commercialization and application of Zn|MnO2 batteries 12 Conclusion Acknowledgments References

792 792 794 794 795 796 799 803 808 811 812 815 822 824 824

Abstract From a historical and commercial standpoint zinc|manganese dioxide (Zn|MnO2) batteries are one of the most important primary battery chemistries. Its development is tied to famous historical moments and spurred economic development of countries as well. From an electrochemical standpoint, it is probably the first battery system where an intercalation process was identified as the main capacity mechanism for its first electron reaction. Its rechargeability was studied in the 1970s and 1980s but problems related to its electrochemistry dented its promising rise and led to the dominance of lithium-ion batteries. In the 21st century, decarbonization has become an important issue in many countries and spurred ambitious goals of decarbonizing the electric grid by 2050. Integration of renewable sources of energy has also ignited development of energy storage systems to reduce the intermittency of these variable sources of energy. Zn|MnO2 batteries have been identified as promising energy storage system because of its low cost, safe and non-toxic raw materials, and raw material abundancy, thus precluding it from any supply chain issues plaguing the world at this time. This chapter covers historical background of this promising chemistry, its fundamental electrochemistry and the challenges that have plagued its rechargeability. In this chapter, we also cover the giant strides that have been made to making this battery rechargeable in the past decade and its reclassification into three different types depending on its electrochemistry. Commercialization of these batteries is also covered and its applications in promising use cases which it can serve as a 100% drop-in replacement for lead acid batteries and also challenge lithium-ion chemistry for the applications of the future.

Key points

• • • • • • • • •

Introduce the chemistry of zinc|manganese dioxide batteries from a historical viewpoint Delve deeply into the electrochemistry and materials science of zinc and manganese dioxide electrodes Expand on the electrochemical mechanisms of proton-insertion (one electron) and conversion-based second electron manganese dioxide electrode Present failure mechanisms seen in the zinc (Zn)|manganese dioxide (MnO2) battery and the solutions developed to address these failure mechanisms Introduce and discuss the scientific principles behind the future generations Zn|MnO2 battery, which are the conversion-based second electron battery and the high voltage (2.45–3 V) battery Discuss on large-scale manufacturing process of commercially available rechargeable Zn|MnO2 batteries Show commercially installed Zn|MnO2 systems and explain their use cases Introduce concept of gelled Zn|MnO2 batteries and show its application in solar microgrid use-case Present a perspective and justify the reclassification of Zn|MnO2 batteries into three types—Proton-insertion, Conversion-based and High Voltage

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00103-1

791

792

1

Batteries – Battery Types – Zinc Batteries | Zinc–Manganese

Introduction

This chapter will provide a detailed background, chemistry, and perspective of the current state of the art zinc (Zn)|manganese dioxide (MnO2) batteries. This chemistry has a long history with it still being primarily used for single discharge applications in home appliances. It has found a new renaissance as rechargeable batteries for use in grid storage applications. This renaissance has been driven by the need to find batteries with low cost and widely available raw materials. We will first start with briefly detailing the history behind this chemistry from its inception to various developments leading to its current form. We will then cover the electrochemistry of its raw materials, the chemistry’s theoretical energy density and the practical energy density achievable. Briefly, we will also detail the primary battery development and their electrochemical characteristics. Finally, we will report on the current state of the art and its various on-field demonstrations.

2

Brief history

The first Zn|MnO2 battery was serendipitously invented in around 1866 by Georges Leclanché.1 It was serendipitous because in Leclanché’s quest for making a high-performance zinc-air battery, he unintentionally discovered much improved cell performance when he used a cathode mixture of manganese dioxide and carbon that was partially soaked in electrolyte of ammonium chloride and mercury chloride. The manganese dioxide acted as the catalyst to give an air cathode that could discharge for longer periods of time. This air cathode was unintentionally created because of the cathode’s exposure to air which created a three-phase boundary of gas, liquid and solid. On completely wetting the cathode in electrolyte the battery was in essence the first Zn|MnO2 battery where the capacity or energy delivered was because of the electrochemical reactions of Zn and MnO2. This Zn|MnO2 battery at the time was the best performing primary battery as it outperformed other chemistries on power and discharge time performance. An interesting fact is that this battery was used to power the first telephones.1,2 However, its popularity waned as other chemistries were discovered that had some rechargeable characteristics with better power densities. Battery development progressed in the early 20th century because of the World Wars. The World Wars drove the U.S. government to look for cheaper batteries with power characteristics that could match or outperform the best batteries of the time. In terms of Zn|MnO2 development there were two critical inventions that led to its dominance in the next century—(1) synthesis of electrolytic MnO2 and (2) the replacement of the Leclanché electrolyte with an alkaline electrolyte.2 The invention of a new synthesis method was more a product of necessity than anything else as during Leclanché’s time the MnO2 was imported from Africa and Ukraine. These were called natural MnO2 and there was no control over the quality or grade of MnO2 that was procured. During the war time, there were concerns over trade disruption of the MnO2 supply which forced the mining companies in the U.S. to make use of their Mn ores and produce MnO2 for their own needs. This led to G.D. Van Arsdale and C.G. Maier to invent the electrosynthesis procedure to make MnO2 which is now called electrolytic MnO2 or EMD.1,2 In scientific literature this MnO2 is often denoted as g-MnO2. This EMD or g-MnO2 is now produced in large quantities for all the primary batteries sold in the world. In the late 1940s, there was push again to improve the discharge power characteristics of the battery. Before this period acidic or neutral aqueous electrolytes were primarily used. However, these electrolytes affected the shelf life and discharge duration was poor. To address these problems Lewis Urry, P.A. Marshall and Karl Kordesch at Union Carbide used alkaline electrolyte which they found further improved the discharge duration of EMD, reduce the internal resistance, eliminated leaks and also increased shelf life compared to the Leclanché battery.3 Karl Kordesch and other researchers further improved the power characteristics of the battery by using high surface area powdered Zn particles instead of Zn cans or Zn foils and using high surface area carbons that improved the conductivity of the EMD cathode. This newly developed Zn|MnO2 battery was used for many of the new applications of the time like flashlights, cameras, hearing aids, etc. In fact, it is the most popular and dominant primary battery (i.e., single discharge) till today. The modern primary battery has a volumetric energy density >420 Wh/L at a cost of $30–50/kWh depending on the brand.4,5 In the 1960s and 1970s, there was a push to reduce the waste generated by the disposal of primary batteries (Fig. 1). At the time, mercury additives were used in the Zn anodes to address some of its problems like self-corrosion. However, the disposal of the primary batteries in landfills and improper recycling led to mercury contamination. This led to research on finding alternative additives and making the Zn|MnO2 chemistry rechargeable. Karl Kordesch, Akiya Kozawa and other stalwarts were at the forefront of this research as well. Karl Kordesch and his team introduced the concept of rechargeability by limiting the depth of discharge (DOD) of both, Zn and MnO2.6 Rechargeability was an issue in this chemistry as trying to access the complete theoretical 2e capacity of Zn (820 mAh/g) and MnO2 (617 mAh/g) would lead to electrode and crystal structure degradation. The degradation mechanisms for each of these electrodes will be covered in the chemistry section. Suffice it to say that rechargeability was a major issue which Karl Kordesch proposed circumventing this problem by limiting the accessibility of their theoretical capacity and thus, improve cycle life. This, however, meant that energy density would be severely curtailed, and the cost of the battery would increase because of limited capacity utilization. Interestingly, this idea would be researched again in the late 2000s and 2010s, where a team of researchers led by Sanjoy Banerjee from City College of New York (CCNY) would find this concept useful for introduction of rechargeable of Zn|MnO2 for use in grid storage applications as low cost and safer alternative to incumbents lead acid and lithium-ion technology.7 This will be again covered in detail later in the chapter.

Fig. 1 Historical timeline of Zn|MnO2 batteries summarized my main events.

794

Batteries – Battery Types – Zinc Batteries | Zinc–Manganese

As research was being conducted in the 1970s to make Zn|MnO2 rechargeable, there was an oil crisis at the same time that further drove efforts to find alternative sources of energy. This included the need to find energy-dense batteries for use in electric vehicles which spurred research into lithium metal and lithium-ion batteries at companies like Exxon Mobil. However, Ford Motor’s had identified that if 70–80% of the 2e capacity of MnO2 could be accessed then there was a viability for use of Zn|MnO2 batteries for used in electric vehicles. Halina Wroblowa and her team at Ford had identified that EMD could not reversibly deliver the 2e capacity unless an additive was used with the cathode. They discovered that using lead oxide (PbO) or bismuth oxide (Bi2O3) intermixed with EMD and carbon could lead to higher utilization or DOD of MnO2 against a reference electrode.8,9 They showed that when MnO2 was mixed with either of these additives and cycled against a reference electrode instead of Zn anode the cathode would deliver >70–80% of the 2e capacity of MnO2 >1000 times at low areal capacity (1000 times.10 The Cu addition would reduce the charge transfer resistance for the main Mn reactions in alkaline electrolyte and reduce any side reactions that could form inactive materials. This cathode would also work with a Zn anode to deliver the complete 2e capacity for >500 times; however, dissolved Zn ions could still react to form a resistive material called chalcophanite that would reduce the energy delivered by the battery.12 The cycle life was limited in this battery because of a Zn short developing after 500 cycles. This Zn short was a result of Zn redistribution in the cell that would redeposit onto the cellophane separator. To address the resistive material formation and Zn shorting issue Gautam G. Yadav and his team including Jinchao Huang developed a calcium hydroxide separator, which would react with the dissolved Zn ions to form calcium zincate.12,13 The calcium zincate was found to be layered structure where dissolved Zn ions could intercalate within the calcium hydroxide interlayers. This calcium zincate separator was reversible where the dissolved Zn ions would deintercalated to plate Zn on charge and the separator would form calcium hydroxide again. This prevented any deleterious side reaction on the MnO2 cathode and also prevented shorts developing after 500 cycles. The best cycle life reported for a 2e Zn|d-MnO2 is >900 cycles by Yadav and his team. Aqueous Zn batteries deliver their capacities 3 V by altering the concentration of the acid and alkaline electrolyte.15 However, the development of this chemistry also entails the development of innovative separators that prevent the neutralization of acid and base electrolyte. The key historical moments reported above relate to the technological development of this promising chemistry. In the coming sections, the chemistry will be reported in detail with key historical figures being identified for their roles in elucidating the electrochemistry behind Zn and MnO2.

3 3.1

Electrochemistry Electrolyte

As noted in the history section, earlier Zn|MnO2 batteries used the Leclanché electrolyte which was a mixture of ammonium chloride and zinc chloride. Mercury chloride was also added to reduce the corrosion of the zinc anodes. In the early 1900s till 1950s electrolytes made of zinc sulfate and manganese sulfate were also tried. There were a number of studies on the electrochemical mechanism of MnO2 and Zn in these electrolytes led by Vosburgh. There were reports on use of alkaline electrolytes like sodium hydroxide for MnO2 reactions in 1882 by Leuchs and potassium hydroxide was tried by Yai in 1903.1 However, the use of alkaline electrolytes along with use of powdered Zn anodes made a big breakthrough in the 1950s, which was reported by Urry, Marshall and Kordesch at Union Carbide. Lately, research on new electrolytes has been reignited, where zinc and manganese triflates and waterin-salt type electrolytes have been tried.16,17 Organic electrolytes have also been tried with Zn anodes.18 In general, aqueous electrolytes are most popular because it is safe, low cost, easy to handle and has very high ionic conductivity. These aqueous electrolytes can be classified into three different types: (1) pH < 7 (acidic), (2) pH 7 (neutral) and (3) pH > 7 (alkaline). Many review papers have been recently published on acidic and neutral or sometimes called slightly acidic electrolytes.19

Batteries – Battery Types – Zinc Batteries | Zinc–Manganese

795

The reader is suggested to look at these reviews. This chapter will focus mostly on alkaline electrolyte. There will be some discussion on acid electrolyte at the end of the chapter but for use in dual electrolyte batteries. Alkaline electrolytes are popular in commercial batteries because they have much higher conductivities compared to acidic or neutral electrolytes. They are readily available as high-grade bulk chemicals with little impurities. They are highly soluble and have very low freezing temperatures. Both potassium hydroxide and sodium hydroxide have been used in literature and commercial batteries. Sodium hydroxide may be preferred because of its low cost compared to potassium hydroxide. However, potassium hydroxide has higher conductivities, can dissolve zinc ions easily and has a lower freezing point. The rest of the chapter will focus on use of potassium hydroxide in Zn|MnO2 batteries.

4

The MnO2 electrode

The MnO2 cathode electrochemical undergoes two electron reactions in aqueous electrolytes. For the EMD electrode each electron’s reaction mechanism is different—the first electron is a homogenous solid-state reaction while the second electron is a heterogenous reaction. The first electron reaction (308 mAh/g) is a proton (H+) insertion reaction and the second electron reaction (about 308 mAh/g) is a dissolution-precipitation reaction, where Mn ions dissolve and precipitate on the electrode. The combined 2 electron (e−) capacity of 617 mAh/g makes the MnO2 cathode very energy dense. The electrochemical mechanism of these two electrons was proposed by Kozawa and his coworkers in the 1960s, which has become widely adopted in alkaline battery literature.6,20,21 The schematic discharge process of the EMD cathode and its corresponding discharge curve is shown in Fig. 2a. First e− reaction is a proton intercalation process in the unique EMD crystal structure which is a mixture of ramsdellite and pyrolusite phases of MnO2.

Fig. 2 (a) This schematic shows the operation of an Zn|MnO2 battery. (b) The voltage curves show the first and second electron reactions of EMD, as well as the detrimental impact of zincate (formed from Zn reaction with OH−, black curve) on the flat voltage (red curve) range after 1 V versus Zinc, where energy is lost.

796

Batteries – Battery Types – Zinc Batteries | Zinc–Manganese

These are 1  1 and 1  2 tunneled structures which form into a disordered structure to make up the EMD crystal structure. As these protons intercalate into these tunneled structures the EMD lattice expands due to the reduction of Mn4+ to Mn3+ ions which eventually forms the groutite phase of a-MnOOH. The sigmoidal shaped discharge curve in Fig. 2b is an indication of this 1e− proton insertion process. The Mn3+ from the a-MnOOH is soluble in alkaline electrolyte, which leads to the 2nd e− reaction. In the 2nd e− reaction, the Mn3+ ions complex with hydroxyl (OH−) ions in the electrolyte to form Mn(OH)63−, which reacts with the 2nd e− to form Mn(OH)42−, which is a tetrahedral complex. This complex eventually precipitates to form Mn(OH)2 as shown in Fig. 2a. The flat shaped discharge curve in Fig. 2b is an indication of the 2nd e− dissolution-precipitation process. The primary Zn|MnO2 batteries that are sold in the commercial market utilize only the 1st e− range of the EMD reaction. The energy density of this 1e− discharge is 420 Wh/L. The reason for limiting to the 1e− capacity is because the end of this reaction is 0.9–1 V versus Zn. The capacity available below 0.8 V is non-useable as most electronics limit their voltage to this value. The 2nd electron reaction delivers its capacity 0.9 V versus Zn and technically it should be possible to use this capacity or energy in the primary batteries, thus significantly increasing the energy density of the battery. However, deleterious side reactions take place in the 2nd electron discharge in the presence of Zn anodes. This will be covered in detail in the Zn anode section. Briefly, dissolved Zn ions react with the dissolved Mn3+ ions to form a spinel compound which is resistive and prevents the 2nd electron reaction. This can be seen in Fig. 2b, where a MnO2 cathode discharged versus a nickel hydroxide counter electrode and its voltage monitored versus a mercury (Hg)|mercury oxide (HgO) reference electrode delivers the 2nd electron capacity as a flat potential. The Hg|HgO voltage is converted to a Zn potential and its around 0.9 V as shown in Fig. 2b (red line). In the presence of a Zn anode, the dissolved Zn ions react with the dissolved Mn ions to form a spinel compound called hetaerolite (ZnMn2O4), which is resistive and creates a sloping potential instead of a flat potential and reduces the energy delivered as shown in Fig. 2b (black curve). Thus, blocking these Zn ions from interacting with the MnO2 cathode is one of the important areas of research in alkaline Zn|MnO2 batteries. In terms of rechargeability of MnO2 cathodes, this has been area of much interest for over 70 years. Achieving the complete 2e− rechargeability has been the holy grail. However, this has been very difficult because of the dissolution-precipitation reactions of MnO2. When the EMD cathode undergoes the 1e− solid-state proton insertion reaction followed by the 2nd e− dissolutionprecipitation reaction to form pyrochroite or Mn(OH)2, there are side reactions taking place in these processes. For example, in the 1e− process, the proton insertion results in lattice expansion which leads to sometimes the breakdown of the crystal structure and formation of inactive spinel structures like hausmannite (Mn3O4) or sometimes Mn2O3. In the 2nd e− process, the octahedral [Mn(OH)63−] and tetrahedral [Mn(OH)42−] complexes react together to form spinel structures of hausmannite. When Zn anodes are used, the dissolved Zn ions called zincate [Zn(OH)42−] react with the Mn complexes to form hetaerolite (ZnMn2O4). During the charge process, the pyrochroite goes through an electrochemical dissolution process to form the Mn complexes again, which results in the formation of groutite.22,23 The groutite again goes through an electrochemical oxidation process to form birnessite or d-MnO2. The most stable form of MnO2 in alkaline electrolyte is the layered birnessite phase. The EMD is a defect structure which is formed only in acid electrolyte under strict conditions. Therefore, once the EMD is discharged to its complete two electron capacity this results in the loss of the defect tunneled structure and the formation of the layered structure. The birnessite on discharge undergoes a direct dissolution-precipitation process to form pyrochroite and on recharge it forms birnessite again. However, the dissolution-precipitation process results in side reactions as aforementioned which results in loss of capacity. In the presence of Zn anodes, this recharge process does not last more than 1 or 2 cycles. The entire process that is described in the preceding paragraph is nicely captured through a cyclic voltammetry (CV) experiment on a EMD or MnO2 cathode as shown in Fig. 3. In Fig. 3a, the first cycle CV curve is shown for a EMD cathode, where negative currents indicate the reduction reaction or a discharge while the positive current indicate the oxidation reaction or a charge. On discharge for a EMD cathode, there are two peaks seen where the first broader peak indicates the proton insertion reaction to form the groutite phase. This groutite phase electrochemically undergoes a dissolution-precipitation reaction as indicated by the narrow long peak to form the pyrochroite phase. This pyrochroite phase electrochemically undergoes oxidation processes on charge as shown by the two peaks to eventually form the birnessite phase. This birnessite phase in the second cycle (Fig. 3b) reduces directly through a dissolution-precipitation reaction to form pyrochroite. The oxidation process for the 2nd cycle is similar to the 1st cycle. On the 10th cycle, as shown in Fig. 3c, the cathode suffers from deleterious side reactions which results in reduced peak area (or capacity loss). The reduced peak area is because of the formation of inactive spinels called hausmannite. In the presence of a Zn anode, the spinel formed would be hetaerolite.

5

First electron or proton insertion battery

The complexity of these issues, especially the problems faced when entering the 2nd electron region, led to early researchers concentrating on only the rechargeability of the 1st electron capacity. Karl Kordesch and his team did much of the pioneering work on investigating the rechargeability of the 1st electron reaction. In one of their seminal papers, they described the dimensional changes that would take place in the EMD cathode during charge and discharge which would affect cycle life.6 In a cylindrical geometry, like the bobbin cell, the electrode expansion could be maintained by the cell wall cans which could mitigate the effect of electrode expansion. However, in a flat plate or prismatic geometry cell, if proper pressure was not maintained, this would lead to electrode material loss and thus rapid capacity loss. The expansion of the electrodes increased with increasing utilization or

Batteries – Battery Types – Zinc Batteries | Zinc–Manganese

797

Fig. 3 (a) Cyclic voltammetry (CV) curve of a EMD or MnO2 cathode in its 1st cycle. Negative current indicates reduction reactions or a discharge, while positive current indicates oxidation reactions or a charge. (b) Second cycle CV curve where the MnO2 starts from the birnessite phase. The second cycle is shown in dark black color while the first cycle is shaded to indicate the difference in electrochemical reactions. (c) Tenth cycle CV curve where the area under the peaks has considerably reduced compared to first and second cycles indicating the formation of inactive spinel phases. Without a Zn anode the spinel phase is hausmannite, while in the presence of Zn anode the spinel phase is hetaerolite. The data presented is the personal data of Gautam G. Yadav.

798

Batteries – Battery Types – Zinc Batteries | Zinc–Manganese

This line represents MnO2 cathodes cycled with excess Zn (low Zn utilization) from Kordesch’s Experiments. This line will change as Zn utilization is increased.

300 200 150

141

Cycle Life

100 50

44

30 20 15

13

17

10

6

5

linear, Depth of Discharge, % 15

25

20

30

35

40

45

50

55

60

Utilization or Depth of Discharge (%) of MnO2 1e capacity

Fig. 4 Cycle life versus utilization or DOD of MnO2 1e capacity where the Zn used is at low utilization. This plot is modified from Kordesch, K.; Gsellmann, J.; Peri, M.; Tomantschger, K.; Chemelli, R. Electrochim. Acta 1981, 26 (10), 1495–1504.

depth-of-discharge of the MnO2 electrode. They showed that maintaining pressure could limit this capacity fade. They also showed the dependance of the cycle number on the DOD of MnO2’s 1e capacity (Fig. 4). Interestingly, they found that the dependence was logarithmic, which was used by E. Voss and G. Huster in 1966 to describe the capacity fade in a lead acid battery.24 Voss and Huster had found that lead acid batteries lost a certain amount of capacity because of the loss of active material due to shedding, sulfation, etc. Kordesch and his team found the same reasons for MnO2 with increasing DOD of the 1e capacity. They achieved 160 cycles at 20%DOD and 40 cycles at 35%DOD of 1e capacity of MnO2. Kordesch and his team further improved on this performance in their other works and wrote on the factors affecting rechargeability of MnO2 and how it can be further improved in terms of retaining capacity for long cycle life. Many other stalwarts like Conway, Donne and others worked on the rechargeability of the MnO2 in the 1e region.25,26 They found additives like bismuth doping, barium hydroxide, titanium dioxide, strontium hydroxide, etc. to help retain capacity for longer cycle life. However, there was no report of a cycle life >200 cycles at >20%DOD of MnO2 1e capacity. Sanjoy Banerjee and his team researched into extending the cycle life of MnO2 cathodes in the 1e regime for grid storage applications in the 2010s. If the cycle life of the MnO2 cathodes could be improved then the cost of energy stored per cycle life ($/kWh/cycle life) would be the lowest among other battery chemistries which would make it ideal for grid storage. They showed that altering the electrode structure and using novel graphites they could achieve very long cycle life (see Fig. 5). At 10% DOD they

(b)

new electrode

(a)

electrode after 3004 cycles

(b)

(a)

Discharge End Voltage (V)

1.5 1.4 1.3 1.2 TM3000_1027

1.1

(1) (2) (3) (6)

1 0.9 0.8 0

1000

2000 3000 Cycle Number

4000

2013/03/30 12:35 HL D4.7

100 um TM3000_2790

2014/09/10 15:40 H

electrode after 3004 cycles EDS mapping for Zn

100 um

adherent film

(c)

5000

Fig. 5 (a) Fade in end of discharge voltage versus cycle life in limited DOD cycled Zn|MnO2 cells. (b) Characterization of the failed cells in the limited DOD cycled experiments where the failure was identified because of spinel formation on the cathode surface. Plots modified from Ingale, N. D.; Gallaway, J. W.; Nyce, M.; Couzis, A.; Banerjee, S. J. Power Sources 2015, 276, 7–18.

Batteries – Battery Types – Zinc Batteries | Zinc–Manganese

799

achieved 3000 cycle life and at 20% DOD they achieved 1000 cycle life against a Zn anode (Fig. 5a), which became the highest reported cycle life for proton insertion battery.7 In Fig. 5a, the end of discharge voltage against a Zn anode is shown to represent degradation in cycle life. As limited DOD is being accessed the best representation of data for failure of electrodes is a degradation of the end of discharge voltage versus cycle life rather than capacity fade as capacity fade does not reveal any valuable or hidden information about the cell or the electrode performance. This degradation of end of discharge voltage with cycle life or time can mean many things like spinel formation, active material shedding from the electrode, carbon oxidation, zinc failure, etc. In Fig. 5a, the authors cycled the MnO2 cathode against a very low utilization of Zn anode capacity (1 order of magnitude lower than organic electrolytes. This difference becomes more significant at low temperatures. Therefore, a solid electrolyte battery is not suitable for high-output-power uses due to its high internal resistance. With regard to fabrication of batteries, physical contact of solid electrolyte and solid electrodes is more difficult than solid–liquid contact. As mentioned in the section “Iodine”, a commercially successful example of solid electrolyte battery is the Li–I2 cell using iodine as I2–P2VP charge transfer complex. Ionic conductivity of lithium iodide is about 10−7 S cm−1 at room temperature but a very thin lithium iodide layer is formed in situ by direct contact of lithium and iodine. This cell is mainly used as a power source of a cardiac pacemaker, which is operated at a constant temperature of human body and needs small output current in the order of microamperes.

Molten salts Molten salts with high lithium ion conductivity can be used as electrolytes as well. Most of the lithium salts, however, are not suitable for lithium batteries due to their high melting point. Therefore, molten salts are mainly used in reserve batteries in which the electrolyte is in solid state during storage. When the battery is used, the electrolyte is heated and becomes ionically conductive liquid. This kind of batteries can be stored for a very long period but the operating period is short. Ionic liquids are a kind of molten salts that are liquid state salts at room temperature. A very low vapor pressure, nonflammability, and a wide potential window are expected for this type of electrolyte, which are desirable properties for the safety of batteries.

Nonaqueous electrolyte solutions Nonaqueous electrolyte solutions are divided into inorganic and organic solutions. An example of inorganic electrolyte is lithium aluminum chloride (LiAlCl4) dissolved in thionyl chloride, which is used in the Li–SOCl2 battery. As mentioned above, the electrolyte solvent also serves as the positive active material in a battery. Therefore, in the initial stage, positive and negative electrodes are directly contacted, but lithium chloride (LiCl) layer spontaneously formed on the surface of lithium prevents short circuit of the battery. The property of the lithium chloride layer, which is the real electrolyte in this system, has a strong influence on the performance of the battery. An organic electrolyte is a solution consisting of a lithium salt and an organic solvent, and this kind of electrolyte is most commonly used for both primary and secondary lithium batteries. The organic solvent must be aprotic in order to avoid reaction with lithium. It is also desirable to have a high dielectric constant to dissolve sufficient amount of lithium salt. Mixtures of two or more organic solvents are often used to obtain good solvent properties. Typical lithium salts are lithium perchlorate (LiClO4), lithium tetrafluoroborate (LiBF4), lithium trifluoromethane sulfonate (LiCF3SO3), and so on.

Separator Requirements of separators for lithium primary batteries The function of a separator in a lithium primary battery is basically the same as other batteries: prevention of internal short-circuit by separation of positive and negative electrodes, and retention of an electrolyte solution. However, due to the different properties of electrolytes, requirement for separators for lithium batteries are not the same as those for aqueous electrolyte batteries. Requirements for separators for a lithium primary battery using organic electrolyte are as follows: (a) Sufficient mechanical strength: Electrodes for real batteries do not have an ideally smooth surface. There may be some small particles of electrode components that lost contact from one electrode and can migrate to other electrode. It is necessary to avoid internal short-circuit even in these cases. Besides, during production process of a battery, the separator must endure tensile and compression forces. (b) Ability to retain sufficient amount of electrolyte: This property is related to the wettability and swelling of each polymer material. The desirable property is different according to the cell system. For a lithium primary cell using organic electrolyte, separator material should be basically lipophilic. (c) Low electric resistance: In order to reduce the internal resistance of a cell, the thickness of electrolyte, namely thickness of the separator, must be thin. Separator materials must have strength to be made in a film with minimum thickness. A high ionic conductivity of a separator layer is desirable, but the material should be electronically insulating. (d) Chemical and electrochemical stability in battery: A separator must be stable against both electrodes. A lithium battery employs lithium, which is an extremely strong reducing agent, as a negative electrode, and strong oxidizing agent as a positive electrode. A separator must also be stable in an organic or inorganic solvent. (e) Adequate thermal properties: Temperature of a cell may become higher during operation and by an abuse or an accident. It is essential for safety of a cell that the separator does not melt at an operation temperature. On the contrary, some batteries use

Battery Types – Lithium Batteries – Lithium Primary Batteries | Overview

67

‘shutdown’ effect of a microporous separator as a safety mechanism. In this case, an unusually high electric current is stopped by blockade of micropores in the separator due to melting. This kind of separator is required to have a melting point below the melting point of lithium (180
C).

Separators for lithium primary batteries Due to the reasons mentioned above, nonwoven fabrics and microporous separators made from polyolefin, such as polyethylene or polypropylene, are used in a lithium battery with an organic electrolyte. A nonwoven fabric is polymer made into a sheet without weaving. A microporous separator is a thinner film of polymer with many submicron-sized pores. Both of these separators can be simultaneously used to obtain desirable properties. In a Li–SOCl2 battery, which uses a liquid positive electrode material, the true separator is the lithium chloride solid electrolyte film formed on the surface of lithium. A glass separator is also used for prevention of short-circuit considering the stability in the corrosive liquid positive electrode.

Cell Structure Lithium primary batteries have been commercialized in many shapes and sizes. Detailed cell designs are described in the article for each system. Shapes of batteries for consumer uses are roughly classified as follows: (a) Coin type: This type of cell, also called as flat type, disk type, or button type, has a simple structure that is composed of a lithium negative electrode and a positive electrode, with a separator soaked with electrolyte solution between them. Common sealing method is the crimp seal using a plastic insulator between positive and negative cases. Diameters of batteries are typically 10–30 mm, with thickness of 1.2–3 mm. This type of battery has a relatively small capacity and used for low-drain-current applications. (b) Cylindrical type: Cylindrical cells can be divided into bobbin (inside-out) and spirally wound types according to the internal structure. A bobbin-type cell has cylinder-shaped positive and negative electrodes, and a separator between them. This configuration enables efficient packing of electrode materials and is suitable for high energy density, but not suitable for high output power uses because of relatively small electrode surface area. On the contrary, a spirally wound type has a larger electrode surface area with thinner wound electrodes and are used for high output power uses. Either crimp seal or laser welding seal is employed for these cells. (c) Other types: Some other structures like pin-type, prismatic-type, and battery packs are manufactured for each purpose.

Manufacturing Process A dry atmosphere is indispensable for the manufacturing process of lithium primary batteries; this is a major difference in the manufacturing process compared with that of primary batteries with aqueous electrolytes. Lithium vigorously reacts with water and it is readily corroded by moisture in ambient air. Therefore, handling of lithium metal is done in a dry air or a dry inert gas atmosphere. All battery parts, such as positive electrode, electrolyte, separator, and current collector, should be dried before battery assembly. Any residual water in a cell may cause self-discharge, deterioration of performance, or gassing during operation or storage. A typical cell assembly procedure for lithium primary batteries is shown in Figure 1. Lithium metal foil is cut into a prescribed size and pressed on to a current collector to form a negative electrode. Positive electrode material is mixed with a conducting agent and a polymer binder. For a coin-type battery, the positive electrode mixture is molded as a pellet. For a larger electrode, the mixture is made into a sheet on a proper current collector metal. The negative and positive electrodes are assembled with a separator, inserted into a battery can, and, then, electrolyte solution is injected and sealed.

Handling and Transportation General cautions in the handling of lithium primary batteries are similar to those for other primary batteries using aqueous electrolytes, but special care should be taken for them because they have a very reactive metal, lithium, as a component. Besides, organic electrolyte-type batteries contain combustible organic solvents. In order to avoid thermal runaway, it is essential to avoid the battery reaching the melting point of lithium (180
C). Important precautions are to avoid charging, short circuit, or forced discharge in the electric circuit. Also, heating, incineration, disassembly, or mechanical deformation must be avoided. These misuses or abuses may cause leakage of electrolyte, overheating, combustion, or explosion of batteries. The phenomena occurring in case of misuses and abuses are dependent on the chemistry, capacity, size, shape, and so on, and users should be careful to follow the cautions for each battery. Transport of lithium primary batteries are regulated by International Civil Aviation Organization (ICAO), International Air Transport Association (IATA), International Maritime Organization (IMO), and each government based on the recommendations by United Nations. In these regulations, packaging and transportation are specified according to the weight of lithium metal contained in a lithium battery. Lithium batteries that meet the Special Provision A45 of IATA are not subject to the regulations.

68

Battery Types – Lithium Batteries – Lithium Primary Batteries | Overview

Positive active material Conducting agent (carbon)

[Mixing, pressing]

Binder (polymer)

Positive electrode

Positive can Separator

[Pressing]

Lithium metal sheet

Negative electrode [Assembly]

Negative can Organic solvent Lithium salt

[Mixing]

Electrolyte solution

[Electrolyte impregnation]

[Sealing]

Cell

Figure 1 A typical cell assembly procedure of a lithium primary cell.

Concluding Remarks In the initial stage of commercialization of lithium primary batteries, increase of application of this battery was rather slow due to incompatibility in voltage to conventional primary batteries. In the 1980s, the advantages of lithium primary batteries, such as a high operating voltage, a high specific energy and high energy density, and excellent storage characteristics, became important for many electronic devices and the production increased very rapidly. Currently, lithium and zinc are two major negative electrode materials for primary batteries. Application areas have expanded to consumer electronics equipments, automatic and digital cameras, computer memory backups, and various measurement systems. As a high-performance and environmentally compatible battery system, lithium primary battery is expected to increase its applications in the future.

Further Reading 1. Besenhard, J.O.; Eichinger, G. High energy density lithium cells: Part I. Electrolytes and anodes. Journal of Electroanalytical Chemistry 1976, 68, 1–18. 2. Brodd, R.J.; Bullock, K.R.; Leising, R.A.; Middaugh, R.L.; Miller, J.R.; Takeuchi, E. Batteries, 1977–2002. Journal of the Electrochemical Society 2004, 151, K1–K11. 3. Cignini, P.; Icovi, M.; Panero, S.; Pistoia, G.; Temperoni, C. Non-stoichiometric molybdenum oxides as cathodes for lithium cells: Part I. Primary batteries. Journal of Electroanalytical Chemistry 1979, 102, 333–341. 4. Cisak, A.; Werblan, L. High-Energy Non-aqueous Batteries. Ellis Horwood: Chichester, UK, 1993. 5. Cooper, J.F.; Hosmer, P.K.; Hosmy, R.V. The anodic behavior of lithium in aqueous lithium hydroxide solutions. Journal of the Electrochemical Society 1978, 125, 1–7. 6. Dey, A.N. Lithium anode film and organic and inorganic electrolyte batteries. Thin Solid Films 1977, 43, 131–171. 7. Dey, A.N.; Bro, P. Primary Li/SOCl2 cells: IV. Cathode reaction profiles. Journal of the Electrochemical Society 1978, 125, 1574–1578. 8. Di Pietro, B.; Scrosati, B.; Bonino, F.; Lazzari, M. Primary lithium-metallic oxysalt organic electrolyte batteries. Journal of the Electrochemical Society 1979, 126, 729–731. 9. Eichinger, G.; Besenhard, J.O. High energy density lithium cells: Part II. Cathodes and complete cells. Journal of Electroanalytical Chemistry 1976, 72, 1–31. 10. Gabano, J.P., Ed.; In Lithium Batteries; Academic Press: London, 1983. 11. Hughes, M.; Hampson, N.A.; Karunathilaka, S. A. G. R. A review of cells based on lithium negative electrodes (anodes). Journal of Power Sources 1984, 12, 83–144. 12. Jackson, G.W.; Blomgren, G.E. Lithium anode properties in a nonaqueous cell. Journal of the Electrochemical Society 1969, 116, 1483–1487. 13. Linden, D.; Reddy, T.B. Lithium batteries. In Handbook of Batteries, 3rd edn; Linden, D., Reddy, T.B., Eds.; McGraw-Hill: New York, 2002;; pp 14.1–14.106. 14. Nishio, K.; Furukawa, N. Practical batteries. In Handbook of Battery Materials; Besenhard, J.O., Ed.; Wiley-VCH: Weinheim, Germany, 1999;; pp 19–61. 15. Owens, B.B.; Skarstad, P.M.; Untereker, D.F.; Passerini, S. Solid electrolyte batteries. In Handbook of Batteries, 3rd edn; Linden, D., Reddy, T.B., Eds.; McGraw-Hill: New York, 2002;; pp 15.1–15.25. 16. Pasquali, M.; Pistoia, G. Primary 1.5 V lithium cells with BiVO4 cathodes. Journal of Power Sources 1989, 27, 29–34. 17. Pistoia, G.; Pasquali, M.; Rodante, F. Button cells based on the Li/Bi2O3 couple. Journal of Power Sources 1985, 16, 263–269. 18. Read, J. Characterization of the lithium/oxygen organic electrolyte battery. Journal of the Electrochemical Society 2002, 149, A1190–A1195. 19. Schlaikjer, C.R. Liquid cathode primary batteries. Journal of Power Sources 1985, 14, 111–122. 20. Scrosati, B. Non aqueous lithium cells. Electrochimica Acta 1981, 26, 1559–1567. 21. Urquidi-Macdonald, M.; Flores, J.; Macdonald, D.D.; Pensado-Rodriguez, O.; Van Voorhis, D. Lithium/water system: Primary batteries. Electrochimica Acta 1998, 43, 3069–3077.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–IodinePolyvinylpyridine CF Holmes, Greatbatch Inc., Clarence, NY, United States © 2009 Elsevier B.V. All rights reserved. This is a reproduction of C.F. Holmes, PRIMARY BATTERIES – NONAQUEOUS SYSTEMS | Lithium–Iodine-Polyvinylpyridine, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 76–82, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00106-4.

Introduction Cell Chemistry Cell Design Cell Performance Electrical Discharge Testing Electrical Test Results Performance and Reliability Summary Further Reading

70 71 71 73 73 73 75 75 75

Abstract For the last 35 years, the lithium/iodine-polyvinylpyridine (PVP) battery system has provided a highly reliable power source for implantable cardiac pacemakers. Literally millions of these cells have been implanted since 1972, and the system remains the most widely used pacemaker battery today. The basic chemical reaction of the cell is the simple combination of elemental lithium and iodine to form lithium iodide, the self-forming solid electrolyte/separator of the system. The battery exhibits high internal impedance, which increases as the cell is discharged. This feature renders it unattractive for most battery applications, but in the pacemaker application, which requires only microampere-level current capability, it has proven to be a reliable and predictable system. Advances in cell design have led to an improvement in both the current-delivery capability of the system and the energy density, the latter parameter having been tripled from early models to today’s designs. Although some modern pacemakers with advanced features are beginning to use lower impedance lithium batteries, the lithium/iodine-PVP system remains the most commonly used power source for pacemakers and is likely to remain so for many years to come.

Glossary Cardiac pacemaker An implantable device used to treat a slow or irregular heart beat. Case-grounded design A battery design in which a central lithium anode is surrounded by the cathode material, which is in contact with the metallic case of the battery. The battery case becomes the positive terminal of the battery. Open-circuit voltage The voltage of a battery when no current is being extracted from it. Operating voltage The voltage of a battery when under a load. PVP Polyvinylpyridine, a material used in the lithium/iodine battery. Qualification testing A series of operational and abusive tests designed to demonstrate the suitability of a new battery model for implantable use. Selim-Bro plot A graph that presents the capacity achieved by the battery as a function of the negative of the logarithm of the current drain under which the battery was discharged. Volumetric energy density A measure of the energy per unit volume of a battery.

Nomenclature

Symbols and Units Eo F n DG

equilibrium voltage Faraday constant number of electrons transferred in the cell reaction Gibbs free energy

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00352-2

69

70

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Iodine-Polyvinylpyridine

Abbreviations and Acronyms AC ADD BoL OCV PVP

alternating current accelerated discharge data beginning-of-life (battery) open-circuit voltage polyvinylpyridine

Introduction The first successful cardiac pacemaker in the United States was implanted in Buffalo, New York, in 1960 (an earlier implant in Sweden lasted only a few days before failing). The Buffalo unit, along with most pacemakers implanted from 1960 through about 1976, was powered with zinc/mercuric oxide aqueous electrolyte batteries. Although these batteries helped to make the use of pacemakers possible in these early years, they exhibited significant drawbacks. They did not have particularly high energy density, they evolved hydrogen gas, preventing hermetic sealing of the battery, they failed catastrophically (i.e., with no ability to detect end of service beforehand), and they were rather heavy. Several individual cells (as many as 10 in some units) were required to attain the required voltage and current. Thus pacemakers using this technology were large, sealed in a plastic polymeric material, and normally failed after 24–40 months. Clearly a better power source was needed. Nuclear batteries were developed for this application. Although these power sources provided excellent longevity and reliability, the use of plutonium in these cells forced very cumbersome regulatory and documentation requirements that limited the widespread use of these cells. No cells using this technology are produced today. In the early 1970s, the development of batteries using an anode of elemental lithium was progressing well in the international battery community. Indeed, one of the first candidate uses of lithium cells was in implantable devices. Lithium-powered pacemakers quickly replaced units using the older zinc/mercuric oxide system in the marketplace. Several cathode chemistries were developed and used in pacemakers, including lithium/cupric sulfide, lithium/silver chromate, lithium/thionyl chloride, and lithium/iodine-polyvinylpyridine (PVP). Of these systems, the lithium/iodine-PVP eventually became the most widely used system employed in pacemakers, and it remains so today. Although some recently developed pacemakers with features requiring higher current-delivery capacity are now using liquid electrolyte lithium batteries, the great majority of pacemakers are still powered with the lithium/iodine-PVP chemistry. Pacemakers using lithium-based systems offered significant advantages over the previously used systems. They permitted the development of much smaller pacemakers that showed significantly greater longevity. They permitted the pacemakers to be hermetically sealed, and pacemaker enclosures using titanium or stainless steel cases with ceramic-to-metal hermetic seals replaced the older, more cumbersome units sealed in polyester. Moreover, the use of lithium-based systems led to significantly greater reliability and the ability to detect the approach of the unit to the end-of-service point with much greater accuracy than was the case with the zinc/mercuric oxide system. The lithium/iodine-PVP battery was invented in the late 1960s by Alan Schneider and James Moser of the Catalyst Research Corporation. Their invention was based on earlier work performed by Gutmann, Herman, and Rembaum of the Jet Propulsion Laboratory. Many improvements to the system were developed by Greatbatch, Mead, Rudolph, and others from the Greatbatch, Inc. laboratories. First implanted in Italy in 1972, this battery rapidly became the power source of choice for the implantable pacemaker, and by the early 1980s the majority of all pacemakers produced worldwide used this system. The system is arguably the first commercially successful application of lithium battery technology, with significant sales occurring during the latter half of the 1970s. It remains the most widely used battery system for powering pacemakers today. Several million have been implanted, and the system is likely to be used for years to come. Over the years, many improvements have been made to the system. Cell design advances have led to a tripling of the energy density, from 0.3 Wh cm−3 to more than 0.9 Wh cm−3. The development of the coating of the anode with PVP has led to a significant increase in the current-delivery capability of the system. Increases in the ratio of iodine to PVP in the cathode material have increased the energy density. There is an inherent reliability in this system. The solid electrolyte/separator, lithium iodide, is self-forming and self-healing. The high impedance of the system ensures that overheating of the system or ‘thermal runaway’ owing to internal or external short-circuiting cannot happen. The discharge curve is well understood and predictable, allowing pacemakers to predict with accuracy the longevity of their products and the detection of the onset of end of service. It could be argued that, by most common measures of battery capability, the lithium/iodine-PVP battery is not a very ‘good’ system. It does not work well at low temperatures (i.e., at or below room temperature), it is permanently and significantly degraded if exposed to temperatures above 65
C for any significant length of time, and it can deliver only microampere-level currents. It has no other commercially successful applications in the battery industry. However, placed at 37
C and asked to provide between 5 and 100 mA of current, it can do so with high energy density, excellent reliability, and predictability of performance.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Iodine-Polyvinylpyridine

71

Cell Chemistry The basic chemical reaction of the lithium/iodine-PVP system is the simple combination of elemental lithium and elemental iodine to form the ionic salt lithium iodide: 2Li +I2 ! 2LiI

[I]

−1

The Gibbs free energy of this reaction is −64.451 kcal mol , and therefore the equilibrium voltage of the cell, calculated from the equation DG ¼ −nFE∘

[1]

where DG is the Gibbs free energy, n is the number of moles transferred in the reaction, F is the Faraday constant, and Eo is the equilibrium voltage is 2.8 V. This is also measured open-circuit voltage (OCV) of the system. The cathode material is formed by an exothermic reaction between iodine and PVP. The material is synthesized by heating the two constituents to a temperature above the melting point of iodine (113.5
C). The ratio of iodine to PVP is between 30/1 and 50/1. Thus most of the cathode is elemental iodine. The remainder is reacted with the PVP. One molecule of iodine is bonded to the nitrogen of the pyridine ring and another atom replaces the alpha hydrogen of the polymer chain. This cathode exhibits an interesting property that permits the functioning of the battery – it is an electronic conductor. Its conductivity is a function of the I2/PVP ratio. The curve is inverted U shaped, with maximum at the I2/PVP weight ratio of approximately 8/1. The existence of unpaired electrons in the material has been confirmed by electron spin resonance spectroscopy, which shows a single sharp signal with a g value of 2.002. As the cell is discharged, the electronic conductivity increases as the I2/PVP weight ratio approaches the value of 8/1, and then begins to decrease as the weight ratio becomes lower than that value. This decrease in conductivity, combined with the increase in electrolyte resistance as the lithium iodide layer thickens during the cell reaction, leads to a lowering of the operating voltage as the cell is discharged. This gradual approach to the ‘elective replacement voltage’ of the battery results in a predictable and detectable means of determining when the pacemaker should be replaced. The reaction product, lithium iodide (LiI), is formed at the interface between the lithium anode and the cathode material. When the cell is manufactured, the molten cathode material is poured into a cell housing containing a central anode of elemental lithium. A layer of lithium iodide is instantly formed between anode and cathode. This layer acts as both the separator, insulating the cathode from the anode, and the solid electrolyte, through which lithium ions migrate to the electrolyte/cathode interface. As the cell reaction proceeds, the lithium iodide layer thickens as more molecules of lithium iodide are formed, causing the internal resistance of the cell to increase. An important feature of the batteries produced by Greatbatch, Inc. is the technique of coating the anode of the cell with PVP before adding the cathode mixture. When first developed, the coating was accomplished by literally painting the PVP, dissolved in a volatile liquid, onto the anode of the cell using small camel’s hair paint brushes. Later, a substrate coating technique was developed and used. In this process, an inert substrate, a polymeric material, is submerged in a solution of PVP in a volatile solvent, thus impregnating the substrate material with PVP. After drying, anode coating blanks of the impregnated material are punched out in the exact shape of the anode itself, and they are pressed onto the lithium anode in the process in which the two lithium blanks are pressed onto the anode current collector. This process results in much more uniform application of the PVP material onto the lithium anode and leads to more consistent performance and lower self-discharge. The anode coating produces profound differences in the electrical performance of the cell and in the morphology of the lithium iodide formed during the cell reaction. Studies of the lithium iodide formed in cells made with and without the anode coating show macroscopic and microscopic differences in the formation of the discharge product. Uncoated-anode cells show a hard, regular, flat layer of lithium iodide after cell discharge. Coated-anode cells show a much more irregular macroscopic structure, with some distortion of the anode and a much more columnar structure of lithium iodide. There is a noticeably tighter bond between the anode and the lithium iodide. At a microscopic level, scanning electron microscopic studies have shown that the lithium iodide formed from discharge of coated-anode cells consists of well-formed individual granular crystallites. In contrast, lithium iodide formed in cells without the anode coating is a much less granular structure having the appearance of a mass of small granules in a hard, amorphous matrix. Cells with the coated anodes demonstrate a much higher voltage throughout cell discharge, and the internal resistance of the cell is much lower than that of an otherwise identical cell with an uncoated anode. The current-delivery capability is much higher, and the ability of the cell to deliver higher-current pulses during operation is greatly enhanced. All lithium/iodine cells produced by Greatbatch, Inc. since 1975 have contained anodes with this coating. Several important properties of the lithium/iodine-PVP battery are shown in Table 1.

Cell Design The design of lithium/iodine-PVP cells has undergone a considerable evolution since its first use in 1972. Unneeded inert material has been removed from the cells, and energy density has greatly increased.

72

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Iodine-Polyvinylpyridine Table 1

Properties of the lithium/iodine-PVP battery

Anode Cathode Electrolyte/separator Application Operating temperature Normal current drain Open-circuit voltage Operating voltage Cell internal resistance

Elemental lithium, coated with PVP Iodine thermally reacted with PVP Lithium iodide, formed by the cell reaction Power source for cardiac pacemakers 37
C 5–100 mA 2.8 V 2.78–2.0 V 50 O at the beginning of life, increasing to several thousand ohms as cell is discharged

PVP, polyvinylpyridine.

The first commercially successful cell, rather large by today’s standards, was rectangular in shape with nominal dimensions of 14 mm45 mm52 mm. The internal construction consisted of a central lithium anode surrounded on both sides by the iodine-PVP material. The weight ratio of iodine to PVP was 10/1. The active cell components were enclosed in a fluoropolymer support and then encased in a polyester potting material. This structure was then sealed into a case made of alloy 304 stainless steel, and the electrical connections were brought out of the interior by glass-to-metal seals. The case itself was electrically neutral. The next widely used cell design reversed the locations of anode and cathode. The anode consisted of two ‘cups’ of lithium material that enclosed the centrally located cathode material. The weight ratio of iodine to PVP was increased to 15/1. Again, the active components were enclosed in a polyester potting material, insulating them from the stainless steel case. This cell model was considerably smaller than the first design, and had a rounded shape more amenable to the construction of smaller pacemakers. This design concept was also employed in a widely used two-cell battery. Two such central cathode cells were potted together in a polyester potting material and connected in series. This structure was then sealed into a stainless steel case, and the electrical connections were brought out via two glass-to-metal seals. The enclosing of the case materials in plastic materials and the isolation of these materials from the case itself were done because it was feared that the iodine/PVP material might react with the stainless steel case material and cause corrosion of the case and leakage of the iodine material out of the battery and into the pacemaker, a disastrous situation that would cause the unit to fail catastrophically. This turned out not to be the case. Extensive material compatibility studies demonstrated that, in cells constructed in an extremely dry environment (as must be the case for all lithium anode cells), the iodine/PVP material was unreactive to the stainless steel case material. It was subsequently conclusively demonstrated by macroscopic and microscopic analysis that there was no significant reaction between the cathode material and the case of the battery even after years of testing at body temperature. This discovery led to the development of the so-called case-grounded design. In this design, the unneeded plastic materials were eliminated. The design consisted of a centrally located lithium anode, pressed around a nickel current collector attached to a glassto-metal seal on the lid of the battery. The molten iodine/PVP material was poured into the battery, surrounding the anode on both sides. The case itself acted as the cathode current collector. The ratio of iodine to PVP was increased to 20/1 in the first versions of the case-grounded design. The first case-grounded model was introduced in 1975. The use of this cell design increased the volumetric energy density of the system by a factor of 3, and all lithium/iodine-PVP cells produced today feature this design concept. It has proven to be a very reliable and useful design concept. Figure 1 shows a cutaway view of a typical case-grounded cell design. Stainless steel feedthrough pin (negative polarity) Insulative glass-to-metal seal

Depolarized fill hole and final close weld

Stainless steel case and lid (positive polarity)

Central lithium anode

Iodine/PVP depolarizer PVP-substrate anode coating

Figure 1 Cutaway view of a case-grounded lithium/iodine-polyvinylpyridine (PVP) battery. Reproduced from Greatbatch Website, www.Greatbatch.com.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Iodine-Polyvinylpyridine

73

Further design improvements have been made since the introduction of the case-grounded system. These improvements, although not as significant as the above-mentioned innovations, have led to better cell performance and higher energy density. In the late 1970s, cells with corrugated anodes were introduced. The corrugations were made by an anode pressing fixture designed with such corrugations. The corrugated anode increased the anode surface area, thus increasing the current-delivery capability of the cell. Moreover, the corrugated anodes did not demonstrate significant distortion of the anode during cell discharge, as was seen in the flat-anode designs. Today all cells contain corrugated anodes. Another modification to the anode structure was the removal of a Halar™ strap that was placed around the anode in earlier designs. The strap was introduced to ensure that the lithium anode did not touch the electrically active case. It was found that this strap was unnecessary, and the lithium anode on cells produced today is held in place by a simple plastic clip designed to center the cell within the case wall. The clip holds the anode in place and touches the bottom of the battery case. This feature has the additional advantage of allowing more of the lithium to be exposed to the cathode material, increasing the effective anode surface area. Further increases in the iodine/PVP ratio have been introduced. Today cells are made with iodine/PVP ratios ranging from 30/1 to 50/1. It should be pointed out that whereas it might be concluded that the difference between 30/1 and 50/1 is large, it is in fact only the difference between 97% iodine and 98% iodine. The method and mechanism of sealing the battery were also improved in the late 1970s. In the first case-grounded designs, the cathode material was poured into the battery before the lid was welded to the case. The lid was then placed on the battery and welded to the case. It was found that this design could lead to contamination of the lid-to-case weld with iodine, causing weld failure. In subsequent designs, the lid, with the anode structure attached to the glass-to-metal seal, was welded to the case before the iodine material was introduced. The cathode material was then introduced into the battery through a ‘fill hole’ via a vacuum-filling technique. The fill hole is then sealed by a redundant (i.e., a two-step) sealing mechanism including a press-fit inner ball seal and an outer seal made by welding a sealing plug to the case, closing the battery.

Cell Performance The discharge characteristics of the lithium/iodine-PVP battery are well understood and well documented. Years of in-house electrical data are available, and field performance of pacemakers powered by this system has been documented and discussed.

Electrical Discharge Testing One of the challenges facing manufacturers and developers of implantable batteries is the fact that to determine electrical performance in real time requires years of discharge data at very light current drains. This problem means that accelerated testing and data modeling techniques must be developed to be able to predict with accuracy the long-term performance of a battery system designed for 6–10 years of discharge use. In the Greatbatch, Inc. laboratories, the electrical performance of all lithium battery models produced by the company is characterized in two ways. First, a select group of batteries are subjected to a series of constant resistive load discharge tests at a variety of loads, ranging from loads much heavier (i.e., currents much higher) than would be seen in a pacemaker to very light loads representing the microampere-level currents demanded by the pacemaker application. These tests are referred to as ‘ADD’ tests (for ‘accelerated discharge data’). The second method of testing is known as life testing. In this test, a running sample of batteries that would otherwise be supplied to pacemaker manufacturers is placed on a constant resistive load of 100 kO and an electrical measurement of voltage and impedance (1000 Hz AC (alternating current)) is taken every 8 weeks. All tests are conducted at body temperature (37
C). The purposes of these two testing procedures are similar but different. The accelerated discharge testing produces data that can be used soon after the development of a new cell to predict real-time performance by modeling techniques. As more discharge data become available as lighter-current tests are completed, the prediction of real-time performance can be enhanced and refined with the passage of time. Users of the cells are provided with regular updates of the accelerated discharge tests. The purposes of life testing are twofold. First, it makes available discharge data at current drains similar to those demanded by the pacemaker application. Second, because it is a running sample taken throughout the production life of the particular cell model, it provides a method of monitoring cell performance and the ability to detect a potential problem long before it becomes apparent in the field. In this testing method, voltage and impedance measurements are taken every 8 weeks during the cell’s discharge. A computer program compares the results with predicted values determined from the modeling described above. If the voltage of any cell falls below the lower bound of the expected value, an automatic alert is distributed and an investigation ensues. The pacemaker manufacturers are provided with regular reports of the life-test data, and any untoward test results are brought to the attention of potentially affected manufacturers immediately.

Electrical Test Results Many differently shaped models of the lithium/iodine-PVP system have been developed in the last 35 years. New shapes are routinely designed, qualified, and produced as new shapes of pacemakers are developed. Accordingly, a vast amount of electrical data is available, as samples of each separate model are subjected to the above-mentioned electrical discharge testing.

74

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Iodine-Polyvinylpyridine 2800

2600

2400

2200

2000

1800

4.53 k: 5.49 k:

Voltage (mVdc)

1600

7.50 k: 10.7 k:

1400

21.0 k: 1200

42.2 k: 73.2 k:

1000

100 k: 140 k:

800

600

400

200

0 0

200

400

600

800

1000

1200

1400

1600

1800

Capacity (mAh)

Figure 2 Discharge curves of a lithium/iodine-polyvinylpyridine (PVP) battery at 37
C under the constant resistive loads shown in the legend.

The graph shown in Figure 2 presents discharge curves for a particular model of lithium/iodine-PVP cell. This cell, first produced in 1989, has seen extensive use in cardiac pacemakers. The cell’s shape is prismatic, with straight side walls and a full-radius rounded bottom geometry. The nominal dimensions of the cell are 3.05 cm2.73 cm0.60 cm. Figure 2 presents average electrical discharge data from groups of 12 cells each discharged at constant resistive loads ranging from 4.53 to 140 kO. The voltage under the given load is plotted as a function of the capacity (in milliampere hours) extracted from the cell. All testing was performed at 37
C. It can be seen from the graph that the approach to end-of-service voltage is reasonably gradual, which is desired in this application. A convenient and frequently used representation of the results of electrical discharge testing of batteries is by using the so-called Selim-Bro plot. This plot presents the capacity achieved by the battery as a function of negative of the logarithm of the current drain under which the battery was discharged. This curve is normally inverted U-shaped. The capacity is lower at higher current drains because of polarization of the cell, i.e., because the resistance of the cell causes the voltage to be lower at the heavier loads. The capacity is lower at the very low current drains because of self-discharge, a process in which the anode and cathode of the battery react with each other internally, without producing an external electrical current. Figure 3 shows the Selim-Bro plot summarizing the data shown in Figure 2. The capacity achieved to a 1.8 V cutoff voltage is plotted as a function of the logarithm of the beginning-of-life (BoL) current drain. The inverted U shape of the curve is seen in the plot.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Iodine-Polyvinylpyridine

75

1.8 1.6 1.4

Capacity (Ah)

1.2 1 0.8 0.6 0.4 0.2 0

0

0.5

1

1.5

2 2.5 ln(BoL current)

3

3.5

4

4.5

Figure 3 Selim-Bro curve for the discharge curves shown in Figure 1. The curve plots the capacity of the battery as a function of the logarithm of the beginningof-life (BoL) current drain.

Performance and Reliability Since the first implant in 1972, several million lithium/iodine-PVP batteries have been produced and implanted in cardiac pacemaker patients. The system has shown an impressive record of reliability, predictability, and safety. Sudden failures of the battery system have been extremely rare, and the predictability of the discharge characteristics have allowed manufacturers and physicians to understand, document, and detect the onset of battery depletion well before the unit ceases to operate. Certainly, this performance record can be attributed in part to the inherent reliability and discharge characteristics of the system as described in the preceding sections. However, the excellent field performance record of the system is due in very large part to the care with which the batteries are designed, manufactured, and tested before shipment to device manufacturers. Stringent design rules have been formulated for use in creating new models of the battery. Careful selection of such parameters as lithium anode thickness, the ratio of anode to cathode, and the details of internal cell design must be made for new cell models. As new models are designed, a detailed and well-documented model qualification is performed. This qualification includes electrical characterization, environmental testing, abuse testing (e.g., short circuiting, cell crushing, and cell puncturing), and exposure to high and low temperatures. The results of this testing are documented in an official qualification report that is distributed to users of the cell. This report may be provided to regulatory agencies as part of the submission of new pacemaker models for approval by such agencies. These efforts, coupled with the life-testing regime described above, are designed to ensure that cells used in this life-saving application will perform as expected and will do so in a safe, reliable manner.

Summary The lithium/iodine-PVP battery system has proven to be a reliable power source for the implantable pacemaker. Millions of cells have been implanted in patients since 1972, and the system has amassed an excellent record of reliability and longevity. The details of the chemical system are well understood. The procedures used to design, manufacture, qualify, and test the batteries have shown to be effective in preventing significant field problems and ensuring reliability. Although advanced versions of modern pacemakers will need lithium cells with higher current-delivery capability, there is no doubt that the lithium/iodine-PVP system will see significant clinical use for many years to come.

Further Reading 1. Brennen, K.R.; Untereker, D.F. Iodine utilization in li/iodine (poly-2-vinylpyridine) batteries. In Power Sources for Biomedical Implantable Devices; Owens, B.B., Margalit, N., Eds.; The Electrochemical Society: Pennington, NJ, 1980;; pp 161–173. 2. Greatbatch, W.; Holmes, C.F. History of implantable devices. IEEE Engineering in Medicine and Biology 1991, 91, 38–49. 3. Greatbatch, W.; Holmes, C.F. The lithium/iodine battery: A historical perspective. Pacing and Clinical Electrophysiology 1992, 15, 2034–2036. 4. Greatbatch W, Mead R, and Rudolph F (1976) Lithium/Iodine Battery Having Coated Anode. US Patent 3,957,533, 18 May 1976.

76 5. 6. 7. 8. 9. 10. 11. 12.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Iodine-Polyvinylpyridine Holmes, C.F. Lithium/halogen batteries. In Batteries for Implantable Medical Devices; Owen, B.B., Ed.; Plenum Press: New York, 1986;; pp 133–180. Holmes, C.F. Implantable lithium power sources. In Lithium Batteries: New Materials, Developments, and Perspectives; Pistoia, G., Ed.; Elsevier: Amsterdam, 1994;; pp 377–416. Holmes, C.F. The role of lithium batteries in modern health care. Journal of Power Sources 2001, 97, 739–741. Liang, C.C.; Holmes, C.F. Performance and reliability of the lithium/iodine battery. Journal of Power Sources 1980, 5, 3–13. Moser JR (1972) Solid State Lithium Iodine Battery. US Patent 3,660,163, 2 May 1972. Moser JR and Schneider AA (1972) Primary Cells and Iodine-Containing Cathodes Therefore. US Patent 3,674,562, 4 July 1972. Selim, R.; Bro, P. Performance domain analysis of primary batteries. Journal of the Electrochemical Society 1971, 118, 829–834. Visbisky, M.; Stinebring, R.C.; Holmes, C.F. An approach to the reliability of implantable lithium batteries. Journal of Power Sources 1989, 26, 185–194.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide K Nishio, Kyoto University, Kyoto, Japan © 2009 Elsevier B.V. All rights reserved. This is a reproduction of K. Nishio, PRIMARY BATTERIES – NONAQUEOUS SYSTEMS | Lithium–Manganese Dioxide, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 83–92, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00115-5.

Introduction Principles of a Li–MnO2 Battery Materials Positive Electrode Negative Electrode Electrolyte Separator Some Studies on the Materials of Li–MnO2 Batteries Positive Electrode Negative Electrode Electrolytes Structures and Assembly of Li–MnO2 Batteries Advantages of Li–MnO2 Batteries Specifications of Some Commercially Available Li–MnO2 Batteries coin-Type Cells Cylindrical, Inside-Out-Type Cells Cylindrical, Spiral-Type Cells Li–MnO2 Cell Packs Cell pack 2CR5 for automatic cameras Cell pack CR-V3 for digital cameras Applications Further Reading

78 78 78 78 79 80 80 80 80 80 81 83 84 85 86 86 87 87 87 87 88 88

Abstract A lithium–manganese dioxide (Li–MnO2) primary cell has many advantages over conventional primary cells, such as a high voltage, a high energy density, a high output power, a low self-discharge rate, and a long storage life. The excellent performance of these cells has been accepted by users, and cells of suitable type and capacity are used as a major or backup power source in various electronic and electric appliances. The Li–MnO2 system was commercialized as a result of the development of heat-treated MnO2 material, which has a suitable crystal structure for reaction with lithium in a nonaqueous electrolyte. This article describes the reaction principle, materials, and structures of Li–MnO2 batteries, as well as some examples of discharge characteristics of commercial batteries.

Glossary Aprotic solvent An organic solvent that cannot donate hydrogen. Crimp seal A method for sealing of a cell by mechanical crimping of a metal container. Gasket An insulating material made of plastics used for sealing of a cell and separation of positively and negatively charged parts. Laser seal A method for sealing of a cell using laser welding technique. Specific surface area Total surface area of a material divided by the mass of the material.

Nomenclature

Abbreviations and Acronyms CMD DEE DEM DME

chemically prepared manganese dioxide diethoxyethane diethoxymethane dimethoxyethane

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00353-4

77

78

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide

DMM EMD EME IEC LED NMD PC PTC SHE

dimethoxymethane electrolytic manganese dioxide ethoxymethoxyethane International Electrotechnical Commission light-emitting diode natural manganese dioxide propylene carbonate positive temperature coefficient standard hydrogen electrode

Introduction Lithium primary batteries are currently one of the most widely used batteries, although they are newly developed battery systems compared to many other primary battery systems with aqueous electrolytes. Research and development of a cell using lithium as negative electrode started in the United States in the 1960s mainly for space applications, and later, development of batteries for consumer use became active also in Europe and Japan. An electrochemical cell using lithium and a nonaqueous electrolyte was a totally new system, and both a new positive electrode material with good discharge performance in a nonaqueous electrolyte and materials for the electrolyte had to be developed. Many new positive electrode materials such as metal oxides, metal sulfides, halogenide compounds, and organic compounds were developed and tested, and lithium cells using iodine complex, polycarbon monofluoride (CFn), and manganese dioxide were commercialized in the 1970s. The lithium–manganese dioxide (Li–MnO2) system was first commercialized in Japan as a coin-type and later it became available as cylindrical cells with a high power or a high energy density.

Principles of a Li–MnO2 Battery The following reaction mechanisms were proposed for a Li–MnO2 cell: Positive electrode reaction: MnðIVÞ O2 + Li + + e − ! MnðIIIÞ O2 ðLi + Þ Negative electrode reaction: Li ! Li + + e − Overall battery reaction: MnðIVÞ O2 + Li ! MnðIIIÞ O2 ðLi + Þ The reaction at the positive electrode is insertion of lithium into the crystal lattice of manganese dioxide, which accompanies reduction of Mn(IV) to Mn(III), and the reaction at the negative electrode is dissolution of lithium into the electrolyte solution. The X-ray diffraction patterns after discharge showed a peak shift to a lower angle, demonstrating an expansion of the crystal lattice. Another simple cell reaction for lithium and manganese dioxide was proposed as follows: 2Li + 2MnO2 ! Mn2 O3 + Li2 O This reaction, however, does not seem to take place, because neither Mn2O3 nor Li2O was found in the reaction products.

Materials Positive Electrode Manganese dioxide is a very common material in the battery field, which has been used for a long time as the positive electrode active material in zinc–carbon dry cells and alkaline manganese dioxide cells. It was also regarded as one of the most promising positive electrode materials for lithium cells, because it was a strong oxidant and less expensive than many other materials. It became clear, however, that manganese dioxide materials used in aqueous electrolyte batteries did not show good discharge performance in a nonaqueous electrolyte. The two major requirements for the manganese dioxide active material in Li–MnO2 cells are as follows: (i) it must be anhydrous; and (ii) it must have a structure suitable for the diffusion of Li+ ions into the manganese dioxide crystal lattice. Manganese dioxide is classified as natural manganese dioxide (NMD) and synthetic manganese dioxide the latter can be chemically prepared manganese dioxide (CMD) or electrolytic manganese dioxide (EMD) according to the preparation method. EMD after heat treatment is used as a positive electrode material in a commercial lithium battery. There have been many studies by many researchers on the discharge performances of manganese dioxide. In early works in Japan, relationships between heat treatment

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide

79

(a)

(b)

(c)

Figure 1 The crystal structures of electrolytic MnO2 (EMD) heat treated at different temperatures (a) 250  C, (b) 375  C, and (c) 450  C.

temperatures, crystal structures, chemical composition, and discharge characteristics were studied. Their findings of a manganese dioxide material with a desirable crystal structure for lithium diffusion became the basis for the first commercial Li–MnO2 battery in the late 1970s. Heat treatment has a significant influence on the crystal structure of EMD. It is basically understood as shown in Figure 1, although it should be noted that there have been some disputes about the crystal structure change. The crystal structure of as-prepared EMD is known as g-MnO2 and it consists of major ramsdellite domain with pyrolusite domain as impurity. Pure ramsdellite has one-dimensional 12 tunnels formed by chains of edge-sharing MnO6 octahedra. Crystal structures of EMD gradually change as the heat treatment temperature increases. By heat treatment of EMD below 250  C, the crystal structure remains as g-MnO2. By heat treatment from about 270 to 400  C, g-b-MnO2, which has both 11 and 12 tunnels, is formed. When the heat treatment temperature is higher than 420  C, b-MnO2 (pyrolusite) with 11 tunnels is obtained. Among these materials, EMD heat treated from 370 to 400  C, which exhibits both the highest discharge capacity and best storage characteristics, is used for the commercial Li–MnO2 batteries.

Negative Electrode Lithium metal is used in a lithium primary cell. The electrode potential of lithium is −3.04 V versus standard hydrogen electrode (SHE), which is the lowest value among all metal electrodes. Lithium has the lowest density (0.53 g cm−3) and the lowest electrochemical equivalent (0.259 g Ah−1) of all solids. As a result of these physical properties, nonaqueous electrolyte batteries using lithium offer the possibility of a high voltage and a high energy density. In a real cell, lithium is mostly used as a foil having a thickness of several tens of micrometers and cut in a proper size. It is not difficult to cut and change the shape of lithium because it is a very soft and flexible metal. It is, however, necessary to pay special attention to the handling of lithium metal. Lithium readily reacts with water to form hydrogen gas. It even reacts with moisture in the atmosphere to form lithium hydroxide on its surface. Therefore, the presence of water should be avoided in all production processes.

80

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide

Electrolyte An electrolyte for a Li–MnO2 cell should possess the following properties: (i) high ionic conductivity in a wide range of temperature; (ii) chemical and electrochemical stability for a long period; (iii) no harmful effects on the positive and negative electrode reactions. In a Li–MnO2 cell, polar aprotic solvents have to be used, as lithium metal reacts with protic solvents such as water and alcohols. The ionic conductivity of an organic electrolyte is generally 1–2 orders of magnitude lower than that of an aqueous electrolyte. Therefore, the composition of electrolytes for higher ionic conductivity is a major research target in nonaqueous systems. In order to dissolve sufficient amounts of solutes in an electrolyte solution, it is desirable that the organic solvent has a high dielectric constant. A high dielectric constant solvent, however, usually has a high viscosity, which is an undesirable property for a solvent. In practical Li–MnO2 cells, mixed solvents consisting of a high- dielectric-constant–high-viscosity solvent and a lowdielectric-constant–low-viscosity solvent are usually employed. A typical mixed solvent is a mixture (50:50 or 70:30, v/v) of propylene carbonate (PC) and 1,2-dimethoxyethane (1,2-DME). It is also important that the organic solvents and solutes do not contain water. It is essential for the good long-term discharge and storage performances of a cell that the electrolyte is stable under the strong oxidizing and reducing environment at the positive and negative electrodes, respectively. An organic solvent that has a low reduction potential and a high oxidation potential, that is, with a wide potential window, should be chosen.

Separator The main functions of a separator in a cell are protection against short circuit caused by direct contact of the positive and negative electrodes and retention of electrolyte solution. A suitable separator should be used for each cell considering the characteristics of the battery. For a Li–MnO2 cell, the separator should have the following properties: (i) (ii) (iii) (iv)

low electric resistance in an electrolyte; retention of a large amount of electrolyte solution; chemical stability in an organic solvent; and durability against both oxidation by manganese dioxide and reduction by lithium.

Separators made from polyethylene or polypropylene, which can meet the above-mentioned requirements, are mainly used for lithium batteries. They are used in the form of a nonwoven cloth or a thin film with micropores, considering the property and cost requirement of each type of battery.

Some Studies on the Materials of Li–MnO2 Batteries As mentioned in the previous sections, performances of a Li–MnO2 cell are strongly dependent on the materials, especially electrodes and electrolyte materials. Some studies on Li–MnO2 cells are presented here for better understanding of this system, although the results are not directly connected with the materials used in currently available batteries.

Positive Electrode It is essential that no water exists in the positive electrode materials, but EMD before heat treatment contains about 5% of water. Heat treatment at 750  C is required to completely remove the water. By a heat treatment at 350–450  C, EMD still contains about 1–2% water, but it is believed that this water is strongly combined in the crystal and does not affect the storage characteristics of a battery. This presumption was supported by a storage test. A test cell using manganese dioxide heat treated at 375  C showed more than 90% of its initial capacity after an 11-month storage at 60  C. Commercialization of Li–MnO2 batteries started with low-power coin-type cells, but the cells soon came to be used in some applications with higher power consumptions. For high-rate as well as low-temperature uses, a large surface area is required for the manganese dioxide material. Figure 2 shows the influence of specific surface area of manganese dioxide on the operating voltage of the test cells at low-temperature pulse discharge. The operating voltage of test cells using manganese dioxide with a high specific surface area was higher than that of test cells using manganese dioxide with a low surface area.

Negative Electrode Lithium is known to form alloys with some metals and the properties of lithium metal are modified by the alloying process. For a secondary cell using metallic lithium as the negative electrode, lithium alloy has been extensively studied to improve the cycling capability. Also in a primary cell, the discharge performance of a lithium electrode is influenced by the addition of a small amount

Pulse discharge voltage at 10th cycle (V)

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide

81

1.9

1.8

1.7

1.6 0 0

20

40

30

50

Specific surface area (m2 g1)

Figure 2 The influence of specific surface area of electrolytic MnO2 on the operating voltage of the test cells at low-temperature pulse discharge. Reproduced from Nohma T, Yoshimura S, Nishio K, and Saito T (1994) Commercial cells based on MnO2 and MnO2-related cathodes. In: Pistoia G (ed.) Lithium Batteries, pp. 417–456. Amsterdam: Elsevier. 100

Capacity retention (%)

80

60

40

20

0

0.01

0.05

0.1

0.5

1

2

Al content of LiAl alloy (wt.%)

Figure 3 Capacity retention of test cells (CR15400 size) using Li and Li–Al alloys after a 40-day storage at 60  C. Reproduced with permission from Nohma T, Yoshimura S, Nishio K, and Saito T (1994) Commercial cells based on MnO2 and MnO2-related cathodes. In: Pistoia G (ed.) Lithium Batteries, pp. 417–456. Amsterdam: Elsevier.

of aluminum. Figure 3 shows the capacity retention of test cells using Li–Al alloys in which the content of aluminum ranged from 0.05% to 2%. Lithium with 0.5–2% aluminum showed good storage characteristics. However, the operating voltage after storage became lower when aluminum content was high, as shown in Figure 4. It was considered that good storage performance was achieved with a proper amount of aluminum in lithium.

Electrolytes Discharge characteristics of a Li–MnO2 cell are greatly influenced by the properties of the electrolyte. Figure 5 shows the relationship between discharge capacity and conductivity of electrolyte of test cells using some two-component solvent systems. The test cells were discharged at a high rate of 560 O. In this test, mixed solvents consisting of PC and some ethers were used. The conductivities of the electrolytes varied from 4.6 to 13.3 mS cm−1. At a high-rate condition, the conductivity of the electrolyte had a strong influence on the discharge performance. The properties of an electrolyte solution are also dependent on the solutes. Table 1 shows the physical properties of organic electrolytes using lithium perchlorate (LiClO4), lithium trifluoromethanesulfonate (LiCF3SO3), lithium tetrafluoroborate (LiBF4), lithium hexafluorophosphate (LiPF6), and lithium hexafluoroarsenide (LiAsF6) as solutes. The solvent is a PC–DME mixture. Conductivities of these electrolytes are around 10−2 S cm−1, which is about 2 orders of magnitude lower than that of aqueous electrolytes. Discharge curves of test cells at 25 and −20  C are shown in Figure 6. The LiClO4/PC–DME electrolyte, which shows a high conductivity and low viscosity at both 25 and −20  C, is one of the most typical electrolytes. Only the LiCF3SO3/PC–DME electrolyte has a high solubility of solute at −20  C and shows excellent low-temperature characteristics. Therefore, LiCF3SO3 is widely used in commercial batteries in spite of its comparatively low conductivity. LiAsF6/PC–DME has good properties as an electrolyte, but it is not used in consumer batteries because of the toxicity of arsenic.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide Pulse discharge voltage at 1st cycle (V)

82

2.2

2.0

1.8

1.6 0

0.05

0.01

0.1

0.5

1

2

Al content of LiAl alloy (wt.%)

Discharge capacity (mA h)

Figure 4 Pulse discharge voltage of test cells (CR15400 size) using Li and Li–Al. •: initial, ○: after a 40-day storage at 60  C. Reproduced with permission from Nohma T, Yoshimura S, Nishio K, and Saito T (1994) Commercial cells based on MnO2 and MnO2-related cathodes. In: Pistoia G (ed.) Lithium Batteries, pp. 417–456. Amsterdam: Elsevier.

PC1, 2-DME

100

PCEME

80 60

PC1, 2-DEE

40

PCDMM

20 0

PC1, 1-DME PCDEM 0

5 10 Specific conductivity (103 S cm1)

15

Figure 5 High-rate discharge capacity versus conductivity of electrolyte solution. Solvent: mixture of diether and propylene carbonate (1:1 volume ratio); solute: 1 mol dm−3 LiClO4. Discharge conditions: temperature¼25  C, and load¼560 O. DME, dimethoxyethane; PC, proylene carbonate. Reproduced with permission from Nishio K, Yoshimura S, and Saito T (1995). Discharge characteristics of manganese dioxide/lithium cells in various electrolyte solutions. Journal of Power Sources 55: 115–117.

Table 1

Physical properties of some organic electrolytes 25  C

LiClO4 LiCF3SO3 LiBF4 LiPF6 LiAsF6

−20  C

Specific conductivity (mS cm−1)

Viscosity (mPa s)

Specific conductivity (mS cm−1)

Viscosity (mPa s)

Solubility (mol dm−3)

12.3 5.9 8.2 17.0 16.3

1.3 1.2 1.0 1.6 1.3

7.1 3.5 5.2 7.6 7.3

3.2 2.4 2.0 3.1 2.7

1.2 >6.0 2.4 1.2 1.0

Source: Reproduced with permission from Takahashi M, Yoshimura S, Nakane I, et al. (1993) A study on electrolytes for manganese dioxide–lithium cells. Journal of Power Sources 43–44: 253–258.

Storage characteristics of Li–MnO2 test cells were studied using mixed solvents of PC and 1,2-DME and some solutes. After a 40-day storage at 60  C, which corresponds to about a 2-year storage at room temperature, test cells using LiClO4, LiCF3SO3, and LiAsF6 showed good storage characteristics with 93% capacity retention. A cell using LiPF6 showed a large decrease in the discharge capacity, and diglyme, triglyme, and tetraglyme were detected in the electrolyte after the test. These glymes were considered to be produced by the reaction of two to four molecules of DME, which was initiated by the presence of PF5.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide

83

Cell voltage (V)

4 3

LiAsF6 LiPF6 LiClO4 LiCF3SO3 LiBF4

2 1 0

0

30

(a)

60

90

120

150

Discharge capacity (mAh)

Cell voltage (V)

4 3 2 1 0 (b)

LiPF6 LiClO4 LiAsF6 LiBF4 0

30

LiCF3SO3

60

90

120

150

Discharge capacity (mAh)

Figure 6 Discharge characteristics of a coin-type test cell using various electrolytes. Solvent: mixture of propylene carbonate (PC) and 1,2-dimethoxyethane (1,2-DME) (1:1 volume ratio); discharge conditions: (a) 25  C, 0.5 mA cm−2; (b) −20  C, 0.5 mA cm−2. Reproduced with permission from Takahashi M, Yoshimura S, Nakane I, et al. (1993) A study on electrolytes for manganese dioxide–lithium cells. Journal of Power Sources 43–44: 253–258.

Negative electrode can

Negative electrode (Li) Separator Gasket Positive electrode (MnO2) Positive electrode can

Figure 7 Structure of a coin-type cell. Courtesy of Sanyo Electric Co., Ltd.

The deterioration of discharge capacity was inferred to be the result of reaction between decomposition products and the lithium electrode. The storage characteristics of a test cell using LiBF4 were found to be worst among all solutes. In this case, manganese was observed on the surface of the negative electrode by an elemental analysis after the storage test. It was considered that decomposition of manganese dioxide had occurred in the electrolyte system using LiBF4.

Structures and Assembly of Li–MnO2 Batteries Lithium–manganese dioxide batteries are classified according to their shapes and structures. Figure 7 shows the structure of a coin-type cell that was commercialized in the early stage. The positive electrode of coin-type cells consists of manganese dioxide with the addition of a conductive material and binder. The negative electrode consists of a lithium metal disk, which is pressed onto the stainless steel can. The typical separator material is a nonwoven cloth made of polypropylene, which is placed between the cloth and the anode. Cylindrical cells can be classified into two basic types: one with an inside-out structure and one with a spiral structure. The former is constructed by pressing the positive electrode mixture into a high-density cylindrical form. The latter consists of a wound, thin positive electrode and a lithium negative electrode with a separator in between. Cells with the inside-out construction are suitable for high energy density, and those with the spiral construction are suitable for high-rate drain. Figure 8 shows an inside-out structure cell with laser sealing. Figure 9 shows two types of spiral structure cells with crimp and laser sealing.

84

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide Gasket

Terminal ()

Laser seal Current collector

Positive electrode (MnO2) Separator Negative electrode (Li)

Positive electrode can

Figure 8 Structure of a cylindrical cell with an inside-out structure. Courtesy of Sanyo Electric Co., Ltd.

Terminal (+)

PTC device

Terminal ()

Laser seal

Gasket

Tab ()

Tab (+)

Tab ()

Negative electrode (Li)

Separator

Positive electrode (MnO2)

Positive electrode (MnO2)

Separator Negative electrode (Li)

Insulator

Tab () (a)

Gasket

Insulator

Can () (b)

Terminal ()

Figure 9 Structures of spiral-type cylindrical cells: (a) crimp and (b) laser sealing. PTC, positive temperature coefficient. Courtesy of Sanyo Electric Co., Ltd.

The nominal voltage of a Li–MnO2 cell is 3 V, which is about double that of other primary cells. Discharge voltages of manganese dry cell, alkaline manganese cell, zinc–silver oxide cell, mercury oxide cell, and zinc–air primary cell are about 1.5 V or lower. Therefore, the usage of a Li–MnO2 cell in place of other primary cells may result in damages. In order to avoid such misuses, Li–MnO2 cells are usually designed in different shapes. As mentioned above, lithium metal is very sensitive to water and it must be handled in a dry atmosphere. All components have to be dried before the battery assembly, because any residual water may cause deterioration of performance and pressure rise owing to reaction of lithium and water. The cell assembly process, in which lithium and other cell components after drying are handled, is conducted in a dry air room or a glove box filled with dry argon gas.

Advantages of Li–MnO2 Batteries The general advantages of the Li–MnO2 battery system are as follows: (1) High voltage Lithium–manganese dioxide cells are capable of maintaining a stable voltage of 3 V, which is about twice that of a conventional dry cell. The discharge profile is much flatter than conventional dry cell or alkaline batteries. Because of this advantage, a single Li–MnO2 cell can replace two conventional cells.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide

85

(2) High energy density Energy densities and specific energies of Li–MnO2 batteries are highest among primary batteries. This is an important merit for many applications, which enables longer usage of appliances without battery replacement. The theoretical specific energy of a Li–MnO2 cell is 856 Wh kg−1, which is much higher than that of aqueous electrolyte primary cells. Energy density values of real cells vary according to the types and sizes of the cells. Cylindrical cells with an inside-out structure have high specific energy of about 280–360 Wh kg−1 and energy density of 600–740 Wh L−1. (3) Excellent discharge characteristics As Li–MnO2 cells are capable of maintaining stable voltage levels throughout long periods of discharge, a single cell can be used as the internal power source throughout the operational lifetime of a given equipment. In addition, cells using a cylindrical, spiral structure can be used to provide high current discharge for a wide variety of applications. (4) Superior leakage resistance The use of an organic solvent rather than an alkaline aqueous solution for the electrolyte results in significantly reduced corrosion and a much lower possibility of electrolyte leakage. (5) Superior storage characteristics Lithium–manganese dioxide cells employing manganese dioxide, lithium, and a stable electrolyte exhibit a very low tendency toward self-discharge. Average self-discharge rate exhibited by Li–MnO2 cells stored at room temperature is as follows: Crimp-sealed cells: 1% per year Laser-sealed cells: 0.5% per year (6) A wide operating temperature range Because Li–MnO2 cells use an organic electrolyte with a very low freezing point, lithium batteries can be operated at extremely low temperatures. Moreover, they demonstrate superior characteristics over a wide range of temperature from cold to hot, as follows: Crimp-sealed cells: −20 to +70  C Laser-sealed cells: −40 to +85  C (7) A high degree of stability and safety As Li–MnO2 cells do not contain toxic liquids or gases, they pose no pollution problems. They are suitable for appliances for general consumers.

Specifications of Some Commercially Available Li–MnO2 Batteries Lithium–manganese dioxide batteries are available in a variety of sizes and specifications. In this section, some examples of specifications of Li–MnO2 batteries are shown. Table 2 shows specifications of commercially available coin-type cells; cylindrical, Table 2

Specifications of some commercial Li–MnO2 batteries

Structure

Coin (crimp seal)

Cylindrical, inside-out (laser seal) Cylindrical, spiral (crimp seal) Cylindrical, spiral (laser seal) Battery pack

a

CR11108. CR15H270. c CR17345. d CP3152. e 2CP4036. f 2CP3945. Courtesy of Sanyo Electric Co., Ltd. b

Model

CR1220 CR2016 CR2430 CR2450 CR17335SE CR17450SE CR23500SE CR-1/3Na CR2b CR123Ac CR17335E-R CR17450E-R CR-V3d CR-P2e 2CR5f

Nominal voltage (V) 3 3 3 3 3 3 3 3 3 3 3 3 3 6 6

Nominal capacity (mA h) 36 80 280 610 1800 2500 5000 160 850 1400 1600 2400 3300 1400 1400

Dimensions (mm)

Weight (g)

Diameter

Height

12.5 20.0 24.5 24.5 17.0 17.0 23.0 11.6 15.6 17.0 17.0 17.0 29.0(L)14.5(W)52.0(H) 34.8(L)19.5(W)35.8(H) 34(L)17(W)45(H)

2.0 1.6 3.0 5.0 33.5 45.0 50.0 10.8 27.0 34.5 33.5 45.0

0.8 1.7 4.0 6.9 17 22 42 3.3 11 17 17 23 38 37 40

86

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide

inside-out-type cells; cylindrical, spiral-type cells; and Li–MnO2 batteries. Specifications are not always the same among manufacturers, and they are sometimes changed by improvement.

coin-Type Cells Coin-type cells are mostly used for low-capacity and low current drain applications. Most of the coin-type Li–MnO2 cells have a model name like ‘CR2024’ (C: International Electrotechnical Commission (IEC) code for Li–MnO2 system; R: round; 20: diameter, 20 mm; 24: height, 2.4 mm).

Cylindrical, Inside-Out-Type Cells Figure 10 shows discharge characteristics of a CR17450SE cell at a standard discharge current. This cell shows a very flat discharge profile. In a temperature range between −40 and +20  C, discharge profiles are flat over a long period of about 10 years, although discharge time becomes shorter at +60  C, which is a severe discharge condition. The actual discharge capacity is influenced by the discharge load (discharge current) and ambient temperature, as summarized in Figure 11. This type of battery shows excellent storage characteristics, which is important for a battery used or stored for a long period. After a 50-day storage at 80  C, which corresponds to about 10-year storage at room temperature, this cell retains more than 97% of the initial capacity.

3.5 Load: 2.7 kohm (= 1 mA)

60 qC 23 qC

Cell voltage (V)

3.0

2.5

0 qC

2.0

1.5

1.0

0

10

20

30

40

50

70

60

80

90

100

110

120

Discharge time (day)

Figure 10 Discharge characteristics of a CR17450SE cell at a standard discharge current: load: 2.7 kO, temperature: 0, 23, and 60  C. Courtesy of Sanyo Electric Co., Ltd.

3000

Capacity (mAh)

2500

80 qC

2000 60 qC

1500

20 qC

23 qC

40 qC

1000 500 0

End voltage: 2.0 V 0.1

0.3

1

3

10

30

100

Discharge load (k:)

Figure 11 Relationship between the discharge load and the discharge capacity of a CR17450SE cell. Courtesy of Sanyo Electric Co., Ltd.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide

87

3000

Capacity (mAh)

2500 2000 85 qC 60 qC 1500

20 qC

23 qC

1000 500 0

40 qC 5

10

50

End voltage = 2.0 V 100

500

1000

Discharge load (:)

Figure 12 Relationship between the discharge load and the discharge capacity of a CR17450E-R cell. Courtesy of Sanyo Electric Co., Ltd.

Cylindrical, Spiral-Type Cells A cylindrical, spiral-type cell CR17450E-R can be operated in a wide range of temperature from −40 to + 85  C. This cell exhibits a very flat discharge profile even at −40  C. Influences of discharge load and temperature on the discharge capacity of a CR17450E-R are summarized in Figure 12.

Li–MnO2 Cell Packs For automatic and digital cameras, Li–MnO2 cells are used as cell packs with a nominal voltage of 3 or 6 V, as shown in Table 2.

Cell pack 2CR5 for automatic cameras The 2CR5 consists of two CR15400 cells connected in series and the nominal voltage is 6 V. A Li–MnO2 cell with a spiral structure is suitable for a fully automatic camera because high output current is needed for film winding, quick charging of strobe light, etc. This battery was developed in the mid-1980s as a user-replaceable lithium battery. At that time, most cylindrical lithium cells were equipped in appliances by the manufacturers. This cell pack is designed with special safety measures. Individual cells of 2CR5 are encapsulated in a plastic container and designed in an asymmetric shape that will prevent misuse. When a 2CR5 pack is short-circuited, a positive temperature coefficient (PTC) thermistor prevents the battery from overheating by substantially increasing the resistance. When the short circuit is removed, the 2CR5 operates normally. The battery case is usually equipped with the PTC thermistor, and hence the ability of the 2CR5 to deliver high current is not impeded. Heating of a Li–MnO2 battery by unusually high current or other reasons brings it to a dangerous state. The battery becomes especially dangerous and may give rise to firing or explosion when the temperature exceeds the melting point of lithium (180  C). However, upon usual temperature rise, the polyolefin separator melts and shuts down the current flow by closing its micropores. The battery cap is equipped with a safety vent and operates under high pressure.

Cell pack CR-V3 for digital cameras A CR-V3 cell pack, which consists of two CR14500 cells in parallel, has a nominal voltage of 3 V and is designed mainly for use in digital cameras. A CR14500 cell has the same dimensions as an AA-size cell; hence, this pack is compatible with two AA-size alkaline manganese cells in series. The average operating voltage of a Li–MnO2 cell is about double that of other primary cells, such as alkaline manganese cells and dry cells. If a lithium cell is used in place of a low-voltage cell, it might cause damages to the electronic circuit. In order to avoid this kind of misuse, dimensions of most of the Li–MnO2 cells are different from those of other cell systems. The individual cell of the CR-V3, namely CR14500 (diameter: 14 mm, height: 50 mm), however, is an exception, which has almost the same size as an AA cell. Therefore, the package of a CR-V3 pack is designed so that it is not easily disassembled. A digital camera needs both a large output power and a high energy density for its power source. It is necessary to make electrodes thinner and longer to achieve high output power with a battery of the same size, but this generally causes lowering of the energy density. As a result of optimization of thicknesses of electrodes, a CR-V3 exhibits good high-rate and low-temperature discharge characteristics without sacrificing the energy density. It was reported that a CR-V3 pack is more advantageous in high-rate, pulse, and low-temperature characteristics than alkaline manganese cells.

88

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Manganese Dioxide

Applications Lithium–manganese dioxide batteries are used in a wide variety of applications according to their size and performances. Coin-type cells are used in applications that generally need low output power. In the 1980s, they made great contributions to the popularization of electronic watches and small-sized calculators. They are used in electronic diaries, PC cards, card-type radios, light-emitting diode (LED) lights, and remote controls of many appliances. Memory backups in many personal computers and office automation equipment are also important applications. Cylindrical cells with an inside-out structure are used for high energy density, low current drain applications, such as electronic meters and memory backups. In these applications, battery life up to 10 years and superior reliability are required. Cylindrical cells with a spiral structure are used for high output power applications. They are used in an exposure meter for cameras and as power sources for fully automatic cameras and digital cameras.

Further Reading 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

Bowden, W.; Bofinger, T.; Zhang, F.; et al. New manganese dioxides for lithium batteries. Journal of Power Sources 2007, 165, 609–615. Ikeda, H. Lithium–manganese dioxide cells. In Lithium Batteries; Gabano, J.P., Ed.; Academic Press: London, 1983;; pp 169–210. Ikeda H, Saito T, and Tamura H (1977) Lithium–manganese dioxide cell. Electrochemistry 45: 314–318 (in Japanese). Ikeda H, Ueno S, Saito T, Nakaido S, and Tamura H (1977) Lithium–manganese dioxide cell. Electrochemistry 45: 391–395 (in Japanese). Johnson, C.S. Development and utility of manganese oxides as cathodes in lithium batteries. Journal of Power Sources 2007, 165, 559–565. Kurimoto, H.; Suzuoka, K.; Murakami, T. High surface area electromigration damage for 3 V Li-cell cathodes. Journal of the Electrochemical Society 1995, 142, 2156–2162. Manev, V.; Ilchev, N.; Nassalevska, A. The lithium–manganese dioxide cell. I. Oxygen and water release during the thermal treatment of MnO2. Journal of Power Sources 1989, 25, 167–175. Nagao, M.; Pitteloud, C.; Kamiyama, T.; et al. Further understanding of reaction processes in electrolytic manganese dioxide electrodes for lithium cells. Journal of the Electrochemical Society 2005, 152, E230–E237. Nishio, K.; Furukawa, N. Practical batteries. In Handbook of Battery Materials; Besenhard, J.O., Ed.; Wiley-VCH: Weinheim, 1999;; pp 19–61. Nishio, K.; Yoshimura, S.; Saito, T. Discharge characteristics of manganese dioxide/lithium cells in various electrolyte solutions. Journal of Power Sources 1995, 55, 115–117. Nohma, T.; Yoshimura, S.; Nishio, K.; Saito, T. Commercial cells based on MnO2 and MnO2-related cathodes. In Lithium Batteries; Pistoia, G., Ed.; Elsevier: Amsterdam, 1994;; pp 417–456. Ohzuku, T.; Kitagawa, M.; Hirai, T. Electrochemistry of manganese dioxide in lithium nonaqueous cell I. X-ray diffractional study on the reduction of electrolytic manganese dioxide. Journal of the Electrochemical Society 1989, 136, 3169–3174. Pistoia, G. Some restatements on the nature and behavior of MnO2 for Li batteries. Journal of the Electrochemical Society 1982, 129, 1861–1865. Shao-Horn, Y.; Hackney, S.A.; Cornilsen, B.C. Structural characterization of heat-treated electrolytic manganese dioxide and topotactic transformation of discharge products in the Li–MnO2 cells. Journal of the Electrochemical Society 1997, 144, 3147–3153. Takahashi, M.; Yoshimura, S.; Nakane, I.; et al. A study on electrolytes for manganese dioxide–lithium cells. Journal of Power Sources 1993, 43–44, 253–258. Thackeray, M.M.; Rossouw, M.H.; de Kock, A.; et al. The versatility of MnO2 for lithium battery applications. Journal of Power Sources 1993, 43–44, 289–300. Urushihara, K.; Tanaka, A.; Nishitani, T.; Morita, S.; Fujimoto, M. High-power primary MnO2/lithium battery CR-V3. Sanyo Technical Review 2002, 34 (1), 106–110.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium-Carbon Fluoride Battery Guiming Zhonga and Weimin Zhaob, aDalian Institute of Chemical Physics Chinese Academy of Sciences, Dalian, Liaoning, China; bBinzhou University, Binzhou, Shandong, China © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is a update of R. Yazami, H. Touhara, PRIMARY BATTERIES – NONAQUEOUS SYSTEMS | Lithium–Polycarbon Monofluoride, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 93–99, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00114-3.

1 2 3 3.1 3.2 3.3 4 4.1 4.2 5 6 References

Introduction Fundamental reaction principles Carbon fluoride Synthesis of CFx Structure and physicochemical properties of CFx Design strategies of CFx structure Electrolytes for Li/CFx battery Lithium salts and solvents Low-temperature electrolytes for Li/CFx battery Cell designs and main applications Outlook

90 90 91 91 92 93 95 95 95 96 97 97

Abstract Li-carbon fluoride battery represents a groundbreaking achievement as the world’s inaugural lithium battery for consumer products, and renowned for its exceptional energy density (
1000 Wh kg−1), long shelf life (
10 years), wide operating range (−60  C–120  C), and high voltage (2.4–3.0 V). Its appeal spans diverse civil and military applications. This chapter focuses on the battery’s fundamental reaction principles, design strategies for carbon fluorides and electrolytes, even addressing low-temperature operation. By unraveling these mechanisms, insights into its unique attributes and potential applications are provided.

Glossary Lithium polysulfide the compound between lithium and sulfur, i.e., Li2Sx (x>1) Shuttle effect the transport of lithium polysufides through the separator between sulfur cathode and lithium anode Lean electrolyte low amount of electrolyte per unit mass of sulfur, i.e., low electrolyte per sulfur (E/S)

Key points

• • • • • •

The chapter expects to elucidate the critical issues for the Li/CFx battery: underlying reaction principles structure and C-F bonding nature of CFx high-energy-density design strategies development of rate capability electrolytes for low-temperature application

Abbreviations and acronyms CFx DME DMI GBL

Carbon fluorides 1,2-Dimethoxyethane 1,3-Dimethyl-2-imidazolidinone Gamma-butyrolactone

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00110-9

89

90

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium-Carbon Fluoride Battery

GIC LGE MB Me2O OCV PC S SFCF TTE z DG

1

Graphite intercalation compound Liquefied gas electrolytes Methyl butyrate Dimethyl ether Open-circuit voltage Propylene carbonate Solvent molecule with a salvation number z Sub-fluorinated carbon fluoride 1,1,2,2-Tetrafluoroethyl-2,2,3,3-tetrafluoropropyl ether Salvation number Change in the Gibbs free energy

Introduction

In pursuit of high-energy-density power sources, Li/F2 system emerges as the ideal prospect with the highest theoretical voltage of 5.92 V and a significant theoretical mass-specific capacity. However, it’s practical realization is impeded by the inherent reactivity and corrosiveness of fluorine. Many kinds of fluorides have been explored, yet none yielded satisfactory cell performance until the advent of carbon monofluoride as a cathode material for primary lithium battery, employing an aprotic solvent–electrolyte system. The Li-carbon fluoride battery (Li/CFx) battery was initially commercialized by Matsushita Electric Co. Ltd. in 1971, guided by the visionary insight of N. Watanabe (Kyoto University, Japan) and M. Fukuda (Matsushita Electric Co. Ltd. Japan).1 This milestone hinged on the application of coke-based carbon fluoride with F/C ratio of 1 as the cathode. The invention attracted much attention, greatly promoted the application of carbon materials, graphite intercalation compounds (GICs), and further catalyzing the evolution of high-energy-density lithium batteries in the subsequent years. Li/CFx battery holds the distinction of delivering the highest-energy-density among commercial lithium batteries until the recent emergence of the commercialized Li-sulfur battery. Nevertheless, it exhibits exceptional volume-energy-density and rate performance among high-energy-density (
500 Wh kg−1) primary lithium batteries, thus captivating interest for the applications that demand both high-energy-density and promising power density. The inherent versatility of the CFx structure further renders the battery capable of achieving an ultrahigh-energy-density of
1000 Wh kg−1. Anchored by the advantages of high-energy-density, long shelf life (
10 years), wide operating temperature range (− 60–120  C), and elevated operational voltage window (2.4–3.0 V), the Li/CFx battery exhibits greatly appealing for numerous civil and military domains. The evolution of the Li/CFx battery prominently depends on the development and utilization of novel CFx structures, electrolyte, binder, conductive agent, etc. The selection of materials and configurations is tailored to the specific application scenarios, guided by fundamental mechanisms. This chapter will focus on the underlying principles in developing Li/CFx battery with ultrahigh-energy-density, ultrahigh-power-density, or operated under extreme conditions. The chapter commences with an exploration of the reaction mechanism intrinsic to Li/CFx, followed by a close examination of the pivotal attributes characterizing CFx and its compatible electrolytes. Additionally, intricate facets of commercial cell design strategies is also depicted, offering a comprehensive tapestry of insights.

2

Fundamental reaction principles

Li/CFx battery is basically composed of a metallic lithium anode, a CFx cathode, and a nonaqueous electrolyte. Among, these elements, the CFx materials and the electrolyte are the pivotal determinants of performance. N. Watanabe and his co-workers firstly investigated and proposed the reaction mechanism of Li/CFx, which is expressed as Eq. [I],1 suggesting the formation of crystalline LiF and carbon upon discharge. CFx + x Li ! C + x LiF

[I]

However, while this equation calculates a Gibbs free energy (DG298) of −410 kJ mol−1 and a theoretical voltage of 4.57 V, these values markedly surpass the observed open-circuit voltage (OCV) values of 3.1–3.8 V. Stanley Whittingham then suggested the existence of an intermediate state of a non-stoichiometric graphite intercalation compound CLixF to elucidate the discrepancy.2 This hypothesis posits the initial formation of this intermediate, which subsequently undergoes disproportionation to form LiF and carbon. The overall reaction mechanism proposed is shown as Eqs. [II] and [III]. x Li + CF ! CLixF

[II]

! C + x LiF

[III]

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium-Carbon Fluoride Battery

91

Fig. 1 Understanding on the reaction mechanism of Li/CFx battery.

Later, Watanabe et al. embarked on a comprehensive investigation of the Gibbs free energies associated with Li+ solvation across diverse solvents. Their results unveiled a striking correlation: an elevated discharge plateau was manifestly aligned with higher solvation energies of Li+, thereby underscoring the discernible influence of solvents on the discharge process. They proposed the lithiation reaction of CFx carried out through an intermediate of solvated phase CF−Li+zS, as expressed in Eqs. [IV]–[VII].3 Anodic reaction: Li + zS ! Li+zS + e−

[IV]

Cathodic reaction: CF + Li+zS + e− ! CF− Li+zS

[V]

−

Overall reaction: CF + Li + zS ! CF Li zS

[VI]

! C + LiF + zS

[VII]

+

where S is a solvent molecule with a solvation number z. Consequently, the electrode potential is determined by the intermediate phase that is formed on the edge plane of layered CFx as a diffusion layer, and the OCV values correspond to the change in the Gibbs free energy of Eq. [V]. The mechanism finds empirical validation in the conspicuous impact of solvents on both OCV and discharge plateau. Nonetheless, it worth noting that the presence of CF− Li+zS intermediate within organic electrolyte, encounters problems as extrapolated to solid-state batteries, which precludes the involvement of reaction pathways predicated upon the formation of solvated intermediate phase. Application of advanced techniques and calculations deepened the understanding (Fig. 1). In 2021, Leung et al. proposed an intermediate phase associated with a CFx-edge propagation Li-insertion mechanism based on density functional theory calculations results.4 Li intercalates at the zigzag edge boundary between fluorinated and defluorinated regions. It is suggested that such intermediate undergoes an edge-propagation mechanism rather than a bulk-phase reaction pathway. Recent works have ventured even deeper, by utilizing cryogenic scanning transmission electron microscopy and x-ray photoelectron spectroscopy. The results depicted a three-region discharge mechanism in Li/CFx systems through the depth of discharge as organic electrolyte was used.5 It is also evidenced that anion in the organic electrolyte played an important role in the reaction,6 affecting the breakdown of C-F bonds. Notably, recent work of high-resolution 19F NMR (nuclear magnetic resonance) spectroscopy unveiled that a scant residue of fluorine retained in carbon layers during lithiation,7 leading to the formation of carbon fluoride with very low fluorination level. Beyond the production of intermediate during the discharge process of CFx, researchers have illuminated the significance of LiF. The intrinsic nature of LiF as an electronic insulator and non-ionic conductor, introduce the potential for the higher resistance upon LiF is produced. In 2009, Sheng S. Zhang et al. proposed a “core-shell” model to illustrate the morphological evolution occurring during discharge and its subsequent effect on the electrochemical profile.8 The results suggested a “shrinking core” of CFx, and a product shell consisting of GIC intermediate, carbon and LiF. The progressive growth of the shell caused an increasing charge-transfer resistance. Further investigations employing in-situ NMR and TEM (transmission electron microscope) techniques demonstrated that the resistance increased as surface of particles was saturated with LiF nuclei and grains.7,9 Results also indicated that the morphological evolution of LiF as using solid-state electrolyte is different. Nano-sized or amorphous LiF covering on the particles were mainly formed in solid-state battery. In contrast, spherical LiF grains with LiF nuclei peppered on the surface were favored in an aprotic solvent–electrolyte system. This distinct phenomena is conceivably rooted in differing reaction mechanisms, and the confined interplay between active material and solid-state electrolytes.

3 3.1

Carbon fluoride Synthesis of CFx

Carbon fluoride is the most important component governing the performance of the Li/CFx battery. Typically, a one-step fluorine-gas fluorination method is harnessed for the production of carbon fluorides (Fig. 2),10 utilizing a fluorine-nitrogen gas mixture to decrease the concentration of fluorine thus enhancing the safety. Fluorination temperature, reaction duration, gas flow

92

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium-Carbon Fluoride Battery

Fig. 2 Schematic of fluorination device to synthesize carbon fluoride materials. 1. reaction vessel; 2. reaction tube; 3. supporting rod of reaction vessel; 4. cooling fan; 5. fin made of aluminum; 6. support pole of reaction tube; 7. flange; 8. toothed wheel; 9. chain; 10. motor; 11. bearing (mechanical seal); 12. thermocouple for measurement; 13. thermocouple for controller; 14. furnace; 15. flow meter for argon; 16. orifice-type flow meter for fluorine; 17. regulator cock; 18. Hg manometer; 19. seal; 20–22. F2 absorber (soda lime pellets); 23. HF absorber (NaF pellets heated to 100  C); 24. F2 gas; 25. Ar gas; 26. recorder; 27. temperature controller. Adapted with permission from Ref. Kita, Y.; Watanabe, N.; Fujii, Y. Chemical Composition and Crystal Structure of Graphite Fluoride. J. Am. Chem. Soc. 1979, 101, 3832–3841. Copyright 2009 American Chemical Society.

rate, and pressure are the decisive factors for the fluorination level and structure of the resultant product. Illustratively, when considering graphite micro sheets, the fluorination reaction of the material initiates at an approximate temperature of 300  C. And the optimum fluorination temperature is between 450– 500  C, while venturing beyond the threshold of 550  C precipitates an escalation in CF2 content on the surface and even cause partial gaseous C-F compounds during reaction. It is noticed that the optimum fluorination temperature depends on the carbon source. The method is facile to synthesize the CFx (x ¼ 0.2– 1) materials and stages 1 to 2 fluorine-graphite intercalation compounds, readily attainable by regulating temperature and the reaction duration. It needs to be mentioned that the F/C ratio of CFx can be swiftly estimated through the comparative pre- and post-fluorination weighting of materials during production. The deployment of 13C nuclear magnetic resonance spectroscopy and elemental analysis yields a more accurate evaluation. Alternative volatile fluorides including HF, IF5, SbF5, or their blends, have been used as fluorine sources and catalyst in the fluorine gas to the synthesis of CFx materials in a mild reaction condition.11 Fluorination of carbon can be achieved even at an ambient temperature, propelled by the presence of volatile fluoride in the reaction atmosphere. This reaction gives rise to the formation of low-temperature carbon fluorides. CFx materials with low x value of 0.05, stages 1–4 fluorine-graphite intercalation compounds, and that with even high F/C ratio of
0.8 have been successfully synthesized by the method.

3.2

Structure and physicochemical properties of CFx

The distinctive attributes of CFx as a cathode material are profoundly associated with the nature of carbon source, fluorination level (value of x), and structures of bulk-phase and surface. Among, these, fluorine-graphite intercalation compounds with F/C ratio of less than 0.1 unveil an extraordinary electrical conductivity of 2  104 S cm−1 than pristine carbon materials, in start contrast to the insulator of graphite fluoride of CF1 characterized by a very low conductivity of 10−12 S cm−1 or even lower. The insulating property stems from the formation of strong covalent C-F bonds with a pronounced charge localization.12 The long-term ordering and the local structure including the C-F bonding natures are two essential elements accounting for the properties. Take graphite fluorides that have been extensively investigated as a prime illustration, where graphite fluoride of CF1 is a stage-1 GIC compound and can be expressed as C1F. Many works considered that it had the structure of the chair-type crystal with AB stacking (Fig. 3A).13 For graphite as the pristine carbon material, stage-1 C1F can be obtained at 600–640  C, while stage-2 C2F forms at a lower-temperature of 350–400  C. In both C1F and C2F compounds, strong covalent C-F bonds with sp3 hybridization of the carbon atoms induce charge localization, leading to very poor electric conductivity. Another structural feature that determines the physical properties and performance of CFx is C-F bonding nature. C-F bonds are essentially ionic as x < 0.1 in CFx, and are covalent with higher x. It is estimated that C-F bond length decreases from 1.47 A˚ for C16F structure to 1.38 A˚ for C1F and C2F structures (Fig. 3B). Besides, there is a type of C-F bonds called “semi-ionic” CF bonds in CFx. The bonds are essentially covalent but with delocalization due to the surrounding conjugated C-C bonds (Fig. 3C), and the bond length is slightly elongated (1.40 A˚ )

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium-Carbon Fluoride Battery

93

Fig. 3 (A) Crystalline structure of CF1 with chair-type crystal and AB stacking. Schematics of (B) covalent and (C) “semi-ionic” C-F bonds. Silver and gray balls indicate fluorine and carbon atoms.

in comparison to that in C1F and C2F structures (1.38 A˚ ).14 Notably, the content of “semi-ionic” C-F bonds is highly important to achieve the high-power-density CFx materials without compromised energy-density performance.15

3.3

Design strategies of CFx structure

The quest for the ultimate energy and power density stands as dual guides in the evolution of the Li/CFx battery. For the highenergy-density CFx materials, two potent strategies surface: elevating the F/C ratio and modulating the operating plateau. Harnessing the intrinsic electrochemical activity of dCF2 in CFx,7 raising the dCF2 group content on the surface or making defects within pristine carbon materials (e.g., hard carbon) to foster dCF2 exhibit effectiveto achieve the high-energy-density CFx materials with high F/C ratio of >1. CFx with F/C ratio of 1.3 has been prepared using boron-doped graphene source,16 delivering a practical specific capacity of 1204 mAh g−1 and energy density of 2974 Wh kg−1 calculated based on the active material (Fig. 4). According to the Eq. [V]. CFx plays a critical role in determining the change in the Gibbs free energy beyond the solvent and the solvation structure. Regulating the structure and F/C of CFx has exhibited great effectiveness to raise the discharge plateaus. Lowering the F/C ratios leads to an elevated discharge voltage. Many works exposed that CFx materials based on hard carbon or graphene pristine materials displayed higher discharge voltage than highly-ordered carbon counterparts of graphite, carbon nanotubes. This is likely due to the structural discrepancy thus the different Gibbs free energy. Markedly, fluorinated graphite and graphene with F/C ratio of 1.0 delivered the operating voltage of 2.4 V and 2.7 V at a low current density,1,15 respectively.

Fig. 4 Discharge profiles of Li/CF1.3 cells. 1C indicates 1.0 A g−1. Reproduced with permission from Ref. Wang, K.; Feng, Y.; Kong, L.; Peng, C.; Hu, Y.; Li, W.; Li, Y.; Feng, W. The Fluorination of Boron-Doped Graphene for CFx Cathode with Ultrahigh Energy Density. Energy Environ. Mater. 2022, e12437. Copyright 2022 John Wiley and Sons.

94

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium-Carbon Fluoride Battery

For the high-power-density application, it is significant to enhance the electronic conductivity and the interfacial charge transfer kinetics. Due to the insulating nature of CFx (x
1) materials, the stage-1 sub-fluorination CFx materials are primarily investigated for the high-power-density Li/CFx battery. The present of “semi-ionic” C-F bonds enhanced the conductivity thus rate capability. It is disclosed that F/C ratio of 0.8–0.9 is the optimum value for high-rate CFx. The CF0.8 of fluorinated graphene even can achieve a remarkable power density of >20, 000 W kg−1 (calculated based on cathode active mass), with a high capacity of 470 mAh g−1 and discharge plateau of 2.2 V at current density of 10 A g−1 (Figs. 5 and 6).15 The work uncovered a profound revelation: the proportion of “semi-ionic” CF bond in all C-F bonds presents a critical impact on rate performance of the CFx. The proportion of “semi-ionic” CF up to 39% of the total fluorine emerges as important threshold. Many types of carbon fluorides and composites based on different carbon sources have been developed for high-power-density Li/CFx battery (Fig. 6). Beyond, the generation of chemical heat is a consequential challenge, particularly in scenarios demanding high-power application.22 As a result of the remarkable change in bond energy between CFx and produced LiF, the high-rate discharging battery would cause a serious temperature rise and make the battery explosion. What’s more, the lower shelf life for sub-fluorination CFx materials should be noticed. Other strategies, such as mixing CFx and oxides (e.g., MnO2, silver vanadium oxides) have been proposed to improve the power capability, also reduce the heat generation.17,23

Fig. 5 Rate capability discharge profiles of Li/CF0.8 (fluorinated graphene) cells by Ref. 15, used under CC BY 4.0/reproduced from original.

Fig. 6 Ragone plots of Li/CFx cells based on several high-power-density CFx materials and composites.15–21 Here, HT and BM indicate hydrothermal and ball-milling treatment.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium-Carbon Fluoride Battery

4

95

Electrolytes for Li/CFx battery

As depicted in the “Fundamental Reaction Principles” section, the solvents in organic electrolytes play an essential role in the lithiation reaction of CFx. Adjusting the solvents would modulate the nature of intermediate and the de-solvation energy, thus possibly elevating the discharge plateau (i.e., energy-density) and enhancing the rate and low-temperature performance.

4.1

Lithium salts and solvents

LiClO4. LiBF4 and LiPF6 are three general lithium salts for Li/CFx battery, the selection of which signifies a delicate balance between conductivity, solubility, and safety. LiClO4 exhibits superior Li+ conductivity and solubility. However, the strongly oxidizing potential raises concerns over reactivity with organic species under high-temperature or high current density. LiPF6 for Li/CFx battery has problem of the decomposition with traces of water to form undesirable HF, which is very detrimental to the shelf life. LiBF4 presents advantages of less toxicity and higher safety, but is restrained by the property of moderate ionic conductivity. Despite that, recent work revealed that the moderate donicity of BF−4 anion facilitates the breakdown of C-F bonds thus raising the capacity in comparison to other salts (Fig. 7A).6 Several new lithium salts were developed in recent years to enhance the dissociation of lithium ion, such as bis(trifluoromethane)sulfonimide lithium salt (LiTFSI) and bis(fluorosulfonyl)imide Lithium (LiFSI). Besides, co-dissolution of multiple lithium salts in solvents has been verified to improve the performance Li/CFx battery. For the solvents, propylene carbonate (PC) was the first focus as it was observed that lithium could be electrodeposited from a solution of LiClO4 in PC in 1958. The low freezing temperature, high dielectric constant, and stability with lithium made it a preferred solvent. Afterwards, ethers like 1.2-dimethoxyethane (DME) and gamma-butyrolactone (GBL) were extensively investigated as alternative candidates for their low viscosity and promising high ionic conductivity, in despite of the safety concern over the high vapor pressure of diethyl ether. Additionally, some solvents have been developed for raising the discharge voltages. It is shown that electrolyte of LiFSI in 1,3-dimethyl-2-imidazolidinone (DMI) and DME (1:1) achieved a higher plateau of 0.2 V than electrolyte of LiPF6 in EC:DEC (Fig. 7B).24

4.2

Low-temperature electrolytes for Li/CFx battery

Enabling robust low-temperature performance of Li/CFx is highly crucial for aerospace applications. The low freezing points, low-viscosity, and low de-solvation energy for reduced charge-transfer resistance are the key points. An electrolyte consisting of 1 M LiBF4 with DME:PC (4:1) with advantage of low-viscosity and freezing temperature was first reported by NASA’s Jet Propulsion Laboratory for low-temperature application, which delivered a high capacity of >600 mAh g−1 at C/40 rate under −40  C. The electrolyte with a crown ether addition has been further explored,25 exhibiting higher conductivity through a more compact Li+/15crown-5 complexes than the complex of Li+ with four solvent molecules, and yielded solid electrolyte layers, which thus increases the discharge capacity at −50 С three- or fourfold in an electrolyte of 1 M LiBF4 in GBL. Afterwards, researchers focus on the weak solvating solvents to reduce the de-solvation energy. E.g., a weak solvating methyl butyrate (MB) or 1.1.2.2-tetrafluoroethyl2.2.3.3-tetrafluoropropyl ether (TTE) combined with DME solvent (Fig. 8A and B),26 resulting in enhanced high-rate and

Fig. 7 Comparative electrochemical profiles of Li/CFx cells using different lithium salts and solvents. Reproduced with permission from Ref. Fu, A.; Xiao, Y.; Jian, J.; Huang, L.; Tang, C.; Chen, X.; Zou, Y.; Wang, J.; Yang, Y.; Zheng, J. Boosting the Energy Density of Li || CFx Primary Batteries Using a 1,3-Dimethyl2-Imidazolidinone-Based Electrolyte. ACS Appl. Mater. Interfaces 2021, 13, 57470–57480. Copyright 2009 American Chemical Society, and Ref. Wang, X.; Song, Z.; Wu, H.; Yu, H.; Feng, W.; Armand, M.; Huang, X.; Zhou, Z.; Zhang, H. Anion Donicity of Liquid Electrolytes for Lithium Carbon Fluoride Batteries. Angew. Chem. Int. Ed. 2022, 61, e202211623.

96

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium-Carbon Fluoride Battery

(A)

(C)

40°C 50°C 60°C

10 Voltage (V)

Conductivity (mS cm1)

2

PM-1 PM-2

1

PM-3

70°C

1.5

1

PM-4 PM-5 COE

0.1 –80

–60

–40 –20 0 Temperature (°C)

20

40

0

250

500

750

Specific Capacity (mAh g-1)

(B)

(D)

3.5

1

0.1 1 M LiBF4-Me2O-PC 1 M LiBF4-DME-PC 1 M LiBF4-Me2O 1 M LiBF4-DME 1 M LiBF4-PC

0.01

Discharge Voltage (V)

10

Conductivity (mS/cm)

0.1 C

0.5

1 M LiBF4-Me2O-PC 10 mA/g

3.0

55 °C 23 60 70

2.5

2.0

1.5 –80

–60 –40

–20 0 20 40 Temperature (°C)

60

0 100 200 300 400 500 600 700 800 9001000 Discharge capacity (mAh*g-1)

Fig. 8 (A, B) The conductivities and (C, D) low-temperature performance of Li/CFx cells using PM electrolytes (1 M LiFSI in PC-MB co-solvent, PM-1 indicates PC: MB ¼ 1:1) and liquefied-gas-based electrolytes (1 M LiBF4 in Me2O-PC). Reproduced with permission from Ref. Yin, Y.; Holoubek, J.; Liu, A.; Sayahpour, B.; Raghavendran, G.; Cai, G.; Han, B.; Mayer, M.; Schorr, N. B.; Lambert, T. N. et al. Ultralow-Temperature Li/CFx Batteries Enabled by Fast-Transport and Anion-Pairing Liquefied Gas Electrolytes. Adv. Mater. 2023, 35, 2207932. Copyright 2023 John Wiley and Sons, and Ref. Fang, Z.; Yang, Y.; Zheng, T.; Wang, N.; Wang, C.; Dong, X.; Wang, Y.; Xia, Y. An all-Climate CFx/Li Battery with Mechanism-Guided Electrolyte. Energy Stor. Mater. 2021, 42, 477–483. Copyright 2021 Elsevier.

low-temperature performance under −70  C, albeit with modest ionic conductivity of 3.5 mS cm−1 between −70  C and 60  C (Fig. 8C). And the weak solvating energy of Me2O would reduce the charge-transfer resistance. The Li/CFx battery even delivered a remarkable capacity of 780 mAh g−1 at 10 mA g−1 under −60  C (Fig. 8D). What’s more, some additives are developed to regulate the formation of LiF and thus the rate performance. A gaseous electrolyte additive BF3 has been proposed recently,28 which can react with LiF to form LiBF4 thus enhancing the interfacial process. At the additive concentration of 0.01 M, the CF1.15 cathode delivered a discharge capacity of 415.5 mAh g−1 with an outstanding power density of 23,040 W kg−1 at a discharge rate of 15C.

5

Cell designs and main applications

The present commercialized Li/CFx battery encompass diverse configurations, such as pouch, Cylindrical, pin, and coin-types (Fig. 9), each tailored to specific application conditions including the energy and power densities, working currents, and operation temperatures. Among these types, the pouch-type Li/CFx battery has garnered increasing prominence for future applications, owing to the advantages of lightweight and inherent flexibility. In a typical Li/CFx cell, the cathode includes carbon fluoride as the active material, carbon black, carbon nanotubes or graphene as the conductive agent, and polyvinylidene fluoride (PVDF) as the binder. LiClO4 or LiBF4 is generally applied as the electrolyte salts for Li/CFx battery, the concentration of which is 0.5– 1.5 M in a binary mixture of carbonate solvent PC and ether DME. The ratio of cyclic carbonate and ether is adjusted according to the requirements of battery discharge current and operating temperature.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium-Carbon Fluoride Battery

97

Fig. 9 Structures of Li/CFx cells and Battery pack. (A) SNL-Built 18,650 Li/(CFx)n Cell; (B) Photograph showing the appearance and dimensions of a spiral-wound cell; (C) Battery pack complete with lid.

In constructing high-performance CFx electrodes, discharge expansion, coverage of LiF product on the surface upon discharge are two critical challenges that should be addressed. Dramatic temperature rise is another serious issue for the high-power-density operation. Constructing a multi-dimensional conductive network with rich pore structure to reserve space for the products thus will reduce the discharge expansion. The carbon nanotubes and carbon black can be evenly distributed on the surface and form a three-dimensional conductive network with better conductivity between active material particles; Simultaneously, the electrode structure also facilitates the ionic transfer process and lifts heat dissipation, which is very important to improve the performance under high-current-density or at low-temperature operation. Besides, the binder with certain ionic conductivity such as guar gum, carboxymethyl chitosan, and agarose are utilized recently to solve the problems by achieving a higher adhesion. The containing polar groups of - COOH or - NH2 in these compounds can realize a strong adhesion between the active materials and the fluid collector, reducing the negative impact of discharge expansion. Recently, carbon-coated aluminum foil as the positive current collector is applied to reduce the overheating problem caused by the increasing internal resistance of the point battery.

6

Outlook

The exploration of high-energy-density and high-power-density power systems remains an ongoing endeavor. Li/CFx as one of the highest-energy-density primary battery has been extensively investigated. This chapter expects to establish a fundamental comprehension of the Li/CFx battery. The intrinsic reaction mechanism of the battery involving the formation and disproportion of the intermediate offers the essential theoretical foundation to construct CFx, electrolyte, etc. for high-performance battery. Exploring new types of carbon materials for CFx is still very important beyond the regulation of the F/C ratio and the content of “semi-ionic” C-F bonds. Any possibility to change the local structure of C-F bonds would be appreciated. Besides, as an essential component to the intermediate product, solvent even possibly plays a decisive role in the discharge voltage and low-temperature performance. Recent application of weak solvating solvents leads to a remarkable low-temperature performance. Furthermore, the intriguing possibility of realizing a rechargeable Li/CFx battery has garnered significant attention, spurring a demand for a deeper understanding on the microscopic process involved. This pursuit necessitates further explorations to unravel intricate mechanisms and inspire innovative approaches to material development and cell design.

References 1. Watanabe, N.; Fukuda, M. Primary Cell for Electric Batteries; U. S. Patent 3,536,532, 1970; p. 27. 2. Whittingham, M. S. Mechanism of Reduction of the Fluorographite Cathode. J. Electrochem. Soc. 1975, 122, 526. 3. Watanabe, N.; Nakajima, T.; Hagiwara, R. Discharge Reaction and Overpotential of the Graphite Fluoride Cathode in a Nonaqueous Lithium Cell. J. Power Sources 1987, 20, 87–92. 4. Leung, K.; Schorr, N. B.; Mayer, M.; Lambert, T. N.; Meng, Y. S.; Harrison, K. L. Edge-Propagation Discharge Mechanism in CFx Batteries—A First-Principles and Experimental Study. Chem. Mater. 2021, 33, 1760–1770. 5. Sayahpour, B.; Hirsh, H.; Bai, S.; Schorr, N. B.; Lambert, T. N.; Mayer, M.; Bao, W.; Cheng, D.; Zhang, M.; Leung, K.; et al. Revisiting Discharge Mechanism of CFx as a High Energy Density Cathode Material for Lithium Primary Battery. Adv. Energy Mater. 2022, 12, 2103196.

98

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium-Carbon Fluoride Battery

6. Wang, X.; Song, Z.; Wu, H.; Yu, H.; Feng, W.; Armand, M.; Huang, X.; Zhou, Z.; Zhang, H. Anion Donicity of Liquid Electrolytes for Lithium Carbon Fluoride Batteries. Angew. Chem. Int. Ed. 2022, 61, e202211623. 7. Zhong, G.; Chen, H.; Cheng, Y.; Meng, L.; Liu, H.; Liu, Z.; Zheng, G.; Xiang, Y.; Liu, X.; Li, Q.; et al. Insights into the Lithiation Mechanism of CFx by a Joint High-Resolution 19F NMR, In Situ TEM and 7Li NMR Approach. J. Mater. Chem. A 2019, 7, 19793–19799. 8. Zhang, S. S.; Foster, D.; Wolfenstine, J.; Read, J. Electrochemical Characteristic and Discharge Mechanism of a Primary Li/CFx Cell. J. Power Sources 2009, 187, 233–237. 9. Ding, Z.; Yang, C.; Zou, J.; Chen, S.; Qu, K.; Ma, X.; Zhang, J.; Lu, J.; Wei, W.; Gao, P.; et al. Reaction Mechanism and Structural Evolution of Fluorographite Cathodes in Solid-State K/Na/Li Batteries. Adv. Mater. 2021, 33, 2006118. 10. Kita, Y.; Watanabe, N.; Fujii, Y. Chemical Composition and Crystal Structure of Graphite Fluoride. J. Am. Chem. Soc. 1979, 101, 3832–3841. 11. Guérin, K.; Pinheiro, J. P.; Dubois, M.; Fawal, Z.; Masin, F.; Yazami, R.; Hamwi, A. Synthesis and Characterization of Highly Fluorinated Graphite Containing sp2 and sp3 Carbon. Chem. Mater. 2004, 16, 1786–1792. 12. Nakajima, T.; Watanabe, N.; Kameda, I.; Endo, M. Preparation and Electrical Conductivity of Fluorine-Graphite fiber Intercalation Compound. Carbon 1986, 24, 343–351. 13. Han, S. S.; Yu, T. H.; Merinov, B. V.; van Duin, A. C. T.; Yazami, R.; Goddard, W. A. Unraveling Structural Models of Graphite Fluorides by Density Functional Theory Calculations. Chem. Mater. 2010, 22, 2142–2154. 14. Hamwi, A. Fluorine Reactivity with Graphite and Fullerenes. Fluoride Derivatives and some Practical Electrochemical Applications. J. Phys. Chem. Solid 1996, 57, 677–688. 15. Zhong, G.; Chen, H.; Huang, X.; Yue, H.; Lu, C. High-Power-Density, High-Energy-Density Fluorinated Graphene for Primary Lithium Batteries. Front. Chem. 2018, 2018 (6), 00050. 16. Wang, K.; Feng, Y.; Kong, L.; Peng, C.; Hu, Y.; Li, W.; Li, Y.; Feng, W. The Fluorination of Boron-Doped Graphene for CFx Cathode with Ultrahigh Energy Density. Energy Environ. Mater. 2022, e12437. 17. Meduri, P.; Chen, H.; Chen, X.; Xiao, J.; Gross, M. E.; Carlson, T. J.; Zhang, J.-G.; Deng, Z. D. Hybrid CFx–Ag2V4O11 as a High-Energy, Power Density Cathode for Application in an Underwater Acoustic Microtransmitter. Electrochem. Commun. 2011, 13, 1344–1348. 18. Lam, P.; Yazami, R. Physical Characteristics and Rate Performance of (CFx)n (0.33 < x < 0.66) in Lithium Batteries. J. Power Sources 2006, 153, 354–359. 19. Yazami, R.; Hamwi, A.; Guérin, K.; Ozawa, Y.; Dubois, M.; Giraudet, J.; Masin, F. Fluorinated Carbon Nanofibres for High Energy and High Power Densities Primary Lithium Batteries. Electrochem. Commun. 2007, 9, 1850–1855. 20. Dai, Y.; Cai, S.; Wu, L.; Yang, W.; Xie, J.; Wen, W.; Zheng, J.-C.; Zhu, Y. Surface Modified CFx Cathode Material for Ultrafast Discharge and High Energy Density. J. Mater. Chem. A 2014, 2, 20896–20901. 21. Reddy, M. A.; Breitung, B.; Fichtner, M. Improving the Energy Density and Power Density of CFx by Mechanical Milling: A Primary lithium Battery Electrode. ACS Appl. Mater. Interfaces 2013, 5, 11207–11211. 22. Read, J.; Collins, E.; Piekarski, B.; Zhang, S. LiF Formation and Cathode Swelling in the Li/CFx Battery. J. Electrochem. Soc. 2011, 158, A504–A510. 23. Kozawa, A. Lithium - MnO2 Cells Containing CFx or C2F in the Cathode. J. Electrochem. Soc. 1987, 134, 780–782. 24. Fu, A.; Xiao, Y.; Jian, J.; Huang, L.; Tang, C.; Chen, X.; Zou, Y.; Wang, J.; Yang, Y.; Zheng, J. Boosting the Energy Density of Li || CFx Primary Batteries Using a 1,3-Dimethyl-2-Imidazolidinone-Based Electrolyte. ACS Appl. Mater. Interfaces 2021, 13, 57470–57480. 25. Ignatova, A. A.; Yarmolenko, O. V.; Tulibaeva, G. Z.; Shestakov, A. F.; Fateev, S. A. Influence of 15-Crown-5 Additive to a Liquid Electrolyte on the Performance of Li/CFx – Systems at Temperatures up to −50  C. J. Power Sources 2016, 309, 116–121. 26. Fang, Z.; Yang, Y.; Zheng, T.; Wang, N.; Wang, C.; Dong, X.; Wang, Y.; Xia, Y. An all-Climate CFx/Li Battery with Mechanism-Guided Electrolyte. Energy Stor. Mater. 2021, 42, 477–483. 27. Yin, Y.; Holoubek, J.; Liu, A.; Sayahpour, B.; Raghavendran, G.; Cai, G.; Han, B.; Mayer, M.; Schorr, N. B.; Lambert, T. N.; et al. Ultralow-Temperature Li/CFx Batteries Enabled by Fast-Transport and Anion-Pairing Liquefied Gas Electrolytes. Adv. Mater. 2023, 35, 2207932. 28. Li, Q.; Xue, W.; Sun, X.; Yu, X.; Li, H.; Chen, L. Gaseous Electrolyte Additive BF3 for High-Power Li/CFx Primary Batteries. Energy Stor. Mater. 2021, 38, 482–488.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Vanadium/Silver Oxides ES Takeuchi, KJ Takeuchi, and AC Marschilok, University at Buffalo (SUNY), Buffalo, NY, United States © 2009 Elsevier B.V. All rights reserved. This is a reproduction of E.S. Takeuchi, K.J. Takeuchi, A.C. Marschilok, PRIMARY BATTERIES – NONAQUEOUS SYSTEMS | Lithium–Vanadium/Silver Oxides, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 100–110, ISBN 9780444527455, https://doi.org/10.1016/ B978-044452745-5.00111-8.

Lithium–Vanadium Oxide Battery Vanadium Oxides Synthesis Structure Electrochemistry Practical Battery Capacity and Voltage Battery Design Lithium–Silver Vanadium Oxide Battery Silver Vanadium Oxide Synthesis and composition Phase diagram and structure Impact of materials synthesis Battery Design and Application Analysis of Discharge Reactions Conclusions Further Reading

100 100 100 100 100 102 103 104 104 104 105 107 108 109 109 109

Abstract This article contains an overview of primary lithium/vanadium oxide battery technology, and the chemistry of some of the relevant vanadium oxides, especially a family of materials containing vanadium, oxygen, and silver in various stoichiometric and nonstoichiometric ratios. First, the structural characteristics of the relevant vanadium oxide materials are described, including some phase diagrams of the materials. Next, electrochemistry and battery applications of these materials are discussed. Although some vanadium oxides have been widely studied as electrode materials for secondary (rechargeable batteries), only primary (nonrechargeable) battery applications will be considered here. Finally, various synthetic routes to electrochemical grade silver vanadium oxide are summarized, including both older methods and more recent synthetic innovations as well as battery performance results.

Glossary Active battery Contains all components of a typical battery, where upon final assembly of the battery it is ready for delivery of electrical energy. Electrolyte-activated battery As first assembled, the battery cathode and anode are kept dry with no exposure to electrolyte. Electrolyte is introduced to the system when the battery needs to be activated. Primary battery Single-use, nonrechargeable battery. Reserve battery Contains all components of a typical battery; however, at least one of the key components is isolated until the battery energy needs to be released by an activation step. Secondary battery Rechargeable battery.

Nomenclature

Symbols and Units a, b, c n T b Eo

unit cell lengths numbers of units involved in a chemical reaction temperature angle between a and c in the unit cell standard electrode potential

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00354-6

99

100

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Vanadium/Silver Oxides

Abbreviations and Acronyms CS HREM ICD m.p. MF OCV PC SVO TGA XRD

crystallographic shear high-resolution electron microscopy implantable cardiac defibrillator melting point methyl formate open-circuit voltage propylene carbonate silver vanadium oxide Thermogravimetric Analysis X-ray diffraction

Lithium–Vanadium Oxide Battery Vanadium Oxides Synthesis Vanadium pentoxide is synthesized commercially, typically from the processing of vanadium-containing metallurgical slags or wastes from power plants. This process involves pyrometallurgical treatment, including high-temperature roasting of the vanadium-containing raw materials with sodium- or calcium-containing compounds. The resulting water-soluble sodium or calcium vanadates can then be dissolved in sulfuric acid solutions. The solutions are then treated, resulting in a selective precipitation of the dissolved vanadium as hydrated V2O5.

Structure The vanadium ions in V2O5 can be described as having a distorted trigonal bipyramidal geometry, a distorted tetragonal pyramidal geometry, or a distorted octahedral geometry. When considered as a distorted tetragonal pyramidal geometry, V2O5 has four equatorial V–O bonds of similar length (
2 Å), whereas the apical V–O bond is considerably shorter (
1.5 Å). Because of the shorter bond length, and the high V–O stretching frequency, the apical bond is often described as a double bond. When considered as a distorted octahedron, the sixth and longest V–O bond (2.8 Å) corresponds to the V2O5 interlayer spacing, accounting for the ability to readily intercalate species into the V2O5 lattice. The V2O5 lattice has been described as an orthorhombic layered structure (Figure 1) composed of edge- and corner-shared distorted octahedra, which form a series of channels intersecting in three perpendicular directions, creating perovskite like cavities for lithium-ion intercalation. Within the V2O5 lattice, there are three inequivalent types of oxygen atoms: terminal vanadyl oxygen atoms O(1) coordinated to one vanadium atom, bridging oxygen atoms O(2) coordinated to two vanadium atoms, and bridging oxygen atoms O(3) coordinated to three vanadium atoms. As the interaction between the V2O5 layers is weak, the crystals cleave easily along the [001] plane.

O(1)

O(2)

O(3)

73 Å

V2O5 unit cell

519

Å

3.5 6

11.

b=

a=

4Å

c = 4.3

V2O5(001)

V

Figure 1 Structure of V2O5. Reproduced from Tepper B, Richter B, Dupuis AC, et al. (2002) Adsorption of molecular and atomic hydrogen on vacuum-cleaved V2O5(0 0 1). Surface Science 496(1–2): 64–72.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Vanadium/Silver Oxides

101

Electrochemistry xLi +V 2 O5 ! Lix V2 O5 Vanadium oxides display a rich electrochemistry, in large part owing to the variable formal oxidation states (II–V) of vanadium cations within vanadium oxides. Specifically, the open-circuit voltage (OCV) of the Li/V2O5 cell versus depth of discharge is shown (Figure 2). Initially, the OCV of the system is 3.4 V. After 0.5 F mol−1 of V2O5 has passed through a cell, the voltage shifts to a 3.2 V plateau. This is then followed by a voltage plateau at 2.4 V. C. Delmas and coworkers have reported a phase diagram for the LixV2O5 structure (Figure 3), which defines the room temperature intercalation of lithium in three phases: a (x3 to 2.4 V is used primarily for battery end of life indication.

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Vanadium/Silver Oxides

13.0

5.0 C 4.5

12.5

4.0

12.0 B 11.5

3.5

Lattice parameter for A (Å)

Lattice parameter for B and C (Å)

103

A 11.0

3.0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

x in LixV2O5

Figure 6 Lattice parameters of LixV2O5. Adapted from Prasad SS (1995) Insertion materials for rechargeable lithium batteries. In: Munshi MZA (ed.) Handbook of Solid State Batteries and Capacitors, pp. 467–513. New Jersey: World Scientific Publishing Co.; figure made in Microsoft Excel using data from Table 3(b), page 484.

Figure 7 Structure of g-LiV2O5. Reproduced from Delmas C, Cognac-Auradou H, Cocciantelli JM, Menetrier M, and Doumerc JP (1994) The LixV2O5 system: An overview of the structure modification induced by the lithium intercalation. Solid State Ionics 69: 257–264.

Battery Design Honeywell Inc. was a main commercial developer of primary Li/V2O5 battery technology. Additional work on Li/V2O5 was reported by both Duracell International Inc. and Eagle-Pitcher Industries. The designs for primary Li/V2O5 cells include a ‘DD’ size, a 5.7 cm-diameter button type cell, and an 11 cm3 prismatic cell design. For a typical cell design, the lithium anode is pressed onto a nickel or stainless-steel current collector. The cathode, usually consisting of a 9:1 mixture of V2O5 and graphite, with a few percent of a microthene power, is pressed onto both sides of a stainless-steel screen. Porous polypropylene is used for the separator, and the preferred case material is 316 or 304 stainless steel. Lithium metal has a standard potential of −3.05 V at 25  C and an electrochemical equivalence of 3860 mAh g−1. As a result, advanced batteries with lithium metal anodes possess high energy density, generating high coulombic output per weight of

104

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Vanadium/Silver Oxides

material. However, owing to the reactivity of lithium metal with water, nonaqueous electrolytes are required, where the lower conductivity of organic-based electrolytes compared to aqueous electrolytes can limit the rate capability that a primary Li/V2O5 cell can provide. The organic electrolyte solution for Li/V2O5 batteries was designed to maximize conductivity. This electrolyte system consists of a 2 mol L−1 lithium hexafluoroarsenate (LiAsF6) and 0.4 mol L−1 lithium hexafluoroborate (LiBF4) solution in methyl formate (MF; HCOOCH3). Three key requirements have been described by C. R. Walk to reduce the propensity for undesirable side reactions in MF-based electrolytes: (1) use high-purity solvent (MF, free of methanol and formic acid) and high-purity electrolyte salt (LiAsF6, free of LiAsF5OH), (2) add a strong Lewis base (such as LiF, present in LiBF4) to neutralize acidic impurities and maintain the basicity of the solution, and (3) use lithium metal with a sufficient reactive surface (approximately 2.1 mg cm−3 of electrolyte solution) to react with the protic materials and maintain the solution basicity. In contrast to typical active batteries, reserve batteries have at least one key component segregated from the other components during storage. This increases the battery shelf life and prevents undesirable battery self-discharge owing to parasitic reactions. Reserve cell designs have seen prevalent use of Li/V2O5 technology. Lithium/vanadium pentoxide reserve cells are manufactured using the same materials as lithium/vanadium pentoxide active cells, but with some key differences in the design of the electrolyte housing. As the electrolyte solution is not in contact with the Li anode prior to activation in a reserve cell, Li rods are placed in the solution to prevent generation of CO gas during storage. Two specific reserve cell designs have been described by C. R. Walk. In the first design, the electrolyte is housed within a glass ampoule in a plastic cup in the center of the cell. Reserve cell activation occurs in less than 5 s via a sharp directed force to the bottom of the cell. Holes in the lithium electrode and the ampoule-containing plastic cup allow the electrolyte to flood the separator and cathode when the battery is thus activated. A plastic support ring protects the ampoule tip to prevent premature battery activation. In an alternative design, the electrolyte is housed in a collapsible foil tube constrained by a release pin. In this design, the removal of the release pin frees a spring, which pushes the tube into a hollow needle, puncturing a septum and allowing electrolyte to enter the cell. As D. Linden has noted, lithium anodes create both significant advantages and significant challenges in reserve battery design. Lithium reserve batteries have excellent storability, with essentially no capacity loss after >10 years of storage. However, there is a substantial loss of design efficiency and simplicity owing to the electrolyte reservoir. It has been estimated that the electrolyte housing and device for electrolyte release cause a penalty of 50% in specific energy and energy density.

Lithium–Silver Vanadium Oxide Battery Silver Vanadium Oxide Synthesis and composition Because of the variety of oxidation states available to vanadium, silver vanadium oxide (SVO) can exist in many different phases, including both stoichiometric and nonstoichiometric phases. Preparation of SVO can occur via large variations in reaction conditions, starting materials, and reagent stoichiometries, where the resulting SVO products can have different structures and electrochemical properties. In this section, some of the early syntheses of SVO are described, as the synthesis and characterization of SVO materials in these earlier studies provided an important foundation for the later successful commercial use of SVO. Although several reports of reactivity studies and catalytic studies involving SVO are also available, these are outside the scope of this article and are not discussed here. Finally, some of the later synthetic work was associated more closely with the electrochemistry, and therefore is discussed in the electrochemistry section. H. T. S. Britton and R. A. Robinson published some of the first work on SVOs, with a series of reports in the 1930s. Silver-containing vanadates of various Ag:V ratios were precipitated via electrochemical titrations of aqueous silver nitrate with aqueous sodium vanadate. Isolation of the stoichiometric solids AgVO3, Ag4V2O7, and Ag3VO4 by various precipitation methods from cold AgNO3/alkali vanadate solutions was also reported, where the importance of aging or boiling the aqueous reactant solutions in order to obtain stoichiometric solids was emphasized. The next series of reports associated with SVO appeared in the mid-1960s. Preparation of three phases of SVO by the reaction of silver powder with V2O5 powder under vacuum or air atmosphere was reported by A. Deschanvres and B. Raveau. Distinct phases and phase mixtures were defined for AgxV2O5 over several ranges of x. A homogeneous nonstoichiometric b phase was observed from 0.17  x  0.45. A homogeneous nonstoichiometric d phase was observed from 0.6  x  0.8, and a biphasic region of b plus d was observed from 0.45  x  0.6. In the presence of air, a nonstoichiometric e phase was generated, which was homogeneous from 1  x1.15. Specifically where x¼1, the stoichiometric product from the air reaction was described as Ag2V4O11. At still higher levels of silver, AgVO3 was obtained. Finally, it was noted that the d phase undergoes a complex air oxidation at 500  C, yielding a mixture of b and e phases. Research associated with SVO synthesis continued in the 1980s. V. L. Volkov and E. I. Andreikov heated 12 different mixtures of silver nitrate and vanadium oxide solids in air at 750  C, producing several stoichiometric and nonstoichiometric phases of SVO. When silver concentration was low (AgxV2O5, 0  x  0.02) an a phase material was formed, whereas for higher silver concentrations (AgxV6O15, 0.85  x  1.0) a b phase material was obtained. Silver vanadates Ag1.2V3O7(g) and Ag2V4O10.5(e) as well as the silver metavanadate AgVO3 were also observed in some cases. More recently, sol–gel methods have been employed in the synthesis of SVOs. L. Znaidi and coworkers synthesized monoclinic b phase SVO via a sol–gel process involving V2O5 xerogels. Vanadium pentoxide gels were formed via acidification of aqueous

Battery Types – Lithium Batteries – Lithium Primary Batteries | Lithium–Vanadium/Silver Oxides

105

vanadium, then the resulting fibrous V2O5 gels were dried to form a thin xerogel film, which was then soaked in a solution containing silver cations. Dehydration and subsequent heating of the intercalated xerogel produced the Ag0.36V2O5 bronze. Organic templates have also been used for the synthesis of SVO. J. Zhu and colleagues reported hydrothermal synthesis of single-crystalline Ag2V4O11 nanobelts from the heating of an aqueous mixture of AgNO3, V2O5, and 1,6-hexanediamine. A. Ramanan and colleagues reported ion exchange synthesis of silver vanadates from organically templated layered vanadates ((org)xV2O5). It was suggested from X-ray diffraction (XRD) data that the parent layered structure of (org)xV2O5 was destroyed within a few hours of mixing with silver nitrate with a gradual evolution of a b-AgVO3 phase after 36–48 h. Transmission electron microscopy indicated AgVO3 nanorods covered with spherical silver nanoparticles.

Phase diagram and structure P. Hagenmuller and coworkers prepared and reported on a series of vanadium bronzes, including AgxV2O5. Four phases were reported, including an orthorhombic a phase (0  x  0.01), a monoclinic b phase (0.29  x  0.41), and a monoclinic d phase (0.67 Li2 Sn5 > Li3 Bi; Li3 As > Li3 Sb > LiAl and LiZn  LiC6 : Lithium-tin alloys have reached a commercial state of the art.47 Lithium aluminum alloys48 (LiAl) expand by 94% during the phase transition. Al + Li ! AlLi. Despite good capacity, intermetallic alloys suffer from poor cyclability already in the first cycle. Sufficient cycle life requires an excess of lithium because lithium is consumed by 1% with each cycle. The higher electrode potential (0.3. . .1.0 V vs. Li|Li+) in comparison to graphite (0.1 V) causes a lower usable cell voltage. Thackeray and co-workers found intermetallic antimony alloys (Cu6Sn5, InSb, Cu2Sb) with structural similarity to their lithiated products (Li2CuSn, Li3Sb). Indium antimony (InSb) and copper antimony (Cu2Sb) offer a stable, face-centered antimony host lattice for reversible lithium intercalation. In LixIn1–ySb (0  x  3, 0  y  1), the volume of the antimony matrix changes isotropic (the same way in all directions) by 4%, while the total electrode volume changes by expands by 46% with the escape of indium. The alloy that is most actively developed today is Li-Si (Table 3). It promises a maximum theoretical capacity of 3577 Ah/kg which is comparable to Li-metal and about 10-times the specific capacity of graphite. However, the expansion of the silicon is about 300% when fully reacted with the Li from the cathode at the end of charge. This expansion can be reduced by oversizing the anode relative to the cathode such that the amount of Li available from the cathode limits the Li-concentration in the Si and thus the maximum expansion. As mentioned in Section 2.2, one of the issues of a Li-metal anode is the restriction on charge rate to avoid Li-dendrite growth. The Li-Si with its higher potential relative to Li compared to graphite allows for higher charging rates not only relative to Li-metal anodes but also relative to graphite. In a graphite-Si composite anode this high rate would be restricted to the lower SOC range.

2.3.2 Silicon alloys and nanocomposites Lithium silicon alloys have the largest specific capacitance to date. Nanoparticles limit mechanical destruction due to changes in volume during charging and discharging. The lithium-silicon alloy Li4.4Si (0.047 V vs. Li|Li+) promises a theoretical capacity of 4212 Ah/kg of silicon, compared to 3861 Ah/kg for metallic lithium. At present, the high capacity is paid with limited lifetime of at most 750 cycles, which allows only certain consumer electronics applications. Lithium intercalation in silicon is achieved via defined two-phase plateaus: Si/Li12Si7 (332 mV vs. Li|Li+); Li12Si7/Li7Si3 (288 mV); Li7Si3/Li13Si4 (158 mV); Li13Si4/Li21Si5 (44 mV). During alloying, a metastable

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview

129

amorphous Li-Si phase is initially formed that can be stabilized in a narrow potential window, and then a moderately stable, crystalline, intermetallic equilibrium phase.49 Unfortunately, the energy density is low due to voids in the structure, the low tap density of nanomaterials, and the addition of carbon. Nanofilms, better than nanocrystals, improve specific capacitance and fast lithium-ion transport during intercalation and deintercalation. Electrodes made of nanofilms (e.g., from Sanyo), nanoparticles (e.g., Sony), nanowires, nanotubes, and hollow particles are expected to improve cycling stability during charging and discharging by minimizing volume changes, albeit at a higher cost. Silicon nanowires on stainless steel were proposed by researchers at Stanford University (2007). Inactive current conductors and unused porosity limit the energy density; cyclability and stress cracking resistance increase with reduced particle size. Risks exist with electrode potentials outside the permissible stability window of conventional electrolytes and by the growth of large passivation layers (SEI). Instead of elemental silicon, silicon suboxides (SiOx, x < 2) can also be used as anode materials.50 During the first charge of a cell the oxide is reduced to silicon metal which acts as the active anode material. The byproduct is lithium silicate that is electrochemically non-active. As some of the Li from the cathode is bound in this silicate, the irreversible capacity is increased. Silicon graphite composites51 roughly achieve up to 1500 Ah/kg at potentials below 0.6 V vs. Li|Li + and combine the advantages of both materials: silicon for high capacitance, and graphite for electronic conductivity and mechanical elasticity during volume changes. The composites are produced by milling or coating of nanostructures. Lifetime and performance depend on structure, morphology, and composition. The binder determines the cycling stability of the C/Si electrode in the order: Sodium carboxymethyl cellulose (Na-CMC) > Hydroxyethyl cellulose > Cyanoethyl cellulose > PVdF-HFP.

2.3.3 Tin composite electrodes (TCO) In the 1980s, Matsushita introduced a cell based on wood metal (a low-melting bismuth-lead-tin-cadmium alloy), which cycling capability deteriorated with increasing depth of discharge (DoD). In 1997, FUJI introduced a technology (STALION) based on tin oxide (TCO). An amorphous composite electrode (ATCO, SnB0.56P0.4Al0.42O3.6) reacts reversibly with lithium at approximately at about 0.5 V and provides twice the capacity of graphite (600 Ah/kg, 2200 Ah/L). X-ray structure (XRD) analyses prove the deposition of metallic tin and the gradual formation of Li4.4Sn nanodomains in a matrix of lithium oxide (Li2O); the cycling stability is limited and the irreversible loss of capacitance in the first cycle is dramatic. Metal deposition : ðÞ

SnO2 + 4 Li + ! 2 Li2 O + Sn

Charging :

Sn + x Li + +x e ! Lix Sn

The Li2O matrix buffers the volume expansion of the reactants. For a mixture of two alloys with different potentials, the active phase is embedded in an electrochemically inactive matrix. The intercalation of lithium in tin takes place in several potential steps with an undesired volume expansion. The theoretical specific capacity for Li4.4Sn is 993 Ah/kg.

2.3.4 Transition metal vanadates Metal oxides that reversibly form lithium oxide demonstrate that intercalation properties and structural integrity are not the only decision criteria for high performance electrode materials. The large capacity of transition metal vanadates LiMVO4 (M ¼ Mn, Fe, Co, Ni, Cu, Cd, Zn) and magnesium vanadate (MnV2O6) is not based on lithium intercalation or alloy formation. Metal nanoparticles, formed during the initial discharge, enable the formation and decomposition of lithium oxide in subsequent cycles (first proposed by Fuji Co.). Metal deposition : ðÞ Charging

MVO4 + 4 Li ! 2 Li2 O + M + VO2 M + x Li + +x e ! Lix M

Indium and iron vanadates (MVO4) and some molybdates exhibit an unusually large affinity for lithium. The combination of vanadium oxide (−) and lithium cobalt oxide (+) is expected to yield 745 Wh/L; this is two to three times more than the energy density of graphite|LiCoO2. 100% capacity retention has been demonstrated for over 100 cycles.

2.3.5 Other reducible metal oxides Metal oxides as anode materials in lithium-ion batteries are mostly expensive (750 Ah/kg, 0.8–1.6 V vs. Li|Li+). Cobalt oxide (CoO, Co3O4), nickel oxide (NiO), copper oxide (CuO), manganese oxide (MnO), and ruthenium oxide (RuO2) can be dispersed in a matrix of amorphous lithium oxide (Li2O). At the nanoscale, the electrochemical reaction pathways in inorganic materials often change.

2.3.6 Lithium metal nitrides The weaker M-X bonds in non-oxides produce higher anode potentials: For example, manganese and iron fluorides compared to oxides, sulfides and nitrides. Lithium cobalt nitride Li3xCoxN (600 Ah/kg) as a dispersion phase promises a large, stable and reversible capacity. The sensitivity to moisture requires restrictive manufacturing conditions.

130

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview

Table 5

Overview of present non-carbon negative electrode (anode) chemistries for Lithium batteries.

Material family Lithium titanate (LTO)

Titanium oxide (TiO2) Silicon alloys

Silicon oxides SiOx Lithium alloys (Sn, Pb, Sb. Bi)

Lithium metal

Conversion reaction materials

Advantages

Challenges +

−

Charge reaction: Li4/3Ti5/4O4 (or Li4Ti5O12) + 3Li + e ! Li7Ti5O12. No volume change (zero strain material). Excellent lifetime (10,000 cycles, 10 years) even at 45  C. No SEI formation and serious electrolyte decomposition. Commercial by Toshiba. Improvement of conductivity by nanostructure, surface coating, and composites. xLi+ + TiO2 + xe− $ LixTiO2 (0  x  1). Nanomaterials required. Intermetallic compounds: Li4.4Si: high capacity in theory: 3578 Ah/kg (Li3.75Si), 4200 Ah/kg (Li34.4Si), 2386 Ah/L. For comparison: lithium metal (3860 Ah/kg, 2087 Ah/L). Working potential (0.4 V vs. Li + |Li). Reversible lithiation: amorphous Si ! intermediates ! Li15Si4. Silicon(II)-oxide (SiO in graphite) forms nanosized Si an inactive phase of Li4SiO4,which suppresses pulverization. Binary phases are known from lithium-tin (Li2Sn5, LiSn, Li7Sn3, Li5Sn2, Li13Sn5, Li7Sn2, Li17Sn4, 1000 Ah/kg in theory), lithium-lead (LiPb, Li3Pb2, Li5Pb2, Li3Pb, Li7Pb2, Li9Pb2, 582 Ah/kg), lithium-antimony (Li2Sb, Li3Sb, 665 Ah/kg), lithium-bismuth (LiBi, Li3Bi, 385 Ah/kg). Example: nanosized tin dispersed in glass matrix Maximum energy density (3.04 V SHE, 3860 Ah/kg in theory),

(a) Metal oxides: 2Li+ + 2 e− + MOx $ xLi2O + M (M ¼ Mn, Fe, Co, Ni, Cu, Cr, Mo, Sn). High theoretical capacity (in Ah/kg): MnO (756), Mn3O4 (937), Mn2O3, (1018), MnO2 (1232). (b) Metal sulfides: 2Li+ + 2x e− + MSx $ xLi2S + M (M ¼ Mn, Fe, Co, Ni, Mo, Sn) are less kinetically inhibited than oxides.

Working potential: 1.55 V vs. Li|Li+. Low capacity: 175 Ag/kg in theory.

Low Li+ conductivity depending on structure (rutile, anatase). Low capacity: < 100 Ah/kg. Extreme volume changes (412%), contact problems, poor reversible capacity and cyclability Low volume change. Volume expansion during lithiation, structural degradation, poor cycle performances. toxicity Unsafe. Low melting temperature (180  C), Li dendrites, Poor cycle performance and coulombic efficiency; inhomogeneous surface with ‘dead’ lithium, electrolyte consumption. Large volume changes. Low voltage (V vs. Li|Li+): MnOx: 1.2, FeOx 1.7, MSx < 1.5

Niobium oxide nitride (NbON) is prepared by thermal decomposition of ammoniacal NbOCl3 precursors under an inert gas atmosphere. The NbO1.3N0.7 nanoparticles show a larger storage capacity for LiC (250 Ah/kg) compared to Nb2O5. A summarizing overview of state-of-the-art anode material sis given in Table 5.

2.4

Li-metal anodes with synthetic SEI

The break-up of at least part of the SEI is responsible for both the loss of electrolyte and active Li. A synthetic SEI to replace the SEI formed by reaction of the Li with the liquid electrolyte must allow for Li-ion transport and block anion and electron transport. This requirement is fulfilled by solid electrolytes developed for solid state Li-batteries. Its mechanical properties must be such that the layer does not break-up when the Li-anode is stripped or plated non-uniformly. This SEI the layer must be thin enough to not impact energy density. The cathode side of such a cell would use a liquid electrolyte. This layer can be inorganic, organic, or inorganic/organic hybrid in nature. In general, the inorganic layer has better Li-ion conductivity, the organic layer better mechanical properties. A hybrid layer that combines these attributes is therefore proposed.52 SEI layers can be created for example by the reaction of a precursor in solution with the Li-surface prior to cell assembly.53

2.5

Zero lithium

The high capacity of Li-metal per weight and volume allows for an excess of anode capacity compared to cathode capacity. When this additional Li is brought into the cell by a Li-metal layer, the ratio of anode to cathode capacity can be chosen freely as it was the case in early Li-metal anode cells. It also allows to choose cathode materials that are either assembled in the discharged or charged state. Cathode materials for Li-ion cells are in the discharged state when assembled and are the source of all the cyclable Li in the cell such as LiNi0.8Co0.1Mn0.1O2 (NCM 811). Using these cathode types as the sole source of Li in a Li-metal anode cell represents the “Zero Lithium” concept (ZLiC).54 For this type of cell, the loss of cyclable Li translates directly into a loss of cell capacity. Therefore, the loss of cyclable Li per cycle must be 0.02% to achieve 1000 cycles to 80% capacity remaining. One way to achieve reasonable cycling efficiency are dual salt electrolytes.55 As the salt is used up in the passivation of the plated Li-deposit during each cycle, the amount of electrolyte in the cell has an impact on the cycle life. On the other hand, the amount of electrolyte also effects the energy density of the cell. This is a

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview

131

general concern for all the concepts discussed here. The ZLiC has the maximum possible theoretical specific energy of all concepts. However, measures required such as an electrolyte reservoir or other concepts discussed below will reduce this advantage over Li-ion. Several approaches are being pursued to improve the reversibility of the Li-anode. For example, two layers are deposited on the current collector. The first layer on top of the current collector is electronically conductive, the second layer is only a Li-ion conductor. On charging, the Li + is reduced to the metal in between the two layers and deposited there. It was also proposed to have a porous conductive layer in between as a 3D substrate for Li-deposition.

2.6

Three-dimensional Li-anodes

The limit of charge current density for planar Li-anodes was discussed above. A way to increase the charge rate is the design of three-dimensional (3D) electrodes in combination with the Zero Lithium concept. For this, a 3D open substrate is required which allows the Li to be plated onto its surface generating a 3D electrode. As a result, the local current density at the same recharge rate is much lower compared to a conventional flat (2D) electrode allowing for faster charge.56 It is obvious that the weight and volume of the anode compared to a 2D Li-foil increase and at least some of the energy density advantages of Li-metal anodes compared to Li-graphite intercalation anodes are lost.

3 3.1

Cathodes Cathodes without Li-source

The first rechargeable lithium batteries had a Li-metal foil or a Li-alloy such as LiAl as anodes and an oxide or sulfide cathode. The cells were assembled in the charged state. All the Li for discharging and subsequent cycling was introduced during cell assembly by the anode. Due to the high specific capacity of Li-metal, Li in excess to the capacity of the cathode could be used to compensate Li-losses on cycling without a large penalty in energy density. The disulfides MS2 (M ¼ Ti, Nb, Ta, Mo, W) and the oxides MO2 (M ¼ V, Cr, Fe, Co, Ni) have been researched since the 1970s as cathode materials (+) for lithium metal anode batteries. Titanium disulfide can store lithium in the intercalation range up to LiTiS2 with only 10% lattice expansion: TiS2 + x Li+ + x e– ! LixTiS2. The diffusion coefficient is D(Li+) ¼ 10–12 m2 s−1. The average discharge voltage of 2.1 V delivers a specific energy of 450 Wh/kg for a rechargeable cell with metallic lithium anode (−). More than 400 charge-discharge cycles with 20% loss of material utilization have been described. Molybdenum disulfide (MoS2) is not stable at a depth of discharge below 10%. The intercalation of lithium ions is accompanied by a structural change and the formation of new phases. More than one Li per MoS2 can be reversible cycled.57 Amorphous molybdenum disulfide is more stable than the crystalline material. Framework structures of vanadium, chromium, molybdenum, tungsten or nickel oxides in which the metal can reversibly switch between two oxidation states, provide sufficient space for the intercalation of lithium ions. Vanadium oxides have been used intensively since 1975. The polycrystalline powder vanadium pentoxide V2O5 stores more than one lithium cation per vanadium atom; vanadium is oxidized up to a mean oxidation state of 3.5 in Li3V2O5 or 3.67 in Li5V3O8. LixV2O5/Li-cells delivered 100 cycles with more than 450 Wh/kg (3.4–1.9 V). Rechargeable polymer electrolyte (Li-SPE) cells use lithium-free V2O5 or its derivatives as the positive electrode. The layered structure of LixVO2 (320 Ah/kg) decomposes upon intercalation with lithium. With the composition Li0.3VO2, one third of the vanadium ions migrate into the layer emptied of lithium, destroys the plane for 2D lithium diffusion and produces a low active defect rock salt lattice. Above 66  C, the semiconductor VO2 changes to metallic conductivity. Another vanadium oxide, V6O13, reacts reversibly with lithium to form LixV6O13 in the range of 0 < x < 8. This corresponds to a high cathode capacity per unit weight of 417 Ah/g. The material discharges along several voltage plateaus suggesting multiple phase transitions. High material utilization as well as good cycling performance has been demonstrated.58 Recent research is evaluating the properties of V6O13 doped for example with Cu, Al, and Cr as well as new synthesis methods. Niobium triselenide, NbSe3, is a cathode material with a theoretical specific capacity of 244 Ah/kg with three Li per NbSe3. It is a metallic conductor and appears in a fibrous form. It was extensively studied in the 1980ties. Manganese oxides are well known as cathodes materials for alkaline and Li-primary batteries. They have also been studies as cathode materials for rechargeable Li-batteries. The material is an electric insulator, and the cathode requires a conductive network, usually carbon based. Rechargeable coin cells are for example offered by Panasonic. Conductive aerogels with low density, small particles, and large surface area permit the rapid diffusion of lithium with moderate volume changes. V2O5 and MnO2 provide (in addition to the faraday currents of the battery processes) the capacitive double layer charging currents at the huge surface of nano-dimensional zones in the highly porous empty volume. Lithium cells with a vanadium oxide xerogel as the positive electrode deliver 200 Ah/kg and 700 Wh/kg; however, the framework structure collapses during charge-discharge processes.

132 3.2

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview Cathodes with Li-source

Cathode materials that are synthesized in a discharged state, that is providing the lithium required for cycling, made the Li-ion battery concept using a Li-free anode material possible. The first commercially successful material was lithium cobalt oxide, LiCoO2, discovered by Goodenough in 1980. This was the first cathode material used in Li-ion batteries by Sony in 1989. The negative electrode material used are different forms of carbon such as graphite or carbon-silicon composites. From a manufacturing point of view, the Li-ion concept has the advantage that the active materials are stable in air which is not the case for lithium or lithium alloys. The protection from moisture during cell manufacturing and operation is still required as the electrolyte salt will hydrolyze and the charged anode material, LixC6, will react with water. Many cathode materials also react even with water in the as-synthesized state. The Zero-Lithium concepts described in Section 2.5 makes use of these cathode materials as the cyclable Li source of the cell. Solid-state electrolyte cells often use Li-metal anodes in place to achieve a higher energy density. To compensate for loss of cyclable Li, cells are built with a Li-metal anode already in place. Due to the high capacity per weight and volume of Li, the penalty is relatively small. Table 2 shows the most commercially used cathode materials, lithium cobalt oxide (LiCoO2), lithium iron phosphate (LiFePO4), and lithium nickel cobalt manganese oxide (LiNixCoyMn1-x-yO2). Compared to the Li-containing cathodes their reversible capacity is comparable. Their advantage is the higher running voltage which is responsible for the higher energy density at cell level.

3.3

Air cathodes

Primary cells that use the oxygen out of the air have been developed for Zinc anode cell. These cells are commercially produced and are mostly used as button cells in hearing aids. Their energy density is with up to 1400 Wh/L very high. The electrolyte is an aqueous solution of KOH. The oxygen is reduced to OH− at a three-phase boundary of the liquid electrolyte, the solid catalyst, and air. Zinc-air cells are stored with a temporary seal to cut off air access. Once opened they are activated and have a limited storage life, for example by the CO2 from the air reacting with the KOH to form the carbonate. Due to the high energy density and the low material cost, R&D efforts persist in developing rechargeable systems with zinc and aluminum anodes.59 Even higher energy densities can be achieved theoretically with Li-air batteries. The capacity of the Li anode is 3860 mAh g−1, the cell voltage depends on the reaction path and the composition of the final product but is at least 3 V. The theoretical specific energy thus exceeds 10,000 Wh/kg without accounting for the oxygen that ideally comes from the air. The reaction of Li with the oxygen occurs along several steps.60 One proposed reaction sequence is first the formation of lithium superoxide, then lithium peroxide and finally lithium oxide. Li  e − +O2 + e− − > LiO2 + Li+ + e− − > Li2 O2 + 2Li + +2e− − > 2Li2 O In some of the past cell concepts, the reaction did not go to the oxide but stopped at the peroxide which is a stable solid at ambient conditions. Experimental cells have used both organic liquid, ionic liquid, and solid electrolytes. Aqueous electrolytes consisting of a saturated LiCl solution have also been evaluated. The stability of the electrolyte with respect to the peroxide and its exposure to air at the cathode are issues to be solved. Catalysts at the air electrode must not only have an acceptable lifetime but also a low overvoltage for the oxygen reduction during discharge respectively the oxidation of the oxide on charge. Deposition of the Li metal on charge suffers the same issues described in 2.2. Judging by the number of publications, R&D in rechargeable lithium air systems has had its peak in the 2010-years.

4 4.1

Electrolytes Electrolyte types

Liquid electrolytes are held in place by capillary forces in the pores of the electrodes and the separator. Lithium-ion polymer electrolyte batteries are sometimes called solid-state cells, as they eliminate free electrolyte within the cell. A gel that conducts lithium ions can also be used, which, however, is close to liquid electrolytes in its properties. Solid polymer cells do not require heavy protective cases. In the absence of a free liquid, flexible foils of any shape suffice as simple and lightweight battery housings. These flat cells are also known as pouch cells. M. Armand replaced the electrolyte solution in the 1980s with a dry polymer, originally polyethylene oxide (PEO) doped with lithium salts, and thus established the solid polymer electrolyte (Li-SPE) technology. The low conductivity at room temperature forces operating temperatures around 80  C for sufficient power currents. Later, lithium hybrid polymer electrolytes (Li-HPE), a polymer matrix was allowed to swell in a solvent, and a conducting salt was added. However, lithium metal dendrites were still a safety issue. In 1994, Bellcore’s plastic lithium ion (PLI) technology made possible the first reliable rechargeable thin-film lithium-ion battery. Polymeric electrolytes in a lithium-ion system offered shape versatility and flexibility. Low temperature performance improvements of PEO based electrolytes are still a development consideration. One proposed solution is a combination of polymer and solid electrolytes.61 A summary overview of the state-of-the-art in electrolyte systems is given in Table 6.

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview Table 6

133

Overview of present electrolyte systems for Lithium-ion batteries.

Material class

Advantages

Challenges

Solution in organic solvent

Alkyl carbonates and conducting salt: high ionic conductivity, configurable composition, compatible with carbon and metal oxides. Additives: 2% vinylene carbonate for graphite anodes; fluoroethylene carbonate for Si-based anodes; sulfolane and tris(trifluoroethyl) phosphate for cathodes. Salt solution in swollen host polymer matrix, e.g. LiCF3SO3 in poly(ethylene oxide): good conductivity, configurable composition, flexible, commercial lithium-polymer battery cells. Li+ complex in polymer matrix: flexible cell, reliable, no flash point, no electrolyte leakage. Beneficial: low glass-transition temperature, branched structure of polymer. Molten salts at room-temperature and dissolved lithium salt: moderate conductivity (mS/cm. . . mS/cm), nonflammable, wide temperature range, not volatile. Wide voltage window (6 V in theory). Cations: imidazolium, ammonium, pyrrolidinium. Anions: sulfonyl imides (TFSI), triflates, BF−4 , dicyanamide (NC)2N, Cl−. Li salt in water or organic solvent ( 1 mol/L): High conductivity, flexible composition (solvent and salt), nonflammable. Anion ridges compensate the lack of solvent molecules in the solvation shell around the Li+ cation. Dimethoxy oligoethers (glymes) for “solvated ionic liquids”. Li+-containing glasses or metal oxides, e.g. LISICON, LIPON: no leakage, nonflammable, not volatile, high energy density, safe all-solid battery, transport number: t(Li+) ¼ 1.

Not heat resistant (>60  C), flammable, electrolyte leakage, electrolysis (CH3CN: 3.8, EC: 5.2 V vs. Li|Li+). LiPF6 reacts with EC.

Polymer gel Solid polymer Ionic liquid

Super-concentrated solutions Inorganic solid

4.2

Same as organic solution Low Li+ mobility and conductivity (0.01. . .10 mS/cm), not useful at room temperature, operating at elevated temperatures High viscosity, low wettability, low Li + transport number (also counter-ions move). Narrow voltage window due to impurities, reduction at the cathode, expensive. Expensive, high viscosity, high mass

Poor Li+ mobility and conductivity (0.001. . .1 mS/cm), narrow potential window (e.g. Li3N), bad electrode/ electrolyte interface, redox activity. Only for thin film cells commercial.

Liquid electrolytes

Liquid electrolytes are commonly used in commercial Li-ion cells and were used in the past for Li-metal anode cells. The electrolyte must be stable over a wide temperature and voltage range. It should be non-toxic and non-flammable and have at least a high flash point and a low heat of combustion. Nonaqueous electrolytes can operate outside their window of thermodynamic stability (3.5–5.5 V vs. Li), because electrolyte decomposition is kinetically controlled. Early Li-metal cells used simple electrolyte formulations such as one molar solution of lithium hexafluoroarsenate (LiAsF6) in propylene carbonate (PC). Most electrolyte solvents for Li-ion cells consist of mixtures of cyclic and linear carbonates (Table 7). The choice of salts has also increased significantly (Table 8). The solvent choice for cells with a non-graphite anode depends both on the anode and the cathode chemistry. The latter is in many cases identical to Li-ion cells, however, the non-carbon anodes will require specific electrolyte formulations to achieve the desired performance.62,63 A typical Li-ion battery electrolyte is a mixture of 20–50% ethylene carbonate (EC), organic carbonates or esters, a conducting salt, and as few additives as possible to achieve the desired characteristics.

Table 7

Solvents for lithium batteries.

Compound

Formula

Density 20  C (g cm−3)

Melting point ( C)

Boiling point ( C)

Ethylene carbonate (EC) Propylene carbonate (PC) Dimethyl carbonate (DMC) Diethyl carbonate (DEC) 1,2-Dimethoxyethane (DME, glyme) 1,2-Dioxolane (DIOX)

(CH2O)2C¼O

1.32 (40  C)

+36

CH3(C3H3O3)

1.21

(CH3O)2C¼O

1.07

Refractive index (20  C)

Permittivity er

Solubility in water 20  C (g/L)

248 (decomposition) 160

1.415 (50  C)

90 (25  C)

214 Ethanol

−48.8

242

132

1.419

65

240

0.5. . .4.7

90

15

1.369

3.1

139

(C2H5O)2C¼O 0.970 (25  C)

−43

126

25

1.384

(CH3OCH2)2

0.87

−58

84

0

1.377

2.8 Ethanol, ether, CHCl3 7.1 miscible

C3H6O2 cycle

1.06

−26

74

–

1.401

–

Flashpoint ( C)

miscible

134

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview Table 8

Examples for electrolytes in lithium-ion batteries and laboratory cells (1 mol/L).

Conducting salt

Formula

Solvent mixture 1:1 (salt in mol/L)

Conductivity at 25  C (mS cm−1)

Lithium hexafluoridophosphate

LiPF6

Lithium bis(trifluoromethane) sulfonimide (LiFTSI)

Li[(CF3SO2)2N]

Lithium bis(fluorosulfonyl)imide (LiFSI) Lithium bis(oxalato)borate (LiBOB) Lithium tris(pentafluoroethyl) trifluorophosphate (LiFAP) Lithium difluoro(oxalato)borate (LiDFOB). Lithium tetrafluoroborate Lithium hexafluoroarsenate

Li[(FSO2)2N] Li[B(C2O4)2] LiPF3(C2F5)3

PC (1) EC/EMC (0.75) EC/DMC (1) PC (1) EC/DMC (1) EC/DMC (0.85) EC/DMC EC/DMC (0.8)

5.8 9.7 10.7 5.1 9.0 12 7.5 9.0

Lithium hexafluorotantalate Lithium perchlorate

LiTaF6 LiClO4

EC/DMC (1) PC (1) PC (1) EC/DMC PC (1) PC (1) EC/DMC (1)

4.0 3.4 5.7 11 5.2 5.6 8.4

Li[F2B(C2O4)]) LiBF4 LiAsF6

Decomposition limit (V vs. Li|Li+)

4.0. . .4.8 4.5. . . 5 >5V > 4.7 > 4.7 > 6.7

EC forms a protective layer (SEI) of presumably lithium ethylene dicarbonate on the graphite surface, which limits the further decomposition of the electrolyte and limits the self-discharge of the battery. Ethylene carbonate is temperature stable, has a high permittivity, is reduced below 0.9 V vs. Li|Li + (on glassy carbon) or 0.8 V (on graphite), and is oxidized above 6.2 V vs. Li|Li + (on glassy carbon in lithium-free solution), thus offering a stability window of roughly 5 V. Lithium ion intercalation in graphite works well from electrolyte mixtures with the conducting salt lithium hexafluorophosphate (LiPF6) in EC/DEC or DMC. Propylene carbonate (carbonic acid propylene glycol ester) is oxidized above 6.6 V vs. Li|Li+. In the long-term, Li2CO3 and propene, linear decarbonates and polymeric decomposition products are formed at the negative electrode. At the positive pole, linear carbonates, CO2, alkenes and polyether carbonates are formed by electrochemical oxidation and thermal decomposition. g-Butyrolactone is stable between reduction (0 V) and oxidation (8.2 V vs. Li|Li+). It is used in 3-V primary batteries with the conducting salt lithium tetrafluoroborate (LiBF4). Stability of solvents. Ethers and carboxylic acid diesters have high permittivity (er ¼ 3. . .7) but exhibit dangerous low flame temperatures (< 30  C). Linear carbonates form carboxylic acids, CO2, alcohols, ethers and alkenes at the negative electrode during aging. The stability against electrochemical oxidation increases in this order: water ethers (DME, DEE, THF), ketones < linear carbonates (EMC, DMC, DEC), acetonitrile, ethyl acetate < cyclic carbonates (PC, EC) and their mixtures < nitro compounds. The stability against electrochemical reduction increases in this order: water, nitro compounds ketones, PC, ethyl acetate < acetonitrile < linear and mixed carbonates, EC < ethers. As mentioned in Section 2.2, pressure on the Li-metal electrode is a key factor influencing Li-metal cycling efficiency. The amount of pressure required is also dependent on the choice of electrolyte and the type of SEI formed at the electrolyte-Li interphase.64

4.3

All-solid-state batteries

Solid-state batteries do not use liquid electrolytes but require solid electrolytes. Recent developments have resulted in materials with satisfactory conductivity at room temperature. Records materials are the metastable Li7P3S11 (LPS, 0.017 S/cm) and Li10GeP2S12 (0.012 S/cm). The lithium-iodine battery is an early development of solid-state technology for cardiac pacemakers (since 1972). Non-conductive iodine forms a conductive charge-transfer complex. A thin layer of lithium iodide on the lithium electrode forms the solid electrolyte: 2 Li + I2 ! 2 LiI ( 10−7 S/cm). Although long-term stability is good, only low current (< 0.1 mA) is delivered, because the cell resistance grows from 50 O to kiloohms during discharge. The framework structure of vanadium oxides (e.g., V6O13) confuted the previous belief that only low-dimensional materials allow sufficient lithium-ion diffusion. Hitachi’s practical solid-state thin-film battery of 1982 was fabricated by vacuum vapor deposition and sputtering techniques and comprised either a titanium sulfide or WO3/V2O5 positive electrode, a Li3.6Si0.6P0.4O4 glass electrolyte, and a lithium metal negative electrode. NTT Co. in Japan used Li3.4V0.6Si0.4O4 glass electrolyte (1989). Eveready Battery in USA (1980) used sulfide glass of Li4P2S7 or Li3PO4/P2S5; Bellcore employed as well lithium borophosphate (LiBP) glass and lithium phosphorous oxynitride (LiPON) glass. Thanks to the US Oak Ridge National Laboratory (ORNL), LiPON was recognized as a standard solid electrolyte for thin-film lithium batteries, which have been commercialized since 2002. In an all-solid-state thin-film battery, the solid films of the negative electrode material, the solid electrolyte, and the positive electrode material are layered on top of each other. Solid electrolytes allow film batteries of less than 0.3 mm thickness. However,

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview

135

slow mass transport limits the energy and power of just a few micrometer thick lithium insertion materials. Lithium metal is deposited mostly by vacuum thermal vapor deposition; solid electrolytes and oxide electrodes are prepared by radio frequency (RF) sputtering, chemical vapor deposition (CVD), electrostatic spray deposition (ESD), pulsed laser deposition (PLD), or sol–gel processes. Typical solid electrolytes are LiPON and lithium glasses (e.g., Li3.6Si0.6P0.4O4, Li6.1V0.61Si0.39O5.36, Li2SO4/Li2O/B2O3, Li2S/SiS2/P2S5, LiI/Li2S/P2S5/P2O5).

5

Separators

The separator in liquid electrolyte batteries has to prevent electronic contact of the electrodes and enable free ionic transport. The separator material must be chemically stable, about 8–25 mm in thickness, and exhibit a porosity of typically 40% with an average pore size well below 1 mm, to hold sufficient liquid electrolyte and to prevent penetration of particles from the electrodes. To ensure a uniform current distribution and to prevent the growth of dendrites on the negative electrode, the permeability of the separator must be uniform. The resistance of a separator to ion transport depends on the thickness, porosity, and tortuosity. As a measure of its transport properties, the air permeability (Gurley value) is determined as the time required for a specific amount of air to pass through the separator material (of given porosity, thickness, and cross-sectional area) under a specific pressure. The mechanical strength is described by Young’s modulus; a 25 mm-thick separator should withstand a tensile strength of 1000 kg cm−2 in the machine direction and a puncture stress according to the maximum load of 0.3 kg on a specified needle. The separator should wet easily, lay flat, and be dimensionally stable. The thermal shrinkage is required to be less than 5% (after 60 min at 90  C); polyethylene (PE) membranes can shrink by 10% during 10 min at 120  C. Shrinkage and softening of the separator at elevated temperatures can cause an internal short and thermal runaway. Common PE–polypropylene (PP) bilayer separators in lithium-ion batteries shut down at about 130  C without loss of mechanical integrity (melting temperature of PP is about165 C). As discussed, Li-metal anodes have the tendency to develop a dendritic deposit. The dendrites growing from the anode surface should not penetrate the separator. This will lead to local short circuits. If the resistance of the short is high, the short current will be small and there will be no thermal damage to the separator. The short will not be stable as the Li will dissolve at the cathode potential. The result will be short dips in the cell voltage during charging. If the resistance will be sufficiently low, a local hotspot will develop, and the cell can enter into a thermal runaway. The properties of different types of separators are shown in Table 9.

Table 9

Separators in liquid electrolyte lithium-ion batteries.

Material

Example

Fabrication and properties

1. Microporous polymer membrane

1. and 2. Semicrystalline polyolefin (PE, PP, high-density polyethylene (HDPE), UHMW-PE, PE-PP, PS-PP, PET-PP blends); polyoxymethylene, poly(4-methyl-1-pentene) 3. PVDF, polyacrylonitrile (PAN); microporous gel polymer electrolyte

2. Nonwoven fabric mat

Cellulose; polyolefin, polyamide, PTFE, PVDF-HFP, PVC, polyester

3. Inorganic composite separator

1. and 2. Metal oxide powders (TiO2, ZrO2, LiAlO2, Al2O3, MgO, CaCO3) in a polymer matrix (PVDF-HFP, PTFE) 3. AlO(OH)/polyvinylalcohol (PVA) on PET Modified polyolefin

1. Dry process (melt-extruding of a precursor film, annealing to improve crystallinity, uni- and biaxial cold and hot stretching): generates distinct slit-pore and straight microstructure 2. Wet process (extrusion of a polymer resin in paraffin oil, stretching, extraction of the oil): generates interconnected spherical or elliptical pores, high tortuosity 3. Phase inversion method (precipitating the dissolved polymer by solvent exchange, casting on a flat substrate): generates an undesired asymmetrical sponge-like structure with open pores on the side facing air and a dense surface on the bottom side Fibrous webs (from a paper-making process, solution extrusion, wet-laid method, melt blowing, etc.) are bonded by (1) resin (using a foreign adhesive) or (2) thermoplastic fibers. High porosity (60–80%), labyrinth-like pores, large pore size (20–50 mm) 1. Phase inversion method (see above) 2. Thermal pressing or calendaring 3. Sol–gel method: multiple coating on a substrate 1. Wetting agents (surfactants) 2. Grafting of hydrophilic functional groups (plasma treatment in a gas atmosphere, sulfonation, fluorination, grafting polymerization with glycidyl- or methyl methacrylate) 1. Dipping or spraying; adhesive bonding of separator and electrodes 2. Phase inversion method (e.g., acrylonitrile-methyl methacrylate copolymer in dimethylformamide (DMF) on a PE membrane is immersed in a water bath) 1. Dipping and evaporation of the solvent 2. In situ polymerization, e.g., polyoxyethylene

4. Surface modification 5. Surface polymer coating

Gel-like polymer film (PEO, PVDF-HFP) on microporous membranes

6. Impregnation of gel polymer electrolyte

Polymer electrolyte in microporous membranes

136

• • •

6 6.1

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview

Polyolefin separators made by the dry process (e.g., Celgard, PP–PE–PP) have an open and straight pore structure, whereas separators from the ‘wax blend’ wet process (e.g., Exxon Mobile ‘Tonen’, PE) show a tortuous pore structure. Inorganic fillers may favor wettability and liquid retention. Separators as thin as 10 mm are being used to optimize cell energy density. At around 130  C, a shutdown separator stops all cell reactions at a temperature below the melting point of lithium (182  C). Nonwoven separators are mats of numerous natural or synthetic fibers bound together by resins or thermoplastic fibers. As thicknesses of 20 mm and less cannot be easily produced and owing to the rough surface, nonwoven separators have only been used as the supporting framework to make gel polymer electrolyte batteries. Inorganic composite separators, containing ultrafine metal oxide particles in a polymer matrix, provide improved wettability for alkylene carbonates, g-butyrolactone and other organic solvents. They show good thermal stability and nearly no shrinkage at high temperatures. Degussa’s ‘Separion’ separator combines a flexible perforated nonwoven polymer mat of Polyethylene terephthalate (PET) with porous ceramic coatings (Al2O3/SiO2) on both sides. As an inorganic binder sol to suspend aluminum oxide powders on a PET substrate, a mixture of tetraethoxysilane, methyltriethoxysilane, and (3-glycidyloxypropyl) trimethoxysilane in the presence of hydrochloric acid was reported in the literature.

Cells Cell design

Basic cell designs of rechargeable lithium-ion batteries are shown in Fig. 3. All four types of completely sealed cell construction are offered commercially with a variety of electrode materials and with both liquid and gelled electrolytes. Except for coin cells, thin, flexible electrodes are being used. Thin metal foils or grids form the electrode substrates and current collectors. Aluminum is the preferred material for the positive electrode and copper is preferred for the negative electrode. Cells with a Li, a Li-alloy or a Li non-carbon intercalation compound can use the same cell designs. Cell design can on one hand be optimized to achieve a high specific energy at cell level by optimizing the fraction of active materials in a cell, for example by using electrodes with a high loading of active materials per unit electrode area and a low porosity. On the other hand, if high-rate capability is required, electrodes will have a lower loading and a higher porosity. High specific energy lithium-ion batteries (up to 270 Wh/kg at cell level) have been used in electric vehicles and deliver moderate rates of up to 6C pulses or 1.6 kW/kg. SAFT has demonstrated very high power (VHP) lithium-ion technology capable of 8 kW/kg (for 2 s) and 12 kW/kg (for a millisecond) in the 270-V emergency battery for an F-35 aircraft. The development of batteries with anode materials will likely follow similar patterns.

Fig. 3 Cell types: (a) wound cylindrical cell. (b) Coin cell with stacked electrodes. (c) Prismatic cell with flat-wound electrodes. (d) Pouch cell with stacked electrodes.

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview

137

In a typical Li-ion cell design, the weight of the positive electrode material constitutes a major part (46%) of the total mass of the cell, followed by the negative electrode (18%), electrolyte (15%), collectors (11%), electrode additives (8%), and separator (2%). Most of the battery volume is consumed by the electrolyte (32%), followed by the electrode active materials (each 25%), electrode additives (10%), collectors (6%), and separator (5%). Thus, reduction of the electrolyte volume and electrode porosity seems to be beneficial to the energy content of the cell. The trade-off is lower rate capability. Particle sizes, electrode porosity, and electrode additives must be optimized to gain the best particle contact and electrolyte wetting. Replacing the graphite anode for example with a Li-metal or silicon anode, the weight and volume fraction of the anode is reduced resulting in higher specific energy at cell level. However, it must be considered that the negative graphite anode is already a relatively small portion of the weight of the cell materials in an Li-ion cell. Therefor the possible gain of specific energy (energy per weight) by for example replacing graphite with lithium is limited. As the graphite has a relatively low density of about 2,2 g cm−3 compared to roughly double the density of the cathode materials, the gain in energy density (energy per volume) is considerably larger.

6.2

Charging of lithium batteries

Lithium secondary batteries including Li-ion batteries and those discussed in this chapter should be charged carefully; the upper voltage limit should not be exceeded to avoid overcharge. Lithium batteries should not be subjected to trickle charging (float charging), i.e., maintaining them in a fully charged condition by continuous charging and balancing self-discharge, interrupted by occasional discharge only. A typical charging method for cells is the CCCV protocol. Charging switches from constant current (CC) to constant voltage (CV) before the cell voltage reaches its upper limit (Fig. 4a). The charge voltage, after rising during constant current charging, is maintained at the upper voltage limit. At this point, the charge current drops. At a predetermined minimum current, the fully charged battery is cut off. The charge rate of Li-the upper charge voltage limit is not reached, lithium-ion cells can be charged at relatively high rates; currents up to 1C rate (current corresponding to rated Ah capacity) are common. If charge rates are too high, lithium metal can be plated at the negative electrode, which leads to rapid cell capacity fade. For the same reason, charge rates should be reduced at lower temperatures. The maximum allowable charge rate is always lower than the maximum allowable discharge rate. The charge rate for the chemistries discussed in this chapter depend on the type of anode used. 2D Li-metal anodes are prone to dendrite formation at high charge rates. 3D Li-anodes and Li-alloy anodes are capable of higher charge rates. The available capacity (stored electric charge) of conventional batteries heavily depends on the discharge rate; therefore, the remaining capacity during a discharge with varying currents can hardly be predicted. Fortunately, the available capacity of lithium cells typically does not decrease significantly with higher discharge rates, although the shape of the discharge characteristics (cell voltage vs. discharged capacity) depends on the discharge rate (Fig. 4b). The state-of-charge (SoC) is the percentage of the maximum available charge that is present inside the battery, SoC(t) ¼ Q(t)/Q0 (previous full charge). The state-of-health (SoH) is the ability of an aged battery to deliver the rated performance compared to the fresh battery; with respect to voltage, self-discharge rate, charge acceptance, and internal resistance, SoH(t) ¼ Q0(t)/QN (rated capacity). The voltage difference between a fully charged cell and a discharged cell can be more than 0.5 V, so OCV is a rough measure of SoC. For Li-metal cells, the voltage dependence is a function of the change of the cathode voltage with SoC. For chemistries with a

Fig. 4 Charge and discharge. (a) Qualitative charging characteristics of a lithium secondary battery. (b) Discharge characteristics at different discharge currents.

138

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview

low voltage dependence of SoC (e.g. LiFePO4), techniques such as Coulomb counting are used. If lithium batteries are not fully discharged and recharged frequently (deep cycle), recalibration of an external electronic fuel meter, such as an SoC meter, could be required after approximately every 30th cycle.

6.3

Cell balancing

Lithium cells must never be overcharged, because there are no side reactions that could consume the excessive energy without damaging the cell. With conventional battery technologies, overcharging a battery with multiple cells connected in series and/or parallel at low rates is used to equalize the SoC between cells. In lithium batteries, active equalization circuitry is needed to keep all the cells within a pack in a similar SoC. The battery monitoring system (BMS) eliminates mismatches of single cells in series or parallel through different currents to individual cells in a series string. Cells connected in series will receive the same charging current; balancing for example can be accomplished by connecting high-impedance shunts across the cells with the highest SoC. Moreover, cell balancing in batteries with multiple cells improves efficiency and prolongs battery life. Some applications such as cell phones require only a single cell; balancing is therefore not required. 1. SoC mismatch: Even with strict process controls, individual cells in a battery will have differences in the rate of self-discharge. Temperatures in a battery can also differ from cell to cell causing different rates of self-discharge. On discharge, the cell with the lowest SoC will determine battery capacity. On charge, the cell with the highest SoC will determine the end of charge. In an unbalanced battery, the capacity will always be less than the capacity of the weakest cell. Cells having the same capacity are considered balanced when they have the same open-circuit voltage (OCV), as a good measure of the relative SoC. (a) By current shunting the charging current across each cell in a series combination, the fully charged cell will not receive any further current, and cells can be balanced at the fully charged state. Current shunting EEPROMs do not protect any cell from too low voltages, and the shunted current is dissipated as heat. (b) Low-power dissipation is a low-cost option and the only option for high-power batteries, because the small equalization currents can be shunted through the cell voltage measurement cables. The charging current must be reduced if one of the individual cells reaches the charging voltage before the others. This method can also be used to balance cells in a battery at rest; however, this increases the self-discharge of the battery. 2. Capacity/energy (C/E) mismatch: This requires balancing during both charge and discharge periods. In a battery pack with individual cells of different capacities, the cells are kept at the same SoC by applying differential amounts of current on every charge/discharge cycle. The ‘energy distribution’ does not dissipate heat but transfers excessive energy from one cell across the other cells where required. Less expensive heat sinks are required, and the available energy is no longer limited by the weakest cell in the battery pack. This, however, requires much more complex circuitry.

6.4

Manufacture

The center of global Li-ion battery manufacturing is Japan, China, Taiwan, and Korea, followed by the USA, Canada, France and Germany.65 Joint ventures between automotive manufacturers and battery suppliers are known from Toyota with Panasonic and Sanyo, Nissan/NEC, Honda/GSY, Ford/Magna, Mitsubishi/GSY, Daimler/BYD, BOSCH/SDI, Continental/ENAX and JC/SAFT. The difference in manufacturing of the type of cells described here will depend on the technology chosen. In addition to a different process for anode manufacturing, additional changes may be required. The manufacture of lithium cells requires dry rooms with relative humidity preferably less than 1%, because water reacts violently with lithium and even traces of moisture are detrimental to the long-term cell performance. Some liquid electrolyte components are highly flammable, and active materials and additives consist of very fine particles and can pose a safety hazard. Serious factory fires have been reported in the past few decades when the materials were unintentionally mishandled. Lithium-ion batteries are assembled in the discharged state and then subjected to an activation process called formation, which typically consists of several charge/discharge cycles. During formation, the cells become live and must be handled with care. Depending on the cathode material used in cells with a Li-containing anode, these cells may be assembled in a charged state just like primary batteries. The cell will contain stored energy that can be released for example by an accidental short. Safety precautions to prevent fires must be taken starting with the electrolyte filling.

6.5

Recycling

Even though Li-ion batteries are produced at a very large scale, recycling66,67 of these batteries and related components is still in its fledgling stage. The impact of cells on recycling described in this chapter will depend very much on the specific cell chemistry including the nature of electrolytes. A truly circular battery economy is still a vision, even for Li-ion that have a history of over 30 years. In the future, all materials should be preserved with as little loss of quality as possible, and environmental pollution caused by excessive raw material mining and massive burning of fossil fuels should be avoided. EU legislation has set binding targets, mass fractions and quotas for battery recycling, although this low-grade recycling is far from the ideal of a circular economy. With the development of new cell chemistries not only the changes in mass manufacturing compared to Li-ion but also the recycling processes should be taken into account.

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Overview

139

In the currently prevailing pyrometallurgical process, all types of lithium-ion batteries are placed in a furnace to reduce cobalt, nickel and copper to a metal alloy, while lithium, aluminum, manganese and other metals end up in furnace slag. The recycled “slag” product is used as a filler for concrete or road surfacing. Hydrometallurgy and mechanical separation achieve higher recycling efficiencies and recover aluminum, manganese, graphite and electrolytes of sufficient quality. These processes are costly, less robust, and limited to well-sorted battery material, which is not currently supported by battery collection and labeling. To maximize recovery and environmental benefits, only one cell chemistry is processed at a time. In the medium term, recycling will not be able to meet the annual demand for battery materials, especially since vehicle and stationary batteries will not return for at least a decade. R&D activities focus on reducing battery masses and the content of expensive or scarce metals. As recovered products have low value, battery recycling is not economically self-sustaining through the sale of recycled products. Incentives for recovery, labeling of cells for recycling, quality targets for recovered products, and reasonable recycling rates are needed.

7

Summary and outlook

Lithium primary batteries with a lithium metal or lithium alloy anode are in use for over 60 years. Initial efforts to develop rechargeable lithium metal anode systems with liquid organic electrolytes have failed due to safety issues. Lithium-ion cells with graphite intercalation anodes have been commercially successful starting in the 1990s. However, the development activities of lithium metal and lithium alloy anodes have been renewed and intensified in the last decade driven by the promise of higher energy density using both liquid and solid electrolytes. The development of solid electrolytes has progressed to the point that their conductivity is comparable or even better than that of liquid electrolytes. For the same cathode and anode materials as in Li-ion cells they promise an improved safety but no improvement in energy density. Higher energy density and higher specific energy is only possible using a Li-metal or a Li-alloy anode such as Li-Si alloys. Several issues must be resolved before a large-scale commercial use, for example in electric vehicles is possible. One development topic is the cycle life of anodes not based on carbon. Another issue is the interface between the solid electrolyte and the cathode material which expands and contracts during cycling. As a possible solution, hybrid systems with a solid electrolyte at the anode side and a liquid electrolyte at the cathode are under development.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

Harris, W. S. Electrochemical Studies in Cyclic Esters; Thesis, Univ. of California: Berkeley, CA, United States, 1958. Whittingham, M. S. Chalcogenide Batteries; Patent US 4009052, 1977. Haering, R. R.; Stiles, J. A. R.; Brandt, K. Lithium Molybdenum Disulfide Battery Cathode; Patent US 4224390, 1980. Fenton, D. E.; Parker, J. M.; Wright, P. V. Complexes of Alkali Metal Ions with Poly(Ethylene Oxide). Polymer 1973, 14, 589. Ozawa, K. Brief Survey on the Historical Development of LIBs. In Encyclopedia of Electrochemistry: Batteries; Passerini, S., Bresser, D., Moretti, A., Varzi, A., Eds.; Wiley-VCH: Weinheim, 2020. Mizushima, K.; Jones, P. C.; Wiseman, P. J.; Goodenough, J. B. LixCoO2 (01000 mAh g−1), even after 150 cycles at 0.5C, which was found to be superior to any other Li/S batteries comprising elemental sulfur and polymer binders. Achieving high area-specific capacity and capacity retention at a higher C-rate is one of the development directions for future R&D studies on lithium-sulfur batteries. The quantity of binders reported in the literatures is more than five wt%, while the binder content in commercial electrodes is restricted lower than 3 wt%. Therefore, exploring new-type multifunctional binders which simultaneously alleviates or overcomes multiple distinctive issues. The selection of polymer binders is a trial-and-error process, and their design, engineering, and synthesis are very tedious.13 More importantly, understanding of the function mechanism of binders in Li/S batteries can be understood by in situ or operando detection to reveal the interaction behaviors between binder molecules and active sulfur intermediates. A combination of molecular simulation calculation is advisable, which can offer new horizons and new inspirations at the molecular and atomic scale for the rational design of functional binders. In a recent attempt Senthil et al.14 developed aqueous processable tragacanth gum-based binder to hold a high sulfur loading over 12 mg cm−2 without sacrificing the sulfur utility and reversibility. The intrinsic rod and sphere-like saccharidic conformal fraction’s multifunctional polar units acted as active

Lithium Batteries – Lithium Secondary Batteries – Li-Metal Battery | Lithium–Sulfur

145

channels to reach the sulfur particles. The binder confined polysulfides with 46% improvement and prevented the volume changes within 16% even at a 4C-rate. The highly flexible LidS battery delivered a gravimetric energy density of 243 Wh kg−1, demonstrating high reactivity of sulfur along with appreciable conformality.

3

Electrolyte

The electrolyte plays a key role in the development of high-performance Li/S batteries. Li/S cell can reversibly discharge/charge in various electrolytes such as sulfide-based solid electrolytes (Thio-LISICON), polymer electrolytes (PEO, PAN, etc.), ionic liquids, and liquid electrolytes (ether, carbonate-based electrolyte). Although each electrolyte has a problem, most Li/S batteries have been studied using ether-based electrolyte.15 Especially, DOL/dimethyl ether (DME)-based binary solvent system has been generally used in Li/S batteries along with Li salts, such as lithium bis(trifluoromethanesulfonyl) imide (LiTFSI) or lithium bis(fluorosulfonyl) imide (LiFSI), and LiNO3 as additive. The linear ether solvent, DME, has a relatively high solubility of lithium polysulfides, which can provide fast conversion kinetics for the redox reactions of polysulfides. On the other hand, the cyclic ether solvent of DOL with low viscosity (0.6 mPa s−1) contributes significantly to forming protective SEI layers on the surface of the Li metal anode.16 To ensure high ion conductivity, Li salts must possess high chemical/electrochemical stability, high solubility, and a high degree of dissociation. Among various Li salts, the LiTFSI has been widely used due to its high thermal stability and favorable compatibility. The additive LiNO3 has been widely used and is generally believed to facilitate the formation of a robust SEI layer on Li metal anodes by decomposing the nitrate products. However, the generation of gases in the lithium/sulfur pouch cells still limits their applications. Despite these advantages, the extensive utilization of electrolyte systems containing the DOL/DME binary solvent, LiTFSI salt, and LiNO3, still many challenges are to be resolved. For example, ether-based electrolytes have a relatively poor anti-oxidation ability and are prone to have an anodic decomposition, which can lead to increased overpotential. Further, ether-based electrolytes can be also reduced by the highly reactive Li metal anode.17 In some cases, the resultant inhomogeneous and unstable SEI layers can continuously expose the fresh Li active sites, resulting in continuous electrolyte consumption, thus increasing polarization. It is worth mentioning that the evolution of gases such as CH4 and H2 induced by side reactions between the Li metal anode and the organic solvent or Li salts can cause an enormous internal pressure in the cell devices. Eventually, this can cause damage to the electrode interface structure, electrolyte leakage, and corrosion of electrode surfaces. The most severe problem is the dissolution of LiPS. Especially, the large amount of electrolyte is required for the good electrochemical property, which is a critical issue for the commercialization of Li/S batteries. Although there was no successful method to reduce the critically E/S ratio, many kinds of research have been reported for the lean electrolyte. For example, the concept of sparingly solvating electrolytes has been recently proposed, in theory, as a pathway to control reactant distribution owing to the very low solubility of polysulfides in such systems and thus their short residence time in the solvated state. Many well-known precipitation-dissolution secondary electrodes operate successfully using the sparingly soluble concept, including Pb/PbSO4, Ag/AgCl, and Fe/FeCl2. Solvates, ionic liquids, and super concentrated electrolytes with reduced sulfur and polysulfide solubility are widely explored for analyzing the performance of Li/S cells. The solubility of polysulfides in Li/S batteries can be suppressed through properly designed electrolyte systems. For example, very high salt concentrations in the electrolyte mean that most solvent molecules strongly interact with the Li salt, leaving minimal or none to solubilize polysulfides. A diluent, typically a hydrofluorinated ether (HFE), is often used to reduce the high viscosity of the electrolytes while reducing polysulfide solubility further. High salt concentration electrolytes based on DOL:DME, glymes, and acetonitrile (ACN) exhibit good polysulfide solubility control and electrochemical performance in Li/S batteries. In addition to the prevailing lithium dendrite growth issue on the lithium metal anode, another major challenge includes the anodic stability of electrolyte vs. cathode, transition metal ion (TM) dissolution from the cathode and their migration to the lithium anode via the electrolyte medium and its possible contamination of the mitigating strategies have been adopted. Among these, the use of additives (usually r(Ni3+)). As a result, the Li-O distance decreases, the Ni-O distance increases, and the a and c lattice constants also increase.3,8 Neutron powder diffraction and magnetic properties can be used to confirm the presence of a few percent Ni2+. In LiNiO2, the stoichiometric composition is closely related to the battery properties. On the other hand, the compounds with lithium excess phase with [Li]3b[LizNi1-z]3a[O2]6c have Li in the Ni layer.9,10

3

Formation diagram

Fig. 2 shows the formation diagram of LiNiO2.11 To obtain the stoichiometric composition of LiNiO2, it is necessary to keep Ni in a divalent state during the synthesis process. Reactions in an oxygen atmosphere or at relatively low temperatures are used, or fine particles are used to increase reactivity. Fig. 3 shows the example of the synthesis conditions for obtaining the stoichiometric composition of LiNiO2 from Li2O2 and NiO, indicating that the synthesis temperatures strongly affect the LiNiO2 stoichiometry. The lithium excess composition is used during synthesis to prevent Ni2+ insertion into the lithium layer. A phase relation diagram can be used to consider the reactions during the synthesis and this consideration is a common issue for all nickel-containing layered rock salt-type oxides.

Fig. 1 Structure of LiNiO2. The a-NaFeO2-type structure (R3− m), structure based on the rock salt structure and two-dimensional triangular Ni lattice. A hexagonal unit cell is shown with the solid lines. The lattice constants are a ¼ 2.876 A˚ and c ¼ 14.191 A˚ . The NiO6 octahedra of trigonal symmetry share their edges to form a triangular Ni lattice. The spinel structure for LiNi2O4 (Li0.5NiO2) is also indicated in the figure.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Nickel Oxide

229

Fig. 2 The formation diagram of Li2O-NiO-1/2O2 system. The compositions, Li2NiO3, Li2NiO2, NiO2, and NiO exist around LiNiO2.11 Figure from Bianchini, M.; Roca-Ayats, M.; Hartmann, P.; Brezesinski, T.; Janek, J. There and Back Again—The Journey of LiNiO2 as a Cathode Active Material. Angew. Chem. Int. Ed. 2019, 58 (31), 10434–10458. https://doi.org/10.1002/anie.201812472, need permission.

Fig. 3 The relationship between the synthesis temperature and compositions in LiNiO2 prepared from Li2O2 and NiO. Stoichiometric LiNiO2 is obtained around the synthesis temperature of 700  C.

4 4.1

Composition and structure details in the phase relation diagram Li2O-NiO-O2 region (general view)

Information on the related phases around LiNiO2 is important for obtaining LiNiO2 with excellent electrochemical properties. Since Ni can take 2, 3, or 4 valence states, the end members for describing the phase relations are rather complicated. The phase relation diagram shown in Fig. 2 is described by Li2O-NiO-O2 ternary system. The major phases present in this diagram are Li2O and Li2O2, rock-salt NiO, layered LiNiO2 and Li2NiO3, spinel LiNi2O4, and Li2NiO2. NiO2 is a phase in which lithium is deintercalated from LiNiO2 and is produced by the battery reaction.12 Along the NiO-NiO2 tie-line, the existence of new oxides with Ni1.25O2, Ni1.5O2, and Ni1.75O2 compositions has also been suggested.13

230 4.2

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Nickel Oxide Composition region obtained by normal synthesis: Li1-zNi1+zO2

The Li1-zNi1+zO2 composition exists on the LiNiO2-NiO tie-line. Compositions on this tie-line are obtained under normal synthetic conditions. The phase transition from layered to cubic rock-salt type structure occurs around z ¼ 0.3–0.4 (shown in Fig. 1).2,14 There are two distinct regions: a rock-salt type structure (near the compositional region of NiO) and a layered rock-salt type structure (near the region of LiNiO2). The ideal composition of LiNiO2 is z ¼ 0 and is the end member of the Li1-zNi1+zO2 solid solution. Usually, the synthesized sample will be a compound with z ¼ 0.01–0.05. The difficulty in synthesizing the stoichiometric composition of LiNiO2 has prevented its mainstream use, although commercialization of LiNiO2 as a cathode material has been attempted. Non-stoichiometric composition is a common issue for cathode materials with layered rock salt type structure (the CoMnNi ternary system, the high nickel system and the excess lithium system).

4.3

Lithium excess region: Li1-zNi1+zO2

Under synthesis conditions at an oxygen partial pressure of 1 atm, a composition close to z ¼ 0 can be obtained with Li1-zNi1+zO2. On the other hand, under synthesis conditions with higher oxygen partial pressure, compositions in the z < 0 region can be obtained. The end member of the high pressure phase is Li2NiO3 (Ni4+)15 with a layered structure containing excess lithium ions in the nickel layer (z ¼ −1/3 in Li1-zNi1+zO2 corresponds the composition [Li]3b [Li1/3Ni2/3]3aO2). The nickel layer has an ordered Li/Ni arrangement and a monoclinic C2/m symmetry. Compounds in the range −0.11 < z < 0 can also be stabilized.16

4.4

Electrochemical lithium de-intercalated region: LixNiO2

The electrochemical Li de-intercalated region from LiNiO2 corresponds to the charging reaction of the battery cathode. The layered structure of Li0.5NiO2 with 0.5Li deintercalated from LiNiO2 changes to the spinel structure with LiNi2O4 composition upon heating.17 The layered and the spinel structures have the same ccp oxygen sublattice. The change from the layered structure to the spinel structure causes a quarter of the Ni ordered in the six-coordinated octahedral positions between the layers of the ccp arrangement in the layered structure to move to the adjacent Li planes, and all lithium ions are shifted to the tetrahedral positions (shown in Fig. 1). The spinel structure is the ground state of the Li0.5NiO2 composition.15 Moreover, the spinel phase reversibly uptakes Li according to the composition LixNi2O4 (1 < x < 2).8 By heating the corresponding layered compound, the spinel phase LixNi2O4 with x < 0.5 has been obtained.8 Li2NiO2 can also be synthesized.3,18 Li2NiO2 is the isomorphism of Li2MnO2 and Ni(OH)2 with an hcp oxygen arrangement, and Ni occupying the octahedral site and Li the tetrahedral site. When Li intercalates into LiNiO2, Li2NiO2 is formed by a two-phase reaction.

4.5

Physical properties: Electronic properties

NiO is an insulator, and the electrical conductivity increases with the introduction of Li. The electronic conductivity of LiNiO2 with stoichiometric composition is 10−1S cm−1.2,19 Ni is in the low-spin trivalent state in the electron configuration t62g−e1g .19 The Jahn-Teller distortion is induced, which leads to the relaxation of the degenerate ground state. In NaNiO2, cooperative ordering of octahedra ordered along their long axes occurs, and the crystal symmetry is decreased to monoclinic C2/m. Two long (2.09 A˚ ) and four short (1.91 A˚ ) Ni-O bonds have been observed locally in LiNiO2.19–21 However, ab initio calculations do not reproduce the ground state.13,22 The distorted ground state of Jahn-Teller is more stable than the undistorted ground state, but the zigzag arrangement of the Jahn-Teller octahedron in the P21/c space group is more stable than the parallel arrangement. On the other hand, ground states with Ni2+/Ni4+ disproportionation and ground states with Ni trimer have also been suggested.20,22 The discrepancies between calculations and experiments in various reports may be due to differences in the actual stoichiometric composition of LiNiO2. Another issue is whether the actual nickel valence is 3+. It is too simplistic to assume that nickel electrons are completely transferred to oxygen and that Ni3+ or O2− is present, so it is important to know how much actual charge transfer is occurring.23 This regular arrangement of Jahn-Teller ions causes monoclinic structures. A similar monoclinic phase appears when lithium is de-intercalated by 0.5 from LiNiO2. Either the appearance of monoclinic phase is due to the ordered arrangement of lithium vacancies, or the ordered arrangement of Jahn-Teller ions which was prevented for some reason in LiNiO2, is removed by the introduction of lithium vacancies, and the ordered arrangement is achieved. Also, this phenomenon in this material illustrates the insignificance of our thinking of the de-intercalation mechanism in terms of the formal charges of the transition elements.

4.6

Physical properties: Magnetic properties

Magnetic properties are strongly influenced by the stoichiometric composition and the ordered arrangement of Ni.24–26 In particular, the Ni present in the Li layer affects the magnetic behavior.27 LiNiO2, which is close to the stoichiometric composition, shows typical Curie-Weiss temperature dependence.25–28 However, the detailed relationship between magnetic properties and structure is unknown because the actual composition of the materials for which magnetic properties have been measured has not been precisely determined. In the stoichiometric composition (ideal composition LiNiO2), the layers containing the magnetic ion Ni are all separated from the nonmagnetic layers (O-Li-O), so that only in-plane magnetic interactions are possible. The Ni in the Ni layer is arranged in a triangular lattice (shown in Fig. 1). If Ni3+(d7) has antiferromagnetic interactions, the 90 exchange interaction

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Nickel Oxide

231

of Ni3+−O−Ni3+ would be a frustration system not allowing for long-range magnetic ordering. On the other hand, once the composition becomes non-stoichiometric and Ni2+ (S ¼ 1) is located in the Li layer, the 180 Ni3+−O−Ni2+ exchange interaction (ferromagnetism) occurs. The relationship between the composition and magnetic properties is considered as follows. The magnetic properties of LiNiO2, which is very close to the stoichiometric composition, show a spin glass transition when measured at low temperatures in low magnetic fields, and the transition temperature is systematically varied by a few percent Ni disordered arrangement as shown in Fig. 4. When a significant amount of Ni2+ is present in the Li layer, a magnetic transition also appears around 200–300 K.26,29 The Ni2+ present in the Li layer exhibits a cluster glass (z > 0.24) with a 180-degree exchange interaction causing ferromagnetic behavior and magnetic clusters starting from this Ni2+. A further increase in Ni2+ leads to macroscopic ferromagnetic behavior.27,30 Thus, magnetic measurements are the most sensitive method to detect the stoichiometric composition of LiNiO2.

4.7

Synthesis

4.7.1 Solid state reaction The solid state method is used to synthesize LiNiO2. Lithium and nickel precursors are mixed and calcined.8,31,32 During synthesis, an oxygen atmosphere is necessary to keep the valence of nickel at 3+.33 At high temperatures (above 800  C), Ni3+ is reduced to Ni2+ to form Li1-zNi1+zO234 (shown in Fig. 3). At calcination temperatures below 500  C, samples with stoichiometric compositions can be synthesized by using mixtures of NiO and Li2NiO3 or Li2O. Usually, the synthesis can be completed at 650–750  C. To obtain LiNiO2, the decomposition temperature of the lithium precursor is kept as low as possible. Since Li is lost in the high temperature range, lithium hydroxides, nitrides, and oxides/peroxides are used. Li2CO3 should be avoided because it decomposes over 725  C.7 To compensate for the loss of Li during the calcination process, excess Li precursors are used. A high Li/Ni ratio facilitates obtaining a stoichiometric composition (Li1-zNi1+zO2 with z ¼ 0.0). Uniformity of the particle surface and interior is also a issue.35 Utilizing large amounts of excess Li yields products that are located in the region of the phase diagram containing Li2NiO2 and Li2NiO3, depending on the oxidation atmosphere during the synthesis.3

4.7.2 Coprecipitation method The coprecipitation method36 is suitable for industrial production. Aqueous solutions are mixed with nickel sulfate or nitrate, and NaOH is added to precipitate Ni(OH)2. NH4OH solution can also be used as a chelating agent. To obtain a homogeneous hydroxide with controlled particle shape and size, the concentration, temperature, mixing, and pH value of the solution are controlled. The coprecipitated precursor is then mixed with the Li precursor and calcined.

4.7.3 Sol-gel method In the sol-gel method, nickel and lithium precursors (usually nitrate, hydroxide, or acetate) are dissolved in water or a mixture of water and ethanol/methanol. Chelating agents such as adipic acid,37 oxalic acid,38 citric acid,38,39 acrylic acid,38 polyvinyl butyral (PVB),38 and triethanolamine (TEA)38 are added to stabilize the sol. Synthetic methods without the addition of chelating agents have also been reported.40 By the calcination after heating at 400–600  C, stoichiometric materials with small crystallite size are obtained.38,40 (a)

(c)

(b) 700

160 z | 0.08 (900°C)

600

140

z | 0.04 (850°C)

2.5

120

400 0% 400

Q /K

sample I sample II

300

2.0

100

z | 0.00 (650°C)

1.5

80 60 40

100

0

50

100 150

200

T (K)

250

300

0.5

20

H =104 Oe 0

1.0

8%

200

meff /mB

H/M (mole/emu)

z | 0.00 z | 0.01 z | 0.02 z | 0.03

3.0

350

0

0

0.05

0 0.10

z in Li1-zNi1+zO2

Fig. 4 (a) Magnetic properties of Li1-zNi1+zO2. Relationship between spin glass transition and non-stoichiometry is indicated. The glass transition increases with the disorder at the Li 3a site. (b) Sample dependence of H/M at H ¼ 104 Oe. (c) Sample dependence of Y and meff. The Y temperature increases with the non-stoichiometric compositions.27 Higher disordered samples with z > 0.08 provide cluster glass behavior. Part labels (b) and (c) from reference. Yamaura, K.; Takano, M.; Hirano, A.; Kanno, R. Magnetic Properties of Li1-xNi1+xO2 (0  x  0.08). J. Solid State Chem. 1996, 127 (1), 109–118. https://doi.org/10.1006/jssc. 1996.0363, need permission

232

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Nickel Oxide

4.7.4 Pechini method The Pechini method was also used. The precursor is mixed with citric acid and ethylene glycol to form polyesters.41

5

Electrochemistry

Electrochemical deintercalation of lithium from LiNiO2 proceeds around 3.8 V, between 3.0 and 4.3 V, within the potential window of the electrolyte, yielding NiO2.3 The capacity is as high as 275 mAh g−1 (Fig. 5). Capacities of around 200–255 mAh−1 have been obtained for the practical applications.42,43 LiNiO2 has a number of issues such as degradation characteristics, thermal stability, initial efficiency, and capacity loss during cycling. Attempts to stabilize it by adding various additive elements and solid solution systems have been developed.

6

Phase transitions associated with lithium de-intercalation

The lithium de-intercalation from LiNiO2 has been investigated.3,31,43–45 The relationship between structural changes and electrochemical properties is shown in Fig. 6. Since there are various reports of structural changes, a simplified diagram based on the basic concepts is indicated. LiNiO2 has a rhombohedral R3−m symmetry. When lithium is de-intercalated, the amount of lithium changes in rhombohedral symmetry in the initial stage of de-intercalation. (Fig. 5). This region (x < 0.15) is labeled as H1. A monoclinic phase M appears at around x ¼ 0.2 and becomes monophasic at x ¼ 0.25. The M phase is stable up to x ¼ 0.5.45 Further deintercalation leads the rhombohedral structure labeled as H2. As the lithium content decreases to near NiO2 (x ¼ 0.8), a new rhombohedral structure, H3, appears. The a and c lattice constants of this phase are small (a ¼ 2.8154(7) A˚ and c ¼ 13.363(6) A˚ 46). This structural change is similar to that of LiCoO2. The H2-H3 transition is accompanied by shrinkage of the unit cell with a large

Fig. 5 Charge-discharge curves and structure changes of LiNiO2 during lithium deintercalation.44 Derivative dy/dE vs y plot of the discharged curve (above) and the expected structural changes in the active domain.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Nickel Oxide

233

Fig. 6 Schematic drawing of the structure appeared during the deintercalation from Li1-z-xNi1+zO2. Deintercalated structures are drawn for the stoichiometric composition (z  0) and non-stoichiometric composition (z  0.05).

volume reduction, causing significant strain in the material.47 However, the structure change in the region close to the NiO2 composition and the monoclinic phase region in the intermediate region has not been determined. This corresponds to the fact that the stoichiometric composition of LiNiO2 is difficult to obtain and the Ni present in the Li layer has a significant influence on the structural changes in those regions. The stoichiometry dependence of structural changes associated with the de-intercalation is still unclear.47–49 The stoichiometric composition of LiNiO2 changes from a rhombohedral (H1) to monoclinic (M), rhombohedral (H2), rhombohedral (H3) phase with a small c axis. This structural change is associated with the initial composition of LiNiO2, and the monoclinic phase in the intermediate region does not appear in the non-stoichiometric composition of Li1-zNi1+zO249 as shown in Fig. 6. The monoclinic phase is caused by a regular arrangement of lithium vacancies or a Jahn-Teller arrangement of Ni3+; the rhombohedral phase in the NiO2 region is caused by the disappearance of Li from the Li layer and a change in the packing structure from ccp to hcp for the oxide ions. This structural change is affected by the slight presence of Ni in the Li layer. The Ni present in the Li layer acts as a pillar to support the NiO layer and thus prevents structural changes that occur in LiNiO2 with stoichiometric composition.9 In addition, in the region that undergoes large structural changes with the charge-discharge, a region that does not undergo structural changes appears due to the Ni present in the Li layer, which can be a factor leading to degradation of cycling characteristics. On the other hand, the presence of a small amount of Ni as a pillar between the Li layers prevents drastic structural changes, which may have a positive effect on charge-discharge characteristics. Thus, the structural change of LiNiO2 during charge-discharge is very complicated and highly dependent on the small amount of Ni present between the Li layers. How to control this non-stoichiometric character in the synthesis process of LiNiO2 and how to detect the Ni present between the few Li layers is of importance.

7

A deeper consideration of the monoclinic region

The monoclinic phase that appears during the de-intercalation process is caused by the c-plane shift of the lattice of the rhombohedral phase. Fig. 7 shows the relationship between the monoclinic and the hexagonal cell.31,50 In Na0.5NiO2, the Jahn-Teller effect of Ni3+ elongates the major axis of the NiO6 octahedron, and the alignment of the major axes in one direction gives rise to monoclinic symmetry. However, no ordered arrangement of Jahn-Teller ions is observed in LiNiO2, which contains the same Ni3+. Similar monoclinic region appearance is also detected in LiCoO2. The compositional region where monoclinic phase appears is around 0.5 Li composition, and this ordered arrangement of lithium defects is believed to be responsible for the symmetry reduction.51 On the other hand, neutron PDF analysis has revealed the presence of local Jahn-Teller distortions in Ni.20 Theoretical calculations confirm that the monoclinic phase is stable near Li0.75NiO2 and Li0.4NiO2 and that Li vacancy ordering promotes the formation of the monoclinic phase.52 In addition to the ordered arrangement of lithium vacancies and the Jahn-Teller arrangement of Ni3+, there are also longer-range Li-Li interactions through the Ni layer and Ni-Ni interactions through Ni2+ in the Li layer. The relevant interactions are in-plane Li-Li repulsion, but there are also long-range Li-Li attraction interactions across the plane

234

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Nickel Oxide

Li Ni O

Hexagonal cell

Monoclinic cell

chex

cmon

ahex =bhex

amon

Fig. 7 Relationship between the monoclinic and the hexagonal cells in LiNiO2.31 Figure from Ohzuku, T.; Ueda, A.; Nagayama, M. Electrochemistry and Structural Chemistry of LiNiO2 (R3m) for 4 Volt Secondary Lithium Cells. J. Electrochem. Soc. 1993, 140 (7), 1862. https://doi.org/10.1149/1.2220730. Chang, K.; Hallstedt, B.; Music, D. Thermodynamic description of the LiNiO2–NiO2 pseudo-binary system and extrapolation to the Li(Co,Ni)O2–(Co,Ni)O2 system. Calphad 2012, 37, 100–107. https://doi.org/10.1016/j.calphad.2012.02.006, need permission.

through JT-active Ni3+ ions. The phase change in LiNiO2 especially the distortion to the monoclinic symmetry is important to investigate the battery reaction mechanism.

8

Phases that appear near NiO2

The NiO2 phase that is observed at the end of the charge is complicated. In addition to the rhombohedral H3 phase, H4 phase with different symmetry has been indicated.12,46 Fig. 8 shows the structure diagram of both H3 and H4 phases. The H3 phase near Li0.1NiO2, with O3-type oxygen packing and the same structure as H1 and H2, with significantly smaller interlayer distances. As the lithium content decreases, the O arrangement changes from ccp to hcp, resulting in an O1-type (ABAB) H4 phase.53 The H3-H4 transition corresponds to a change in the shear structure, and the small amount of Ni present in the Li layer affects this transition.12,46,54 The phases that appear near NiO2 are thus extremely complex and related to the stoichiometric composition of LiNiO2.

9

Stoichiometry and electrochemical properties

The following is a summary of the relationship between the electrochemical properties and stoichiometric composition of LiNiO2. For Li1-zNi1+zO2 with z > 0, the theoretical capacity decreases due to the low Li content. Ni2+ in the interlayer and a large

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Nickel Oxide

235

Fig. 8 Proposed H4-type structure. The difference between the H3 and H4 types corresponds to the difference in oxygen arrangements. There is also an intermediate state between these two types. It can be regarded as a shear structure. The formation of this shear structure is related to the nickel present in the lithium layer.46 Figure from reference Croguennec, L.; Pouillerie, C.; Mansour, A. N.; Delmas, C. Structural Characterisation of the Highly Deintercalated LiNiO Phases (with  0.30). J. Mater. Chem. 2001, 11 (1), 131–141, https://doi.org/10.1039/B003377O, need permission.

irreversibility in the first cycle, resulting in a lower capacity. The charge-discharge curve of LiNiO2 shows phase transitions, but the curve becomes smoother as z value increases. The monoclinic region around Li0.5-zNi1+zO2 ceases to appear with increasing z. Near the composition Li0.1-zNi1+zO2, the large volume change associated with the H2-H3 transition decreases with increasing z. The H4 phase ceases to appear with increasing z. The control of the stoichiometric composition is essential to obtain a material with excellent charge-discharge properties.

10

Mechanical properties

The H2/H3 phase transition in LiNiO2 (4.15 V vs. Li+/Li) causes a volume change of about 4%.45,47 The strain causes the cracks in the particles during cycling, reducing the contact between primary particles and decreasing their electrochemical properties.55 The electrolyte penetrates the cracks and reacts with the particle surfaces. Cracks increase with cycling, leading to pulverization of secondary particles.56 The expansion and contraction of the particles during charging and discharging must be suppressed.

11

Thermal instability

LiNiO2 has low thermal stability.57,58 De-intercalated samples (x ¼ 0.5 in LixNiO2) decompose in two stages17,57: at around 200  C, the layered structure changes to a spinel phase with an exothermic reaction. At 300  C, the spinel transforms to a rock-salt structure, releasing O2.59,60 During this structural change, Ni migrates to the Li layer, followed by Li migration to the tetrahedral sites, resulting in the formation of a defective spinel structure. As O2 is generated, a disordered layered structure with mixed Li/Ni forms via the

236

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Nickel Oxide

two-phase region, and finally a disordered arrangement of rock-salt structures is formed. In the pyrolysis process of LixNiO2, the electrolyte participating in the reaction is responsible for releasing most of the heat, and the large primary particles are more thermally stable than the smaller particles, and the size of the secondary particles does not affect the process.59,61

12

Material modification

12.1 Doping Cobalt: The disordered arrangement of Ni2+ is reduced and the structure is stabilized.62 The simultaneous addition of Co and Al has been used as a practical material in automotive use. Iron: The addition of iron changes the layered structure to a rock-salt type structure.63 Oxidation of Fe to 4+ is also induced.64 Manganese: Manganese compounds include layered rock-salt type LiMnO2, which is obtained as a metastable state, zigzag layered LiMnO2, which can be synthesized as a stable state, and Li2MnO3, in which Ni4+ is stable. The additive system is therefore complex, as each compound is described as an end member in the Li-Ni-Mn-O system.65 With Mn addition relatively high capacities can be achieved and thermal stability is improved.66 Aluminum: Layered structure is stabilized and thermal stability is improved; co-doping with Co has been developed as a practical material.60 Other doping systems: various doping systems have been investigated: Cu,67 B,68 Ga,69 Y,70 Sb,71 Zn,72 In and Tl,69 Ca and Nb,73 42 W and Zr,73 etc.

12.2 Coating A method of covering the surface of LiNiO2 particles with a thin layer of a stable compound to improve stability in electrochemical reactions. Reactions such as O2 release and SEI formation are controlled. Various metal oxides (MgO, Al2O3, SiO2, TiO2, ZnO, SnO2, ZrO2, Li2O-B2O3-glasses, etc.) have been investigated.74

12.3 Core-shell structure The particle structure has a Ni-Co-Mn layered system near the surface of the particle and encapsulates LiNiO2 inside. The surface is thermally and mechanically stabilized by placing a more stable layered structure on the surface. This type of particle structure can be synthesized by controlling the synthesis conditions of the coprecipitation method.75

13

Conclusion

LiNiO2 is the cathode material with relatively low potential and the highest capacity in the layered rock-salt type structure. The presence of trivalent nickel in the structure poses many challenges for material synthesis and stability. The electronic structure of Ni with spin s ¼ 1/2 exhibits frustrated magnetism. The Li-Ni-O phase diagram should be considered for the synthesis of trivalent nickel ions. With charging, lithium de-intercalates, resulting in lithium defects. There are structural features such as the regular arrangement of the defects and the Jahn-Teller arrangement, which are involved in the battery characteristics but have not been clarified. The divalent nickel present between the lithium layers contributes significantly to the capacity and stability of charge and discharge. These are fundamental phenomena not only for LiNiO2 but also for other layered rock salt cathode materials. The challenge is how to reduce the nickel between the lithium layers and how to ensure the stable presence of trivalent nickel during synthesis. Attempts are being made to solve this problem by exploring various synthesis conditions and element substitution systems.

References 1. Dyer, L. D.; Borie, B. S., Jr.; Smith, G. P. Alkali Metal-Nickel Oxides of the Type MNiO2. J. Am. Chem. Soc. 1954, 76 (6), 1499–1503. DOI: https://doi.org/10.1021/ ja01635a012. 2. Goodenough, J. B.; Wickham, D. G.; Croft, W. J. Some Magnetic and Crystallographic Properties of the System Li+xNi++1−2xni+++ O. J. Phys. Chem. Solid 1958, 5 (1), x 107–116. https://doi.org/10.1016/0022-3697(58)90136-7. 3. Dahn, J. R.; von Sacken, U.; Michal, C. A. Structure and Electrochemistry of Li1yNiO2 and a New Li2NiO2 Phase with the Ni (OH)2 Structure. Solid State Ion. 1990, 44 (1), 87–97. https://doi.org/10.1016/0167-2738(90)90049-W. 4. Van der Ven, A.; Ceder, G. Lithium Diffusion in Layered Lix CoO2. Electrochem. Solid St. 2000, 3 (7), 301. https://doi.org/10.1149/1.1391130. 5. Ohzuku, T.; Ueda, A.; Nagayama, M.; Iwakoshi, Y.; Komori, H. Comparative Study of LiCoO2, LiNi12Co12O2 and LiNiO2 for 4 Volt Secondary lithium Cells. Electrochim. Acta 1993, 38 (9), 1159–1167. https://doi.org/10.1016/0013-4686(93)80046-3. 6. Delmas, C.; Pérès, J. P.; Rougier, A.; Demourgues, A.; Weill, F.; Chadwick, A.; Broussely, M.; Perton, F.; Biensan, P.; Willmann, P. On the Behavior of the LixNiO2 System: An Electrochemical and Structural Overview. J. Power Sources 1997, 68 (1), 120–125. https://doi.org/10.1016/S0378-7753(97)02664-5. 7. Delmas, C.; Ménétrier, M.; Croguennec, L.; Saadoune, I.; Rougier, A.; Pouillerie, C.; Prado, G.; Grüne, M.; Fournès, L. An Overview of the Li(Ni,M)O2 Systems: Syntheses, Structures and Properties. Electrochim. Acta 1999, 45 (1), 243–253. https://doi.org/10.1016/S0013-4686(99)00208-X.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Nickel Oxide

237

8. Kanno, R.; Kubo, H.; Kawamoto, Y.; Kamiyama, T.; Izumi, F.; Takeda, Y.; Takano, M. Phase Relationship and lithium Deintercalation in lithium Nickel Oxides. J. Solid State Chem. 1994, 110 (2), 216–225. https://doi.org/10.1006/jssc.1994.1162. 9. Hirano, A.; Kanno, R.; Kawamoto, Y.; Takeda, Y.; Yamaura, K.; Takano, M.; Ohyama, K.; Ohashi, M.; Yamaguchi, Y. Relationship between Nonstoichiometry and Physical-Properties in LiNiO2. Solid State Ion. 1995, 78 (1–2), 123–131. https://doi.org/10.1016/0167-2738(95)00005-q. 10. Pouillerie, C.; Suard, E.; Delmas, C. Structural Characterization of Li1−z−xNi1+zO2 by Neutron Diffraction. J. Solid State Chem. 2001, 158 (2), 187–197. https://doi.org/10.1006/ jssc.2001.9092. 11. Bianchini, M.; Roca-Ayats, M.; Hartmann, P.; Brezesinski, T.; Janek, J. There and Back Again—The Journey of LiNiO2 as a Cathode Active Material. Angew. Chem. Int. Ed. 2019, 58 (31), 10434–10458. https://doi.org/10.1002/anie.201812472. 12. Tarascon, J. M.; Vaughan, G.; Chabre, Y.; Seguin, L.; Anne, M.; Strobel, P.; Amatucci, G. In Situ Structural and Electrochemical Study of Ni1−xCoxO2 Metastable Oxides Prepared by Soft Chemistry. J. Solid State Chem. 1999, 147 (1), 410–420. https://doi.org/10.1006/jssc.1999.8465. 13. Das, H.; Urban, A.; Huang, W.; Ceder, G. First-Principles Simulation of the (Li–Ni–Vacancy)O Phase Diagram and its Relevance for the Surface Phases in Ni-Rich Li-Ion Cathode Materials. Chem. Mater. 2017, 29 (18), 7840–7851. https://doi.org/10.1021/acs.chemmater.7b02546. 14. Berbenni, V.; Massarotti, V.; Capsoni, D.; Riccardi, R.; Marini, A.; Antolini, E. Structural and Microstructural Study of the Formation of the Solid Solution LixNi1−xO. Solid State Ion. 1991, 48 (1), 101–111. https://doi.org/10.1016/0167-2738(91)90205; Li, W.; Reimers, J. N.; Dahn, J. R. Crystal Structure of LixNi2-xO2 and a Lattice-Gas Model for the Order-Disorder Transition. Phy. Rev. B 1992, 46 (6), 3236–3246. https://doi.org/10.1103/PhysRevB.46.3236. 15. Migeon, H. N.; Zanne, M.; Gleitzer, C.; Aubry, J. The Li2O-NiO-O2 System at 670 C and the Consequences of Non-stoichiometry on the Magnetic Properties of the LixNi1−xO1y Phases. J. Mater. Sci. 1978, 13 (3), 461–466. https://doi.org/10.1007/BF00541793. 16. Tabuchi, M.; Kuriyama, N.; Takamori, K.; Imanari, Y.; Nakane, K. Appearance of Lithium-Excess LiNiO2 with High Cyclability Synthesized by Thermal Decomposition Route from LiNiO2-Li2NiO3 Solid Solution. J. Electrochem. Soc. 2016, 163 (10), A2312. https://doi.org/10.1149/2.0861610jes. 17. Thomas, M. G. S. R.; David, W. I. F.; Goodenough, J. B.; Groves, P. Synthesis and Structural Characterization of the Normal Spinel Li[Ni2]O4. Mater. Res. Bull. 1985, 20 (10), 1137–1146. https://doi.org/10.1016/0025-5408(85)90087-X. 18. Rieck, H.; Hoppe, R. Ein neues Oxoniccolat: Li2NiO2. Z. Anorg. Allg. Chem. 1972, 392 (3), 193–196. https://doi.org/10.1002/zaac.19723920302. 19. Rougier, A.; Delmas, C.; Chadwick, A. V. Non-cooperative Jahn-Teller Effect in LiNiO2: An EXAFS Study. Solid State Commun. 1995, 94 (2), 123–127. https://doi.org/ 10.1016/0038-1098(95)00020-8. 20. Chung, J. H.; Proffen, T.; Shamoto, S.; Ghorayeb, A. M.; Croguennec, L.; Tian, W.; Sales, B. C.; Jin, R.; Mandrus, D.; Egami, T. Local Structure of LiNiO2 Studied by Neutron Diffraction. Phy. Rev. B 2005, 71 (6), 064410. https://doi.org/10.1103/PhysRevB.71.064410. 21. Nakai, I.; Takahashi, K.; Shiraishi, Y.; Nakagome, T.; Nishikawa, F. Study of the Jahn–Teller Distortion in LiNiO2, a Cathode Material in a Rechargeable Lithium Battery, Byin SituX-Ray Absorption Fine Structure Analysis. J. Solid State Chem. 1998, 140 (1), 145–148. https://doi.org/10.1006/jssc.1998.7943. 22. Chen, Z.; Zou, H.; Zhu, X.; Zou, J.; Cao, J. First-Principle Investigation of Jahn–Teller Distortion and Topological Analysis of Chemical Bonds in LiNiO2. J. Solid State Chem. 2011, 184 (7), 1784–1790. https://doi.org/10.1016/j.jssc.2011.05.024. 23. Montoro, L. A.; Abbate, M.; Almeida, E. C.; Rosolen, J. M. Electronic Structure of the Transition Metal Ions in LiCoO2, LiNiO2 and LiCo0.5Ni0.5O2. Chem. Phys. Lett. 1999, 309 (1), 14–18. https://doi.org/10.1016/S0009-2614(99)00650-8. 24. Goodenough, J. B.; Wickham, D. G.; Croft, W. J. Some Ferrimagnetic Properties of the System LixNi1−xO. J. Appl. Phys. 2004, 29 (3), 382–383. https://doi.org/ 10.1063/1.1723144 (acccessed 1/25/2024). 25. Reimers, J. N.; Dahn, J. R.; Greedan, J. E.; Stager, C. V.; Liu, G.; Davidson, I.; Von Sacken, U. Spin Glass Behavior in the Frustrated Antiferromagnetic LiNiO2. J. Solid State Chem. 1993, 102 (2), 542–552. https://doi.org/10.1006/jssc.1993.1065. 26. Shirakami, T.; Takematsu, M.; Hirano, A.; Kanno, R.; Yamaura, K.; Takano, M.; Atake, T. Spin Glass-like Magnetic Properties of LiNiO2. Mater. Sci. Eng. B: Solid-State Mater. Adv. Technol 1998, 54 (1–2), 70–72. https://doi.org/10.1016/s0921-5107(98)00130-5. 27. Yamaura, K.; Takano, M.; Hirano, A.; Kanno, R. Magnetic Properties of Li1-xNi1+xO2 (0  x  0.08). J. Solid State Chem. 1996, 127 (1), 109–118. https://doi.org/10.1006/ jssc.1996.0363. 28. Rougier, A.; Delmas, C.; Chouteau, G. Magnetism of Li1−zNi1+zO2: A Powerful Tool for Structure Determination. J. Phys. Chem. Solid 1996, 57 (6), 1101–1103. https://doi.org/ 10.1016/0022-3697(95)00403-3. 29. Takematsu, M.; Kawaji, H.; Atake, T.; Shirakami, T.; Hirano, A.; Kanno, R. Magnetic Properties and Spin Glass like Behavior of Li1-xNi1+xO2 (x0.03). Electrochemistry 2000, 68 (6), 465–468. https://doi.org/10.5796/electrochemistry.68.465. 30. Mertz, D.; Ksari, Y.; Celestini, F.; Debierre, J. M.; Stepanov, A.; Delmas, C. Magnetic Cluster Formation in LixNi1−xO Compounds: Experiments and Numerical Simulations. Phy. Rev. B 2000, 61 (2), 1240–1245. https://doi.org/10.1103/PhysRevB.61.1240. 31. Ohzuku, T.; Ueda, A.; Nagayama, M. Electrochemistry and Structural Chemistry of LiNiO2 (R3m) for 4 Volt Secondary Lithium Cells. J. Electrochem. Soc. 1993, 140 (7), 1862. https://doi.org/10.1149/1.2220730. 32. Rougier, A.; Gravereau, P.; Delmas, C. Optimization of the Composition of the Li1−zNi1+z O2 Electrode Materials: Structural, Magnetic, and Electrochemical Studies. J. Electrochem. Soc. 1996, 143 (4), 1168. https://doi.org/10.1149/1.1836614. 33. Park, K. S.; Park, S. H.; Sun, Y. K.; Nahm, K. S.; Lee, Y. S.; Yoshio, M. Effect of Oxygen Flow Rate on the Structural and Electrochemical Properties of lithium Nickel Oxides Synthesized by the Sol–Gel Method. J. Appl. Electrochem. 2002, 32 (11), 1229–1233. https://doi.org/10.1023/A:1021606712486. 34. Yamada, S.; Fujiwara, M.; Kanda, M. Synthesis and Properties of LiNiO2 as Cathode Material for Secondary Batteries. J. Power Sources 1995, 54 (2), 209–213. https://doi.org/ 10.1016/0378-7753(94)02068-E. 35. Xu, J.; Lin, F.; Nordlund, D.; Crumlin, E. J.; Wang, F.; Bai, J.; Doeff, M. M.; Tong, W. Elucidation of the Surface Characteristics and Electrochemistry of High-Performance LiNiO2. Chem. Commun. 2016, 52 (22), 4239–4242. https://doi.org/10.1039/C5CC09434H. 36. Garcia, B.; Pereira-Ramos, J. P.; Caurant, D.; Baffier, N. Synthesis and Electrochemical Properties of LiNiO2 Prepared Via an Aqueous Solution Technique. Chem. Lett. 2003, 27 (6), 543–544. https://doi.org/10.1246/cl.1998.543 (acccessed 1/25/2024); Caurant, D.; Baffier, N.; Garcia, B.; Pereira-Ramos, J. P. Synthesis by a Soft Chemistry Route and Characterization of LiNixCo1−xO2 (0  x  1) Cathode Materials. Solid State Ion. 1996, 91 (1), 45–54. https://doi.org/10.1016/S0167-2738(96)00418-3. 37. Chang, C.-C.; Scarr, N.; Kumta, P. N. Synthesis and Electrochemical Characterization of LiMO2 (M¼Ni, Ni0.75Co0.25) for Rechargeable lithium Ion Batteries. Solid State Ion. 1998, 112 (3), 329–344. https://doi.org/10.1016/S0167-2738(98)00183-0. 38. Mehrdad Balandeh, S. A. Synthesis and Characterization of LiNiO2 Nanopowder with Various Chelating Agents. J. Nanomater. 2010, 2010, 6. https://doi.org/10.1155/2010/ 695083. 39. Sun, Y.-K.; Oh, I.-H. Synthesis of LiNiO2 Powders by a Sol-Gel Method. J. Mater. Sci. Lett. 1997, 16 (1), 30–32. 40. Chang, C.-C.; Kim, J. Y.; Kumta, P. N. Influence of Crystallite Size on the Electrochemical Properties of Chemically Synthesized Stoichiometric LiNiO2. J. Electrochem. Soc. 2002, 149 (9). https://doi.org/10.1149/1.1495495. A1114; Chang, C.-C.; Kim, J. Y.; Kumta, P. N. Implications of Reaction Mechanism and Kinetics on the Synthesis of Stoichiometric LiNiO2. J. Electrochem. Soc. 2002, 149 (3), A331. https://doi.org/10.1149/1.1436082. 41. Dearden, C.; Zhu, M.; Wang, B.; Castro, R. H. R. Synthesis, Size Reduction, and Delithiation of Carbonate-Free Nanocrystalline lithium Nickel Oxide. J. Mater. Sci. 2013, 48 (4), 1740–1745. https://doi.org/10.1007/s10853-012-6932-y. 42. Kim, U. H.; Jun, D. W.; Park, K. J.; Zhang, Q.; Kaghazchi, P.; Aurbach, D.; Major, D. T.; Goobes, G.; Dixit, M.; Leifer, N.; et al. Pushing the Limit of Layered Transition Metal Oxide Cathodes for High-Energy Density Rechargeable Li Ion Batteries. Energ. Environ. Sci. 2018, 11 (5), 1271–1279. https://doi.org/10.1039/C8EE00227D. 43. Li, H.; Zhang, N.; Li, J.; Dahn, J. R. Updating the Structure and Electrochemistry of LixNiO2 for 0  x  1. J. Electrochem. Soc. 2018, 165 (13), A2985. https://doi.org/ 10.1149/2.0381813jes.

238

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Nickel Oxide

44. Arai, H.; Okada, S.; Ohtsuka, H.; Ichimura, M.; Yamaki, J. Characterization and Cathode Performance of Li1−xNi1+xO2 Prepared with the Excess lithium Method. Solid State Ion. 1995, 80 (3), 261–269. https://doi.org/10.1016/0167-2738(95)00144-U. 45. Li, W.; Reimers, J. N.; Dahn, J. R. In Situ x-Ray Diffraction and Electrochemical Studies of Li1−xNiO2. Solid State Ion. 1993, 67 (1), 123–130. https://doi.org/10.1016/01672738(93)90317-V. 46. Croguennec, L.; Pouillerie, C.; Mansour, A. N.; Delmas, C. Structural Characterisation of the Highly Deintercalated LiNiO Phases (with  0.30). J. Mater. Chem. 2001, 11 (1), 131–141. https://doi.org/10.1039/B003377O. 47. Yoon, C. S.; Jun, D.-W.; Myung, S.-T.; Sun, Y.-K. Structural Stability of LiNiO2 Cycled above 4.2 V. ACS Energy Lett. 2017, 2 (5), 1150–1155. https://doi.org/10.1021/ acsenergylett.7b00304. 48. Xu, J.; Hu, E.; Nordlund, D.; Mehta, A.; Ehrlich, S. N.; Yang, X.-Q.; Tong, W. Understanding the Degradation Mechanism of Lithium Nickel Oxide Cathodes for Li-Ion Batteries. ACS Appl. Mater. Interfaces 2016, 8 (46), 31677–31683. https://doi.org/10.1021/acsami.6b11111. 49. Yang, X. Q.; Sun, X.; McBreen, J. New Findings on the Phase Transitions in Li1−xNiO2: In Situ Synchrotron X-Ray Diffraction Studies. Electrochem. Commun. 1999, 1 (6), 227–232. https://doi.org/10.1016/S1388-2481(99)00046-6. 50. Chang, K.; Hallstedt, B.; Music, D. Thermodynamic Description of the LiNiO2–NiO2 pseudo-Binary System and Extrapolation to the Li(Co,Ni)O2–(Co,Ni)O2 System. Calphad 2012, 37, 100–107. https://doi.org/10.1016/j.calphad.2012.02.006. 51. Peres, J. P.; Weill, F.; Delmas, C. Lithium/Vacancy Ordering in the Monoclinic LixNiO2 (0.50x0.75) Solid Solution. Solid State Ion. 1999, 116 (1), 19–27. https://doi.org/ 10.1016/S0167-2738(98)00239-2. 52. Kang, K.; Ceder, G. Factors that Affect Li Mobility in Layered lithium Transition Metal Oxides. Phy. Rev. B 2006, 74 (9), 094105. https://doi.org/10.1103/PhysRevB.74.094105; Arroyo y de Dompablo, M. E.; Ceder, G. On the Origin of the Monoclinic Distortion in LixNiO2. Chem. Mater. 2003, 15 (1), 63–67. https://doi.org/10.1021/cm020239+. 53. Amatucci, G. G.; Tarascon, J. M.; Klein, L. C. CoO2, the End Member of the Lix CoO2 Solid Solution. J. Electrochem. Soc. 1996, 143 (3), 1114. https://doi.org/ 10.1149/1.1836594. 54. Croguennec, L.; Pouillerie, C.; Delmas, C. Structural Characterisation of New Metastable NiO2 Phases. Solid State Ion. 2000, 135 (1), 259–266. https://doi.org/10.1016/S01672738(00)00441-0. 55. Sun, H.-H.; Manthiram, A. Impact of Microcrack Generation and Surface Degradation on a Nickel-Rich Layered Li[Ni0.9Co0.05Mn0.05]O2 Cathode for Lithium-Ion Batteries. Chem. Mater. 2017, 29 (19), 8486–8493. https://doi.org/10.1021/acs.chemmater.7b03268. 56. Dokko, K.; Nishizawa, M.; Horikoshi, S.; Itoh, T.; Mohamedi, M.; Uchida, I. In Situ Observation of LiNiO2 Single-Particle Fracture during Li - Ion Extraction and Insertion. Electrochem. Solid St. 2000, 3 (3), 125. https://doi.org/10.1149/1.1390977. 57. Dahn, J. R.; Fuller, E. W.; Obrovac, M.; von Sacken, U. Thermal Stability of LixCoO2, LixNiO2 and l-MnO2 and Consequences for the Safety of Li-Ion Cells. Solid State Ion. 1994, 69 (3), 265–270. https://doi.org/10.1016/0167-2738(94)90415-4. 58. Morales, J.; Pérez-Vicente, C.; Tirado, J. L. Thermal Behaviour of Chemically Deintercalated Li1−1Ni1+xO2. J. Thermal Anal. 1992, 38 (3), 295–301. https://doi.org/10.1007/ BF01915494. 59. Arai, H.; Okada, S.; Sakurai, Y.; Yamaki, J.-I. Thermal Behavior of Li1−yNiO2 and the Decomposition Mechanism. Solid State Ion. 1998, 109 (3), 295–302. https://doi.org/ 10.1016/S0167-2738(98)00075-7. 60. Ohzuku, T.; Ueda, A.; Kouguchi, M. Synthesis and Characterization of LiAl1/4Ni3/4O2 (R 3−m ) for Lithium-Ion (Shuttlecock) Batteries. J. Electrochem. Soc. 1995, 142 (12), 4033. https://doi.org/10.1149/1.2048458. 61. MacNeil, D. D.; Lu, Z.; Chen, Z.; Dahn, J. R. A Comparison of the Electrode/Electrolyte Reaction at Elevated Temperatures for Various Li-Ion Battery Cathodes. J. Power Sources 2002, 108 (1), 8–14. https://doi.org/10.1016/S0378-7753(01)01013-8; Zhang, Z.; Fouchard, D.; Rea, J. R. Differential Scanning Calorimetry Material Studies: Implications for the Safety of lithium-Ion Cells. J. Power Sources 1998, 70 (1), 16–20. https://doi.org/10.1016/S0378-7753(97)02611-6. 62. Cho, J.; Park, B. Preparation and Electrochemical/Thermal Properties of LiNi0.74Co0.26O2 Cathode Material. J. Power Sources 2001, 92 (1), 35–39. https://doi.org/10.1016/ S0378-7753(00)00499-7; Ehrlich, G. M.; Puglia, F. J.; Gitzendanner, R.; Hellen, B.; Marsh, C. Flat Plate Prismatic Li-Ion Cells Using Advanced Cathode Materials. J. Power Sources 1999, 81–82, 863–866. https://doi.org/10.1016/S0378-7753(99)00123-8. 63. Reimers, J. N.; Rossen, E.; Jones, C. D.; Dahn, J. R. Structure and Electrochemistry of LixFeyNi1-yO2. Solid State Ion. 1993, 61 (4), 335–344. https://doi.org/10.1016/01672738(93)90401-N. 64. Mohan, P.; Kalaignan, G. P. Structure and Electrochemical Performance of LiFexNi1-xO2 (0.00  x  0.20) Cathode Materials for Rechargeable lithium-Ion Batteries. J. Electroceram. 2013, 31 (1), 210–217. https://doi.org/10.1007/s10832-013-9815-y. 65. Rossen, E.; Jones, C. D. W.; Dahn, J. R. Structure and Electrochemistry of LixMnyNi1−yO2. Solid State Ion. 1992, 57 (3), 311–318. https://doi.org/10.1016/0167-2738(92) 90164-K. 66. Arai, H.; Okada, S.; Sakurai, Y.; Yamaki, J. I. Electrochemical and Thermal Behavior of LiNi1−z MzO2 (M¼Co , Mn , Ti ). J. Electrochem. Soc. 1997, 144 (9), 3117. https://doi.org/ 10.1149/1.1837968. 67. Kim, J.; Amine, K. A Comparative Study on the Substitution of Divalent, Trivalent and Tetravalent Metal Ions in LiNi1−xMxO2 (M ¼ Cu2+, Al3+ and Ti4+). J. Power Sources 2002, 104 (1), 33–39. https://doi.org/10.1016/S0378-7753(01)00900-4. 68. Julien, C.; Nazri, G. A.; Rougier, A. Electrochemical Performances of Layered LiM1−yMy0 O2 (M¼Ni, Co; M0 ¼Mg, Al, B) Oxides in lithium Batteries. Solid State Ion. 2000, 135 (1), 121–130. https://doi.org/10.1016/S0167-2738(00)00290-3. 69. Song, M.; Kwon, S.; Kwon, I.; Park, H. Variation of LiNiO2 Cathode Properties with Substitution of Ga, in and Tl by the Combustion Method. J. Appl. Electrochem. 2007, 37 (4), 421–427. https://doi.org/10.1007/s10800-006-9273-1. 70. Kudo, T.; Mizutani, N. Effect of Addition of a Foreign to LiNiO2 by Complex Polymerized Method on its Electrochemical Properties. Key Eng. Mater. 1999, 169-170, 217–220. https://doi.org/10.4028/www.scientific.net/KEM.169-170.217. 71. Yang, Z.; Zhang, Z.; Pan, Y.; Zhao, S.; Huang, Y.; Wang, X.; Chen, X.; Wei, S. First-Principles Investigation of the Effects of Sb Doping on the LiNiO2. J. Solid State Chem. 2016, 244, 52–60. https://doi.org/10.1016/j.jssc.2016.09.015. 72. Kwon, I. H.; Park, H. R.; Song, Y. Y. Effects of Zn, Al and Ti Substitution on the Electrochemical Properties of LiNiO2 Synthesized by the Combustion Method. Russ. J. Electrochem. 2013, 49 (3), 221–227. https://doi.org/10.1134/S1023193513030099. 73. Yoon, C. S.; Choi, M.-J.; Jun, D.-W.; Zhang, Q.; Kaghazchi, P.; Kim, K.-H.; Sun, Y.-K. Cation Ordering of Zr-Doped LiNiO2 Cathode for Lithium-Ion Batteries. Chem. Mater. 2018, 30 (5), 1808–1814. https://doi.org/10.1021/acs.chemmater.8b00619. 74. Li, C.; Zhang, H. P.; Fu, L. J.; Liu, H.; Wu, Y. P.; Rahm, E.; Holze, R.; Wu, H. Q. Cathode Materials Modified by Surface Coating for lithium Ion Batteries. Electrochim. Acta 2006, 51 (19), 3872–3883. https://doi.org/10.1016/j.electacta.2005.11.015; Cho, J.; Kim, T.-J.; Kim, Y. J.; Park, B. High-Performance ZrO2-Coated LiNiO2 Cathode Material. Electrochem. Solid St. 2001, 4 (10), A159. https://doi.org/10.1149/1.1398556. 75. Sun, Y.-K.; Chen, Z.; Noh, H.-J.; Lee, D.-J.; Jung, H.-G.; Ren, Y.; Wang, S.; Yoon, C. S.; Myung, S.-T.; Amine, K. Nanostructured High-Energy Cathode Materials for Advanced lithium Batteries. Nat. Mater. 2012, 11 (11), 942–947. https://doi.org/10.1038/nmat3435; Zhong, S.-W.; Zhao, Y.-J.; Lian, F.; Li, Y.; Hu, Y.; Li, P.-Z.; Mei, J.; Liu, Q.-G. Characteristics and Electrochemical Performance of Cathode Material co-Coated LiNiO2 for Li-Ion Batteries. Trans. Nonferrous Met. Soc. Chin. 2006, 16 (1), 137–141. https:// doi.org/10.1016/S1003-6326(06)60024-1.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Manganese Oxides Masaki Okada, Kyoto University, Kyoto, Japan © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of M. Yoshio, H. Noguchi, SECONDARY BATTERIES – LITHIUM RECHARGEABLE SYSTEMS – LITHIUM-ION | Positive Electrode: Manganese Oxides, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 307–317, ISBN 9780444527455, https://doi.org/10. 1016/B978-044452745-5.00200-8. (https://www.sciencedirect.com/science/article/pii/B9780444527455002008)

1 2 3 4 5 6 7 References Further reading

Introduction g-(EMD and CMD) and b-like MnO2 Composite dimensional manganese oxide and its analogs Lithium manganese oxide Lithium manganese oxide spinel Li2MnO3-based positive electrode materials Concluding remarks

239 241 244 246 248 251 252 253 253

Abstract This article has been reviewed MnO2, Li0.33MnO2, LiMnO2, Li-Mn-O compounds, and their metal-ion doped compounds as energy storage materials based on Mn3+/4+ redox couples with respect to manganese-based positive electrode materials for lithium batteries and lithium-ion batteries. Li0.33MnO2, spinel-type Li-Mn-O compounds, Li2MnO3-related compounds, and other practical cathode materials were focused on, and the basic properties and electrochemical reaction mechanisms of the materials have been discussed in detail.

Glossary BET Brunauer, Emmett and Teller theory, is used to measure the surface area of solid particles. CDMO Composite Dimensional Manganese Oxide is based on the composite of Li2MnO3 and g/b-MnO2. EV Electric Vehicle is a vehicle that is powered by electricity and motor without internal combustion engine ICE). CMD/HCMD Chemical Manganese Dioxide/Heat-treated Chemical Manganese Dioxide are synthesized from chemical reaction. EMD/HEMD Electrolytic Manganese Dioxide/Heat-treated Electrolytic Manganese Dioxide are synthesized by electrolysis. HEV Hybrid Electrolytic Vehicle is a vehicle that is powered by both ICE and electric motor. There are various types of HEV, such as series/parallel HEV and plug-in HEV and range extender. OCV Open Circuit Voltage is determined by the electrical potential gap between two electrodes without any load (¼ ideally equip to an infinite resistance). XRD X-ray diffraction (measurement).

Key points

• • •

1

Highlighting manganese based positive active materials. Discussing crystalline structure, physical properties and electrochemical properties of manganese based positive active materials. Reviewing various manganese oxides including MnO2, Li0.33MnO2, LiMnO2, LiMn2O4 and Li-Mn-O compounds.

Introduction

Manganese oxides are widely used not only for aqueous alkaline batteries, but also for non-aqueous primary or secondary lithium batteries. A candidate for positive electrode materials for electric vehicle (EV) or- hybrid electric vehicle (HEV) is the spinel structure Li-Mn-O compounds.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00305-4

239

240

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Manganese Oxides

Manganese oxide-related compounds, such as heat treated electrolytic or chemically synthesized manganese dioxide (HEMD or HCMD), composite dimensional manganese oxide (CDMO), and Li2O • yMnO2, were pioneer positive electrode materials for Lithium battery and are still used as positive electrodes for lithium batteries.1 M. M. Thackeray had reviewed manganese oxide-based positive electrode materials for lithium battery.2 This review mainly describes the progress after 1997 and covers various polymorphs of manganese dioxide, CDMO-type material, which has the chemical formula Li0.33MnO2, 3 and 4 V class spinel Li-Mn-O, and LiMnO2. These materials commonly have the capacity at 3-4 V region and their properties are summarized in Table 1. Furthermore, typical structures of manganese-based positive electrode materials are illustrated in Fig. 1, where orthorhombic lithium manganese dioxide (LiMnO2) with layered zigzag plane and Li0.33MnO2 are excluded. The compounds in Fig. 1 are classified into three groups.3 The smallest unit of structure is built up of MnO6 octahedron, which shares corner and/or edges. The first group has one-dimensional runnels for lithium diffusion. Pyrolusite, Ramsdellite, hollandite, and romanechite belong to this group and they have 1  1, l  2, 2  2, and 2  3 runnels, respectively. The g-MnO2 and Li0.33MnO2 are also classified into this group. The size of the runnel is decided by incorporated cations, such as K+, Pb2+, Ba2+, Mg2+, and Ca2+. The second group consists of birnessite and monoclinic LiMnO2 (m-LiMnO2), which have two-dimensional planar lithium diffusion path transforms into spinel structure after lithium insertion. The last group comprises spinel l-Mn and Li-Mn-O compounds, which have three-dimensional diffusion path.

Table 1

Capacities of Mn-based cathode materials and the shape of discharge curves.

Compounds

Type of tunnel

Shape of initial discharge curve

Specific capacity (mAh g–1)

g-MnO2 b-MnO2 a-MnO2 Romanechite Todorakite Layered MnO2 Li0.33MnO2 m-LiMnO2 o-LiMnO2 Li2O  yMnO2 (y ¼ 4) LiMn2O4

1  1, 1  2 11 22 23 33 1  00 1  1, 1  2 Zigzag 1  1 11 Three-dimensional Three-dimensional

Flat and S type Flat and S type S type Complexed S type (3 plateaus) S type S type S type Transformed to spinel Transformed to spinel Flat Two flat and one S type

300 290 200–220 200 100–130 140–150 180 210 230 170 296a

a

Theoretical value.

a

c

c

a

b (a)

Ba2+ c

a

c

a b

b (c)

c

b

(b)

(d)

Ba2+

a

H2O

b (e)

(f)

Fig. 1 Typical structures of Mn-based cathode materials: (a) pyrolusite (1  1), (b) ramsdellite (1  2), (c) hollandite (2  2), (d) birnessite (1  1), (e) romanechite (2  3), and (f ) spinel (1  1).

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Manganese Oxides

2

241

g-(EMD and CMD) and b-like MnO2

Electrolytic manganese dioxide (EMD) has been widely used as positive electrode materials in alkaline manganese batteries. H. Ikeda and coworkers in Sanyo have used it in lithium primary batteries after heat treatment for removing the combined water, and such EMD is called HEMD. Original EMD is g-type MnO2, which is mainly composed of Ramsdellite and pyrolusite unit (De Wolf defect) and contains a small amount of cation vacancy. Furthermore, another defect called microtwining, which is caused by the change in the direction of growth, also presents in the structure. Such characters of EMD give amorphous-type X-ray diffraction (XRD) patterns as shown in Fig. 2(B). The characterization of g-MnO2 with Pr % (index of De Wolff defect) and Tw % (index of microtwining) has been proposed by Y. Charbe and J.Pannetier.4 The Pr % and Tw % of EMD are generally 30–40% and 0–100%, respectively. The chemically prepared manganese dioxide (MnO2) gives lower Pr % of 20%. The manganese dioxide sample, prepared by acid digestion of Li0.7MnO2 at boiling condition and expressed as curve A in Fig. 2, also gives lower Pr % of 21% and Tw % of 0–2%. Heat treatment of EMD increases Pr % and decreases Tw %. Such tendency is a common character in g-MnO2. Sample of curve C in Fig. 2 is obtained by heating curve A sample at 400  C for 2 h and is transformed to b-like-type profiles; however, its tetragonal (110) peak width is much wider than that of crystalline b-MnO2, (curve D), which is prepared by hydrothermal method from Li0.7MnO2 at 180  C. The Pr % of curve C sample is calculated to be 96%. The heat treatment temperature influences the structure transformation of EMD. Heating at 0.

(see Table 4). The formation of this impurity phase can be viewed as the precipitation of the defects of Eq. (I). This feature is illustrated in Fig. 13, where the sarcopside iron(II) phosphate lattice is illustrated in parallel with that of lithium iron phosphate olivine. In the spirit of Eq. (I), a solid solution in case of lithium excess would have required the formation of the neutral defect. Li0Fe + LiI_

(II)

The fact that we did not observe the defect represented in Eq. (II) shows that its energy of formation is actually so large that it cannot be generated during the synthesis process. The consequence is that the lattice cannot accommodate the excess of lithium and form a solid solution, so the lithium in excess precipitates as the trilithium phosphate impurity phase we have observed. Therefore, these experimental results also clearly evidence a strong dissymmetry in the Li–Fe exchange of ions, namely a few at% iron on lithium sites can be observed, but no lithium on iron sites. This result is in agreement with the prior structural analysis of samples prepared by hydrothermal route. We do not have any evidence of lithium on iron sites either in our samples or in prior works, and this observation makes questionable the existence of any Li–Fe antisite pair although calculations of its energy of formation have suggested it might exist. Let us recall that the limit of solution for the Fe%Li + VLi0 defects is 6 at.%. Above this limit, these defects aggregate to form sarcopside precipitates, which confirms the theoretical calculations of a negative binding energy of the Fe%Li + VLi0 defects. This feature also suggests that, in the range of concentrations for such defects, x < 6 at.%, aggregates of Fe%Li + VLi0 can be formed at the atomic scale as precursor of the larger sarcopside clusters formed with x  6 at.%. In particular, this result provides some insight into the recent observation of the aggregation of Fe%Li at this atomic scale by TEM for a defect concentration of 1 at.%. The complex defect of Eq. (I) strongly degrades the cathode efficiency since the lithium insertion/extraction is essentially a one-dimensional ionic conduction along the lithium channels, that is, along the b-direction (see Fig. 1); any Fe%Li defect blocks the lithium diffusion along the channel where the defect in Eq. (I) is located. This result is consistent with the high migration energy (0.70 eV) of Fe%Li along this direction. Therefore, attention must be then paid to a preparation of lithium iron phosphate that avoids from the introduction of a lithium deficiency.

14

Surface effects

So far, attention has been focused on the bulk properties of the lithium iron phosphate particles. We have also outlined the critical role played by the carbon coat, which is essential to ensure a good electric contact between the particles. However, between the carbon layer and the particle, there is a surface layer that also plays an important role. First, the structural disorder in the surface layer is usually larger than in the bulk of the particle. This property is clearly seen in the HRTEM image in Fig. 4, and the typical thickness of the surface layer is about 3 nm. Since the carbon coat is porous, this layer is in contact with the electrolyte or with the ambient atmosphere and thus is the place where all the chemical reactions occur that impact the electrochemical performance. For instance,

286

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

Table 4 Effect of different carbon coating and preparation methods on LFP, above table reproduced from Our recent review (ref. Batteries 2022). For more details please refer to our reference. Source of carbon

Type of carbon/LFP

Preparation methods

Performance

Graphite

C-rGO/LFP

168 mAh g−1 at 0.05 C

Graphite Graphite

rGO/LFP Graphene/LFP

Graphite Graphite Commercial Commercial Commercial Commercial Commercial Commercial Commercial Commercial Commercial Commercial Commercial Graphene Graphene oxide CNT CNT CNT Glucose Glucose Glucose

Nitrogen doped C/LFP Boron doped C/LFP LFP/MWCNT LFP/MWCNT LFP/MWCNT LFP/MWCNT LFP/MWCNT LFP/MWCNT LFP/MWCNT LFP/CNT 3D LFP/CNT-PVP C-LFP C/LFP G/LFP/C LFP/C/rGO LFP/MWCNT LFP/C/CNT LFP/Core shell LFP/C C/LFP Graphite sheet/N doped C/LFP C/LFP C-LFPsupra balls C/LFP C/LFP C/LFP C/LFP Graphitized LFP/C Co doped C-LFP LFP/C C/LFP Core-shell C/LFP C/LFP

Modified Hummer’s method, In-situ polymerization Ball milling, modified Hummer’s method Graphite oxidation, thermal treatment, and chemical reduction Electrostatic grafting Solgel, Thermal treatment, Modified Hummers method Hydrothermal Hydrothermal, heat treatment Plasma treatment Spray drying Vacuum freeze drying/solvothermal tape-cast fabrication Chemical synthesis Co-precipitation CVD, Vacuum drying Wet processing Solgel Ball mill, solid state reaction Solvothermal, carbon coating 3D printing Sol-gel Solvothermal, liquid deposition CVD, Thermal processing Hydrothermal synthesis Insitu plasma treatment

171.9 mAh g−1 at 0.1 C 162.2 mAh g−1 at 0.1 C 121 mAh g−1 at 1 C 160.3 mAh g−1 at 0.3 C 114 mAh g−1 at 1 C 157.4 mAh g−1 at 0.2 C 152.7 mAh g−1 at 1 C 144.9 mAh g−1 at 0.1 C 192 mAh g-1 at 0.1 C 195 mAh g−1 123 mAh g−1 143.8 mAh g−1 at 1 C 150 mAh g−1 163.8 mAh g−1 at 0.1 C 129 mAh g−1 1.44 mAh cm−2 at 0.5 C 158 mAh g−1 at a rate of 1 C 132.8 mAh g−1 at 0.2 C 89.69 mAh g−1 at 200 C 162 mAh g−1 at 1 C 100.7 mAh g−1 at 150 C

Hydrothermal, calcination Solvothermal, Thermal treatment Hydrothermal Vacuum deposition Freeze drying, thermal processing Ball mill, heat treatment Insitu synthesis Combustion method Solgel Chemical synthesis

170 mAh g−1 162 mAh g−1 at 1 C 98 mAh g−1 at 0.1 C 123.9 mAh g−1 at 5 C 155.5 mAh g−1 at 0.1 C 148.3 mAh g−1 at 0.1 C 164 mAh g−1 at 2 C 49 mAh.g−1 at 2.4 mg.cm−2 93.54 mAh g−1 120.2 mAh g−1

Carbothermal reduction Ball mill, Chemical treatment

Glucose Glucose Fructose Sucrose Chitosan Citric acid Citric acid Citrate Citric acid Polymer Polystyrene xylitol-PVA PVDF Thermoplastic polyurethane/Super P N-methylimidazole butyl-3-methylimidazolium dicyanamide Oleylamine [BMIm]N(CN)2—Ionic liquid [VEIm]NTf2—Ionic liquid Polybenzoxazine Dopamine, polyethylene glycol [BMIM]BF4—Ionic liquid Mxene

158 mAh g−1 at 0.1 C 142 mAh g−1 at 0.1 C

Fluorine doped C-LFP LFP/C C-LFP Nitrogen doped C/LFP

Ball mill, rheological solid state phase method phase separation Colloidal synthesis microwave pyrolysis

147 mAh g−1 at 0.1 C Volumetric energy density— 617.8 Wh L−1 at 10 C 100.2 mAh g−1 at 20 C 153 mAh g−1 at 0.2 C 164 mAh g−1 133.6 mAh g−1

LFP/C C-LFP C/LFP LFP/Nitrogen doped C Nitrogen doped C-LFP Fluorine doped N-C-LFP Mxene/LFP/C

Supercritical alcohol, calcination microwave-assisted pyrolysis Hydrothermal Thermal treatment Spray drying Hydrothermal Electrostatic self-assembly

84 mAh g−1 at 20 C 149.4 mAh g−1 136.4 mAh g− 1 at 0.1 C 156.9 mAh g−1 156 mAh g−1 at 0.1 C 162.2 mAh g−1 at 0.1 C 156.6 mAh g−1 at 1 C

different companies buying the same lithium iron phosphate powder from the same place to make cathode elements of lithium-ion batteries for sale did not find the same performance; we recently understood that this difference in performance actually came from the difference in the storage of a product, which is sensitive to moisture. There are many efforts devoted now to reduction of the size of the particles, which also implies increasingly important surface effects. Therefore, we now turn to surface water and its significance for reduction of particle size.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

b

287

a

a

c

FeO6 PO4 O Li Fe

(a)

(b)

Fig. 13 Structural similarity between (a) LiFePO4 olivine, and (b) sarcopside Fe3(PO4)2, which can also be written as Fe0.5FePO4 to make contact with the formula Li1+2xFe1−xPO4. The dashed arrow represents the direction for Li-ion motion in the channels.

15

Reaction to moisture

To understand the aging of a particle upon exposure to water, the particles can just be dropped into water to see what happens. First, we observe that part of the carbon detaches from the particles and floats at the surface, bringing with them a few particles that are not completely detached like fish caught in a fishing net. This is sufficient to understand that the carbon layer is not waterproof and is useless to protect the particle against exposure to moisture. This was actually expected, since the carbon layer is not well crystallized, and the image of the fishing net at the atomic scale is realistic (probably the reason why the carbon layer does not affect the ionic conductivity). One can learn more on the reaction with water by analysis of the water in which the particles have been immersed. The inductively coupled mass spectroscopy analysis shows that it contains iron and phosphorus, which gives evidence that part of the particles have dissolved into the water. We can then evaporate the water in which the particles are immersed and analyze the deposit at the bottom of the boat. The deposit is white with a black crust of carbon on top of it, and part of the deposit has a blue iridescence. The most efficient tool to analyze the product is again Raman spectroscopy, for the same reason as it was the best tool to characterize the carbon deposit as described in Characterization of the Carbon Coating. The Raman spectra of the black/blue deposit and white deposit have been measured with an He–Ne laser beam as excitation source (wavelength 632.8 nm). The result is illustrated in Fig. 14 for the black/blue part of the deposit. The spectrum of uncoated lithium iron phosphate particles is also reported in Fig. 6 for comparison. While the intrinsic spectrum of lithium iron phosphate is dominated by the peak at 960 cm−1 associated with the stretching mode of the PO4 unit, the Raman spectrum of the part that shows blue iridescence is dominated by the two bands characteristic of the carbon. The structure centered at 960 cm−1, however, is also distinctly seen, so that this part of the material also contains phosphate. The Raman spectrum of the black part that does not show blue iridescence which shows again the dominant bands characteristic of carbon, but it does not show any structure at 960 cm−1, which confirms that the blue iridescence is linked to the presence of phosphate and a phosphatization effect. On the other hand, three additional structures can be seen at the lower frequencies 398, 263, and 219 cm−1, which are characteristic of lithium hydroxide monohydrate LiOHH2O. This was actually expected since lithium is known to be very reactive with water according to the reaction. Li + H2 O ! LiOH +

1 H 2 2

(III)

A broad band in the Raman spectrum of the black part can be seen in the vicinity of 1070 cm−1, which is also detected in the part with blue iridescence. This broad line has also been detected in the Raman spectrum of molten lithium hydroxide (LiOH) and is attributed to the vibration of the CO3 molecular unit, giving evidence of lithium carbonate in addition to the lithium hydroxide. This was also expected, since the lithium carbonate is a by-product of lithium hydroxide reacting with carbon dioxide atmosphere according to the equation.

288

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

10000

Intensity (cps)

8000

Part with blue iridescence

6000

4000

2000

0

Black

0

500

1000

1500

2000

Raman shift (cm–1) Fig. 14 Raman spectrum of the deposit after evaporation of the water in which LiFePO4 has been immersed. Black refers to the black part, and blue to the part that shows blue iridescence. The spectrum of uncoated LiFePO4 is shown for comparison. The vertical broken line points to the position of the stretching mode of PO4 units.

2LiOH + CO2 ! Li2 CO3 + H2 O

(IV)

This is even the reaction at the origin of the use of lithium hydroxide to clean the atmosphere from carbonic gas in a confined atmosphere, as in submarines, for instance. Therefore, the Raman spectra show that the primary reaction of lithium iron phosphate with water is the extraction of the lithium from the surface layer to react with water according to Eq. (III) in the first step and Eq. (IV) as a second step. Indeed, both lithium hydroxide and lithium carbonate are white products, and they are responsible for the white part of the deposit. Moreover, the dissolution of the particle into water is fast within the first hour, but then it stabilizes. Even after 24 h of immersion in water, the quantity of iron and phosphorus in the water only corresponds to the quantity of matter that is contained in the 3-nm-thick surface layer of the particles. It means that the lithium is extracted from the surface layer only, which is fortunate for applications. The reason is that, once the lithium has been removed from the surface layer of lithium iron phosphate, we are left with a layer of iron(III) phosphate (FePO4) that is known to be hydrophobic. Once formed, this iron(III) phosphate layer that is waterproof protects the particle from further damage upon exposure to moisture. Actually, the result is the same as the phosphatation of iron, an industrial process used to passivate the surface of iron compounds by covering them with a thin layer of iron(III) phosphate. (Intermediate steps in the phosphatation of iron can be found for instance on the web site http://www.francetraitement.com/fr/phosphatation_fer.html (in French)). In particular, the iron used in the building industry has usually had this phosphatization treatment, which can be easily recognized by its blue iridescence attributed to the diffraction of light on the ultrathin protective layer. Similarly, the same blue iridescence occurs only in the part of the deposit that is phosphate. In addition, the ratio of concentrations of phosphorus and iron ions in the water after immersion of the particles is [P]/[Fe] ¼ 2.3. This result suggests the presence of Fe(H2PO4)2, which is soluble in water and for which [P]/[Fe] ¼ 2. This compound is also formed during the phosphatization industrial process (see the above reference). Another evidence of the delithiation of the surface layer is provided by magnetic measurements. Before immersion of the particles in water, the effective magnetic moment carried by iron, as deduced from the fit of the magnetic susceptibility by the Curie law in the paramagnetic regime, is 5.36 mB for the sample investigated in Zaghib and coworkers, a typical value for a sample with few lithium vacancies (see Lithium Vacancy: The Magnetic Polaron). After immersion in water, this moment increases to 5.40 mB. This increase comes from the fact that the removal of a fraction x of lithium implies that the same fraction x of Fe2+ (meff ¼ 4.90 mB) has switched to the Fe3+ configuration (magnetic momentum meff ¼ 5.92 mB) to maintain charge neutrality. Therefore, x is readily given by.   (6) ð5:36Þ2 + x ð5:92Þ2 − ð4:90Þ2 ¼ ð5:40Þ2 The solution is x ¼ 0.04. Taking into account the log-normal distribution of the sizes of the particles, we have determined that this fraction x corresponds to a delithiation of a layer of thickness 3.3 nm at the surface of the particles. This is in quantitative agreement with the thickness of the surface layer observed on HRTEM images in Fig. 15. The lithium that has been removed from the surface

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

289

Fig. 15 High-resolution transmission electron microscopy (HRTEM) images of the surface of LiFePO4 particles before carbon coating (left) and after carbon coating (right). The lighter part surrounding the darker particles is the carbon layer, few nanometers thick on the right picture. The lighter parts on the image to the left show residual carbon that remains only at several spots of the surface, most of the carbon having been removed by the ball milling.

layer is the lithium that has been found in the deposit after evaporation of the water in which the particles have been immersed under the form of lithium hydroxide and lithium carbonate. We also recover the result that, after this delithiation upon exposure to water molecules, we are left with a surface layer of hydrophobic iron phosphate that is waterproof and protects the particles against further corrosion. On the other hand, no reaction at the surface of lithium iron phosphate takes place in dry air. Therefore, our results answer the question on the reaction of the material with air: the material is not sensitive to oxygen or air; it only needs protection against moisture. We shall detail this statement in the section devoted to electrochemistry, since it is a major issue for applications.

16

Size effects

We have already outlined the interest in decreasing the size of the particles. Nanosized particles can be obtained by jet-milling the particles of Sample B in a liquid solution containing isopanol under inert atmosphere. The particles obtained after drying can be considered as uncoated particles because of the great damage caused by the milling process. The carbon-coated particles have been obtained with lactose as the carbon precursor in acetone solution according to the following procedure. The uncoated particles were mixed with the carbon precursor. The dry additive corresponded to 5 wt% carbon in lithium iron phosphate. After drying, the blend was heated at 750
C for 4 h under argon atmosphere. A typical TEM image shows that the powders are composed of well-dispersed particles. The average size of the primary particles is found to be about 45 nm. Only a few particles agglomerate, which is remarkable since there is a tendency for small particles to stick together to form aggregates. The particles after milling have a disordered surface layer, about 3 nm thick as usual. The HRTEM image after carbon coating in Fig. 15 shows the same particle, now covered with a 3-nm-thick carbon layer and a less-disordered surface layer. The procedure to characterize these particles is the same as the procedure above described for thicker particles: XRD, FTIR, Raman spectroscopy, and magnetic properties. The full study has been reported elsewhere. Let us review the results that we have obtained, starting with the particles after jet milling, that is, without a carbon coat. The coherence length deduced from the width of the XRD lines by application of Scherrer’s law is d ¼ 35  5 nm, the core of the particles below the surface has a size that is the same as the coherence length deduced from the XRD analysis. Therefore, the particles are crystallites surrounded by a disordered surface layer. Nevertheless, the FTIR spectrum is in between the spectra of well-crystallized Sample B and the glassy sample in Fig. 5, which is evidence that the spectrum is the superposition of the contribution from the well-crystallized core (same contribution as Sample B in Fig. 5) and the disordered surface layer (same contribution as the glass in Fig. 5). The surprise comes from the magnetic properties. The effective magnetic moment per Fe ion, deduced from the fit of w−1(T ) by the Curie–Weiss law in the paramagnetic region where w−1(T ) is linear, is quite small: meff ¼ 4.66 mB. Remember that impurity effects reviewed in a former section lead to a value of the magnetic moment larger than 4.9 mB, so the smaller value of meff in the present sample must then have another origin,

290

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

namely a dramatic decrease of the contribution to w of the iron atoms at the surface of the lithium iron phosphate particles. This effect can be modeled by a decomposition of the experimental value of w(T ) in two components, wðT Þ ¼ y

C0 C0 + ð1 − y Þ T + y0 T

(7)

The first component is that of the core region of lithium iron phosphate particles, and the second term is the contribution of the Fe atoms that are in the surface layer of the particles; y is the fraction of iron ions in the bulk. We know from the XRD and FTIR experiments reported in the previous sections that the core region is not affected by any size effect of the particle, so that the parameters C0 and y0 are those relevant for this material, namely C0 ¼ 3.00 emuKmol−1, and y0 ¼ 101  6 K. The second component is the contribution of Fe spins in the surface layer. If the surface Fe3+ ions are in their low-spin state, which, we will see, turns out to be the case, the frustrated net interatomic exchange interactions would be negligible, so we assume their contribution to the magnetic susceptibility satisfies the Curie law. We are then left with two fitting parameters, namely C0 , which depends on the spin carried by the iron atoms in the surface layer, and the fraction y of iron atoms in the ‘bulk’ of the particles. We have shown elsewhere that the fit of the susceptibility curve by Eq. (7) gives a unique solution for the fitting parameters y and C0 . The result is y ¼ 0.89, C0 ¼ 0.37 emuKmol−1. This value of C0 corresponds to a spin S ¼ 1/2, implying that the iron ions in the surface layer are Fe3+ ions in the low-spin state. Again, we can deduce the thickness of the surface layer corresponding to the fraction (1 − y) of iron ions in the surface layer, following the procedure we have described elsewhere. For this particular size of particles, we recover the thickness 3 nm of the surface layer that had been evidenced on the TEM images. All the structural and physical analyses are then fully self-consistent. The inverse of the magnetic susceptibility as a function of temperature of the coated particles is reported in Fig. 16 together with that of the particles before carbon coating for comparison. The magnetic response of the coated particles is closer to the result predicted for intrinsic lithium iron phosphate (see the w−1(T ) curve after reduction of the surface contribution in Fig. 9) with, however, an effective moment 5.02 mB still slightly larger than the theoretical value 4.90 mB. Therefore, Fe3+ has switched to the high-spin state 5/2, for which the effective moment is 5.92B. Actually, we note that. yð4:9Þ2 + ð1 −yÞð5:92Þ2 ¼ ð5:02Þ2

(8)

which means that the excess in magnetic moment with respect to the theoretical value is entirely attributable to the conversion of Fe3+ (S ¼ 1/2) into Fe3+ (S ¼ 5/2) in the surface layer. We then recover the result we have found for bigger particles, namely exposure to the moisture of ambient air, even for a short period of time, delithiates the surface layer of C-LiFePO4 with the consequence that iron is in the Fe3+ (S ¼ 5/2) in this surface layer just below the carbon layer. This phenomenon also explains the presence of a few percent of iron in the Fe3+ state revealed by Mössbauer experiments, not detected by XRD. In an earlier work, A. S. Andersson and coworkers have already invoked the possibility of an extremely thin, nanocrystalline phase at the boundary of the lithium iron phosphate phase since the work was devoted to the study of partially delithiated samples rather than lithium iron phosphate itself. This hypothesis is now confirmed. Indeed, at the time these authors made these measurements, the sensitivity of the particles to moisture was not yet known, and the particles had most likely been exposed to the moisture of the ambient

140 LiFePO4

H/M (mol Oe emu–1)

120 Jet-milled sample 100

80 After heat treatment with lactose (750 °C)

60

40 0

100

200

300

Temperature (K) Fig. 16 Inverse of the magnetic susceptibility of LiFePO4 prepared by jet milling, before and after carbon coating, as a function of temperature.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

291

atmosphere. The relative concentration of Fe3+ detected by these authors is of the same order of magnitude as in the present sample, but smaller, simply because the particles were bigger. The high-spin or low-spin state of Fe3+ in the surface layer is a probe of the degree of disorder of the layer, since the low-spin state where the surface layer is disordered and the high-spin state where the surface layer is well crystallized. However, the origin of the disorder of the surface layer and the spin state is still questionable. One hypothesis is that the carbon deposit is responsible for this effect. Indeed, the spin transition of iron ions in presence of carbon is commonly observed in biology and is responsible for the efficiency of many proteins that are classified by heme iron. Similar effects have also been observed in ferrite nanoparticles investigated in the context of spintronics. Another hypothesis is that the carbon deposit is not responsible for this effect, the reduction of the disorder being only due to the recrystallization of the surface layer upon heating the particles at 750
C, since the particles have been sintered at this temperature in the 4 h during the carbon-coating process. At the present time, this question is still open.

17

Electron paramagnetic resonance

The derivative signals of the absorption spectra of the carbon-free sample are reported in Fig. 17. Let us recall that no electron paramagnetic resonance (EPR) signal is detected in lithium iron phosphate in the absence of magnetic clusters. The EPR signal is then an evidence of magnetic clustering. For isolated magnetic clusters, one expects a signal characteristics of a gyromagnetic factor g ¼ 2. At the frequency used in the experiments, such a signal is centered at H ¼ 3300G. Indeed, this signal, already detected in other lithium iron phosphate samples that contained ferrimagnetic particles, is also detected in the present work and has a comparable shape. The structure at 3300G has the same width. In addition, a large signal is also detected at smaller magnetic fields; it is associated with a strong rise in the magnetization curves. The spectrum of the carbon-coated sample is reported in Fig. 18; it shows a dramatic decrease in the ESR signal. The integrated strength of the structure at 3300G is 100 times smaller than that in the carbon-free sample. The structure at 3300G is much broader than in the carbon-free sample, which suggests that the origin of the signal is different. This feature is clearly related to the fact that we did not observe any ferrimagnetic nanoparticles in the carbon-coated sample. This very small signal is the signature of uncoupled spins in such a small concentration that it may be due to spins at residual impurities or defects. This kind of spin is very sensitive to its local neighborhood, which might explain the large broadening of the ESR line.

0.01

100 K

300 K

60 K

ESR signal (au)

50 K 0

LiFePO4 Carbon free

–0.01

–0.02

0

2000

4000

6000

8000

Magnetic field (G) Fig. 17 Electron spin resonance (ESR) spectrum for the carbon-free LiFePO4 sample at several temperatures indicated in the figure. Note the unit is arbitrary, but it is the same one as in the next figure, so that the relative intensity between the spectra of the two samples is given by the ratio of the ordinates between the spectra in the two figures.

292

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

260 K

2 u10–6

180 K

ESR signal (au)

140 K 0 100 K

LiFePO4

–2 u10–6

Carbon coated

–4 u10–6 0

2000

4000

6000

8000

Magnetic field (G) Fig. 18 Electron spin resonance (ESR) spectrum for the carbon-coated LiFePO4 sample at several temperatures indicated in the figure.

18

Performance of optimized LiFePO4

Here, we present an overview of the high-temperature performance for an optimized B-type sample, that is, carbon coated (C-LiFePO4). The coffee-bag cell was charged and discharged at C/8 for the first cycle followed by 12 cycles at C/4 with 1 h rest before each charge and discharge. This high-temperature test was made at 60
C, which is the appropriate condition to investigate possible iron dissolution in nonaqueous electrolytes. Fig. 19 shows the XRD patterns of the new generation of lithium iron

LiFePO4

10

20

(311)

30

40

50

(412)/(610) (331) (430)

(022) (131) (222)/(402)/(231)

(321)/(212)

(112)/(202)

Before cycling (121) (410) (102)/(221)/(401)

(301)

(C) (211)/(201)

(011)

(210)

(200)

(111)/(201)

(101)

XRD intensity (au)

(After 200 cycles (*carbon additive)

60

2q (degree) Fig. 19 X-ray diffraction (XRD) pattern of C-LiFePO4 positive electrode before and after 200 cycles. Notice that the olivine framework (Pmna space group) remains intact after cycling at 60
C.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

293

phosphate after 200 cycles (47 days) at 60
C. There is no change in the olivine structure after cycling at 60
C. We observed Bragg lines with the same intensity as that for the pristine material. The capacity loss was below 3% in 100 cycles for this optimized electrode material. A close examination was made for the detection of any iron dissolution that could occur after long-term cycling. The analysis of iron species was investigated at the separator/lithium (SL) interface by SEM cross section (slice view) as shown in Fig. 20(a) and (c). The micrograph (Fig. 20(a)) obtained from evaluation of the earlier-generation material shows the presence of iron islands at the SL interface. Obviously, some iron particles (or ions) migrate through the electrolyte from the LiFePO4-positive electrode to the lithium-negative electrode. The net effect of this migration is a large decrease in capacity retention of the Li//LiFePO4 cell. Fig. 20 (b) shows a micrograph obtained from tests with an optimized electrode in a lithium cell with a lithium-foil anode. In this case, there is no iron detected at the SL interface, which remained intact after 100 cycles. In fact, this high performance was possible not only because of the optimized synthesis of the lithium iron phosphate powders, but also because of strict control of the structural quality of the materials. Several physical methods were utilized to analyze the local structure and the electronic properties of the phospho-olivine framework. Fig. 20(c) shows the SEM picture of graphite taken from a lithium-ion cell after 200 cycles. No iron was observed at the surface of the electrode material. Energy-dispersive X-ray (EDX) spectrometry analysis of the graphite electrode

Fig. 20 Postmortem Scanning electron microscopy (SEM) images of the detection of iron species at the separator lithium (SL) interface. (a) Image showing the formation of iron islands at the interface with an earlier generation of LiFePO4. (b) No iron was detected at the surface of lithium foil with the optimized LiFePO4. (c) Lithium-ion graphite electrode after 200 cycles does not show the presence of iron particle.

294

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

confirms this last observation. Iron dissolution, if it occurred, should be evident in the EDX spectrum on the graphite side. In addition, elemental analysis of the electrolyte in the lithium-ion cell was carried out by the ICP technique to detect any iron after 200 cycles. No iron, even at the ppm level, was found in the electrolyte solution. Thus, all these data converge to the conclusion that lithium iron phosphate is not soluble at 60
C with the optimized material. The electrochemical properties of lithium iron phosphate are known to be sensitive to the mode of preparation and the structural properties. This can be an advantage for potential applications since it allows for an optimization of the material if we can correlate the mode of preparation with the structural and the physical properties. To address this issue, we investigated this relationship in lithium iron phosphate samples that were grown under different conditions. Undesirable impurities in the lattice can be introduced during the growth process. For instance, the presence of iron phosphide can increase the electronic conductivity, but on the other hand it also decreases the ionic conductivity so that both the capacity and cycling rates are degraded with respect to C-LiFePO4. Actually, the presence of any of the impurity phases listed in the section devoted to this subject is damageable to the electrochemical performance. Fortunately, we have also shown in this article that we now can master the synthesis process to avoid them. The carbon deposition process, which was by organic precursors to make C-coated samples, generates a reductive gas such as hydrogen that is more kinetically active and reduces Fe3+ impurities in the 400–700
C temperature range used in our studies. This reduction is also favored by the fact that the organic precursor is usually mixed with the lithium iron phosphate material or with the lithium iron phosphate chemical precursors by solution processes at a molecular-size level. Surface effects are also important. We have already reported the aging of lithium iron phosphate upon exposure to moisture. This effect is illustrated in Fig. 21, where we have reported the evolution of the capacity of a C-LiFePO4/LiPF6–EC–DEC/Li cell as a function of the time that the lithium iron phosphate has spent in dry atmosphere and in ambient atmosphere (55% relative humidity (RH)). No degradation at all is observed in dry atmosphere. On the other hand, when left in ambient air with 55% RH, the degradation increases with temperature and becomes dramatic at the scale of 6 months. Therefore, lithium iron phosphate must be protected against humidity and kept in a glove box for academic research and in a dry room at less than 5% RH. On the other hand, no aging is observed if this recommendation is respected. In case the powder has been left inadvertently exposed to humidity for a few days only, there is no need to panic. We have shown that the damage during this short period of time is reversible, and the initial capacity of the battery is recovered after drying since the first charge/discharge cycle relithiates the surface layer. This is again one fortunate consequence of the phosphatation effect described earlier, as the iron phosphate surface layer is protective. However, this layer only slows down the degradation process and cannot stop it entirely; after 6 months of exposure to 55% RH humidity, the damage is irreversible.

180

Specific capacity (mAh g–1)

160

25 °C

140

120

45 °C Aging in humid atmosphere Aging in dry atmosphere

100

60 °C 80

1

2

3 4 Time (month)

5

6

Fig. 21 Capacity of the C-LiFePO4 (HTR sample)/LiPF6–EC–DEC/Li cells as a function of time spent in dry atmosphere and in ambient atmosphere (55% RH), at three different temperatures. The temperatures at which the full curves (in dry atmosphere) have been obtained can be distinguished by the fact that they do not overlap and the property that the lower the temperature, the higher the capacity is.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

295

Potential (V) versus Li0/Li+

4.5 LiFePO4 Before lactose

4.0

3.5

3.0 Charge 2.5

2.0

Discharge

0

40

80

120

160

–1

Specific capacity (mAh g )

(a)

Potential (V) versus Li0/Li+

4.5

3.5

3.0 Charge 2.5

2.0 (b)

LiFePO4 After lactose

4.0

Discharge

0

40

80

120

160

Specific capacity (mAh g–1)

Fig. 22 Charge–discharge voltage profile of LiFePO4 before (a) and after (b) carbon coating of the cathode material (particle size 45 nm), using LFP/ LiPF6–EC–DEC/Li cell at room temperature. The test was performed galvanostatically at charge–discharge rate C/24.

The disorder in the surface layer also plays an important role. Fig. 22 shows the typical charge–discharge voltage profile of the LiFePO4(LFP)/LiPF6–EC–DEC/Li cell prepared with the lithium iron phosphate cathode material investigated in section titled Size Effects (size of the particles: 45 nm). The test was performed galvanostatically at a charge–discharge rate C/24 in the voltage range 2.2–4.0 V versus Li0/Li+. After carbon coating, the charge–discharge profile appears with the typical voltage plateau at 3.45 V versus Li0/Li+ attributed to the two-phase reaction of the (1 − x)FePO4 + xLiFePO4 system. The carbon-coated material exhibits a reversible capacity of 160 mAhg−1 that amounts to a utilization efficiency of 94%. This is typically the same profile as the one observed for bigger particles prepared under optimized conditions. It is commonly believed that the reduction of the size of the lithium iron phosphate particles may effect the phase diagram of the FePO4/LiFePO4 system and favor the existence of the solid solution LixFePO4. However, we can see from the present study and Fig. 22 that even the lowering of the particle size down to about 45 nm does not reduce the extension of the plateau and then does not affect the phase diagram of FePO4/LiFePO4, which remains a two-phase system.

296

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

On the other hand, before carbon coating, the width of the plateau is reduced. In Fig. 22, there is a significant slope in the voltage-versus-capacity graph (converted to the concentration x of lithium) in the whole range 0.8 < x < 1 during the charge, which extends to the range 0.7 < x < 1 during the discharge, even though the cycles have purposely been done very slowly at a C/24 rate, to be as close as possible to thermodynamic equilibrium. In Fig. 22, the significant slope in the voltage-versus-capacity curve observed in the whole range of composition 0.7 < x < 1 is in agreement with very similar charge–discharge profiles found for particles of similar size reported by different authors, and this effect has been considered as an evidence for the reduction of the miscibility upon decreasing the size of the particle down to 40 nm. Our results, however, are in contradiction with this assumption since the plateau is not significantly reduced in Fig. 22 with respect to the plateau observed for bigger particles, provided that the surface layer is well ordered, which suggests a different interpretation. The solid solution in the range 0.7 < x < 1 (and in the range x < 0.3 on the lowx side of the phase diagram) as revealed by the slope in the charge–discharge curves is located only in the surface layer, not in the ‘bulk’ of the particles. In addition, this solid solution is due to the disorder, since the reduction of the disorder by the carbon-coating process induced a demixing of the two phases and the restoration of the plateau. Recently, a slope in the charge/discharge profiles has been observed over even larger ranges of lithium concentrations x. In that case, however, a very large concentration of defects in the bulk of the particles has been detected. This uncontrolled amount of defects is another type of disorder that, in addition to an eventual strong disorder in the surface layer, is responsible for this effect and a degradation of the electrochemical performance. One important result of the present work is that the jet-milling preparation process can remove these disorder effects so that the full performance and the electrochemical properties are fully recovered, just like in bigger particles. After a recent calculation of the electronic structure of lithium iron phosphate that takes into account the finite size of the particles, a diameter of the particle on the order of 40 nm is still too large to affect the phase diagram, so any solid solution should be still unstable at room temperature. In this calculation, the surface energy is taken into account, but no disorder was introduced, so that the results should be compared with our experiments for particles after carbon coating. Indeed, our results are in agreement with this calculation, since we find that the charge–discharge profile and the plateau that is the signature of the two-phase system is the same as in the case of bigger particles.

19

Electrochemical performance at 60
C

Fig. 23 presents a typical electrochemical performance of C-LiFePO4/LiPF6–EC–DEC/Li cells at 60
C. The charge–discharge curves were obtained by cycling at C/4 rate (about 35 mAg−1) in the voltage range 2.2–4.0 V versus Li0/Li+. The optimized lithium iron phosphate exhibits a reversible capacity that is maintained over many charge–discharge cycles. The 13th and 97th cycles show a similar specific capacity of 160 mAhg−1. These results illustrate the excellent electrochemical performance of the carbon-coated olivine material. The electrode can be fully charged up to 4 V, which is its most reactive state. This remarkable performance is attributed to the optimized carbon-coated particles and their structural integrity under a large current in the electrode. Even at such a high cycling rate, C-LiFePO4 exhibits rapid kinetics of lithium extraction and realizes most of its theoretical capacity (170 mAhg−1). The discharge profile appears with the typical voltage plateau (at 3.45 V vs Li0/Li+) attributed to the two-phase reaction of the (1−x)FePO4 + xLiFePO4 system. Fig. 24 shows the Ragone plots of the Li//LiFePO4 cells cycled at 25 and 60
C. From C/12 to 6C,

4.5

LiFePO4/LiPF6–EC-DEC/Li C/4 at 60 °C

Cell voltage (V)

4.0

3.5

3.0

(D) 97th cycle (C) 97th cycle (C) 13th cycle (D) 13th cycle

2.5

2.0 0

50

100

150

200

–1

Specific capacity (mAh g ) Fig. 23 The charge–discharge profiles of the Li//LiFePO4 cells with optimized carbon-coated electrode (13th and 97th cycle) at 60
C. Experiments were carried out at C/4 rate in the voltage range 4.0–2.2 V. The 13th and 97th cycles show similar specific capacity of 160 mAhg−1.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Lithium Iron Phosphates

297

Specific capacity (mAh g–1)

160

120

25°C

80

60°C

40

LiFePO4/EC–DEC–LiPF6/Li 0 0.01

0.1

1 C-rate

10

100

Fig. 24 Ragone plots of the Li//LiFePO4 cells cycled at 25
C and 60
C.

the capacity is almost maintained constant at 150 mAhg−1; from 6C to 25C the capacity decreases with increasing current density. The cell still has a good performance at 10C with 134 mAhg−1, which represents 86% of the capacity at C/12. A comparison of the specific capacity between A-type and B-type lithium iron phosphate electrodes during long-term cycling has shown that the cycling performance for the new generation of C-LiFePO4 material is excellent at 60
C. After 100 cycles at C/4 rate and with a typical cutoff voltage of 4.0–2.2 V, a constant capacity was observed. This best performance is due to the improved technology used in electrode fabrication, i.e., improvements in the nature and the morphology of the carbon coating and the optimization of the particle size of the olivine phase. These are the two main factors controlling the electrode performance. In our previous exploration of the surface properties of the lithium iron phosphate particle, we have shown by Raman spectroscopy that the deposit is a disordered graphite-type carbon. The small amount of carbon ( + 4.5 V versus lithium. High-voltage spinel Manganese based positive material with the general formula LiMxMn1-xO4 with (M ¼ Cr. Fe, Co, Ni, Cu) where the M2+/M4+ or M3+/M4+ redox couple works at potentials > + 4.5 V versus lithium, the most popular material is LiNi0.5Mn1.5O4. Li-rich layered oxides Positive electrode materials with more Li+ in the layered structure than the stoichiometric amount of typical layered oxides (Li+/TM ¼ 1); general formula Li1+xMO2. These materials can give very high capacities up to >250 mAh g−1. Olivine structure Known as polyanion naturally seen in Fe2SiO4 and MgSiO4 structure. One of them, LiFePO4 (LFP) is a commercially used positive active material. Spinel structure The structure based on the cubic crystal found in the formula MgAl2O4.

Key points

• • • •

Description of structural and electrochemical properties of high voltage spinel type materials with the general formula: LiMxMn1-xO2. Synthesis, characterization and electrochemical performance of LiNi0.5Mn1.5O4. Description of structural an electrochemical properties of high voltage olivines LMPO4 (M ¼ Co, Ni) with olivine structure. General description of HE-NMCs and electrochemical reaction mechanism of HE-NMCs.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00328-5

305

306

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: High-Voltage Materials

Nomenclature HE-NMC LFP LMP LRLO NMC

1

Li- and Mn-rich lithium manganese cobalt layered oxides Lithium iron phosphate LiFePO4. Lithium transition metal phosphate (M ¼ Co, Ni). Lithium and manganese rich layered oxides. Layered LidNidMndCo oxides.

Introduction

Key issues for further development of advanced batteries are increase in energy density, cost reduction and availability of raw materials. Two main concepts can lead to higher energy density, either development of new positive electrode and negative electrode materials with higher capacities or an increase of the cell voltage. For negative electrode materials, an increase of the overall cell voltage is not possible from a physical point of view. In this field, research focuses on materials like silicon or metallic lithium with higher specific capacities. Positive electrode materials are structurally limited in their specific capacity and development efforts also addresses alternative materials with increased discharge potentials. Moreover, a positive electrode material with a higher capacity improves the specific energy of the whole cell only in relation to its weight fraction whereas a positive electrode material with a higher voltage directly increases the energy of the total cell. Positive electrode materials can be termed high-voltage compounds if the discharge curve exhibits a distinct redox step at a potential significantly beyond the mean discharge voltage of conventional materials, such as LiMn2O4, LiCoO2 and layered oxides like LiNi1-x-yCoxAlyO2 (NCA) or LiNi1-x-yMnxCoyO2 (NMC). The high-voltage materials can be divided into two structural categories: LiMxMn1-xO4 (spinel structure) and LiMPO4 (olivine structure) (Table 1). In the spinel type structure the M2+/M4+ or the M3+/M4+ transition is utilized in the high-voltage step to which manganese does not contribute to. In purely oxidic systems the potential of the M2+/M3+ redox couple is located at low voltage levels.1 In LiMPO4 the potential of this redox couple is shifted to higher values due to the higher ionic character of the MdO bond. The so-called Li-rich layered oxides (LRLO, HE-NMC) are an additional class of layered materials with the general formula x Li2MnO3 ∙ (1-x) LiNi1-y-zMnyCozO2 (y + z  1). They can provide very high reversible capacities (>250 mAh g−1) in combination with average discharge potentials over 3.5 V vs. Li+/Li and using a broader potential range compared to conventional used layered oxides like NCA and NMC.2,3 Fig. 1 shows the schematic discharge voltage curves of various positive electrode materials.

2

High-voltage spinels LiMxMn2-xO4

The high-voltage LiMxMn2-xO4 materials are structurally related to the LiMn2O4 spinel. LiMn2O4 crystallizes in the space group Fd-3 m. The oxygen ions form a cubic close-packed arrangement. Manganese ions occupy one half of the octahedral voids, forming a three-dimensional network of edge-sharing MnO6-octahedra (16d site). Electronic conduction proceeds as charge transfer Table 1

High-voltage positive electrode materials for lithium-ion batteries.

Formula

Structure type

˚ Lattice parameters/A

Charge distributions

Redox couple

High-voltage potential vs. Li/Li+/V

LiCr0.5Mn1.5O4

Spinel

a ¼ 8.2083

h i Li Cr3+ 0:5 Mn4+ 1:5 O4

Cr3+/Cr4+

4.8

3+

4+

3+

4+

LiCrMnO4 LiFe0.5Mn1.5O4 LiCo0.5Mn1.5O4

Spinel Spinel Spinel

a ¼ 8.189 a ¼ 8.2483 a ¼ 8.1485

Li[Cr Mn ]O4 Li[Fe3+0.5Mn4+1.5]O4 h i Li Co3+ 0:5 Mn4+ 1:5 O4

Cr /Cr Fe3+/Fe4+ Co3+/Co4+

4.8 4.9 5.0

LiCoMnO4 LiNi0.5Mn1.5O4

Spinel Spinel

a ¼ 8.073 a ¼ 8.169

Li[Co3+Mn4+]O4 h i Li Ni2+ 0:5 Mn4+ 1:5 O4

Co3+/Co4+ Ni2+/Ni4+

5.0 4.7

LiCu0.5Mn1.5O4 LiNiVO4 LiCoPO4

Spinel Inverse spinel Olivine

Li1-xCux[LixCu0.5-xMn1.5]O4 V[Ni2+Li+]O4 LiCo2+PO4

Cu2+/Cu3+ Ni2+/Ni3+ Co2+/Co3+

4.9 4.8 4.8

LiNiPO4

Olivine

a ¼ 8.199 a ¼ 8.2223 a ¼ 10.2001 b ¼ 5.9199 c ¼ 4.69 a ¼ 10.0317 b ¼ 5.8569 c ¼ 4.6768

LiNi2+PO4

Ni2+/Ni3+

5.1–5.3

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: High-Voltage Materials

307

Fig. 1 Typical voltage profiles of an olivine LiFePO4 (black), an high-voltage spinel LiNi0.5Mn1.5O4 (red), a Ni-rich NMC LiNi0.8Mn0.1Co0.1O2 (blue), a Co-free layered oxide LiNi0.5Mn0.5O2 (green), and a Li-rich layered oxide Li1.2Mn0.54Ni0.13Co0.13O2 with Li metal as counter electrode.

Fig. 2 Structural model of LiMn2O4.

between the polyvalent manganese ions via the common edges of the MnO6-octahedra. Lithium ions occupy 1/8 of the tetrahedral sites namely those (8a site) that share faces with vacant octahedral sites (16c site). The LiO4-tetrahedra share corners with the MnO6-octahedra (16d site). The remaining octahedral vacancies (16c site) form the same three-dimensional network as the 16d positions but shifted about ½, ½ and ½ in x, y and z direction. Lithium ions on 8a positions move along the pathway of this network (8a ! 16c ! 8a ! 16c ! . . .) during the topotactical delithiation/lithiation process (Eq. 1). The Mn3+/Mn4+ redox couple is positioned at about 4.1 V vs. Li/Li+. LiMn3+ Mn4+ O4 ⇆ Lix Mn3+ 1−x Mn4+ 1+x O4 + x Li+ + x e−

(1)

In the structural model of LiMn2O4 the blue-colored octahedra represent manganese on the 16d site and the yellow-colored tetrahedra lithium on the 8a site (Fig. 2). To obtain high-voltage materials the introduction of a second redox couple with a higher potential is necessary. The high-voltage spinels LiMxMn2-xO4 contain first row transition metals M that partially replace trivalent manganese ions on the 16d site. By means of spectroscopic investigations the substituents M have been shown to adopt the oxidation state +2 in the case of nickel and copper and +3 in the case of chromium, cobalt and iron. Manganese is present in the tri- and tetravalent state. In the stabilizing spinel matrix the substituents are electrochemically active and can be cycled reversibly at potentials significantly above the Mn3+/Mn4+ level. Two substitution models are discussed. Trivalent substituents M3+ simply replace Mn3+ ions according to the formula 3+ 4+ 3+ 4+ 3+ by Li[M3+ x Mn1-xMn ]O4. The theoretical upper limit of x ¼ 1 is obtained for Li[M Mn ]O4. In contrast, the substitution of Mn divalent substituents M2+ results in the oxidation of equal amounts of Mn3+ to Mn4+. The substitution pattern is described by the 3+ 4+ 2+ 4+ formula Li[M2+ x Mn1-2xMn 1+x]O4. The theoretical upper limit of x ¼ 0.5 is obtained for Li[M0.5Mn1.5]O4. In both models the fully 3+ substituted compounds do not contain any more Mn ions. The electrochemistry is completely governed by the substituent. A single high-voltage step with maximum capacity should be obtained. For x values below the theoretical limit Mn3+ ions remain in the structure and lead to an additional voltage step at 4.1 V vs. Li/Li+

308

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: High-Voltage Materials Principal manganese based high-voltage spinels (fully substituted systems).

Table 2

High-voltage redox couple

Ideal high-voltage redox reaction

M2+ ! M3+ + e−

h i Li M2+ 0:5 Mn4+ 1:5 O4 ! M4+ 0:5 Mn4+ 1:5 O4 + e− + Li+ h i h i Li M2+ 0:5 Mn4+ 1:5 O4 ! Li0:5 M3+ 0:5 Mn4+ 1:5 O4 + 0:5 e− + 0:5 Li+

M3+ ! M4+ + e−

Li[M3+Mn4+]O4 ! [M4+Mn4+]O4 + e− + Li+

M

2+

!M

4+

−

+2e

h

i

Substituent Ni2+ Cu2+ Cr3+, Fe3+ and Co3+

Three principal types of high-voltage redox couple are possible. They are summarized together with the idealized high-voltage spinel composition, the redox reaction and the relevant substituents in Table 2. The high-voltage Li[Ni2+0.5Mn4+1.5]O4 spinel (lithium manganese nickel oxide, LMNO) is the most promising candidate of this group in terms of electrochemical performance and costs. It is a cobalt free material, which works reversibly at high voltages about 4.7 V vs. Li/Li+ without any contribution on the 4 V plateau.4 Li[Co3+Mn4+]O4 comes close to the expected behavior.5 Iron, chromium and copper substituted compounds show distinct contributions in the 4 V region even if fully substituted. Oxygen deficiency, substitutional disorder between lithium (8a) and substituent (16d) as well as phase separation effects cause this behavior.

2.1

Synthesis of LiMxMn2-xO4

LiMxMn2-xO4 materials are synthesized by thermal treatment of precursors containing lithium and transition metals in stoichiometric amounts. The precursors are intimate mixtures of lithium and transition metal compounds. A homogeneous distribution of the component within the precursor is essential. Finely dispersed mixtures can be obtained from co-precipitation, sol-gel synthesis or thermal decomposition of e.g. nitrates. Alternatively, high-energy milling can be applied to coarse-grained mixtures in the case of oxidic starting materials. The thermal treatment is performed under air. Control of the annealing profile is important as the oxygen stoichiometry of the products varies depending on temperature and the nature of the substituent.

2.2

High voltage spinel LiNi0.5Mn1.5O4

LiNixMn2-xO4 forms solid solutions for 0 < x  0.5. Suitable precursors can be prepared by several methods. Special care has to be taken during the thermal treatment. Deviations from the ideal metal:oxygen stoichiometry have been observed. At temperatures above 710  C oxygen-deficient spinel (LiNi0.5Mn1.5O4-d) forms, accompanied by partial reduction of Mn4+ ions. Values for d up to 0.14 have been observed. Segregation of NiO or LizNi1-zO can occur. Stoichiometric LiNi0.5Mn1.5O4 can be prepared either by annealing at 700  C or, if higher temperatures are applied during calcination, by cooling the product very slowly under air to repair the oxygen framework and reoxidize the Mn3+ ions.6 Compounds of the solid-solution series LiNixMn2-xO4 as well as the oxygen-deficient LiNi0.5Mn1.5O4-d crystallize in the cubic face-centered structure (space group Fd-3 m) with a statistical distribution of nickel and manganese ions on the 16d site. In the fully substituted spinel (x ¼ 0.5) additionally an ordering within the transition metal sublattice may occur for the ideal oxygen stoichiometry (d ¼ 0). In the ordered variant Ni2+ occupies the 4b site and Mn4+ the 12b site (Fig. 3). Superstructure reflections in the X-ray diffractograms can be detected. Progressive loss of ordering with increasing temperature is evidenced by neutron scattering between 700  C and 800  C.7 One electron can be exchanged per formula unit LiNixMn1.5-xO4-d corresponding to a specific capacity of approximately 148 mAh g−1. Two potential steps at 4.1 V and 4.7 V vs. Li/Li+ can be observed. The relative length of the individual steps can be 3+ 4+ adjusted by the degree of substitution x and the oxygen deficiency d. From the general formula Li+[Ni2+ x Mn1-2+2dMn 1+x-2d]O4-d 3+ 4+ the length of the steps calculates to (1-2 + 2d) electrons for the Mn /Mn redox couple at 4.1 V and to (2-2d) for the Ni2+/Ni4+ 4+ 2+ redox couple at 4.7 V. The fully delithiated product has the formula [Ni2+dNi4+ for d >0. A single 0.5-dMn1.5]O4-d with remaining Ni high-voltage step with a flat potential behavior is obtained for the fully substituted phase (x ¼ 0.5) with ideal oxygen stoichiometry (d ¼ 0). Slight oxygen-deficiency in the fully substituted phase improves the high-current compatibility and cycle life. This behavior is due to the higher electronic conductivity of the oxygen deficient phase. The increased conductivity is explained by an electron-hopping mechanism between Mn3+ and Mn4+ ions. The electronic conductivity correlates with the Mn3+ content and the lattice parameter a, i.e., that the conductivity can be deduced from structural data. Ni2+ can be partially replaced by divalent and 2+ 4+ trivalent, Mn4+ by tetravalent transition metal ions. Partial substitution of Ni2+ by Mg2+ in LiNi2+ 0.5-xMgx Mn1.5O4 (0 < x < 0.1) 2+ increases the cycling stability. The specific capacity is reduced with x, as Mg ions replace electrochemically active Ni2+ ions. The cycling stability is improved as well by small amounts of Ti4+ ions replacing Mn4+ ions. Partial replacement of Ni2+ by Fe3+ in LiNi0.5-xFexMn1.5O4 leads to three different potential regions attributed to the transitions Mn4+/Mn3+ (4.3–3.9 V), Ni4+/Ni2+ (4.9–4.5 V) and Fe4+/Fe3+ (5.1–4.9 V vs. Li/Li+).8

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: High-Voltage Materials

309

Fig. 3 Cation ordering in stoichiometric LiNi0.5Mn1.5O4 (space group P4332)

M3+ =M4+ : ⋯LiM3+ x Mn2−x O4 Trivalent transition metal ions M3+ can replace Mn3+ ions to form Li[MxMn2-x]O4 with 0 < x  1. High-voltage materials are obtained for the substituents Cr3+, Fe3+ and Co3+. With Cr3+ and Co3+ solid solutions exist up to x ¼ 1, whereas the incorporation of Fe3+ is structurally limited to approximately x < 0.6. Above this limit Fe3+ ions exchange with part of the lithium ions on the 8a site to form (Li1-yFey)[LiyFex-yMn2-x]O4. Phase separation might occur for iron-rich phases.9 With increasing substitution of x in Li[MxMn2-x]O4 the lattice parameter a decreases for Co3+ and Cr3+ while it increases for Fe3+ ions. Substitutional disorder is described to be 2–3 % (x ¼ 0.5) for Co3+ and Fe3+ and 8 % (x ¼ 1.0) for Co3+ ions. Oxygen deficiency may occur. The electrochemical reaction proceeds between M3+ and M4+ at the high-voltage step and Mn3+ and Mn4+ for the 4.1 V step. Including possible oxygen deficiency the materials can be described by the formula Li+[M3+xMn3+1-x + 2dMn4+1-2d]O4-d. The length of the steps calculates to (1-x + 2d) electrons per formula unit for the Mn3+/Mn4+ redox couple at 4.1 V and to (x-2d) for the M3+/M4+ redox couple at the high-voltage step. The fully delithiated product has the formula [M3+2dM4+x-2dMn4+2-x]O4-d with remaining M3+ for d > 0. The potential of the high-voltage step increases in the order Cr3+/Cr4+ (4.8 V) < Fe3+/Fe4+ (4.9 V) < Co3+Co4+ (5.0–5.1 V vs. Li/Li+). The theoretical maximum high-voltage capacity at the substitutional limit of x ¼ 1 is nearly obtained for cobalt. The Cr3+ and Fe3+ analogs (x ¼ 1) do not reach the maximum high-voltage capacity. The Cr3+ analogue cannot be cycled reversibly. For x > 0.6 the irreversible formation of tetrahedrally coordinated Cr6+ during charging is described.10 The Fe3+ analogue with x ¼ 1 is a phase mixture. Generally a higher irreversible capacity can be observed for all the compounds compared to LiNi0.5Mn1.5O4 and is ascribed to electrolyte oxidation at high potentials. M2+ =M3+ : LiCu0:5 Mn1:5 O4 Cu2+ can replace Mn3+ ions to form Li[MxMn2-x]O4 with 0 < x  0.5. The cation distribution in these compounds is assumed to be quite complex with copper and lithium both on 8a and 16d positions. The Cu2+/Cu3+ redox couple is located at 4.9 V vs. Li/Li+. Compared to LiNi0.5Mn1.5O4 only a one-electron step can be utilized. The theoretical high-voltage capacity for x ¼ 0.5 is about 72 mAh g−1. Although the theoretical overall capacity is reached no exclusive high-voltage step is obtained. The major part of the discharge takes place at significantly lower potentials. This behavior is explained by the formation of Mn3+ ions due to pronounced oxygen deficiency.11

2.3

LiNiVO4—Inverse spinel

LiNiVO4 crystallizes in the space group Fd-3 m and belongs to the inverse spinels. The pentavalent vanadium is located on the tetrahedral 8a site, whereas lithium and nickel are randomly distributed on the octahedral 16d site (Fig. 4). A disordered nickel vanadium distribution may occur to form (NixV1-x)[LiNi1-xVx]O4. Among the synthesis techniques used for the manganese based high-voltage spinels a hydrothermal precipitation is possible as well for synthesis of LiNiVO4.12 The electrochemical redox reaction of nickel performs at 4.8 V vs. Li/Li+. However, it is unclear, which oxidation states are involved. The isostructural cobalt analogue does not belong to the high-voltage materials (4.2 V vs. Li/Li+). Lithium transport is assumed to proceed along the pathway (16d ! 8a ! 16d ! . . .). Specific capacities range up to 60 mAh g−1.

310

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: High-Voltage Materials

Fig. 4 Structural model of LiNiVO4 (inverse spinel structure).

Olivine − type Materials : ⋯LiMPO4 with M ¼ Ni, Co Olivine-type materials LiMPO4 (M ¼ Fe, Mn, Co, Ni) can be used as positive electrode materials. The redox step M2+/M3+ is located at 3.5 V (Fe), 4.1 V (Mn), 4.8 V (Co) and 5.1–5.3 V (Ni) vs. Li/Li+. Among the olivine-type materials LiCoPO4 and LiNiPO4 are high-voltage materials. They crystallize in the space group Pnma. The oxygen atoms form a slightly distorted, hexagonal close-packed arrangement with phosphorous ions on tetrahedral sites and transition metal and lithium ions on crystallographically different octahedral sites.13,14 The MO6-octahedra are linked through common corners in the bc-plane to form zigzag chains in two perpendicular directions that are cross-linked via octahedral corners. The LiO6-octahedra form edge-sharing chains along the b-axis. Each FeO6-octahedron has common edges with two LiO6-octahedra and a PO4-tetrahedron. PO4-groups share one edge with a FeO6-octahedron and two edges with LiO6-octahedra (Fig. 5). Electrochemically the redox-couple M2+/M3+ can be utilized according Eq. (2) with theoretical specific capacities between 166 mAh g−1 and 171 mAh g−1 for the one-electron step. LiM2+ PO4 ⇆M3+ PO4 + Li+ + e−

(2)

Lithium diffusion proceeds preferentially along the b-axis between two octahedral sites via the tetrahedral void sharing faces with both. In purely oxidic systems the potential of the M2+/M3+ redox couple is located at a low potential. In these systems a high-voltage level is only obtained if the oxidation state M4+ is involved. In LiMPO4 the potential of the M2+/M3+ redox-couple is shifted to higher values due to the presence of the phosphate group. Compared to oxidic systems the MdO bond is weakened in LiMPO4 by the strong inductive effect of the pentavalent phosphorous ion on the coordinated oxide ions. The resulting higher ionic character of the MdO bond is the reason for the potential shift. Compared to the high-voltage spinels electronic conductivity is poor and lithium diffusion is one-dimensional. To minimize the effect of the low conductivity it is necessary to introduce conductive coatings and to reduce the particle and crystallite size. LiMPO4 materials are synthesized by thermal treatment of precursors containing lithium, transition metals and phosphate in stoichiometric amounts. The mixtures can be obtained from co-precipitation, sol-gel synthesis or thermal decomposition of e.g., nitrates. Direct hydrothermal synthesis is possible. The thermal treatment between 600  C and 800  C can be performed in air or inert atmosphere. If a carbon source is added before the thermal treatment inert atmosphere is necessary. Carbon can be introduced in high dispersion by high-energy milling as well. Mixed phospho-olivines LiM11-xM2xPO4 form solid solutions for 0 < x < 1 and M ¼ Fe, Mn, Co and Ni. A single as well as a two-step mechanism is discussed for the electrochemical reaction of LiCoPO4. Specific capacities up to 125 mAh g−1 have been reported. In case of LiNiPO4 electrochemical activity could be proved only by cyclic voltammetry in 1 M LiPF6/sulfolane electrolyte. Up to now, there is no electrochemical proof using the full theoretical capacity of LiNiPO4.

Fig. 5 Structural model of LiMPO4 (viewing direction close to b).

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: High-Voltage Materials

3

311

Li-rich layered oxide materials

The so-called Li-rich layered oxides (LRLO, HE-NMC) with the general formula x Li2MnO3 ∙ (1-x) LiNi1-y-zMnyCozO2 (y + z  1) are one of the most promising future positive material classes for Li-ion batteries. They can reversibly insert more than one Li+ into the layered structure, which leads to exceptional high specific capacities (>250 mAh g−1). An additional advantage of these materials is the fact that they contain a higher amount of manganese, which is more abundant than cobalt or nickel.2 Also Co-free compositions between the endmembers LiNi0.5Mn0.5O2 and Li2MnO3 exist. The detailed structure of LRLO is still under discussion. In general, there are two main models to describe the structure: a solid solution or a nanocomposite of Li2MnO3 and LiMO2.15 The crystal lattice of monoclinic Li2MnO3 (C/2 m) is very similar to trigonal LiMO2. The Li2MnO3 lattice can be described as a NaFeO2 type structure with additional honeycomb ordering in the transition metal layer (Fig. 6). Lithium cations in the TM layer strongly influence the electrochemical characteristics of the LRLOs.16 Compared to the stoichiometric layered materials with only redox active transition metal cations, LRLOs exhibit a combination of both cationic and anionic redox reactions. In contrast to NCAs and NMCs, the voltage curves of LRLOs start with a sloping voltage profile (redox of transition metal cations), which turns into a plateau at a potential of about 4.5 V during the initial delithiation (Fig. 7). The plateau at 4.5 V is assigned to the activation of Li2MnO3 domains with contribution of an oxygen-anion redox process. The end of the plateau is accompanied by an irreversible oxygen release. So far, the anionic redox reaction and the mechanism of the activation during the first delithiation are not completely understood. Based on extensive experimental and theoretical studies a combination of cationic and anionic redox reactions is assumed. The reversibility of the anionic redox is strongly dependent on the stabilization of the reduction products of the oxide ions. Several groups reported that highly covalent TMdO bonds are able to stabilize peroxide O2− species in the structure and to limit their further reduction to molecular O2.17 In addition, alternative mechanisms are proposed including disproportionation of

Fig. 6 Overview over the structure and the electrochemistry of HE-NMC: (a) layered R m crystal structure of HE-NMC, (b) an exemplary configuration of the in-plane honeycomb-ordering of Li, Mn, Ni and Co in the TM layer reproduced from House, R. A.; Rees, G. J.; Pérez-Osorio, M. A.; Marie, J.-J.; Boivin, E.; Robertson, A. W.; Nag, A.; Garcia-Fernandez, M.; Zhou, K.-J.; Bruce, P. G. Nat. Energy 2020, 5, 777, https://doi.org/10.1038/s41560-020-00697-2.

x in LixNi0.13Co0.13Mn0.54O2 1.2

1.0

0.8

0.6

0.4

0.2

0.0

Voltage versus Li+/Li (V)

5 O oxidation

Ni and Co oxidation

Voltage hysteresis

4

3

Ni, Co and O reduction

2 0

50

100

150

200

250

Capacity (mAh g–1) Fig. 7 Initial charge-discharge cycle for Li1.2Ni0.13Co0.13Mn0.54O2.

300

350

312

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: High-Voltage Materials

nominally O− species into oxide (O2−) and molecular O2 under transition metal migration in the structure. O2 molecules in the surface-near regions can be released, which leads to electrolyte oxidation and a surface densification by transformation from layered via spinel-like to rock-salt structure. This mechanism seems to be the major reason for the high irreversible initial capacity loss. In general, similar synthesis routes used for the synthesis NCAs or NMCs can be used also for the synthesis of LRLO materials. Technically applied synthesis routes start with the coprecipitation of a precursor with defined particle morphology and porosity. The precursor is mixed with a stoichiometric amount of a lithium salt (e.g., LiOH or Li2CO3) and heat treated in a furnace at high temperatures under air. The particle surface of LRLO materials influences significantly the electrochemical performance. Various strategies for surface modifications via post treatment of freshly synthesized LRLOs are reported in the literature including the formation of spinel-like surface layers by washing processes or the application of surface coatings. Typically used post treatment routes are a combination of a chemical treatment, forming a Li+-deficient structure in the surface-near region, and a subsequent heat-treatment step that induces a structural reorganization.

4 4.1

Outlook Development challenges

One of the major challenges of high-voltage materials and Li-rich layered materials is the reactivity of the electrolyte at potentials >4.5 V vs. Li/Li+. Most of high voltage materials suffer from low cycling stability especially at higher temperatures. The most promising high voltage material in terms of cost, raw materials availability and electrochemical performance is the spinel LNMO. Therefore, most of the development efforts for high voltage materials focus on this material. Extensive studies are ongoing to achieve a more deep understanding of chemical and electrochemical processes at the electrode/electrolyte interface of high voltage positive materials, in order to increase the cycling stability and calendar life especially at higher temperatures. More stable electrolyte formulations and more oxidation stable conducting additives have to be developed. Ionic liquids show a high stability at high potentials. However, their stability at low potentials is restricted. Instead of graphite, different negative electrode materials such as lithium titanates or recently lithium niobates are used in combination with LNMO. Another strategy to improve the long-term stability of LNMO is the modification of the particle surface of the positive electrode materials by coating with metal oxides or phosphates. Furthermore, coatings can work as HF scavenger and reduce the decomposition of conducting salts like LiPF6 combined with the dissolution of transition metal ions into the electrolyte that can interact with the negative electrode. For commercial use, several issues, limiting the cycle life of a LMNO containing cell, still have to be solved mainly related to the electrolyte instability of this high operational voltage material, which results in passivating surface films, manganese dissolution, chemical cross-talk and poisoning of the negative electrode. Li-rich layered oxides are very promising positive materials for Li-ion batteries. Based on the manganese dominant composition the amount of critical and expensive transition metals such as Co or nickel can be minimized. In addition, these materials exhibit exceptional high specific discharge capacities. The anionic redox mechanism of these materials can lead to the release of O2 molecules from the surface-near regions, which leads to electrolyte oxidation and a surface densification by transformation from layered via spinel-like to rock-salt structure. This mechanism seems to be the major reason for the high irreversible initial capacity loss. In addition, LRLOs exhibit voltage fading and a large voltage hysteresis between charge and discharge. Both effects are associated with the anionic redox reaction. Strategies to improve the electrochemical performance and the cycling stability include modification of the electrolyte, doping of the bulk to stabilize the host structure and to reduce transition metal migration at high SOCs, and various kinds of surface modification by post treatment or surface coating.

4.2

Competitive advantages

The use of high-voltage positive electrode materials against the standard negative electrode material graphite provides an increased overall cell voltage. The energy density of the cell is enhanced proportionally with the cell voltage. The higher charge potential enables the activation of Li-rich layered oxides resulting in exceptional high reversible capacities. In addition, the estimated immense growth of the lithium ion battery market will increase the demand of raw materials like Li, Mn, Ni and Co drastically. Therefore it is very important to find Co-free alternative positive materials and compositions based on earth-abundant materials to reduce both the costs and the environmental footprint of the LIB. Furthermore the use of high-voltage positive electrode materials allows replacing graphite by alternative negative electrode materials. Negative electrode materials with higher redox potentials compared to graphite can be combined with high-voltage positive electrode materials to obtain an overall cell voltage close to that of conventional lithium-ion batteries. Several improvements can be expected. Graphite works beyond the lower stability limit of the electrolyte. Operation is possible due to the formation of a passivating film (SEI). Safety problems may arise from lithium plating on graphite or from a breakdown of the SEI layer. Alternative negative electrode materials, such as lithium titanates or lithium niobates, work within the stability window of the electrolyte. No SEI formation is necessary and superior safety can be expected.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: High-Voltage Materials

313

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

Kawai, H.; Nagata, M.; Tukamoto, H.; West, A. R. J. Power Sources 1999, 81-82, 67–72. Qiu, B.; Zhang, M.; Wu, L.; Wang, J.; Xia, Y.; Qian, D.; Liu, H.; Hy, S.; Chen, Y.; An, K.; Zhu, Y.; Liu, Z.; Meng, Y. S. Nat. Commun. 2016, 7, 12108. Zuo, W.; Luo, M.; Liu, X.; Wu, J.; Liu, H.; Li, J.; Winter, M.; Fu, R.; Yang, W.; Yang, Y. Energ. Environ. Sci. 2020, 13, 4450. Amdouni, N.; Zaghib, K.; Gendron, F.; Mauger, A.; Julien, C. M. Ionics 2006, 12, 117–126. Kawai, H.; Nagata, M.; Tukamoto, H.; West, A. R. Electrochem. Solid St. 1998, 1, 212. Kunduraci, M.; Al-Sharab, J. F.; Amatucci, G. G. Chem. Mater. 2006, 18, 3585–3592. Kim, J. H.; Myung, S. T.; Yoon, C. S.; Kang, S. G.; Sun, Y. K. Chem. Mater. 2004, 16, 906–914. Amine, K.; Tukamoto, H.; Yasuda, H.; Fujita, Y. J. Power Sources 1997, 68, 604–608. Ohzuku, T.; Ariyoshi, K.; Takeda, S.; Sakai, Y. Electrochim. Acta 2001, 46, 2327–2336. Sigala, C.; Guyomard, D.; Verbaere, A.; Piffard, Y.; Tournoux, M. Solid State Ion. 1995, 81, 167–170. Ein-Eli, Y., Jr.; Howard, W. F.; Lu, S. H.; et al. J. Electrochem. Soc. 1998, 145, 1238–1244. Fey, G. T.-K.; Li, W.; Dahn, J. R. J. Electrochem. Soc. 1994, 141, 2279–2282. Wolfenstine, J.; Read, J.; Allen, J. L. J. Power Sources 2007, 163, 1070–1073. Wolfenstine, J.; Allen, J. J. Power Sources 2005, 142, 389–390. Lu, Z.; Chen, Z.; Dahn, J. R. Chem. Mater. 2003, 15, 3214. House, R. A.; Rees, G. J.; Pérez-Osorio, M. A.; Marie, J.-J.; Boivin, E.; Robertson, A. W.; Nag, A.; Garcia-Fernandez, M.; Zhou, K.-J.; Bruce, P. G. Nat. Energy 2020, 5 (777), 2. Schipper, F.; Nayak, P. K.; Erickson, E. M.; Amalraj, F.; Srur-Lavi, O.; Penki, T. R.; Talianker, M.; Grinblat, J.; Sclar, H.; Breuer, O.; Julien, C. M.; Munichandraiah, N.; Kovacheva, D.; Dixit, M.; Major, D. T.; Markovsky, B.; Aurbach, D. Inorganics 2017, 5, 32.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials Albert W Xiao, Research Division, Advanced Materials and Processing Laboratory, Nissan Motor Company, Ltd., Yokosuka, Kanagawa, Japan © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 3 4 5 6 7 References

Introduction The conversion mechanism Practically relevant conversion chemistries A note about conversion anodes Toward commercial utilization of conversion-type fluoride cathodes Future research directions for conversion-type materials Conclusions

315 316 320 321 322 325 325 326

Abstract The push for higher energy-density lithium-ion batteries has placed a premium on cathode materials that exceed the lithium storage capacity of conventional intercalation type cathode materials (140–200 mAh/g). Among the myriad of next-generation cathode candidates, conversion-type cathodes have attracted great interest due to their relatively high redox potentials, high crystal densities and gravimetric capacities often in excess of 400 mAh/g. Conversion cathodes are ionocovalent transition metal compounds that, upon reaction with lithium, are “converted” into a finely divided mixture of metallic particles embedded in a lithium salt. While conversion compounds comprise nearly all transition metal nitrides, oxides, fluorides, phosphides, sulfides, and chlorides, not all conversion compounds are equally viable as lithium-ion battery cathodes. This chapter will provide a brief scientific history of conversion cathodes and highlight the differences in chemistry and utility between the different classes of conversion compounds. The initial focus of this chapter will be on the mechanistic understanding of the conversion reaction. At face value, conversion reactions appear to be highly disordered, but increasingly advanced characterization and sample preparation techniques have revealed a surprising level of order that often dictates the reversibility of the reaction. In some cases, conversion reactions have even been confirmed to be entirely topotactic. The latter half of this chapter will focus on the specific materials systems that are most promising in terms of energy density, cost, and performance. Promising and pragmatic methods to enable their utilization at the commercial scale will be discussed.

Glossary Compositional inhomogeneity Where a single material contains regions with varying composition. Conversion Lithium storage by reaction to form new materials. Fast Fourier transform (FFT) An image analysis technique that can be used to derive structural information from a high-resolution transmission electron micrograph. In situ TEM An analysis technique where (electro)chemical lithiation is performed and observed within the column of a transmission electron microscope. Intercalation Lithium storage at interstitial lattice sites. Monodisperse When each particle in an ensemble exhibits identical or nearly identical morphological traits. Nanorod A rod-shaped crystallite with nanoscale dimensions. Phase asymmetry Where a different progression of phases is observed between charge and discharge processes. Reconversion The charging half cycle of the conversion reaction. Stack-level Pertaining to a full-layered assembly of the active components of a battery, including current collectors, electrode composites, separators, and electrolytes. Topotactic Where a distinct crystallographic orientation relationship exists between reactant and products of a transformation. Zone axis A high-symmetry orientation of the crystal lattice typically denoted as being parallel to a particular crystallographic direction.

Key points

• • •

314

This chapter aligns recent discoveries in the lithiation mechanisms of different conversion cathode materials and details a paradigm shift in the understanding of the conversion reaction. Transition metal fluorides are defined as the most practically relevant conversion chemistries in terms of energy density and durability. Limitations of transition metal fluorides and strategies used to improve their performance are summarized and discussed.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00134-1

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials

1

315

Introduction

It may already be evident to those more fervent readers of this encyclopedia that the importance of lithium-ion batteries (LIBs) is difficult to overstate. LIBs are already ubiquitous in portable electronic devices, and they have underpinned the full-tilt electrification of the automobile industry that is currently underway. Nevertheless, higher energy densities are still desired for longer-range electric vehicles and required to enable the development of new technologies, such as electric passenger aircraft.1 Unfortunately, the paradigm of charge storage by lithium intercalation is fast approaching the physico-chemical limit of its energy density.2,3 On the anode side, the push beyond intercalation-based chemistries has been embodied by the near commercialization of silicon or semi-silicon anodes as well as the race to enable lithium metal anodes (by solid state techniques or otherwise). However, in order to secure the most significant increases in energy density, higher capacity cathode materials must also be developed. The capacity of conventional intercalation-type cathodes used in LIBs is fundamentally limited, as most chemistries can reversibly accommodate only a fraction of a lithium per formula unit without damage to the host structure.3 Conversion type electrodes have garnered significant interest for their ability to reversibly store multiple lithium per formula unit, yielding nearly three to six times the capacity of intercalation type electrodes.5 While such basic values are factual (at a theoretical level), they are often extrapolated into wildly overpromising blanket claims about overall energy density. Readers who are interested in realistic calculations of energy density should examine the reviews by Wang et al. and Olbrich et al.6,7 Based on very conservative calculations, conversion type cathodes paired with a lithium metal anode could realistically offer 2–3 times the energy density (500–700 Wh/kg) of state-of-the-art intercalation-based LIBs at the stack level.7 However, such values are only achievable through a handful of materials systems and say nothing about the rate capability, cycle life, or energy efficiency of any such hypothetical battery. With the aim of advancing the field of conversion type cathodes, this chapter will pragmatically discuss the capabilities and limitations of the conversion reaction, which materials are practically capable of producing high energy densities, and how improved performance may be achieved. At the simplest level, the conversion reaction refers to the complete reduction of a transition metal compound with lithium, during which the transition metal compound is “converted” into a finely divided mixture of metallic particles and a lithium salt (Fig. 1).8 This can be described generally by the equation below, where M is a transition metal and X can be any of N, O, F, P, S or Cl.5,6,9 c 0 Mcb=a a Xb + cbLi $ aM + bLic X

The conversion reaction is extremely widespread and has, in fact, been known far before the discovery of reversible lithium intercalation.6 Many of the earliest experiments on primary lithium cells utilized a lithium metal anode and a transition metal sulfide cathode.10 Even electrodes that are decidedly categorized as intercalation-type cathodes (LiCoO2, Li2Mn2O4, LiFePO4) will readily undergo a conversion reaction if lithiated beyond their intercalation capacity.11,12 While conversion reactions are not (A)

(B)

Charged

Discharged

Fig. 1 (A) A series of in-situ TEM micrographs depicting the lithiation of an iron fluoride nanoparticle and the conversion of FeF2 to a finely divided mixture of LiF and metallic iron. Produced using data from reference4 with permission of the author. (B) A cartoon illustrating the general transformation between charge and discharge during a typical conversion reaction. Lithium insertion in transition metal oxides, fluorides, and sulfides results in the formation of metallic nanoparticles imbedded in a matrix of Li2O, LiF and Li2S respectively.

316

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials

difficult to find, the renewed interest in their application stems from the fact that several high-capacity conversion type materials exhibit remarkable reversibility.13–15 Both intercalation and conversion reactions are often present in varying degrees in the electrochemistry of a single material (whether lithiated or non-lithiated). However, the two processes have long been considered as distinct and independent phenomenon that occupy separate domains, both at the atomistic level as well as in the battery research community. Only very recently have several astonishing similarities been pointed out between intercalation reactions and the reversible conversion reactions in a handful of transition metal sulfides, oxides, and fluorides.4,13,14,16–18 It is thus timely, for the latest edition of this encyclopedia, to examine how these studies form a basis for a unifying theory of reversible conversion reactions across different classes of materials and highlight how these findings usher in a new paradigm of thinking that blurs the lines between conversion and intercalation reactions. It is perhaps non-sensical to try and pinpoint the inception of conversion electrode technologies, as they pre-date the lithium-ion battery itself and have been known since the beginnings of primary lithium battery research, long before the term “conversion reaction” was coined.19,20 While we often think of intercalation reactions as the quintessential mechanism of lithium batteries, conversion reactions are, in-fact, far more widespread. A number of transition metal oxides and sulfides were considered as primary lithium battery cathodes in 1970s and 80s, most notably FeS2, Fe2O3, and CuS.21–24 CuF2 was studied as a cathode material as early as the 1963.25 While several of these compounds exhibited some reversibility at high temperatures, these materials were generally not considered for application in secondary batteries.26 The commercialization of intercalation-based lithium-ion batteries in 1991 led to a renewed interest in conversion electrodes due to their much higher theoretical capacities, and research on conversion type cathodes was revived in earnest after a high degree of reversibility was discovered in a number of systems. Jean-Marie Tarascon’s group demonstrated through several reports in the 2000s, that conversion reactions in select metal oxides were highly reversible.27–29 Around the same time, the groups led by Amatucci and Maier pioneered the development of reversible transition metal fluoride cathodes.30–33 Transition metal sulfides, and iron sulfides in particular, had been the subject of continuous research as cathodes in high temperature molten salt batteries since the 1980s, and by the turn of the century, several research groups began applying them as reversible cathodes for room temperature LIBs.34–36 Initially, these conversion-type cathode systems were poorly understood, as traditional materials characterization methods (i.e. XRD, macroscale X-ray and magnetic spectroscopy) were insufficient to understand the nanoscale structural evolution involved in the conversion reaction. With only low spatial resolution characterization methods, it was simply concluded that conversion reactions happen via a disordered process that destroys the initial structure of the material.31,32,37,38 In terms of battery performance, the initial conversion cathodes performed very poorly, requiring large amounts of conductive carbon to function, even at vanishingly small current densities.31,32 Over the last two decades, the field of conversion cathodes has evolved with concurrent advancements in materials characterization, battery electrode engineering, and electrolyte science. Developments in nanostructuring of conversion-type active materials enabled access to near theoretical capacities at higher current densities.8,15,39 Improved understanding of cathode-electrolyte interphase (CEI) formation and the tailoring of electrolyte chemistry has led to prolonged cycle life at and above room temperature.40–42 The development of high-resolution analytical transmission electron microscopy characterization techniques has uncovered many of the primary mysteries of the conversion reaction.4,13,43 Finally, in 2023, we are flirting with the development of lithium conversion cathodes with commercially relevant energy density, cycle life, and rate capability.7 In the sections below, we will review the latest understanding of the conversion mechanism and the demonstrated performance capabilities and limitations of conversion cathode materials to inform their use in potential applications.

2

The conversion mechanism

As mentioned above, the conversion reaction describes the reaction of a transition metal salt (MOx, MFx, MSx, etc.) with lithium to form a finely divided mixture of metallic nanoparticles (2.3 V vs. Li+/Li should be considered as cathode materials, essentially limiting the field to a handful of transition metal fluorides.7 One might be tempted to think that compounds with lower potentials could be used as anodes, as has been suggested for a number of transition metal sulfides and even oxides (Cu2S, Co3O4, NiO, etc.); however, the lithiation potentials of these compounds is not truly low but rather significantly depressed by a massive overpotential, meaning the use of these materials as anodes would provide a vanishingly small EMF in a full cell.50,55,56 Accordingly, transition metal conversion compounds are not sensible anodes for lithium-ion batteries. Based on the analysis above, it is already clear that among the simple transition metal compounds, only transition metal fluorides exhibit electrode potentials high enough to be viable lithium-ion battery cathodes. The selection of conversion cathodes for reasonable energy density would be even further limited to fluorides of the later 3d transition metals (i.e. Fe, Co, Ni, Cu).6,7 Aside from energy density considerations, there are also significant practical limitations that preclude the use of other transition metal compounds. Transition metal chlorides are incompatible with conventional liquid electrolytes due to their high solubility in (A)

(B) 1000

Equilibrium Discharge Potential (V vs. Li+/Li)

3.5

Fluoride

3 2.5

Oxide

2 1.5

Sulfide

1 0.5 0 -0.5

Sc3+ Ti2+ Ti3+

V2+

V3+ Cr2+ Cr3+ Mn2+ Fe2+ Co2+ Ni2+ Cu2+ Zn2+

Stack-level Specific Energy (Wh/kg)

4

4.0 V vs. Li+/Li

900

3.5 V

3.0 V 2.5 V

800

2.0 V

700 600 NMC811-lithium 500

1.5 V

1.0 V

400 300

NMC622-graphite

200 100 0 100

300 500 700 Cathode Active Material Capacity (mAh/g)

900

Fig. 3 (A) A plot of the equilibrium discharge potentials for the full conversion with lithium of the lowest oxidation state fluorides, oxides, and sulfides of the 3d transition metals. Potentials are calculated from tabulated values of experimentally measured standard Gibbs free energies of formation. The equilibrium discharge potential for compounds of a certain transition metal generally increase with the electronegativity of the anion, with the fluorides exhibiting significantly higher discharge potentials. (B) A plot of the stack-level specific energy density as a function of capacity for a hypothetical conversion cathode with different average discharge potentials. Specific energy values were calculated using the method reported by Olbrich et al. with a cathode active material mass fraction of 85 wt%.7 The density of the cathode active material was taken to be the same as FeF2. Dashed lines represent the specific energy values of an NMC622-graphite system and a hypothetical NMC811-lithium system for comparison.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials

321

polar solvents.57 In transition metal sulfides, the conversion reaction competes with anionic sulfur redox, which not only destroys the crystal structure but also results in polysulfide dissolution.18,51,57 While transition metal oxides exhibit more favorable materials properties, they tend to suffer from higher overpotentials than transition metal fluorides; as such, their discharge potentials are depressed well below 2 V and in many cases even below 1 V vs. Li+/Li.55 As part of a simple techno-economic analysis, Olbrich et al. modeled the stack level energy density and materials cost of the later 3d transition metal fluorides (FeF2, FeF3, CoF2, NiF2, CuF2) with realistic penalties placed on usable capacity, overpotential, excess lithium, excess electrolyte, and active material mass fraction.7 Even by these more conservative estimates, the authors suggested that all the chemistries in question could be competitive in terms of specific energy, even when compared to a hypothetical NMC 811-lithium cell with a liquid electrolyte. FeF2 and CoF2 exhibited the smallest advantage in terms of specific energy; while FeF2 could be competitive in terms of cost, the cost of CoF2 was prohibitively high. The authors noted that NiF2, CuF2, and FeF3 could all exceed 600 Wh/kg at the stack level, with both CuF2 and FeF3 potentially reaching costs below 40 USD/kWh.7 While this analysis highlights the appeal of these two chemistries, they both have significant limitations. FeF3 requires a 3-electron transfer to achieve its theoretical capacity; this results in greater volume change and a larger proportion of LiF in the discharged state which in turn result in greater irreversible capacity loss and more sluggish kinetics compared to FeF2. CuF2 is perhaps the most attractive conversion material in terms of energy density and cost; however, the conversion reaction in native CuF2 is fundamentally irreversible.58–60 In FeF2 and other transition metal difluorides, very small metallic nanoparticles nucleate at interfaces within the primary particle on discharge. In a landmark 2011 study, Wang et al. revealed that in the case of CuF2, the nucleation of very large metallic Cu particles is observed primarily on the surface of the primary particles.45 The authors hypothesized that this discrepancy was a result of the higher diffusivity of Cu2+ cations compared to Fe2+ cations, which allowed the Cu cations to diffuse further to particle surface before being reduced. However, in the decade since, it has been revealed that the transition metal cations are likely reduced prior to long range diffusion and that the conversion reaction relies almost entirely on the diffusion of transition metal atoms with the F- being essentially immobile.4,43 Furthermore, the reversibility of CoF2 and NiF2 – with more diffusive cations than FeF2 – has been well documented.41,61 An alternative explanation considers that the nucleation of small metallic particles within the fluoride lattice requires lattice matching at low-energy interfaces metal/fluoride interfaces. In this regard, the irreversibility of CuF2 may be a result of its unique Jahn-Teller distorted crystal structure, which provides few low energy facets for the nucleation of initial copper particles. Despite the irreversibility of native CuF2, several strategies have been attempted to enable reversible cycling in modified copper fluorides. Recently Seo et al. suggested that capacity loss in CuF2 was partially a result of the dissolution of metallic Cu from the surface of the converted nanoparticle on charging.60 Indeed, the authors showed that by applying a NiO coating that a very limited reversible capacity (150 mAh/g) could be achieved for several cycles. Unfortunately, it appears that large overpotential induced by the coarser structure of the discharged CuF2 inhibits full reconversion of the CuF2. To date, the most successful report of CuF2 redox is in a solid solution compound, Fe0.5Cu0.5F2, which not only demonstrated a reversible capacity of 500 mAh/g, but also exhibited a cycling hysteresis as low as 148 mV. Wang et al. suggested that the improved performance was a result of a synergistic coupling of the Fe and Cu redox couples.62 Interestingly, these results have not been reproduced by others. A similar composition produced by Gordon et al. using chemical synthesis methods showed little improvements in hysteresis.61 In a recent report revisiting the results of Wang et al., Omenya et al. suggested that in the Fe0.5Cu0.5F2 system, the Cu2+/Cu0 redox capacity was quickly lost on cycling and that this capacity loss was partially compensated by the appearance of a reversible Fe3+/Fe2+ redox couple.59 While the methodology followed by Wang et al. and Omenya et al. are nearly identical, the reversibility observed in the charge/discharge profiles differed significantly enough that the two authors came to different conclusions. Ultimately, neither report investigated how the presence of FeF2 affected the morphology of the discharged CuF2, and it is likely that reversibility of the FexCu1-xF2 systems is influenced significantly by electrolyte particle morphology/crystallinity, electrode construction, and electrolyte choice. While reversible cycling of CuF2 remains unsolved, reversible copper conversion reactions have been observed. A copper conversion reaction at a relatively high potential was reported in the Cu3(PO4)2 system.63 Like CuF2, Cu3(PO4)2 exhibits a high discharge potential resulting from the ionicity of the Cu-OPO3 bond. Furthermore, due to a high Cu content, the theoretical capacity of Cu3(PO4)2 is comparable to transition metal difluorides at 422 mAh/g. Most importantly, Zhong et al. showed that Cu3(PO4)2 could be easily discharged and charged to its theoretical capacity.63 While the capacity fade in this report was rapid, it suggests that polyanionic conversion compounds may represent a viable class of alternative conversion cathodes.

4

A note about conversion anodes

As this chapter primarily focuses on cathode materials, an extended discussion on true conversion anode chemistries will not be included. However, we will briefly describe the general design of potential conversion anode materials and highlight some examples of practical conversion anodes from the literature. As mentioned above, not all conversion chemistries labeled as “anodes” should rightfully be considered as such. Many transition metal oxides and sulfides (CoO, NiO, FeS, etc.) exhibit a low lithiation potential in half-cell tests vs. a lithium metal counter electrode. In many cases, authors use this data to justify labeling these compounds as “anodes.” However, as can be seen from Fig. 3A, the equilibrium potential of these compounds is actually relatively high, and the low observed potential is simply a result of the large overpotential required to drive lithiation. For anode materials, it is in-fact the

322

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials

delithiation potential—not the lithiation potential—that contributes to the cell voltage. As the large overpotentials associated with conversion electrodes further elevates the de-lithiation potential, the equilibrium potential of a conversion anode material should be much less than 1 V vs. Li+/Li in order to generate a meaningful electromotive force. In general, conversion materials that fit the above description would be compounds of more electropositive metals in their lower oxidation states. These may include the alkaline earth metals, rare-earth metals, some early transition metals (e.g., Sc, Ti, V, Y, Zr), and certain main group metals (e.g., Al, Sn). Not all of these elements are viable when considering other important factors such as materials cost. While the anode potential is primarily determined by the reduction potential of these metals, it is also heavily influenced by the anion species. Electronegative anions such as O2− or F− will shift the anode potential upwards; as such, more electropositive anions such as S2−, P3−, and H− would be more favorable in terms of energy density. Selecting anions with lower electronegativity also means that anion redox can occur in a similar potential window as the cation redox. This can lead to much higher capacities than are possible with conversion reactions alone but also typically introduce challenges in terms of reversibility. Likewise, many electropositive metals can undergo alloying reactions with lithium at low potentials further complication reaction mechanisms. In the case of Sn4P3, for example, it has been shown that the conversion of Sn4P3 to Sn and LiP3 can be accompanied by further lithiation of Sn to form Li4Sn and that oxidation of Li3P to LiP occurs on charging.64 Among the compounds that are rightfully considered conversion anodes, metal hydrides are perhaps the most interesting. The extremely low atomic mass of the hydride anion results in high theoretical gravimetric capacities, and the low work function of such compounds results in very low anode potentials. For example, MgH2 has a theoretical capacity of 2038 mAh/g and an anode potential of approximately 0.5 V vs. Li+/Li. Several studies have demonstrated the reversible cycling of MgH2 to form Mg and LiF with reversible capacities above 700 mAh/g.65 El Kharbachi et al. even demonstrated a lower than 100 mV hysteresis in this material.66 Unfortunately, metal hydrides often react with O2 and water, making them difficult to handle. Additionally the evolution of H2 at elevated temperatures and potentials poses additional safety concerns.

5

Toward commercial utilization of conversion-type fluoride cathodes

The previous section examined why transition metal fluorides are the most promising and perhaps the only truly viable conversion type cathode materials. After nearly two decades of research both the mechanistic understanding as well as the achievable performance of transition metal fluorides has reached a point where targeted development could truly enable their use in select battery applications. This section will detail the major developments that have brought about the current level of performance, highlight potential applications for the effective use of transition metal fluoride cathodes, and outline the further research activities required to bring such technologies to maturity. As with mechanistic studies, reports investigating the performance of transition metal fluorides have primarily focused on the simple binary iron fluorides, with reports on FeF2 comprising the vast majority. Transition metal fluorides are particularly challenging materials to utilize due to three major issues: low ionic/electronic conductivity, poor reversibility, and large overpotentials. These obstacles are related not only to the materials properties of transition metal fluorides, but are also significantly influenced by the conversion mechanism outlined in section three above. The low electronic conductivity results from the large band gap originating with ionic nature of the metal-fluoride bond. The low ionic conductivity is related to the relatively small 1D channels for lithium diffusion. However, it is now known that the rate capability is significantly limited by fact that Li insertion/ de-insertion requires cooperative Fe-diffusion.4,43 On charging, the ionic conductivity is even further limited as it requires the migration of Fe/LiF interfaces and the slow diffusion of Fe2+ cations.43 Poor reversibility was previously attributed to volume change and a loss of active material structure; however, as demonstrated by Xiao et al., the morphological and crystallographic transformations of the conversion reaction are inherently reversible, with the original active material structure being largely restored after each cycle.4 As such, the reversibility of the conversion reaction should be primarily influenced by extrinsic factors (electrolyte, cathode construction, etc.). The earliest reports from the groups of Ammatucci and Maier showcased the difficulty of cycling transition metal fluorides.30–32 Badway et al. first prepared FeF3 electrodes using a simple high-energy ball milling technique. Although the full capacity of FeF3 was demonstrated, it required nearly 50 wt% carbon, a cycling temperature of 70  C and an extremely slow C/100 cycling rate. Under these conditions, nearly 30% capacity fade was observed over the first 14 cycles.31 To overcome the poor performance of ball milled samples, later studies focused on the direct chemical synthesis of FeF3 nanoparticles using solvothermal methods. By achieving sufficiently small particle sizes, these techniques enabled room temperature cycling at the theoretical capacity and slightly higher current densities (e.g. C/20); however, they did little to mitigate capacity fade (60% loss after 50 cycles).67,68 Li et al. attributed this capacity fade to an incomplete reconversion reaction, while other authors suggested that it was the result of cathode dissolution.67 Recently, Xiao et al. demonstrated that these two failure mechanisms are actually one and the same.4 The authors revealed that, on discharge, a layer of metallic iron segregates to the surface of the FeF2 particle and is effectively separated from the core of the particle by a second layer of LiF, with which it shares a semi-coherent interface. On charging, the reconversion of this surface metal is inhibited and it is readily dissolved in typical organic carbonate based electrolytes. After many cycles, it was shown that progressive leaching of the Fe from the particle surface results in insufficient Fe to fully reform FeF2 on charging.4 The recent review by Olbrich et al. details the three major failure mechanisms in FeF2 and similar systems (Fig. 4A).7 In addition to cathode dissolution, transition metal fluorides are prone to excessive electrolyte decomposition on discharge, as the surface metallic layer serves as a catalyst. The third failure mechanism was also revealed by Xiao et al. who observed rapid particle coarsening after one

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials

323

Fig. 4 An illustration depicting the three major failure mechanisms in transition metal fluoride conversion cathodes (transition metal dissolution, particle fusing, and electrolyte decompositions) and the methods that have been employed in the literature to mitigate these failure mechanisms. Reproduced with permission from reference Olbrich, L.F.; Xiao, A.W.; Pasta, M. Conversion-Type Fluoride Cathodes: Current State of the Art. Curr. Opin. Electrochem. 2021, 30, 100779. https://doi.org/10.1016/j.coelec.2021.100779. (Copyright Elsevier).

cycle in a conventional electrolyte and further showed, using in-situ TEM lithiation, how lithium diffusion between adjacent FeF2 particles causes them to fuse into a single larger particle.4 In the last decade, many reports of “high-performance” transition metal fluoride cathodes have surfaced.15,39–41,69,70 At face value, many of these reports show stable cycling at near theoretical capacities, with minimal capacity fade for hundreds of cycles. Many of these reports even appear to achieve extremely high rate capabilities, achieving 50–70% of the theoretical capacity at rates above 1C.39,40,70 These studies generally rely on the same principle: heterogeneous nucleation of transition metal fluoride nanoparticles on or within nanostructured carbon scaffolds (Fig. 4B). This trope has been applied to nearly every next-generation non-intercalation LIB electrode from silicon anodes to metal oxides and sulfur cathodes with blanket promises of mitigating any associated failure mechanisms.71 While these nanocomposite materials often perform impressively based on normalized capacity, cycle life, and rate capability; they fall short on major benchmarks of commercial viability such as active material mass fraction, areal capacity, production line compatibility and cost. In other words, the amount of carbon introduced in these nanocomposites is so high (40–60 wt%) that they dilute the energy density of the cathode material while increasing materials cost and manufacturing complexity. One of the most recent, and perhaps the most extreme example, of this strategy was presented by Su et al. who demonstrated stable cycling at close to 500 mAh/g for 1900 cycles (0.5C) as well as extremely high rate cycling at 30C and 60C with a stable capacity of around 150 and 110 mAh/g respectively after 500 cycles.72 In this case however, the active material mass fraction in the electrode was only 50 wt%, resulting in an areal capacity of 0.43 mAh/cm2. The authors claimed that this impressive performance was a result of the encapsulation of FeF2 in a polymer-derived carbon coating and an in-situ oxidation of the metallic surface Fe to Fe3O4.72 The primary unanswered question surrounding such “strongly-coupled” carbon composites is how much, if any, of the performance characteristics are maintained upon increasing the active material mass fraction and areal capacity to practical levels (e.g. >85 wt% and >3 mAh/cm2). The most important contribution of such works is that they demonstrated that capacity fade in transition metal fluorides could be mitigated to a large extent. In the case of Su et al. as well as many others, the mesoporous carbon scaffolds serve as an anchor, preventing the exfoliation of particles and separating adjacent particles to prevent them from fusing. The carbon coating layer further physically passivates the particle surface, mitigating electrolyte decomposition and electrode dissolution. Other reports have demonstrated that the formation of a stable solid electrolyte interphase was also able to serve as an effective passivation layer, preventing cathode dissolution and excess electrolyte decomposition; however, typically anchoring on carbon nanotubes or similar scaffolds was still required to prevent particle fusing and exfoliation.40,41

324

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials

Generally speaking, electrolyte engineering is a more favorable means of achieving stable cycling as it avoids using excessive amounts of carbon and requires no change to the battery manufacture process. As first demonstrated by Maier and Amatucci, transition metal fluoride cathodes exhibit rapid capacity fade in the carbonate-based electrolytes used in conventional lithium-ion batteries.31,32 Wang et al. were the first to demonstrate the systematic effect of varying electrolyte composition on the performance of transition metal fluoride cathodes. In the context of a CoF2-CNT composite cathode, they tested a series of electrolytes based on LiPF6 dissolved in mixtures of ethylmethyl carbonate (EMC) and fluoroethylene carbonate (FEC) and demonstrated that cycling stability increased with increasing FEC content. Upon decomposition, FEC is speculated to form a more robust solid electrolyte interphase that prevents the dissolution of the transition metal on charging.41 More recently, Huang et al. performed an extensive and systematic comparison of electroltyes for FeF2 cathodes. The authors prepared electrolytes of three different salts at three different concentrations, each in one of two different solvent systems.40 The electrolyte that produced the most stable cycling behavior consisted of a highly concentrated (3M) lithium bis(fluorosulfonyl)imide (FSI) salt dissolved in a dimethoxyethane (DME). Xiao et al. utilized an even higher FSI concentration in the form of a LiFSI in N-methyl-N-propylpyrrolidinium bis(fluorosulfonyl)imide (Pyr1,3FSI) ionic liquid electrolyte.4 The IL electrolyte resulted in nearly constant charge transfer resistance over the first 5 cycles, while a cell with a conventional carbonate electrolyte exhibited a rapid impedance rise. These results indicated that after the first cycle, the ionic liquid electrolyte produced a stable SEI layer that was that prevented further electrolyte decomposition. This result was corroborated by ex-situ TEM, showcasing that a dense 10 nm thick SEI layer formed after the first cycle and remained essentially unchanged after many cycles. XPS of the lithium counter electrode confirmed that the SEI layer prevented the dissolution of Fe from the cathode. As mentioned previously, in a conventional carbonate electrolyte, fusing of the FeF2 particles was observed on cycling; however, in the ionic liquid electrolyte, no fusing was observed. In this case, the electrolyte alone was able to prevent the three major failure mechanisms associated with FeF2, allowing the authors to employ electrodes with 70 wt% FeF2, prepared by slurry casting. In their mechanistic study on FeF2, Xiao et al. suggested that high discharge rates could be facilitated by the fast diffusion of Fe0 and the facile nucleation of metallic iron on low energy interfaces, while charging rates would inevitably be limited by the immobile nature of the same interfaces and the slow diffusion of Fe2+. In the same study, the authors experimentally demonstrated

Fig. 5 (A) Nyquist electrochemical impedance plots demonstrating the difference in magnitude and evolution of the charge transfer resistance between an unstable electrolyte (1 M LiPF6 in EC/DMC) and a stable electrolyte (1 M LiFSI in Pyr1,3FSI). (B) A plot of coulombic efficiency and discharge capacity versus cycle number for FeF2 electrodes showing stable cycling at elevated temperatures in an ionic liquid electrolyte. (C) A plot of discharge capacity and coulombic efficiency versus cycle number for FeF2 electrodes cycled at various current densities. The dark purple plot demonstrated the inherent discharge rate capability of a material when a slow charging rate is applied to recover the full capacity. Reproduced with permission from reference Xiao, A.W.; Lee, H.J.; Capone, I.; Robertson, A.; Wi, T.U.; Fawdon, J.; Wheeler, S.; Lee, H.W.; Grobert, N.; Pasta, M. Understanding the Conversion Mechanism and Performance of Monodisperse FeF2 Nanocrystal Cathodes. Nat. Mater. 2020, 19(6), 644–654. https://doi.org/10.1038/s41563-020-0621-z.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials

325

that more than 500 mAh/g could be maintained at discharge rates as high as C/2, while the same capacity could only be charged at a rate of C/20 (Fig. 5).4 The conversion mechanism FeF2 and consequently the capabilities and limitations of FeF2 are now largely understood. At a reasonable active material mass fraction (85 wt%), FeF2 an exceed the energy density of the NMC lithium system, with further capability to reduce materials costs.7 At this active material mass fraction, high capacity at moderate discharge rates (1C) should be achievable; however, the charging rate at room temperature will likely be limited. With further optimization of cathode construction and electrolyte design, 500–1000 cycles at deep discharge may also be possible. With greater oxidative stabilities and lower cell voltages than layered oxides, transition metal fluorides further offer an opportunity for improved safety if paired with temperature stable electrolytes. For these reasons, transition metal fluorides may not be useful for electric vehicles or consumer electronics, which require fast charging, or for grid-scale load-leveling applications where high cycle life is paramount. Instead, transition metal fluoride batteries could play a role in applications where high specific energy is an overriding priority and exchange of batteries between charges can be tolerated. These include electric aircraft, high-altitude pseudo-satellites, and potentially even long-range transit buses or commercial vehicles that are serviced at a central hub. Further research is still required in order to catapult transition metal fluorides to commercial viability. First, low-cost carbon coating techniques should be developed that are functional at low weight percent (5–10 wt%). While the cycling stability that has recently been reported in some FeF2-carbon nanocomposites is impressive, efforts should be focused on translating these results to a more energy dense format and applying them in conjunction with targeted electrolyte design. One of the primary benefits of transition metal fluoride cathodes is a low materials cost (especially in the case of FeF2). While highly concentrated FSI based electrolytes have demonstrated the greatest ability to enable the stable cycling of transition metal fluoride cathodes, such electrolytes exhibit low conductivity (high viscosity) and extremely high cost. Further research should focus on improving conductivity and reducing electrolyte cost, while maintaining the SEI forming ability and thermal stability of high FSI ionic liquids. While transition metal fluorides have been demonstrated to be more thermally stable than conventional cathode materials, stable operation at elevated temperatures (>60  C) could greatly improve the competitiveness of this technology. High temperature cycling could allow for higher power densities and significantly decreased charging time, and would further decrease the overpotentials observed on charge and discharge.

6

Future research directions for conversion-type materials

The previous section outlined the research and development activities achievements required to enable the use of current conversion type cathode technology in commercial applications. In this concluding section, future avenues of fundamental research targeted at the development of new conversion electrodes with improved materials properties are discussed. Despite the improvements in electrochemical performance outlined above, transition metal fluoride conversion cathodes are still plagued by a low ionic/ electronic conductivity and large voltage hysteresis. As these properties are inherent to the materials themselves, they can only be solved by changes in crystal structure and chemistry. More than a decade ago, Pereira et al. demonstrated that oxygen could be substituted at fluorine sites in the FeF2 structure with a concomitant increase in average discharge potential and electronic conductivity.73 Interestingly, higher levels of oxygen substitution generally resulted in lower capacities, while more limited oxygen content (x < 0.4 in FeO2F2-x) resulted in improved performance over FeF2. This results from the fact that O incorporation introduces a one-electron Fe3+/Fe2+ intercalation reaction at 3 V vs. Li+/Li, but depresses the potential of the two-electron Fe2+/Fe0 conversion reaction below the lower voltage limit. Kim et al. later demonstrated that improved cycle life in FeOxF2-x cathodes was a result of oxygen segregation to the particle surface during cycling, forming an electrochemically inactive but more electrically conductive LiFeOF phase.74 Other groups have found success in employing oxygen substitution in more complex phases such as Fan et al. who prepared a Fe0.9Co0.1OF material.54 The authors demonstrated that limiting the extent of the conversion reaction with O substitution could result in stable capacity retention for over 1000 cycles and decreased voltage hysteresis, while the improved electrical conductivity enabled reasonably high-rate capability. While the incorporation of oxygen resulted in reduced capacity, the co-substitution of an element such as cobalt – with a higher 3+/2+ redox potential – increased the average discharge potential to 2.5 V vs. Li+/Li, maintaining high energy density.54 In addition to more varied cation and anion substitution, the effect of epitaxial surface coatings should be explored. In both fluorides and oxyfluorides, coulombic inefficiency and capacity loss can be attributed to the formation of inactive material at the particle surface. Furthermore, as the conversion reaction tends toward lattice coherency between all phases in the active material particle, control of the surface crystal structure may be able to influence the evolution of the conversion reaction.

7

Conclusions

Lithium conversion reactions have a history as long as that of the lithium-ion battery itself. Despite these decades of research, conversion electrodes have yet to find their way into any commercial applications. While this may suggest inherent difficulties in their implementation, the mechanistic understanding of the conversion mechanism has, only very recently, advanced to the point where the fundamental capabilities and limitations of such materials are clear. This significant improvement in mechanistic understanding has resulted from the use of high-resolution analytical transmission electron microscopy techniques in conjunction

326

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials

with single-crystalline active material particles to identify the lattice orientation relationships between the parent phase and the discharged products. Separate investigations into various transition metal fluorides, oxides, and sulfides have congruently reported that reversible conversion reactions proceed through a topotactic transformation, where a common anion sublattice facilitates transformation from the parent transition metal compound to the discharged lithium salt. Such transformations rely on the cooperative movement of lithium and transition metal ions through the anion host lattice, resulting in the displacement of transition metals from their lattice sites and ultimately the nucleation of metallic particles at preferential interfaces within the material on discharge. This mechanistic understanding suggests that conversion compounds are inherently limited in their charging rate and reaction overpotential but are capable of moderate discharge rates and reasonably long cycle life. While certain chemistries—specifically the later 3d transition metal fluorides—can rightfully be considered as high energy density cathode materials on paper, their practical performance has been severely limited until very recently. Through a combination of electrolyte engineering and rational incorporation of carbon additives, high capacity (>500 mAh/g) cycling up to 1900 cycles has been demonstrated for FeF2 at the lab scale. In order to translate such success into a commercially viable product, high-capacity cycling at high active material mass fraction (>85 wt%), high areal capacity, and higher temperatures (> 60  C) should be targeted. In addition, the development of high performance conversion cathodes through rational cation/anion substitution should be further explored.

References 1. Schäfer, A. W.; Barrett, S. R. H.; Doyme, K.; Dray, L. M.; Gnadt, A. R.; Self, R.; O’Sullivan, A.; Synodinos, A. P.; Torija, A. J. Technological, Economic and Environmental Prospects of All-Electric Aircraft. Nat. Energy 2019, 4 (2), 160–166. https://doi.org/10.1038/s41560-018-0294-x. 2. Choi, J. W.; Aurbach, D. Promise and Reality of Post-Lithium-Ion Batteries with High Energy Densities. Nat. Rev. Mater. 2016, 1. https://doi.org/10.1038/natrevmats.2016.13. 3. Whittingham, M. S. Ultimate Limits to Intercalation Reactions for Lithium Batteries. Chem. Rev. 2014, 114 (23), 11414–11443. https://doi.org/10.1021/cr5003003. 4. Xiao, A. W.; Lee, H. J.; Capone, I.; Robertson, A.; Wi, T. U.; Fawdon, J.; Wheeler, S.; Lee, H. W.; Grobert, N.; Pasta, M. Understanding the Conversion Mechanism and Performance of Monodisperse FeF2 Nanocrystal Cathodes. Nat. Mater. 2020, 19 (6), 644–654. https://doi.org/10.1038/s41563-020-0621-z. 5. Yan, D.; Yang, H. Y.; Bai, Y. Tactics to Optimize Conversion-Type Metal Fluoride/Sulfide/Oxide Cathodes toward Advanced Lithium Metal Batteries. Nano Res. 2023,. https://doi. org/10.1007/s12274-023-5427-7. 6. Wang, L.; Wu, Z.; Zou, J.; Gao, P.; Niu, X.; Li, H.; Chen, L. Li-Free Cathode Materials for High Energy Density Lithium Batteries. Joule 2019. https://doi.org/10.1016/j. joule.2019.07.011. 7. Olbrich, L. F.; Xiao, A. W.; Pasta, M. Conversion-Type Fluoride Cathodes: Current State of the Art. Curr. Opin. Electrochem. 2021, 30, 100779. https://doi.org/10.1016/j. coelec.2021.100779. 8. Bruce, P. G.; Scrosati, B.; Tarascon, J.-M. Nanomaterials for Rechargeable Lithium Batteries. Angew. Chem. Int. Ed. 2008, 47 (16), 2930–2946. https://doi.org/10.1002/ anie.200702505. 9. Cabana, J.; Monconduit, L.; Larcher, D.; Palacín, M. R. Beyond Intercalation-Based Li-Ion Batteries: The State of the Art and Challenges of Electrode Materials Reacting through Conversion Reactions. Adv. Mater. 2010, 22 (35), E170–E192. https://doi.org/10.1002/adma.201000717. 10. Whittingham, M. S. Chemistry of Intercalation Compounds: Metal Guests in Chalcogenide Hosts. Progress in Solid State Chemistry 1978, 12. 11. Yu, L.; Tian, Y.; Xing, Y.; Hou, C.; Si, Y.; Lu, H.; Zhao, Y. Unraveling the Overlithiation Mechanism of LiMn2O4 and LiFePO4 Using Lithium-Metal Batteries. Ionics (Kiel) 2021, 27 (12), 5021–5035. https://doi.org/10.1007/s11581-021-04211-w. 12. Shu, J.; Shui, M.; Huang, F.; Ren, Y.; Wang, Q.; Xu, D.; Hou, L. A New Look at Lithium Cobalt Oxide in a Broad Voltage Range for Lithium-Ion Batteries. J. Phys. Chem. C 2010, 114 (7), 3323–3328. https://doi.org/10.1021/jp911994b. 13. Zhang, W.; Li, Y.; Wu, L.; Duan, Y.; Kisslinger, K.; Chen, C.; Bock, D. C.; Pan, F.; Zhu, Y.; Marschilok, A. C.; Takeuchi, E. S.; Takeuchi, K. J.; Wang, F. Multi-Electron Transfer Enabled by Topotactic Reaction in Magnetite. Nat. Commun. 2019, 10 (1), 1972. https://doi.org/10.1038/s41467-019-09528-9. 14. Zou, J.; Yuan, K.; Zhao, J.; Wang, B.; Chen, S.; Huang, J.; Li, H.; Niu, X.; Wang, L. Delithiation-Driven Topotactic Reaction Endows Superior Cycling Performances for High-Energy-Density FeSx (1  x  1.14) Cathodes. Energy Storage Mater. 2021, 43, 579–584. https://doi.org/10.1016/j.ensm.2021.09.036. 15. Gu, W.; Magasinski, A.; Zdyrko, B.; Yushin, G. Metal Fluorides Nanoconfined in Carbon Nanopores as Reversible High Capacity Cathodes for Li and Li-Ion Rechargeable Batteries: FeF2 as an Example. Adv. Energy Mater. 2015, 5 (4), 1–7. https://doi.org/10.1002/aenm.201401148. 16. Qu, K.; Ding, Z.; Wu, M.; Liu, P.; Chen, S.; Zhu, R.; Han, B.; Ma, X.; Gao, P.; Li, J. Unraveling Atomic-Scale Lithiation Mechanisms in a NiO Thin Film Electrode. J. Mater. Chem. A Mater. 2020, 8 (47), 25198–25207. https://doi.org/10.1039/d0ta08415h. 17. Chen, S.; Yang, C.; Shao, R.; Niu, J.; Wu, M.; Cao, J.; Ma, X.; Feng, J.; Wu, X.; Lu, J.; Wang, L.; Qi, J.; Gao, P. Direct Observation of Li Migration into V5S8: Order to Antisite Disorder Intercalation Followed by the Topotactic-Based Conversion Reaction. ACS Appl. Mater. Interfaces 2020, 12 (32), 36320–36328. https://doi.org/10.1021/ acsami.0c08428. 18. Zou, J.; Zhao, J.; Wang, B.; Chen, S.; Chen, P.; Ran, Q.; Li, L.; Wang, X.; Yao, J.; Li, H.; Huang, J.; Niu, X.; Wang, L. Unraveling the Reaction Mechanism of FeS2as a Li-Ion Battery Cathode. ACS Appl. Mater. Interfaces 2020, 12 (40), 44850–44857. https://doi.org/10.1021/acsami.0c14082. 19. Semones, D. E.; Tidwell, T.; McCallum, J. Lithium Battery Research. Annual Report No. 1, 1 April 1968- -31 March 1969. [Li/Halide, Most Promising Were AgF, CuF/Sub 2/and CdCl/Sub 2/]; United States; 1969. https://www.osti.gov/biblio/7367581. 20. Seiger, H. N.; Lyall, A. E. The High Energy Lithium Battery System in Electric Vehicle Propulsion; SAE Technical Paper, 1968. https://doi.org/10.4271/680454. 21. Morzilli, S.; Scrosati, B. Iron Oxide Electrodes in Lithium Organic Electrolyte Rechargeable Batteries. Electrochim. Acta 1985, 30 (10), 1271–1276. https://doi.org/ 10.1016/0013-4686(85)85002-7. 22. Uetani, Y.; Yokoyama, K.; Okamoto, O. Preparation of Iron Sulfides and the Study of their Electrochemical Characteristics for Use in a Nonaqueous—Lithium Battery. J. Power Sources 1980, 5 (1), 89–98. https://doi.org/10.1016/0378-7753(80)80098-X. 23. Godshall, N. A.; Raistrick, I. D.; Huggins, R. A. Thermodynamic Investigations of Ternary Lithium-Transition Metal-Oxygen Cathode Materials. Mater. Res. Bull. 1980, 15 (5), 561–570. 24. Whittingham, M. S. The Role of Ternary Phases in Cathode Reactions. J. Electrochem. Soc. 1976, 123 (3), 315–320. https://doi.org/10.1149/1.2132817. 25. Bauman, H. F.; et al.; Lockheed Missiles and Space Co. Inc., Palo Alto CA, Mechanical and Mathematical Sciences Lab. Lithium Anode Limited Cycle Secondary Battery; Quarterly technical progress rept. no. 2, Defense Technical Information Center: Fort Belvoir, VA, 1963. 26. Thackeray, M. M.; Coetzer, J. A Preliminary Investigation of the Electrochemical Performance of E-Fe2O3 and Fe3O4 Cathodes in High-Temperature Cells. Mater. Res. Bull. 1981, 16. 27. Taberna, P. L.; Mitra, S.; Poizot, P.; Simon, P.; Tarascon, J.-M. High Rate Capabilities Fe3O4-Based Cu Nano-Architectured Electrodes for Lithium-Ion Battery Applications. Nat. Mater. 2006, 5 (7), 567–573. https://doi.org/10.1038/nmat1672.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials

327

28. Poizot, P.; Laruelle, S.; Grugeon, S.; Dupont, L.; Tarascon, J.-M. Nano-Sized Transition-Metal Oxides as Negative-Electrode Materials for Lithium-Ion Batteries. Nature 2000, 407 (6803), 496–499. https://doi.org/10.1038/35035045. 29. Grugeon, S.; Laruelle, S.; Herrera-Urbina, R.; Dupont, L.; Poizot, P.; Tarascon, J.-M. Particle Size Effects on the Electrochemical Performance of Copper Oxides toward Lithium. J. Electrochem. Soc. 2001, 148 (4), A285. https://doi.org/10.1149/1.1353566. 30. Amatucci, G. G.; Pereira, N. Fluoride Based Electrode Materials for Advanced Energy Storage Devices. J. Fluor. Chem. 2007, 128 (4), 243–262. https://doi.org/10.1016/j. jfluchem.2006.11.016. 31. Badway, F.; Cosandey, F.; Pereira, N.; Amatucci, G. G. Carbon Metal Fluoride Nanocomposites. J. Electrochem. Soc. 2003, 150 (10), A1318. https://doi.org/ 10.1149/1.1602454. 32. Li, H.; Balaya, P.; Maier, J. Li-Storage Via Heterogeneous Reaction in Selected Binary Metal Fluorides and Oxides. J. Electrochem. Soc. 2004, 151 (11), A1878. https://doi.org/ 10.1149/1.1801451. 33. Li, H.; Richter, G.; Maier, J. Reversible Formation and Decomposition of LiF Clusters Using Transition Metal Fluorides as Precursors and their Application in Rechargeable Li Batteries. Adv. Mater. 2003, 15 (9), 736–739. https://doi.org/10.1002/adma.200304574. 34. Uetani, Y.; Yokoyama, I.; Okamoto, O. Preparation of Iron Sulfides and the Study of their Electrochemical Characteristics for Use in a Non-Aqueous-Lithium Battery, vol. 5; Elsevier: Sequoia, S.A., 1980. 35. Kim, B.-C.; Takada, K.; Ohta, N.; Seino, Y.; Zhang, L.; Wada, H.; Sasaki, T. All Solid State Li-Ion Secondary Battery with FeS Anode. Solid State Ion. 2005, 176 (31–34), 2383–2387. https://doi.org/10.1016/j.ssi.2005.05.019. 36. Kaun, T. D.; Nelson, P. A.; Redey, L.; Vissers, D. R.; Henriksen, G. L. High Temperature Lithium/Sulfide Batteries. Electrochim. Acta 1993, 38 (9), 1269–1287. https://doi.org/ 10.1016/0013-4686(93)80057-7. 37. Li, L.; Jacobs, R.; Gao, P.; Gan, L.; Wang, F.; Morgan, D.; Jin, S. Origins of Large Voltage Hysteresis in High-Energy-Density Metal Fluoride Lithium-Ion Battery Conversion Electrodes. J. Am. Chem. Soc. 2016, 138 (8), 2838–2848. https://doi.org/10.1021/jacs.6b00061. 38. Yamakawa, N.; Jiang, M.; Grey, C. P. Investigation of the Conversion Reaction Mechanisms for Binarycopper(ii) Compounds by Solid-State Nmr Spectroscopy and X-Ray Diffraction. Chem. Mater. 2009, 21 (14), 3162–3176. https://doi.org/10.1021/cm900581b. 39. Song, H.; Cui, H.; Wang, C. Extremely High-Rate Capacity and Stable Cycling of a Highly Ordered Nanostructured Carbon–FeF 2 Battery Cathode. J. Mater. Chem. A 2015, 3 (44), 22377–22384. https://doi.org/10.1039/C5TA06297G. 40. Huang, Q.; Turcheniuk, K.; Ren, X.; Magasinski, A.; Gordon, D.; Bensalah, N.; Yushin, G. Insights into the Effects of Electrolyte Composition on the Performance and Stability of FeF 2 Conversion-Type Cathodes. Adv. Energy Mater. 2019, 9 (17), 1–11. https://doi.org/10.1002/aenm.201803323. 41. Wang, X.; Gu, W.; Lee, J. T.; Nitta, N.; Benson, J.; Magasinski, A.; Schauer, M. W.; Yushin, G. Carbon Nanotube-CoF2 Multifunctional Cathode for Lithium Ion Batteries: Effect of Electrolyte on Cycle Stability. Small 2015, 11 (38), 5164–5173. https://doi.org/10.1002/smll.201501139. 42. Wu, F.; Borodin, O.; Yushin, G. In Situ Surface Protection for Enhancing Stability and Performance of Conversion-Type Cathodes. MRS Energy Sustainability 2017, 4 (1), 9. https://doi.org/10.1557/mre.2017.11. 43. Karki, K.; Wu, L.; Ma, Y.; Armstrong, M. J.; Holmes, J. D.; Garofalini, S. H.; Zhu, Y.; Stach, E. A.; Wang, F. Revisiting Conversion Reaction Mechanisms in Lithium Batteries: Lithiation-Driven Topotactic Transformation in FeF2. J. Am. Chem. Soc. 2018, 140 (51), 17915–17922. https://doi.org/10.1021/jacs.8b07740. 44. Wang, F.; Yu, H. C.; Chen, M. H.; Wu, L.; Pereira, N.; Thornton, K.; Van Der Ven, A.; Zhu, Y.; Amatucci, G. G.; Graetz, J. Tracking Lithium Transport and Electrochemical Reactions in Nanoparticles. Nat. Commun. 2012, 3, 1201–1208. https://doi.org/10.1038/ncomms2185. 45. Wang, F.; Robert, R.; Chernova, N. A.; Pereira, N.; Omenya, F.; Badway, F.; Hua, X.; Ruotolo, M.; Zhang, R.; Wu, L.; Volkov, V.; Su, D.; Whittingham, M. S.; Grey, C. P.; Amatucci, G. G.; Zhu, Y.; Graetz, J.; Omenya, F.; Badway, F.; et al. Conversion Reaction Mechanisms in Lithium Ion Batteries: Study of the Binary Metal Fluoride Electrodes Conversion Reaction Mechanisms in Lithium Ion Batteries: Study of the Binary Metal Fluoride Electrodes. J. Am. Chem. Soc. 2011, 133, 18828–18836. https://doi.org/10.1021/ ja206268a. 46. Lin, F.; Nordlund, D.; Weng, T. C.; Zhu, Y.; Ban, C.; Richards, R. M.; Xin, H. L. Phase Evolution for Conversion Reaction Electrodes in Lithium-Ion Batteries. Nat. Commun. 2014, 5. https://doi.org/10.1038/ncomms4358. 47. Yamakawa, N.; Jiang, M.; Key, B.; Grey, C. P. Identifying the Local Structures Formed during Lithiation of the Conversion Material, Iron Fluoride, in a Li Ion Battery: A Solid-State NMR, X-Ray Diffraction, and Pair Distribution Function Analysis Study. J. Am. Chem. Soc. 2009, 131 (30), 10525–10536. https://doi.org/10.1021/ja902639w. 48. Ko, J. K.; Wiaderek, K. M.; Pereira, N.; Kinnibrugh, T. L.; Kim, J. R.; Chupas, P. J.; Chapman, K. W.; Amatucci, G. G. Transport, Phase Reactions, and Hysteresis of Iron Fluoride and Oxyfluoride Conversion Electrode Materials for Lithium Batteries. ACS Appl. Mater. Interfaces 2014, 6 (14), 10858–10869. https://doi.org/10.1021/am500538b. 49. Li, C.; Chen, K.; Zhou, X.; Maier, J. Electrochemically Driven Conversion Reaction in Fluoride Electrodes for Energy Storage Devices. NPJ Comput. Mater. 2018, 4 (22). https://doi. org/10.1038/s41524-018-0079-6. 50. Yu, S. H.; Lee, S. H.; Lee, D. J.; Sung, Y. E.; Hyeon, T. Conversion Reaction-Based Oxide Nanomaterials for Lithium Ion Battery Anodes. Small 2016, 12 (16), 2146–2172. https://doi.org/10.1002/smll.201502299. 51. Xu, X.; Liu, W.; Kim, Y.; Cho, J. Nanostructured Transition Metal Sulfides for Lithium Ion Batteries: Progress and Challenges. Nano Today 2014, 604–630. https://doi.org/ 10.1016/j.nantod.2014.09.005. Elsevier B.V. 52. Wang, C. In Situ Transmission Electron Microscopy and Spectroscopy Studies of Rechargeable Batteries under Dynamic Operating Conditions : A Retrospective and Perspective View. J. Mater. Res. 2014, 30 (3), 326–339. https://doi.org/10.1557/jmr.2014.281. 53. Xie, Z.; Jiang, Z.; Zhang, X. Review — Promises and Challenges of In Situ Transmission Electron Microscopy Electrochemical Techniques in the Studies of Lithium Ion Batteries. J. Electrochem. Soc. 2017, 164 (9), 2110–2123. https://doi.org/10.1149/2.1451709jes. 54. Fan, X.; Hu, E.; Ji, X.; Zhu, Y.; Han, F.; Hwang, S.; Liu, J.; Bak, S.; Ma, Z.; Gao, T.; Liou, S. C.; Bai, J.; Yang, X. Q.; Mo, Y.; Xu, K.; Su, D.; Wang, C. High Energy-Density and Reversibility of Iron Fluoride Cathode Enabled Via an Intercalation-Extrusion Reaction. Nat. Commun. 2018, 9 (1), 1–12. https://doi.org/10.1038/s41467-018-04476-2. 55. Evmenenko, G.; Warburton, R. E.; Yildirim, H.; Greeley, J. P.; Chan, M. K. Y.; Buchholz, D. B.; Fenter, P.; Bedzyk, M. J.; Fister, T. T. Understanding the Role of Overpotentials in Lithium Ion Conversion Reactions: Visualizing the Interface. ACS Nano 2019, 13 (7), 7825–7832. https://doi.org/10.1021/acsnano.9b02007. 56. Huang, Z. X.; Wang, Y.; Wong, J. I.; Shi, W. H.; Yang, H. Y. Synthesis of Self-Assembled Cobalt Sulphide Coated Carbon Nanotube and Its Superior Electrochemical Performance as Anodes for Li-Ion Batteries. Electrochim. Acta 2015, 167, 388–395. https://doi.org/10.1016/j.electacta.2015.03.183. 57. Wu, F.; Yushin, G. Conversion Cathodes for Rechargeable Lithium and Lithium-Ion Batteries. Energ. Environ. Sci. 2017, 10 (2), 435–459. https://doi.org/10.1039/C6EE02326F. 58. Hua, X.; Robert, R.; Du, L. S.; Wiaderek, K. M.; Leskes, M.; Chapman, K. W.; Chupas, P. J.; Grey, C. P. Comprehensive Study of the CuF2 Conversion Reaction Mechanism in a Lithium Ion Battery. J. Phys. Chem. C 2014, 118 (28), 15169–15184. https://doi.org/10.1021/jp503902z. 59. Omenya, F.; Zagarella, N. J.; Rana, J.; Zhang, H.; Siu, C.; Zhou, H.; Wen, B.; Chernova, N. A.; Piper, L. F. J.; Zhou, G.; Whittingham, M. S. Intrinsic Challenges to the Electrochemical Reversibility of the High Energy Density Copper (II) Fluoride Cathode Material. ACS Appl. Energy. Mater. 2019, 2 (7), 5243–5253. https://doi.org/10.1021/acsaem.9b00938. 60. Seo, J. K.; Cho, H.-M.; Takahara, K.; Chapman, K. W.; Borkiewicz, O. J.; Sina, M.; Shirley Meng, Y. Revisiting the Conversion Reaction Voltage and the Reversibility of the CuF2 Electrode in Li-Ion Batteries. Nano Res. 2017, 10 (12), 4232–4244. https://doi.org/10.1007/s12274-016-1365-6. 61. Gordon, D.; Huang, Q.; Magasinski, A.; Ramanujapuram, A.; Bensalah, N.; Yushin, G. Mixed Metal Difluorides as High Capacity Conversion-Type Cathodes: Impact of Composition on Stability and Performance. Adv. Energy Mater. 2018, 8 (19), 1–10. https://doi.org/10.1002/aenm.201800213. 62. Wang, F.; Kim, S.-W.; Seo, D.-H.; Kang, K.; Wang, L.; Su, D.; Vajo, J. J.; Wang, J.; Graetz, J. Ternary Metal Fluorides as High-Energy Cathodes with Low Cycling Hysteresis. Nat. Commun. 2015, 6 (1), 6668. https://doi.org/10.1038/ncomms7668. 63. Zhong, G.; Bai, J.; Duchesne, P. N.; McDonald, M. J.; Li, Q.; Hou, X.; Tang, J. A.; Wang, Y.; Zhao, W.; Gong, Z.; Zhang, P.; Fu, R.; Yang, Y. Copper Phosphate as a Cathode Material for Rechargeable Li Batteries and Its Electrochemical Reaction Mechanism. Chem. Mater. 2015, 27 (16), 5736–5744. https://doi.org/10.1021/acs.chemmater.5b02290.

328

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Positive Electrode: Conversion Materials

64. Kim, Y.-U.; Lee, C. K.; Sohn, H.-J.; Kang, T. Reaction Mechanism of Tin Phosphide Anode by Mechanochemical Method for Lithium Secondary Batteries. J. Electrochem. Soc 2004, 151 (6), A933. https://doi.org/10.1149/1.1738679. 65. Bhatt, M. D.; Lee, J. Y. High Capacity Conversion Anodes in Li-Ion Batteries: A Review. Int. J. Hydrogen Energy 2019, 44 (21), 10852–10905. https://doi.org/10.1016/ j.ijhydene.2019.02.015. 66. El Kharbachi, A.; Uesato, H.; Kawai, H.; Wenner, S.; Miyaoka, H.; Sørby, M. H.; Fjellvåg, H.; Ichikawa, T.; Hauback, B. C. MgH2–CoO: A Conversion-Type Composite Electrode for LiBH4-Based All-Solid-State Lithium Ion Batteries. RSC Adv. 2018, 8 (41), 23468–23474. https://doi.org/10.1039/C8RA03340D. 67. Li, L.; Meng, F.; Jin, S. High-Capacity Lithium-Ion Battery Conversion Cathodes Based on Iron Fluoride Nanowires and Insights into the Conversion Mechanism. Nano Lett. 2012, 12 (11), 6030–6037. https://doi.org/10.1021/nl303630p. 68. Li, T.; Li, L.; Cao, Y. L.; Ai, X. P.; Yang, H. X. Reversible Three-Electron Redox Behaviors of FeF3 Nanocrystals as High-Capacity Cathode-Active Materials for Li-Ion Batteries. J. Phys. Chem. C 2010, 114 (7), 3190–3195. https://doi.org/10.1021/jp908741d. 69. Zhou, J.; Zhang, D.; Zhang, X.; Song, H.; Chen, X. Carbon-Nanotube-Encapsulated FeF2 Nanorods for High-Performance Lithium-Ion Cathode Materials. ACS Appl. Mater. Interfaces 2014, 6 (23), 21223–21229. https://doi.org/10.1021/am506236n. 70. Chun, J.; Jo, C.; Sahgong, S.; Kim, M. G.; Lim, E.; Kim, D. H.; Hwang, J.; Kang, E.; Ryu, K. A.; Jung, Y. S.; Kim, Y.; Lee, J. Ammonium Fluoride Mediated Synthesis of Anhydrous Metal Fluoride-Mesoporous Carbon Nanocomposites for High-Performance Lithium Ion Battery Cathodes. ACS Appl. Mater. Interfaces 2016, 8 (51), 35180–35190. https://doi. org/10.1021/acsami.6b10641. 71. Wang, H.; Dai, H. Strongly Coupled Inorganic-Nano-Carbon Hybrid Materials for Energy Storage. Chem. Soc. Rev. 2013, 42, 3088–3113. https://doi.org/10.1039/c2cs35307e. 72. Su, Y.; Chen, J.; Li, H.; Sun, H.; Yang, T.; Liu, Q.; Ichikawa, S.; Zhang, X.; Zhu, D.; Zhao, J.; Geng, L.; Guo, B.; Du, C.; Dai, Q.; Wang, Z.; Li, X.; Ye, H.; Guo, Y.; Li, Y.; Yao, J.; Yan, J.; Luo, Y.; Qiu, H.; Tang, Y.; Zhang, L.; Huang, Q.; Huang, J. Enabling Long Cycle Life and High Rate Iron Difluoride Based Lithium Batteries by in Situ Cathode Surface Modification. Adv. Sci. 2022, 9 (21), 2201419. https://doi.org/10.1002/advs.202201419. 73. Pereira, N.; Badway, F.; Wartelsky, M.; Gunn, S.; Amatucci, G. G. Iron Oxyfluorides as High Capacity Cathode Materials for Lithium Batteries. J. Electrochem. Soc. 2009, 156 (6), A407. https://doi.org/10.1149/1.3106132. 74. Kim, S. W.; Pereira, N.; Chernova, N. A.; Omenya, F.; Gao, P.; Whittingham, M. S.; Amatucci, G. G.; Su, D.; Wang, F. Structure Stabilization by Mixed Anions in Oxyfluoride Cathodes for High-Energy Lithium Batteries. ACS Nano 2015, 9 (10), 10076–10084. https://doi.org/10.1021/acsnano.5b03643.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector Futoshi Matsumoto and Mika Fukunishi, Department of Applied Chemistry, Faculty of Chemistry and Biochemistry, Kanagawa University, Yokohama, Kanagawa, Japan © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 2.1 2.2 2.3 3 4 5 5.1 5.2 5.2.1 5.2.2 5.3 6 7 7.1 7.2 8 8.1 8.2 8.3 8.4 8.5 9 10 References

Introduction Conventional CCs Anodes Cathodes Alternative CC materials Three-dimensional structure of CCs Development of CCs to ensure safety CCs for aqueous batteries Metal CCs Carbon CCs Carbon-coated CCs Carbon film CCs Summary of carbon CCs Flexible CCs CCs for Si and Li deposition anodes CCs for Li deposition anodes CCs for Si anodes CCs for post-LIBs Sodium batteries Magnesium batteries Calcium batteries Aluminum batteries Sulfur batteries CCs for all-solid-state batteries Conclusions

330 330 330 331 332 332 333 336 336 336 336 337 339 339 340 340 341 342 342 342 343 343 343 343 344 344

Abstract Current collectors (CCs) in batteries provide a path through which current flows and hold the electrode active materials in the cathode and anode on its surface. CCs have the role of maintaining electronic contact between the electrode active materials and themselves. Due to the recent demand for improved battery performance, along with the functions mentioned above, the CCs are increasingly involved in improving the battery’s energy and power density, input/output characteristics, and safety. In addition, research on various types of batteries is being conducted to meet the demand for producing batteries using elements that are abundant on the earth, with the aim of improving battery performance and lowering their price. CCs for cathodes and anodes compatible with these batteries need to be considered from various viewpoints. In this chapter, the types and problems of current CCs in lithium-ion batteries presently in use, as well as the requirements for CCs used in next-generation batteries, are discussed.

Glossary All-solid-state batteries As the name suggests, all-solid-state batteries are batteries in which all the components that make up the battery are solid. Secondary batteries basically consist of two metal electrodes (anode and cathode) and an electrolyte that fills the space between them. Conventional secondary batteries use liquid as the electrolyte, but all-solid-state batteries use solid as the electrolyte. Binder Lithium-ion batteries use materials that become the positive and negative electrodes. These are called active materials. The active material is in powder form, and the binder is the "glue" that binds it together and fixes it on a metal foil called a current collector foil (positive electrode: aluminum foil, negative electrode: copper foil). Energy density Energy density is defined as the amount of energy stored per unit volume in a system or space. In batteries, it is defined as the energy stored within the battery. It is expressed as energy density per weight or volume of the battery. Energy is calculated as the product of battery capacity (Ah) and battery voltage (V).

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00171-7

329

330

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

Ionic liquid An ionic liquid is a liquid "salt" consisting only of ions (anions, cations), and liquid compounds in particular are called ionic liquids. Generally, "salts" composed of ions are represented by inorganic salts, and are known to have a high melting point. Since the early 1990s, when "ionic liquids" with low melting points were reported, various compounds have been synthesized and their physical properties have been reported. There is no clear definition, and there are various names for it, such as "room temperature molten salt" and "room temperature molten salt." Dendrite formation Dendrite is a phenomenon in which metal extends from anode toward cathode during charging of a secondary battery. In the anode reactionduring the charging, some metal ions may deposit as metal in electrochemical reduction, the metal form dendrites, which grow and extend as the charge and discharge cycle is repeated. When the formed dendrites peel off, the amount of anode active material decreases, resulting in a decrease in capacity. However, the continued growth of the dendrites is more problematic, as some of the dendrites break through the pores of the porous separator and reach the cathode, shorting the battery. And this may cause a fire or explosion.

Key points

• • • • • • •

1

Summary of the materials used in current collectors mainly for lithium-ion batteries. Characteristics and functions that current collector foil should have. Corrosion of the current collectors in batteries and how to prevent it. Fabrication of the structures and their effects in a three-dimensional current collector for improving the energy density and input/output characteristics of batteries. Structure and concept of a current collector for interrupting short-circuit current when an abnormal situation occurs in a battery. Types of current collectors and the characteristics that they should have for water-based secondary batteries with high environmental adaptability. Explanation of the characteristics that current collectors should have in next-generation batteries and application examples.

Introduction

In primary and secondary batteries, current collectors (CCs) are needed for the anodes and cathodes to allow electronic current flow that is as unhindered as possible to and from the cell terminal. In addition, anode and cathode active materials that store electricity need to be attached to the CCs. If the active materials separate from the CCs, the active materials cannot be charged and discharged thereafter, resulting in a decrease in the capacity of the battery. So, CCs are required to have high electronic conductivity and good adhesion to active materials. In addition, features such as the use of low cost and lightweight materials are needed with consideration of the further use of batteries.1–3 Recently, three-dimensional structuring of CCs has been devised to allow the current to pass through active material layers at high speed and to attach more active materials on the CCs.4 Since there is a demand for inexpensive, lightweight, and corrosion-free CCs, consideration has been given to shifting from conventional metal-based CCs to carbon-based current CCs.5 More recently, there have been many requests for CCs, such as proposals for CC structures that ensure safety against abnormalities in batteries.6 This chapter focuses on lithium-ion batteries (LIBs) and related batteries and describes the characteristics and problems of the CCs used in present practical secondary batteries and next-generation batteries, and the research results conducted to solve those problems are provided.

2

Conventional CCs

In current LIBs that use carbon-based materials, such as graphite as the anode active material and lithium (Li) metal oxides as the cathode active material, copper (Cu) is used as the CC for the anode, and aluminum (Al) is used as the CC for the cathode; these provide electrochemical and chemical stability in LIB cells.7 This combination has not changed since commercially available LIBs were developed. For the composition ratio of the weight of the LIBs, the Cu CC is 8.1% and the Al CC is 6.9%; thus, the CC alone accounts for 15% of the weight.7 As with weight, the thickness of the CC foil causes a reduction in the energy density of the LIBs, and a shift toward using thinner CCs has occurred. The various CCs are introduced below. CCs should have high electronic conductivity; thus, to compare the electronic conductivity of various CCs that are used in actual LIBs and for research development, a summary of the conductivity is provided in Table 1.

2.1

Anodes

When Cu is oxidized in the potential region over 3.5 V (vs. Li/Li+), the Cu surfaces are subjected to over-discharge voltage of the anodes.9 However, in the lower potential region, Cu is stable and does not react with Li to form a Li-Cu alloy. Additionally, Cu is

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

331

Table 1 Selected materials considered for use as lithium positive electrode current collectors and their relative sheet conductivities.8 Material

Relative conductivity per unit volume

Relative conductivity per unit mass

Relative conductivity per unit cost

Ag Cu Au Al Mo W Zn Ni Fe Pt Cr Ta 304SS 316SS Ti SiC Mn C pyrolytic graphite C graphite C black

1.05 1 0.7 0.4 0.31 0.29 0.28 0.24 0.17 0.16 0.13 0.13 0.1 0.1 0.040 0.012 0.009 0.007 0.0003 0.00001

0.9 1 0.33 1.3 0.27 0.13 0.36 0.25 0.20 0.067 0.16 0.072 0.1 0.1 0.079 0.032 0.01 0.03 0.0012 0.00004

0.01 1 0.00008 2 0.01a 0.02 0.8 0.05 2a 0.000008 0.05 0.001a 0.1a 0.07a 0.02a 0.001 0.01 –b 0.0005 0.00002

a

Pure scrap prices. No bulk prices available. Reproduced with permission from Whitehead, A. H.; Schreiber, M. Current Collectors for Positive Electrodes of Lithium-Based Batteries. J. Electrochem. Soc. 2005, 152, A2105–A2113. © 2005 The Electrochemical Society. b

an abundant resource, very cheap in price, and easy to process. These characteristics are the reasons for the selection of Cu as the CC foil for the anodes. The potential region of about 3.5 V (vs. Li/Li+) could be reached if the negative electrode will be completely discharge, i.e., all Li+ are deintercalated. The then dissolved Cu2+ redeposits during the charging process. If the deposits grow as dendrites up to the cathodes, a short circuit of the cell may occur, and the battery may run out of control. Even if the deposits do not grow and Cu is deposited on the surface of the anode surfaces or on the separator, the insertion of lithium ions (Li+), as an example, into the anodes is inhibited, and Li metal is deposited on the anode surfaces. The mechanism of corrosion of the Cu CCs is thoroughly summarized in the review by Ryan et al.10 When water and hydrogen fluoride generated from the decomposition of lithium hexafluorophosphate (LiPF6) in organic solutions are present in the electrolyte solution, copper oxide (CuO) or copper fluoride (CuF2) are formed on the surface of the Cu CCs and form layers, but the layers do not grow thick. Additionally, a report has shown that the layer does not inhibit the dissolution of Cu but rather promotes the dissolution of Cu.11 It has also been reported that the ease of the dissolution of the Cu CC surfaces varies depending on the composition of the organic electrolyte containing the lithium salt used.11 Yashiro et al. summarized in detail the dissolution potential and reaction mechanism on the high potential side in the Cu CCs as along with the Al CCs described later.12 The development of additives for producing a protective layer is also under examination.13 Carbon coating of the surface of the Cu CCs is used to strengthen the adhesion of the anode active materials to the CC surfaces and to reduce the interfacial resistance between the active material layer and the CCs. Examples of carbon materials used are carbon black, graphene and a conducting polymer.14,15 As a coating method, in most cases, a slurry obtained by mixing carbon materials and binders is coated on the surface of the CCs and dried.14

2.2

Cathodes

In most practical batteries, Al is used as a cathode CC because the Al CCs are electrochemically stable even when exposed to a high potential during charging of the cathodes; here, aluminum oxide (Al2O3) is formed as a protective layer on the Al CC surfaces. Above 1 V (vs. Li/Li+), Al2O3 is formed, and in the case of electrolyte solutions in which lithium salts containing fluorine atoms, such as LiPF6, are dissolved and at voltages above 3.5 V, aluminum fluoride (AlF3) is formed on the Al CC surfaces based on the reaction of hydrofluoric acid (HF) with the Al CC surfaces (Eqs. 1–3).15 LiPF6 ⇆ LiF + PF5

(1)

PF5 + H2 O ! POF3 + 2HF

(2)

Al2 O3 + 6HF ! 2AlF3 + 3H2 O

(3)

This AlF3 layer is stable as a protective layer for the Al CCs but may not be protective at either high voltages or long-term operation. Furthermore, HF, which forms the protective layer, has a negative effect on the cathode materials. The replacement of LiPF6

332

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

with another salt is currently under consideration. Without the LiPF6 salt, the protective layer cannot be formed; thus, other methods for preventing dissolution of the Al CC surfaces are being investigated. Research and development on the following methods for protecting Al CC surfaces are under consideration: changing the composition of the electrolyte solutions,16,17 adding additives to form organic passivation films on the Al CC surfaces,18,19 coating highly conductive graphene-like carbon layers on the Al CC surfaces,20 forming solid-state imide salt-based polymer plastic crystal electrolyte membranes on the Al CC surfaces,21 and adding fumed silica nanoparticles to the electrolyte solutions.22 As a remedy for another problem of CCs, various modifications of the CC surface have been made to enhance the retention of the cathode active materials. Acid treatment of the Al surface to form irregularities on the surface of the CCs has been reported to improve the cycle stability of the cathodes.23 In addition, a report has shown that the ohmic resistance of the cathode layer increases when the Al CC surfaces are roughened by surface treatment. This occurs because oxide layers are formed on the surfaces of the Al CCs due to the surface treatment, and no treatment method that reduces the ohmic resistance has been found yet.24 Additionally, carbon coatings were applied on Al CCs to decrease the resistance between the cathode layer and CC surfaces, to enhance the adhesion strength between the cathode layer and CC surfaces and to depress the corrosion of the Al CCs.25–27 A report has also shown that the carbon coating effect did not require a conductive additive in the cathode layers (Fig. 1).28

2.3

Alternative CC materials

In addition to Cu and Al, iron (Fe), nickel (Ni), titanium (Ti), chromium (Cr) and stainless steel (SS) are often used as CCs for anodes or cathodes. A summary of the dissolution behavior of these CCs with the exception of Ni and SS is shown in Fig. 2. Ti and Cr, similar to Al, form a protective layer at high potentials; thus, they can be relatively good CCs for cathodes. Fe tends to dissolve the protective layer at voltages before 4 V. When Fe is used as a high-voltage cathode CC, Fe is considered to be unsuitable for oxidized and dissolved CC surfaces. Specific application examples of the CCs to anodes and cathodes are described below. Since Fe is an inexpensive and abundant metal, every year, there are several reports on the desire to use iron.29,30 Ni can be dissolved at high potentials and is therefore used as a CC for anodes31,32 and low potential cathodes, such as sulfur cathodes.33 Due to its ease of processing, nickel foam is frequently used to produce thick active material layers.34 Ti, similar to Al, forms a stable protection layer of titanium fluoride (TiF4) on Ti surfaces at high voltages, but it is less stable than the protection layer formed on the Al surface35,36; thus, it is not used as a CC for high-voltage cathodes but is used for anodes37 and low-voltage cathodes.38 Recently, studies have also reported on the use of Ti compounds with high electronic conductivity and flexibility, such as TiCx and TiNx, as CCs.39,40 Similar to Al and Ti, Cr also form protective layers at high voltage. No example of using Cr metal as CC foils and foams exists; however, coating the CC surfaces with Cr plating41 and alloying Cr42 produces CCs with higher corrosion resistance than Al or Ni CCs. SS, which are Fe-based alloys and contain Cr, Ni and carbon, are the most likely candidates for Al CCs for cathodes.43 Nickel-free, nitrogen-treated stainless steel could be applied as CCs for a 5 V-class cathode material.44 Studies by Kendrick et al.,7 Schreiber et al.8 and Yashiro et al.,12 all of which have been cited numerous times, thoroughly summarize the reactivities of these CCs in an organic electrolyte solution. Although some of these papers are somewhat older, they provide an overview of the basics.

3

Three-dimensional structure of CCs

Various attempts have been made to create electrodes with thicker active material layers to increase the battery energy density. As the anode and cathode layers become thicker, a structure that allows Li+ and electrons to move smoothly throughout the thick layer needs to be considered. Therefore, the development of three-dimensional CCs to shorten the distance that electrons can move

Fig. 1 Schematic representation of the various types of contact at the electrode—current collector interface for a (left side) conventional cathode and a (right side) new cathode configuration of primer carbon-coated layer/cathode layer without conductive additive.28 Reproduced with permission from Busson, C.; Blin, M.-A.; Guichard, P.; Soudan, P.; Crosnier, O.; Guyomard, D.; Lestriez, B. A Primed Current Collector for High Performance Carbon-Coated LiFePO4 Electrodes With No Carbon Additive. J. Power Sources 2018, 406, 7–17. © 2018 Elsevier B.V.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

333

Fig. 2 Schematic drawings of the passivation for several metals in a nonaqueous alkyl carbonate solution containing LiPF6 salt.12 Reproduced with permission from Myunga, S.-T.; Sasaki, Y.; Sakurada, S.; Sun, Y.-K.; Yashiro, H. Electrochemical Behavior of Current Collectors for Lithium Batteries in Non-aqueous Alkyl Carbonate Solution and Surface Analysis by ToF-SIMS. Electrochim. Acta 2009, 55, 288–297. © 2018 Elsevier B.V.

within the active material layer is needed. In this section, three-dimensional CCs are summarized as classification points regarding the production of three-dimensional CCs and the materials needed for the CCs. Metal foam is a typical example of a three-dimensional CC, and numerous research examples have been reported.45,46 To create three-dimensional CCs, a mold with protrusions is pressed onto a flat metal foil to create protrusions at a pitch of 1–3 mm on the metal CC foil.47 A filter method has been proposed for pouring an electrode active material into a three-dimensional CC foil with these protrusions to create a high-density and thick active material layer (Fig. 3). This method allows the attainment of active material layer thicknesses of 70–400 mm (capacity: 4–20 mAh cm−2). These electrodes showed stable charge–discharge cycles.47 It has been also reported that a CC woven with metal fibers can coat an active material with a coating amount of 180 mg cm−2.48 Other various methods have also been proposed to create three-dimensional frameworks of CCs. A typical example is based on porous alumina. By anodizing Al substrates, holes are created perpendicular to the substrate surface. By coating the inner wall of the holes with carbon, the CC and electrode active material are integrated (Fig. 4).49 By carbonizing natural wood, a three-dimensional structure similar to porous alumina can be obtained. By packing cathode material into these three-dimensional carbon structures, the creation of cathodes with a 500 mm cathode layer thickness is possible, and they are more robust than active material layers built on top of the traditional plate of CCs.50 Other proposed methods for creating a three-dimensional structure of CCs include a template method,51 a method of selectively eluting one metal of an alloy,52 and a method using a three-dimensional printer.53,54 Additionally, three-dimensional CCs created on a larger scale using photolithography have been reported.55 Although the proposed three-dimensional structure of CCs can improve battery performance, problems remain regarding mass production. For mass production, the only method available is the employment of metal foams and the creation of three-dimensional electrodes by adding three-dimensional irregularities to metal foil using a mold. However, further production methods need to be devised from the viewpoint of using simple methods as an important factor.

4

Development of CCs to ensure safety

Batteries are being researched and developed to store more energy in the battery, but more stored energy can be dangerous. A higher density of energy needs more improved safety measures. A typical battery accident is caused by an internal short circuit between the cathode and the anode; this causes a large amount of uncontrolled current to flow in the battery and results in a large amount of heat generation.56 The function of CCs is to retain the active material on CCs and to efficiently collect and distribute electrons from/to the active material layers. For safety, it is necessary to retain the materials at all times. In addition, CCs with a function that

334

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

Fig. 3 Electrode processing. Illustration of the filtration process in two steps: (a) suspension filtration on the CC, (b) filtration cake on the CC, (c) binder impregnation, and (d) electrode obtained. (e) Pierced current collector with burrs and in the inset, a detail of the burrs.47 Reproduced with permission from Zolin, L.; Chandesris, M.; Porcher, W.; Lestriez, B. An Innovative Process for Ultra-Thick Electrodes Elaboration: Toward Low-Cost and High-Energy Batteries. Energy Technol. 2019, 7, 1900025. © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Fig. 4 Schematic illustration of the fabrication process to produce the porous alumina/C anode material.49 Reproduced with permission from Tong, X.; Zhang, F.; Ji, B.; Sheng, M.; Tang, Y. Carbon-Coated Porous Aluminum Foil Anode for High-Rate, Long-Term Cycling Stability, and High Energy Density Dual-Ion Batteries. Adv. Mater. 2016, 28, 9979–9985. © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

335

can interrupt the flow of current when large current flows can contribute to improving safety. In recent years, due to growing interest in the importance of safety in LIBs, many studies have been performed on ensuring safety with CCs. The safety concept is that the CCs will separate into smaller pieces in the event of an anomaly such that they do not carry current to each other. Concavities and convexities are formed on the surface of the CCs from the stage of producing the anodes and cathodes. This unevenness acts as a cut and breaks apart in an abnormal situation. Currently, several CC designs have been proposed that do not ignite during the nail penetration test.57,58 Fig. 5 compares the decomposition of the CC foils that occurs when impact is applied to the center of a conventional CC foil and a CC foil with a processed surface. Furthermore, the charging and discharging characteristics of electrodes prepared with modified CCs do not deteriorate even if the surface of the CC foil is processed.57 In the same way that deformation and decomposition of CCs are used to prevent fires due to battery failure, the use of CCs coated with Al or Cu metal have been proposed. By adopting this separator/CC composite (SCC) structure, current does not flow in the thickness direction of the CC foils. Therefore, when a nail is stuck in the electrode, as shown in Fig. 6, a large short-circuit current flows in the conventional electrode prepared (left) with nonmodified CCs. However, in this structure, the plastic part has the effect of blocking the flow of current (right).59 Methods have been devised in which the polymer material used below the CC is positively altered for battery safety. A shape-memory polymer is used for the base of the CC, and when the internal temperature of the battery reaches 90  C, the polymer material deforms into a protrusion-like shape. The copper layer deposited on the shape-memory polymer layer is divided by the protrusions, thereby interrupting the flow of the current and preventing the battery from running out of control (Fig. 7).60

Fig. 5 Photos of reference (left) and modified (right) CCs after impact testing.57 Reproduced with permission from Wang, M.; Le, A. V.; Shi, Y.; Noelle, D. J.; Qiao, Y. Heterogeneous Current Collector in Lithium-Ion Battery for Thermal-Runaway Mitigation. Appl. Phys. Lett. 2017, 110, 083902. © 2017 AIP Publishing.

Al layers Polymer layer

Al foil Cathode-active layer Separator Anode-active layer Cu foil

Traditionary Li-ion batteries with AI-foil current collector

Li-ion batteries with highly deformable separator/current collector

Fig. 6 Illustration of the working principle of highly stretchable SCC.59 Reproduced with permission from Liu, Z.; Dong, Y.; Qi, X.; Wang, R.; Zhu, Z.; Yan, C.; Jiao, X.; Li, S.; Qie, L.; Li, J.; Huang, Y. Stretchable Separator/Current Collector Composite for Superior Battery Safety. Energy Environ. Sci. 2022, 15, 5313–5323. © 2022 The Royal Society of Chemistry.

336

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

Fig. 7 Design of the micropatterned and shape-memorized current collectors. (a) Self-shutdown performance of safe lithium-ion batteries with smart current collectors before thermal runaway. (b) Safe battery internal structure and the trigger mechanism of the automatic cut-out current collector.60 Reproduced with permission from Jia, J.; Liu, H.; Liao, S.; Liu, K.; Wang, Y. Early Braking of Overwarmed Lithium-Ion Batteries by Shape-Memorized Current Collectors. Nano Lett. 2022, 22, 9122−9130. © 2022 American Chemical Society.

5

CCs for aqueous batteries

By changing the solvents of the electrolyte solutions from organic solvents to water, the expensive organic solvents can be replaced with cheap water, and the price of the battery can be reduced. In addition, safety is also improved because no organic solvent with a risk of ignition is used.61 The use of aqueous solutions not only causes advantages but also reduces the battery voltage, which is constrained by the potential window of water; additionally the electrode performance is reduced due to the reaction between the electrode active material and water. Therefore, there are restrictions on the active materials that can be used in aqueous batteries (ABs), such as Ni-rich cathode materials, because the materials are dissolved when they contact water. The problem with current CCs is that they tend to corrode in aqueous solutions. Furthermore, to widen the potential window and increase the battery voltage, the application of concentrated electrolyte solutions has recently been studied,62,63 but a risk of corrosion of the CC can occur in this concentrated electrolyte solution.

5.1

Metal CCs

Suo et al. presented interesting results on the oxidation behavior of conventional CC materials, such as Ni, Ti, Al and SS, in aqueous solutions (Fig. 8).64 Their results showed that the corrosion potential of conventional CCs was significantly lower than the charge–discharge voltage of cathode materials. Therefore, when these materials were used in ABs, there was a high possibility that CCs would corrode. However, another paper stated that an Al CC could be used in an aqueous solution of 21 mol kg−1 lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) without corrosion of the Al CC surface.65 As shown in many papers, for concentrated electrolyte solutions, materials of Ti, SS and Cr can be used as CCs.66–68 A novel result is the use of silver (Ag) wire foil as the CC. In this case, rather than concentrated electrolyte solutions, a solution classified as a molecular crowding electrolyte of 2 m LiTFSI–94% polyethylene glycol (PEG)–6% H2O is used.

5.2

Carbon CCs

5.2.1 Carbon-coated CCs Although good results have been achieved with the metal CCs, as described above, to prevent the metal CCs from directly contacting aqueous electrolyte solutions, the surfaces of the metal CCs are coated or the metal material is covered with carbon felt.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

337

Fig. 8 Corrosion testing of metal-based CCs (Ni, SS, Ti, and Al) in 1 mol kg−1 aqueous electrolytes (1 mol kg−1 LiNO3, 1 mol kg−1 LiTFSI, 1 mol kg−1 LiCl, and 1 mol kg−1 Li2SO4). Tafel plots of Ni, Ti, Al, and SS in 1 mol kg−1 Li2SO4, LiTFSI, LiNO3, and LiCl solutions and redox potentials of the traditional cathodes versus Li/ Li+ are included.64 Reproduced with permission from Liu, B.; Yue, J.; Lv, T.; Wang, S.; Zhou, A.; Xiong, X.; Suo, L. Sandwich Structure Corrosion-Resistant Current Collector for Aqueous Batteries. ACS Appl. Energy Mater. 2021, 4, 4928−4934. © 2021 American Chemical Society.

Al and SS CCs were sandwiched with two carbon black/polyethylene composite films as cathode CCs for preparing the cathode (carbon layer (carbon black/polyethylene composite (CBPE))-metal-carbon layer (CMC), Fig. 9).64 Spinel lithium manganese oxide (LiMn2O4) cathodes with the carbon-coated SS CC showed extremely high and stable charge–discharge cycle performance compared to the cathode using only SS as the CC. Therefore, this result was caused by the corrosion resistance of the cardon-coated SS CC. By using a slurry method to form a carbon layer on the surface of the SS CC using carbon black or graphite particles and a binder, the corrosion potential of SS shifted to the positive potential side by c.0.2 V.69 A simpler method, uses an Al CC and adds an organic additive to an aqueous electrolyte solution such that the additive is adsorbed on the Al surface to form a protective layer.70

5.2.2 Carbon film CCs There are many examples of CCs made of films or foams made of materials whose main component is carbon materials. Carbon materials have been applied in ABs and are a popular choice for lightweight and flexible CCs. Two representative examples are provided here. The first is a mixture of carbon nanotubes (CNTs) and cellulose nanofibers (CNFs). CNTs contribute to the improvement of the electronic conductivity, and CNFs contribute to the improvement of the mechanical strength of the formed carbon film based on the interaction between CNTs and CNFs (Fig. 10).71 The film production method is a simple method of

338

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

Fig. 9 Design of the CMC CC. (a) Schematic illustration of the CMC CC. (b) Photograph of the CMC CC. (c) Photo of CBPE.64 Reproduced with permission from Liu, B.; Yue, J.; Lv, T.; Wang, S.; Zhou, A.; Xiong, X.; Suo, L. Sandwich Structure Corrosion-Resistant Current Collector for Aqueous Batteries. ACS Appl. Energy Mater. 2021, 4, 4928−4934. © 2021 American Chemical Society.

Fig. 10 (a) Schematic illustrations of the carbon nanotube (CNT)–cellulose nanofiber (CNF), all-fiber-based CCs, where (b) CNTs provide a current pathway and CNFs act as a backbone to enable strength. (c) Association between CNFs and CNTs, which enables the stability of CNTs in the CNF aqueous dispersions and the promising mechanical strength of the CNT–CNF films.71 Reproduced with permission from Luo, W.; Hayden, J.; Jang, S.-H.; Wang, Y.; Zhang, Y.; Kuang, Y.; Wang, Y.; Zhou, Y.; Rubloff, G. W.; Lin, C.-F.; Hu, L. Highly Conductive, Light Weight, Robust, Corrosion Resistant, Scalable, All-Fiber Based Current Collectors for Aqueous Acidic Batteries. Adv. Energy Mater. 2018, 8, 1702615. © 2018 Wiley-VCH Verlag GmbH & Co.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

339

Fig. 11 “Epitaxial welding” for the formation of the welded CNTs (W-CNTs). (a) Schematic of CNTs with a conformal polymer coating and after high-temperature heating. The inset shows a picture of the CNT-polyacrylonitrile (PAN) film under high-temperature (2800 K) Joule heating. SEM images of (b) CNT, (c) CNT-PAN, and (d) W-CNT, showing the microstructure evolution at each step.73 Reproduced with permission from Yao, Y.; Jiang, F.; Yang, C.; Fu, K. K.; Hayden, J.; Lin, C.-F.; Xie, H.; Jiao, M.; Yang, C.; Wang, Y.; He, S.; Xu, F.; Hitz, E.; Gao, T.; Dai, J.; Luo, W.; Rubloff, G.; Wang, C.; Hu, L. Epitaxial Welding of Carbon Nanotube Networks for Aqueous Battery Current Collectors. ACS Nano 2018, 12, 5266−5273. © 2018 American Chemical Society.

vacuum-filtrating the dispersion solution of CNTs and CNFs and hot-pressing the sediment formed on the filter to form a film. Even if the film is bent, it is flexible enough to return to its original shape without breaking. Similarly, a mixture of graphite and amorphous carbon was used to make a carbon CC, which was utilized as a CC in a bipolar-type aqueous rechargeable sodium (Na)-ion battery.72 The second example is based on the structure of CNTs. The CNT structure is very fragile, and its electronic conductivity is low due to the point contact between the CNTs. Therefore, a technique to weld CNTs together with a carbon material has been developed. A polymer is coated on the surface of the carbon nanotubes in the assembled carbon nanotube structure and then baked at a high temperature to carbonize the polymer. By covering the surface inside the structure with the carbon layer, a carbon film with mechanical strength can be obtained (Fig. 11).73

5.3

Summary of carbon CCs

One of the problems with carbon CCs is the low electronic conductivity of the CC. The electrical conductivities of the CCs introduced in this section were 1000 S cm−1 for the Ti-coated CMC CC,64 600 S cm−1 for the CNT/CNF-coated CC,71 and 1300 S cm−1 for the carbon nanotube welding CC.73 In the case of Al, its conductivity is 3774 S cm−17; thus, the electronic conductivity is only approximately half to one third. The improvement of their conductivities is considered necessary. As mentioned above, we do not believe that a conclusion can reached as to whether metal CCs in ABs are fine or whether carbon-based CCs are better, although electron conductivity is reduced. This is because, in each paper, endurance tests of the CCs are conducted to some extent; however, the test results under severe conditions, such as overcharge/overdischarge conditions and low temperature/high temperature conditions, are insufficient to draw solid conclusions. When these data are sufficient, conclusion can then be drawn.

6

Flexible CCs

Flexible batteries are an important factor in high-performance batteries and are expected to have a variety of applications in the future. In flexible batteries, the CCs need to have flexibility. Among flexible CCs, many examples are carbon-based CCs. Carbon-based CCs in water-based batteries can be classified as flexible CCs. In this section, examples of flexible metal CC foils are introduced. Peng et al. proposed using a braided metal wire with a diameter of 50 mm as a CC (Fig. 12).74 Compared to the case where a metal wire with a diameter of approximately 200 mm was used as a CC, the weight energy density was increased by 210%, and the volumetric energy density was increased by 34% due to the enhancement of the transportation of Li+. String-like lithium cobalt

340

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

Fig. 12 Two types of fiber current collectors. (a, b) Schematic of solid fiber current collector (SFCC) and braided fiber current collector (BFCC). (c) Schematic illustration of the preparation of BFCC through a braiding process of multiaxial ultrathin metal wires. (d, e) SEM images showing the top and side views of the SFCC, respectively. (f, g) SEM images showing the top and side views of the BFCC, respectively.74 Reproduced with permission from Huang, X.; Wang, C.; Li, C.; Liao, M.; Li, J.; Jiang, H.; Long, Y.; Cheng, X.; Zhang, K.; Li, P.; Wang, B.; Peng, H. Braided Fiber Current Collectors for High-Energy-Density Fiber Lithium-Ion Batteries. Angew. Chem. Int. Ed. 2023, 62, e202303616. © 2023 Wiley-VCH GmbH.

oxide (LiCoO2) cathodes and graphite anodes were created by filling and coating braided metal wire CCs with active materials. The string-like anode was wrapped with a separator sheet and then twisted with a cathode, resulting in the formation of a string-like battery. The twisted anode/separator/cathode wires were woven into a textile battery. Although this is based on the same concept regarding flexibility, an example of a slightly larger battery is the one proposed by Arias et al.75

7

CCs for Si and Li deposition anodes

Li metal and silicon (Si) have theoretical capacities of 386076 and 420077 mAh g−1, respectively; these values are more than 10 times the capacity of conventionally used graphite, causing them to be very attractive anode materials. There are problems that must be overcome to use these materials as anodes. In Li metal anodes, the problem is dendrite formation during Li deposition (charging process). In the case of silicon anodes, the problem is Si particle shedding; here, the Si particles are removed from the electron conduction path due to volumetric changes caused by swelling and shrinkage of the Si particles during charging and discharging. Various modifications have been applied to the CC surface to prevent dendrite formation and Si particle shedding.

7.1

CCs for Li deposition anodes

One improvement in Li dendrite deposition is to prevent current concentration, that is, to increase the anode areas. However, even if the surface area is increased, dendrite deposition still occurs. In this case, the surface is prone to locally deposit Li. The anode surface material is another important factor. Xie et al. proposed the use of Ni foam for increasing the anode surface area and graphene surfaces that cover the Ni foam surfaces (graphene@Ni foam).78 The graphene coating was prepared by reducing methane gas as a carbon source in a hydrogen gas and high-temperature atmosphere. When comparing the charge–discharge efficiency of the Cu foil, Ni foam, and graphene@Ni foam anode CCs, the efficiency of the Cu foil decreased significantly as the cycles progressed, while the efficiency of the Ni foam gradually decreased as the cycles progressed. However, the graphene@Ni foam anode CC showed a stable efficiency of over 90% for up to 100 cycles, demonstrating its effectiveness (Fig. 13). The suppression of Li dendrite formation in a Cu porous structure formed by dealloying Zn using a Cu-Zn alloy substrate was reported; this structure is an example a three-dimensional CC.79 A laminate of Cu microparticles was also used as the

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

341

Fig. 13 (Up) Schematic illustrations of the comparison of Li deposition on Cu foil, Ni foam and graphene@Ni foam surfaces. (Below) Coulombic efficiency of Li deposition vs. cycle number on Cu foil, Ni foam and graphene@Ni foam surfaces.78 Reproduced with permission from Xie, K.; Wei, W.; Yuan, K.; Lu, W.; Guo, M.; Li, Z.; Song, Q.; Liu, X.; Wang, J.-G.; Shen, C. Toward Dendrite-Free Lithium Deposition via Structural and Interfacial Synergistic Effects of 3D Graphene@Ni Scaffold. ACS Appl. Mater. Interfaces 2016, 8, 26091−26097. © 2016 American Chemical Society.

three-dimensional structure for anodes.80 Examples of inhibition by surface properties that inhibit the formation of Li dendrites include surface modification with organic molecules, N-doped graphene,81 black phosphorous82 and polydopamine,83 surface modification with inorganic materials and structures, Cu-CuO-Ni hybrid structures,84 silver nanoparticle layers on the Cu foils85 and dependence of the formation of Li dendrites on the crystal orientation of the Cu CCs.86

7.2

CCs for Si anodes

Two methods for CCs have been investigated for stabilizing the high charge/discharge capacity of Si anodes. The first method is to strengthen the bond by modifying the surface of the CC to maintain the electronic connection between the Si anode layer and the CC. The second point for the methods is the use of CCs to create a three-dimensional structure that allows volume changes such that the Si anode layer does not collapse due to volume changes that accompany the swelling and contraction of the Si anode layer. Li-substituted polyacrylic acid (LiPAA) was used to form the interface layer for binding the Si anode layer and CC surface, and the interface layer was created in advance by making a LiPAA slurry; next, the slurry was coated on the surface of the Cu CC foil and dried. Finally, the Si anode layer was added to the interface layer.87 To increase the adhesion area between the Cu CC and the Si layer, a three-dimensional structure was created on the surface of the Cu CC with Cu microparticles.88 The basic concept of a three-dimensional structure that allows the Si active material to swell and contract and to exhibit stable charge/discharge cycles is to secure space for swelling in advance. Swelling of the Si active material causes deformation of the active material layer, which destroys the electron transmission path and leads to a significant reduction in performance. Wang et al. proposed a method in which the CC itself became swollen and contracted in response to the volume change accompanying the swelling and contraction of the silicon layer during charging and discharging.89 A highly elastic silicone rubber (SR) substrate served as the base material for CCs. After coating a silicon film with polydopamine (PDA), it became a deformable SR-Copper (DSRC) CC for the Si anode by coating it with a Cu layer (Fig. 14(a)). During the charging process, the deformable SR layer in the DSRC CC contracted when the Si active material layer was swollen. However, during the discharging process after charging, voids were created in the Si active material layer, which disrupted the electron conduction path in the Si active material layer. However, as the deformable SR layer in the DSRC CC became swollen, the Si active material layer was compressed, and it became a compressed Si active material layer before charging. The deformable SR layer played the role of a spring to maintain the electronic connection in the Si active material layer (Fig. 14(b)).

342

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

Fig. 14 Schematic of the fabrication process and working principle for the deformable SR-Copper (DSRC) current collector. (a) Current collector design and the consequent fabrication procedures from a silicone rubber substrate to a silicone rubber-PDA to an SR-Copper. (b) Schematic showing the intercalate/deintercalate Li process: DSRC can effectively resist anode volume expansion and contraction, maintaining tight electrical contact between the anode and separator/SSE.89 Reproduced with permission from Liang, P.; Huang, Z.; Chen, L.; Shao, G.; Wang, H.; Sun, H.; Wang, C.-A. Highly Elastic and Low Resistance Deformable Current Collectors for Safe and High-Performance Silicon and Metallic Lithium Anodes. J. Power Sources 2021, 511, 230418. © 2021 Elsevier B.V.

8 8.1

CCs for post-LIBs Sodium batteries

In Na batteries, the dissolution and deposition of Na metal is used as the anode reaction. Since Na does not alloy with Al, it can be used in place of a Cu CC as the anode CC. Since Al has a lower specific gravity and is cheaper than Cu, changing the anode CC materials from Cu to Al is effective in terms of cost and weight energy density in practical batteries.90 By using Al as a CC for a Na metal anode, it is possible to use a structure that can further suppress the formation of dendrites during Na deposition by utilizing the porous structure of the Al surface.91 NaxSn-based anodes are used in Na-ion batteries, which are made by depositing tin on Al CCs.92 Al CCs can be used as the CCs for cathodes. However, the use of Al CCs is limited by the selection of optimal solvents and sodium salts.93,94

8.2

Magnesium batteries

Magnesium (Mg) batteries have a higher volumetric energy density than LIBs; since Mg are metals that are stable in the air, they are also highly safe. Mg is 24 times less expensive than Li. If Mg metal is used for anodes, almost no dendrite formation of Mg occurs. Various promising Mg batteries have been proposed.95 For the practical use of Mg batteries, the problem of corrosion of the CCs

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

343

needs to be solved. This occurs because the organic electrolytes in Mg batteries contain high concentrations of chloride ions (Cl−); thus, the CCs are easily corroded due to the strong corrosive nature of Cl−. Due to the problems with Al as CCS for cathodes,96 studies are being conducted on alternative CCs.97,98 The results of using Ti,99 niobium (Nb),100 tungsten (W), and molybdenum (Mo)101 as CC materials for the cathodes of Mg batteries have been reported. It has also been proposed that conventional CC materials can be used by adding additives102 or changing the electrolyte.103

8.3

Calcium batteries

Similar to the abovementioned post-LIBs, calcium (Ca) ion batteries and Ca batteries also have a problem in which there is a risk of corrosion of the CCs depending on the type of electrolyte, and improvements are being investigated. Ca also does not form an alloy with Al; thus, it is theoretically possible to use Al as anode CCs. Al, similar to other post-LIBs, has a high risk of corrosion by the Ca salts. Although SS, which is conventionally used as a CC, also has corrosion problems when used as a CC for cathodes, it has been reported that graphite sheets are suitable CCs.104 Graphite CC foils have also been proposed because metal oxides on the surface of the CC act as catalysts for the corrosion reactions of the CC surfaces when metal CCs are used. Since there is still some uncertainty regarding the electrolyte solution for the Ca ions and Ca batteries, the optimal CCs for the batteries have not yet been determined. Although the prevention of corrosion of the CCs is possible by forming a protective film on the surface of the CCs using additives, the electrolyte solution for this purpose needs to be determined first.

8.4

Aluminum batteries

An Al battery consists of an Al metal anode, chloroaluminate ionic liquid electrolytes, and a cathode into which Al ion complexes are intercalated and desorbed to/from cathode materials in discharging and charging processes, respectively. Since chloroaluminate ionic liquid electrolytes are used as the electrolyte solution, the problem of corrosion of cathode CCs is a major barrier to practical application. Conventional Al and SS CCs rapidly corrode during charging and discharging. For this reason, research results have been reported using CCs made of expensive metals that do not corrode, such as Mo,105 W106 and TiNx.107 Although these metals do not corrode, they are expensive and therefore increase the price of the Al battery itself.108 For the purpose of preventing corrosion and cutting the cost of CCs, many studies have been reported on carbon-based CCs.109,110

8.5

Sulfur batteries

Sulfur batteries are defined as those that use the sulfur redox reaction as a cathode reaction. These batteries are attracting attention because they are anticipated to have charge/discharge capacities that greatly exceed the charge/discharge capacity of conventional lithium metal oxide cathodes when using the redox reaction of sulfur.111 Problems with sulfur cathodes include the low conductivity of the sulfur electrode, a decrease in the capacity due to the active material polysulfide dissolution from the cathodes and shuttling; these lead to reduced cycle life and stability. 3D CCs, such as Ni foam,33 are used to keep the active materials in the cathodes and to maintain as much contact between the electron-conductive CC surface and the cathode material as possible. As an alternative to 3D CCs, the use of the cathode layers containing a high proportion of carbon nanofibers mixed with cathode active materials has also been proposed.112 Methods have also been reported in which the cathode active materials are retained in the cathodes by utilizing a special interaction between the CC surfaces and the cathode active material. Lu et al. reported that by immersing a 3D Cu CC in a polysulfide solution, nanoflake-like Cu2S was formed on the surface of the CC, resulting in stable charge–discharge test results for the cathode.113 Mitra et al. proposed the use of carbon cloth CC doped with nitrogen and sulfur for immobilization of cathode active material in Mg-S batteries. The Mg-S battery could maintain a capacity of 388 mAh g−1 after 40 cycles in charge/discharge cycle tests.114

9

CCs for all-solid-state batteries

Batteries in which the conventional liquid electrolyte solution is replaced with solid electrolytes having ion conductivity are called all-solid-state batteries. All-solid-state batteries are expected to be developed because they provide significant improvements over conventional liquid-based batteries in terms of battery performance, safety, and performance stability.115,116 A type of thin film all-solid-state battery used a cathode layer, a solid electrolyte layer, an anode layer; these are stacked one after another using a vapor deposition method or a similar method. Another type builds up layers consisting of the cathode material particles, solid electrolyte particles, and anode material particles (or negative electrode metal). Each layer produced by a vapor deposition method or a similar method needs to be frequently subjected to a crystallization treatment for the anode, cathode and solid-electrolyte materials to impart a function that allows charging and discharging reactions to occur. When heat treatment is performed, elements in the CCs may react with the active materials, or conversely, elements in the active materials may diffuse into the CCs. As a result, problems occur, such as a decrease in charge/discharge capacity and an increase in resistance at the interface between the active materials and the CCs. As a method for preventing these problems, the development of CCs with a composition that does not react with the cathode and anode active materials during heat treatment is needed. The use of CCs with compositions, such as CrxN117 and

344

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

Ni-Al-Cr,42 has been proposed. Some researchers reported that by crystallizing the cathode active material using a firing method called flash lamp annealing, a battery could be produced without the firing temperature being transmitted to the Al CC on which cathode material layers were formed.118 In another method, the particles of a cathode material, an anode material, and a solid electrolyte are stacked up, and a cathode CC and an anode CC are needed. When conventional metals are used as the CCs, the CCs and the anode and cathode materials come into direct contact, which causes problems in low electronic conductivity. To increase the number of contact points, the sandwich of the cathode layer, solid electrolyte layer, and anode layer is pressed under high pressure to increase the number of contact points. Furthermore, to increase the number of contact points, carbon-based CCs were used, which were known to have deformation behavior.119 One of the advantages of using all-solid-state batteries is that bipolar batteries can be assembled. In the bipolar battery, cathode active material and anode active material layers are formed back to back on both sides of a CC.120 A battery was fabricated using LiMn0.85Fe0.1Mg0.05PO4 as the cathode active material and Li4Ti5O12 as the anode active material on both sides of an Al CC foil along with cubic garnet-type Li7La3Zr2O12 and a gel polymer as a hybrid electrolyte. A 12 V-class battery could be created by stacking five bipolar electrodes with electrolyte layers.121 Conventionally, an Al CC is used for the cathode, and a Cu CC is used for the anode; thus, as an example, a cladding material of Al and Cu was used as the CCs of the bipolar electrode.122 When preparing a bipolar electrode in combination with a Li metal anode, an SUS CC foil can be used; here, Li metal foil is attached to the SUS CC foil, and a sulfur cathode layer is formed on the other surface of the SUS CC foil.123

10

Conclusions

The CCs for the cathode and anode electrodes are conventionally made of Al and Cu metals. Foils have been used as CCs; however, since foil materials cannot store electricity, the trend for years has been to produce thinner CCs to improve the energy density. Carbon-based CCs are being actively studied to reduce the price and weight of conventional metal-based CC foils. Carbon-based CCs are also being studied to avoid corrosion; this is a problem that CCs always have. Until now, CCs have played an unacknowledged role with the aim of improving the energy density and safety of batteries; improvements need to consider three-dimensional CCs and electrode safety. There is a growing trend toward proactively changing fundamental aspects, such as the structure of the CCs and modification of the surface of the CCs. On the other hand, the development of water-based batteries, post-LIBs, and all-solid-state batteries is progressing with the aim of improving the battery energy density and input/output characteristics, along with lowering prices. Major changes are needed in the characteristics and structure of the CCs compatible with these batteries. Among these, the CC characteristics are greatly involved in battery performance, and the importance of CCs is increasing. For future directions, the CCs will change from metal CCs to carbon-based CCs; eventually, they will be incorporated into the structure of the cathodes and anodes, such that it is unclear if they can be classified as CCs. To improve battery performance, various expensive materials are used, and the CC structure is complicated. The performance of these batteries is very attractive when compared to conventional batteries; however, their practical use seems unlikely since the mass production of these batteries is not easily accomplished. The resolution of these disadvantages is the major challenge moving forward.

References 1. Li, H.; Wang, L.; Song, Y.; Zhang, Z.; Zhang, H.; Du, A.; He, X. Significance of Current Collectors for High Performance Conventional Lithium-Ion Batteries: A Review. Adv. Funct. Mater. 2023, 33, 2305515. 2. Yamada, M.; Watanabe, T.; Gunji, T.; Wu, J.; Matsumoto, F. Review of the Design of Current Collectors for Improving the Battery Performance in Lithium-Ion and Post-lithium-Ion Batteries. Electrochem 2020, 2, 124–159. 3. Lamiel, C.; Hussain, I.; Ma, X.; Zhang, K. Properties, Functions, and Challenges: Current Collectors. Mater. Today Chem. 2022, 26, 101152. 4. Jin, S.; Jiang, Y.; Ji, H.; Yu, Y. Advanced 3D Current Collectors for Lithium-Based Batteries. Adv. Mater. 2018, 30, 1802014. 5. Issatayev, N.; Nuspeissova, A.; Kalimuldina, G.; Bakenov, Z. Three-Dimensional Foam-Type Current Collectors for Rechargeable Batteries: A Short Review. J. Power Sources Adv. 2021, 10, 100065. 6. Zhang, Z.; Song, Y.; Zhang, B.; Wang, L.; He, X. Metallized Plastic Foils: A Promising Solution for High-Energy Lithium-Ion Battery Current Collectors. Adv. Energy Mater. 2023, 13, 2302134. 7. Zhu, P.; Gastol, D.; Marshall, J.; Sommerville, R.; Goodship, V.; Kendrick, E. A Review of Current Collectors for Lithium-Ion Batteries. J. Power Sources 2021, 485, 229321. 8. Whitehead, A. H.; Schreiber, M. Current Collectors for Positive Electrodes of Lithium-Based Batteries. J. Electrochem. Soc. 2005, 152, A2105–A2113. 9. Hendricks, C. E.; Mansour, A. N.; Fuentevilla, D. A.; Waller, G. H.; Ko, J. K.; Pecht, M. G. Copper Dissolution in Overdischarged Lithium-Ion Cells: X-Ray Photoelectron Spectroscopy and X-Ray Absorption Fine Structure Analysis. J. Electrochem. Soc. 2020, 167, 090501. 10. Guo, L.; Thornton, D. B.; Koronfel, M. A.; Stephens, I. E. L.; Ryan, M. P. Degradation in Lithium Ion Battery Current Collectors. J. Phys. Energy 2021, 3, 032015. 11. Zhao, M.; Kariuki, S.; Dewald, H. D.; Lemke, F. R.; Staniewicz, R. J.; Plichta, E. J.; Marsh, R. A. Electrochemical Stability of Copper in Lithium-Ion Battery Electrolytes. J. Electrochem. Soc. 2000, 147, 2874–2879. 12. Myunga, S.-T.; Sasaki, Y.; Sakurada, S.; Sun, Y.-K.; Yashiro, H. Electrochemical Behavior of Current Collectors for Lithium Batteries in Non-aqueous Alkyl Carbonate Solution and Surface Analysis by ToF-SIMS. Electrochim. Acta 2009, 55, 288–297. 13. Kim, Y.-S.; Lee, S.-H.; Son, M.-Y.; Jung, Y. M.; Song, H.-K.; Lee, H. Succinonitrile as a Corrosion Inhibitor of Copper Current Collectors for Overdischarge Protection of Lithium Ion Batteries. ACS Appl. Mater. Interfaces 2014, 6, 2039–2043. 14. Chen, K.-T.; Yang, Y.-C.; Yi, Y.-H.; Zheng, X.-T.; Tuan, H.-Y. A Carbon Ink for Use in Thin, Conductive, Non Peelable, Amphiphilic, Antioxidant, and Large-Area Current Collector Coating with Enhanced Lithium Ion Battery Performance. J. Colloid Interface Sci. 2021, 598, 155–165. 15. Jiang, X.; Chen, J.; Yang, Y.; Lv, Y.; Ren, Y.; Li, W.; Li, C. Corrosion Protection of Copper Current Collector of Lithium Ion Batteries by Doped Polypyrrole Coatings. Int. J. Electrochem. Sci. 2020, 15, 2667–2676.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

345

16. Park, K.; Yu, S.; Lee, C.; Lee, H. Comparative Study on Lithium Borates as Corrosion Inhibitors of Aluminum Current Collector in Lithium Bis(Fluorosulfonyl)Imide Electrolytes. J. Power Sources 2015, 296, 197–203. 17. Liu, X.; Shen, C.; Gao, N.; Hou, Q.; Song, F.; Tian, X.; He, Y.; Huang, J.; Fang, Z.; Xie, K. Concentrated Electrolytes Based on Dual Salts of LiFSI and LiODFB for Lithium-Metal Battery. Electrochim. Acta 2018, 289, 422–427. 18. Kawamura, T.; Tanaka, T.; Egashira, M.; Watanabe, I.; Okada, S.; Yamaki, J. Methyl Difluoroacetate Inhibits Corrosion of Aluminum Cathode Current Collector for Lithium Ion Cells. ECS Solid State Lett. 2005, 8, A459–A463. 19. Shkrob, I. A.; Pupek, K. Z.; Abraham, D. P. Allotropic Control: How Certain Fluorinated Carbonate Electrolytes Protect Aluminum Current Collectors by Promoting the Formation of Insoluble Coordination Polymers. J. Phys. Chem. C 2016, 120, 18435–18444. 20. Li, X.; Deng, S.; Banis, M. N.; Davis, K. D.; Zhang, D.; Zhang, T.; Yang, J.; Divigalpitiya, R.; Brandys, F.; Li, R.; Sun, X. Suppressing Corrosion of Aluminum Foils Via Highly Conductive Graphene-like Carbon Coating in High-Performance Lithium-Based Batteries. ACS Appl. Mater. Interfaces 2019, 11, 32826–32832. 21. Zhou, Y.; Hu, J.; He, P.; Zhang, Y.; Xu, J.; Wu, X. Corrosion Suppression of Aluminum Metal by Optimizing Lithium Salt Concentration in Solid-State Imide Salt-Based Polymer Plastic Crystal Electrolyte Membrane. ACS Appl. Energy Mater. 2018, 1, 7022–7027. 22. Li, Y.; Zhang, X.-W.; Khan, S. A.; Fedkiw, P. S. Attenuation of Aluminum Current Collector Corrosion in LiTFSI Electrolytes Using Fumed Silica Nanoparticles. ECS Solid State Lett. 2004, 7, A228–A230. 23. Loghavi, M. M.; Askari, M.; Babaiee, M.; Ghasemi, A. Improvement of the Cyclability of Li-Ion Battery Cathode Using a Chemical Modified Current Collector. J. Electroanal. Chem. 2019, 841, 107–110. 24. Nakamura, T.; Okano, S.; Yaguma, N.; Morinaga, Y.; Takahara, H.; Yamada, Y. Electrochemical Performance of Cathodes Prepared on Current Collector with Different Surface Morphologies. J. Power Sources 2013, 244, 532–537. 25. Wu, H.-C.; Wu, H.-C.; Lee, E.; Wu, N.-L. High-Temperature Carbon-Coated Aluminum Current Collector for Enhanced Power Performance of LiFePO4 Electrode of Li-Ion Batteries. Electrochem. Commun. 2010, 12, 488–491. 26. Swain, P.; Viji, M.; Mocherla, P. S. V.; Sudakar, C. Carbon Coating on the Current Collector and LiFePO4 Nanoparticles - Influence of sp2 and sp3-like Disordered Carbon on the Electrochemical Properties. J. Power Sources 2015, 293, 613–625. 27. Liu, T.; Cao, F.; Ren, L.; Li, X.; Sun, S.; Sun, X.; Zang, Z.; Niu, Q.; Wu, J. A Theoretical Study of Different Carbon Coatings Effect on the Depolarization Effect and Electrochemical Performance of LiFePO4 Cathode. J. Electroanal. Chem. 2017, 807, 52–58. 28. Busson, C.; Blin, M.-A.; Guichard, P.; Soudan, P.; Crosnier, O.; Guyomard, D.; Lestriez, B. A Primed Current Collector for High Performance Carbon-Coated LiFePO4 Electrodes with no Carbon Additive. J. Power Sources 2018, 406, 7–17. 29. Ding, Y.; Li, J.; Zhao, Y.; Guan, L. Direct Synthesis of Iron Oxide Nanoparticles on an Iron Current Collector as Binder-Free Anode Materials for Lithium-Ion Batteries. Mater. Lett. 2012, 81, 105–107. 30. Kim, J.; Lee, S.; Kim, K. W.; Son, J.; Mun, J. Electrochemical Corrosion Behavior of Iron in Lithium-Ion Battery Electrolyte. J. Electrochem. Sci. Technol. 2021, 12, 424–430. 31. Geaney, H.; McNulty, D.; O’Connell, J.; Holmes, J. D.; O’Dwyer, C. Assessing Charge Contribution From Thermally Treated Ni Foam as Current Collectors for Li-Ion Batteries. J. Electrochem. Soc. 2016, 163, A1805–A1811. 32. Sa, Q.; Wang, Y. Ni foam as the current collector for high capacity C–Si composite electrode. J. Power Sources 2012, 208, 46–51. 33. Liu, L.-J.; Chen, Y.; Zhang, Z.-F.; You, X.-L.; Walle, M. D.; Li, Y.-J.; Liu, Y.-N. Electrochemical Reaction of Sulfur Cathodes with Ni Foam Current Collector in Li-S Batteries. J. Power Sources 2016, 325, 301–305. 34. Zhou, Q.-Y.; Tan, L.; Lv, T.-B.; Li, M.-C.; Zhang, J.-J.; Zhao, Z.-Q.; Jin, X.-J.; Liu, Z.; Hou, P.-P.; Zeng, Z.; Deng, S.; Dai, G.-P. Nickel Foam Coated by Ni Nanoparticle-Decorated 3D Nanocarbons as a Freestanding Host for High-Performance Lithium−Sulfur Batteries. ACS Appl. Mater. Interfaces 2023, 15, 3037–3046. 35. Kanamura, K.; Hoshikawa, W.; Umegaki, T. Electrochemical Characteristics of LiNi0.5Mn1.5O4 Cathodes with Ti or Al Current Collectors. J. Electrochem. Soc. 2002, 149, A339–A345. 36. Piao, N.; Wang, L.; He, X. Anodic Stabilities of Various Metals as the Current Collector in High Concentration Electrolytes for Lithium Batteries. J. Electrochem. Soc. 2021, 168, 030509. 37. Choi, H.; Park, H.; Um, J. H.; Yoon, W.-S.; Choe, H. Processing and Characterization of Titanium Dioxide Grown on Titanium Foam for Potential Use as Li-Ion Electrode. Appl. Surf. Sci. 2017, 411, 363–367. 38. Hua, K.; Fang, D.; Bao, R.; You, X.; Tao, J.; Li, C.; Liu, Y.; Luo, Z.; Yi, J.; Shu, Y.; Sun, B. Porous iron Vanadate Nanowire Arrays on Ti Foil as a High-Performance Lithium-Ion Battery. Appl. Surf. Sci. 2019, 465, 1047–1054. 39. Wang, C.-H.; Kurra, N.; Alhabeb, M.; Chang, J.-K.; Alshareef, H. N.; Gogotsi, Y. Titanium Carbide (MXene) as a Current Collector for Lithium-Ion Batteries. ACS Omega 2018, 3, 12489–12494. 40. Zhao, H.; Yang, P.; Du, M.; He, W.; Liu, M.; Yan, W.; Zhou, L.; Liu, J.; Gu, Z.; Zou, Z. An Ultrathin Surface-Nitrided Porous Titanium Sheet as a Current Collector-Free Sulfur Host for High-Gravimetric-Capacity Lithium–Sulfur Batteries. Chem. Commun. 2019, 55, 1655–1658. 41. Cho, K.; Baek, J.; Balamurugan, C.; Im, H.; Kim, H.-J. Corrosion Study of Nickel-Coated Copper and Chromate-Coated Aluminum for Corrosion-Resistant Lithium-Ion Battery Lead-Tab. J. Ind. Eng. Chem. 2022, 106, 537–545. 42. Filippin, A. N.; Lin, T.-Y.; Rawlence, M.; Zünd, T.; Kravchyk, K.; Pellicer, J. S.; Haass, S. G.; Wäckerlin, A.; Kovalenko, M. V.; Buecheler, S. Ni–Al–Cr Superalloy as High Temperature Cathode Current Collector for Advanced Thin Film Li Batteries. RSC Adv. 2018, 8, 20304–20313. 43. Furukawa, K.; Yoshimoto, N.; Egashira, M.; Morita, M. Anodic Behavior of Stainless-Steel Substrate in Organic Electrolyte Solutions Containing Different Lithium Salts. Electrochim. Acta 2014, 140, 125–131. 44. Shironita, S.; Ihsan, N.; Konakawa, K.; Souma, K.; Umeda, M. Investigation of Nitriding Treated Ni-Free Stainless Steel as Current Collector for 5 V-Class Li-Ion Secondary Cell. Electrochim. Acta 2019, 295, 1052–1056. 45. Ni, S.; Li, T.; Lv, X.; Yang, X.; Zhang, L. Designed Constitution of NiO/Ni Nanostructured Electrode for High Performance Lithium Ion Battery. Electrochim. Acta 2013, 91, 267–274. 46. Yao, M.; Okuno, K.; Iwaki, T.; Kato, M.; Tanase, S.; Emura, K.; Sakai, T. LiFePO4-Based Electrode Using Micro-Porous Current Collector for High Power Lithium Ion Battery. J. Power Sources 2007, 173, 545–549. 47. Zolin, L.; Chandesris, M.; Porcher, W.; Lestriez, B. An innovative process for ultra-thick electrodes elaboration: toward low-cost and high-energy batteries. Energy Technol. 2019, 7, 1900025. 48. Aguayo, N. A.; Espiñeira, P. P.; Manian, A. P.; Bechtold, T. Three-Dimensional Embroidered Current Collectors for Ultra-Thick Electrodes in Batteries. RSC Adv. 2016, 6, 69685. 49. Tong, X.; Zhang, F.; Ji, B.; Sheng, M.; Tang, Y. Carbon-Coated Porous Aluminum Foil Anode for High-Rate, Long-Term Cycling Stability, and High Energy Density Dual-Ion Batteries. Adv. Mater. 2016, 28, 9979–9985. 50. Chen, C.; Zhang, Y.; Li, Y.; Kuang, Y.; Song, J.; Luo, W.; Wang, Y.; Yao, Y.; Pastel, G.; Xie, J.; Hu, L. Highly Conductive, Lightweight, Low-Tortuosity Carbon Frameworks as Ultrathick 3D Current Collectors. Adv. Energy Mater. 2017, 7, 1700595. 51. Zhang, D.; Dai, A.; Fan, B.; Li, Y.; Shen, K.; Xiao, T.; Hou, G.; Cao, H.; Tao, X. Y.; Tang, Y. Three-Dimensional Ordered Macro/Mesoporous Cu/Zn as a Lithiophilic Current Collector for Dendrite-Free Lithium Metal Anode. ACS Appl. Mater. Interfaces 2020, 12, 31542–31551. 52. Liu, X.; Zhang, R.; Yu, W.; Yang, Y.; Wang, Z.; Zhang, C.; Bando, Y.; Golberg, D.; Wang, X.; Ding, Y. Three-Dimensional Electrode with Conductive Cu Framework for Stable and Fast Li-Ion Storage. Energy Stor. Mater. 2018, 11, 83–90. 53. Rocha, V. G.; Tuñón, E. G.; Botas, C.; Markoulidis, F.; Feilden, E.; D’Elia, E.; Ni, N.; Shaffer, M.; Saiz, E. Multimaterial 3D Printing of Graphene-Based Electrodes for Electrochemical Energy Storage Using Thermoresponsive Inks. ACS Appl. Mater. Interfaces 2017, 9, 37136–37145.

346

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

54. Zhang, C.; Shen, K.; Li, B.; Li, S.; Yang, S. Continuously 3D Printed Quantum Dot-Based Electrodes for Lithium Storage with Ultrahigh Capacities. J. Mater. Chem. A 2018, 6, 19960. 55. Choi, S.; Kim, T.-H.; Lee, J.-I.; Kim, J.; Song, H.-K.; Park, S. General Approach for High-Power Li-Ion Batteries: Multiscale Lithographic Patterning of Electrodes. ChemSusChem 2014, 7, 3483–3490. 56. Wu, X.; Song, K.; Zhang, X.; Hu, N.; Li, L.; Li, W.; Zhang, L.; Zhang, H. Safety Issues in Lithium Ion Batteries: Materials and Cell Design. Front. Energy Res. 2019, 7, 65. 57. Wang, M.; Le, A. V.; Shi, Y.; Noelle, D. J.; Qiao, Y. Heterogeneous Current Collector in Lithium-Ion Battery for Thermal-Runaway Mitigation. Appl. Phys. Lett. 2017, 110, 083902. 58. Naguib, M.; Allu, S.; Simunovic, S.; Li, J.; Wang, H.; Dudney, N. J. Limiting Internal Short-Circuit Damage by Electrode Partition for Impact-Tolerant Li-Ion Batteries. Joule 2018, 2, 155–167. 59. Liu, Z.; Dong, Y.; Qi, X.; Wang, R.; Zhu, Z.; Yan, C.; Jiao, X.; Li, S.; Qie, L.; Li, J.; Huang, Y. Stretchable Separator/Current Collector Composite for Superior Battery Safety. Energy Environ. Sci. 2022, 15, 5313–5323. 60. Jia, J.; Liu, H.; Liao, S.; Liu, K.; Wang, Y. Early Braking of Overwarmed Lithium-Ion Batteries by Shape-Memorized Current Collectors. Nano Lett. 2022, 22, 9122–9130. 61. Bin, D.; Wen, Y.; Wang, Y.; Xia, Y. The Development in Aqueous Lithium-Ion Batteries. J. Energy Chem. 2018, 27, 1521–1535. 62. Zhang, H.; Liu, X.; Li, H.; Hasa, I.; Passerini, S. Challenges and Strategies for High-Energy Aqueous Electrolyte Rechargeable Batteries. Angew. Chem. Int. Ed. 2021, 60, 598–616. 63. Chao, D.; Qiao, S.-Z. Toward High-Voltage Aqueous Batteries: Super- or Low-Concentrated Electrolyte? Joule 2020, 4, 1839–1851. 64. Liu, B.; Yue, J.; Lv, T.; Wang, S.; Zhou, A.; Xiong, X.; Suo, L. Sandwich Structure Corrosion-Resistant Current Collector for Aqueous Batteries. ACS Appl. Energy Mater. 2021, 4, 4928–4934. 65. Kühnel, R.-S.; Reber, D.; Remhof, A.; Figi, R.; Bleiner, D.; Battaglia, C. “Water-in-Salt” Electrolytes Enable the Use of Cost-Effective Aluminum Current Collectors for Aqueous High-Voltage Batteries. Chem. Commun. 2016, 52, 10435–10438. 66. Zhang, G.; Zhou, W.; Chen, M.; Wang, Q.; Li, A.; Xu, J.; Chen, J. Electrochemical Stability of Steel, Ti, and Cu Current Collectors in Water-in-Salt Electrolyte for Green Batteries and Supercapacitors. J. Solid State Electrochem. 2022, 26, 85–95. 67. Zhang, G.; Zhou, W.; Chen, M.; Wang, Q.; Li, A.; Xu, J.; Chen, J. Stainless Steel Foil: A More Appropriate Current Collector than Titanium Foil for the Cathodes of Aqueous Zinc Ion Batteries. Electrochim. Acta 2023, 437, 141519. 68. Gheytani, S.; Liang, Y.; Jing, Y.; Xu, J. Q.; Yao, Y. Chromate Conversion Coated Aluminium as a Lightweight and Corrosion-Resistant Current Collector for Aqueous Lithium-Ion Batteries. J. Mater. Chem. A 2016, 4, 395–399. 69. Wen, Y. H.; Shao, L.; Zhao, P. C.; Wang, B. Y.; Cao, G. P.; Yang, Y. S. Carbon Coated Stainless Steel Mesh as a Low-Cost and Corrosion-Resistant Current Collector for Aqueous Rechargeable Batteries. J. Mater. Chem. A 2017, 5 (15), 15752–15758. 70. Tekaligne, T. M.; Merso, S. K.; Yang, S.-C.; Liao, S.-C.; Tsai, F.-Y.; Fenta, F. W.; Bezabih, H. K.; Shitaw, K. N.; Jiang, S.-K.; Wang, C.-H.; Wu, S.-H.; Su, W.-N.; Hwang, B. J. Corrosion Inhibition of Aluminum Current Collector by a Newly Synthesized 5-Formyl-8-Hydroxyquinoline for Aqueous-Based Battery. J. Power Sources 2022, 550, 232142. 71. Luo, W.; Hayden, J.; Jang, S.-H.; Wang, Y.; Zhang, Y.; Kuang, Y.; Wang, Y.; Zhou, Y.; Rubloff, G. W.; Lin, C.-F.; Hu, L. Highly Conductive, Light Weight, Robust, Corrosion Resistant, Scalable, All-Fiber Based Current Collectors for Aqueous Acidic Batteries. Adv. Energy Mater. 2018, 8, 1702615. 72. Hou, Z.; Mao, W.; Zhang, Z.; Chen, J.; Ao, H.; Qian, Y. Bipolar Electrode Architecture Enables High-Energy Aqueous Rechargeable Sodium Ion Battery. Nano Res. 2022, 15, 5072–5080. 73. Yao, Y.; Jiang, F.; Yang, C.; Fu, K. K.; Hayden, J.; Lin, C.-F.; Xie, H.; Jiao, M.; Yang, C.; Wang, Y.; He, S.; Xu, F.; Hitz, E.; Gao, T.; Dai, J.; Luo, W.; Rubloff, G.; Wang, C.; Hu, L. Epitaxial Welding of Carbon Nanotube Networks for Aqueous Battery Current Collectors. ACS Nano 2018, 12, 5266–5273. 74. Huang, X.; Wang, C.; Li, C.; Liao, M.; Li, J.; Jiang, H.; Long, Y.; Cheng, X.; Zhang, K.; Li, P.; Wang, B.; Peng, H. Braided Fiber Current Collectors for High-Energy-Density Fiber Lithium-Ion Batteries. Angew. Chem. Int. Ed. 2023, 62, e202303616. 75. Zamarayeva, A. M.; Ostfeld, A. E.; Wang, M.; Duey, J. K.; Deckman, I.; Lechêne, B. P.; Davies, G.; Steingart, D. A.; Arias, A. C. Flexible and Stretchable Power Sources for Wearable Electronics. Sci. Adv. 2017, 3, e1602051. 76. Cheng, X.-B.; Zhang, R.; Zhao, C.-Z.; Zhang, Q. Toward Safe Lithium Metal Anode in Rechargeable Batteries: A Review. Chem. Rev. 2017, 117, 10403–10473. 77. Park, C.-M.; Kim, J.-H.; Kim, H.; Sohn, H.-J. Li-Alloy Based Anode Materials for Li Secondary Batteries. Chem. Soc. Rev. 2010, 39, 3115–3141. 78. Xie, K.; Wei, W.; Yuan, K.; Lu, W.; Guo, M.; Li, Z.; Song, Q.; Liu, X.; Wang, J.-G.; Shen, C. Toward Dendrite-Free Lithium Deposition Via Structural and Interfacial Synergistic Effects of 3D Graphene@Ni Scaffold. ACS Appl. Mater. Interfaces 2016, 8, 26091–26097. 79. Zhao, H.; Lei, D.; He, Y.-B.; Yuan, Y.; Yun, Q.; Ni, B.; Lv, W.; Li, B.; Yang, Q.-H.; Kang, F.; Lu, J. Compact 3D Copper With Uniform Porous Structure Derived by Electrochemical Dealloying as Dendrite-Free Lithium Metal Anode Current Collector. Adv. Energy Mater. 2018, 8, 1800266. 80. Chen, J.; Zhao, J.; Lei, L.; Li, P.; Chen, J.; Zhang, Y.; Wang, Y.; Ma, Y.; Wang, D. Dynamic Intelligent Cu Current Collectors for Ultrastable lithium Metal Anodes. Nano Lett. 2020, 20, 3403–3410. 81. Zhang, R.; Wen, S.; Wang, N.; Qin, K.; Liu, E.; Shi, C.; Zhao, N. N-Doped Graphene Modified 3D Porous Cu Current Collector Toward Microscale Homogeneous Li Deposition for Li Metal Anodes. Adv. Energy Mater. 2018, 8, 1800914. 82. Zhao, Y.; Ye, H.; Zhang, H.; Zhao, D.; Huang, L.; Lee, J. Y. The Beneficial Effects of Black Phosphorous Modification of the Anode Current Collector in Li-Metal Free Li2S-Based Batteries. Mater. Today Energy 2022, 30, 101179. 83. He, Y.; Xu, H.; Shi, J.; Liu, P.; Tian, Z.; Dong, N.; Luo, K.; Zhou, X.; Liu, Z. Polydopamine Coating Layer Modified Current Collector for Dendrite-Free Li Metal Anode. Energy Stor. Mater. 2019, 23, 418–426. 84. Wu, S.; Zhang, Z.; Lan, M.; Yang, S.; Cheng, J.; Cai, J.; Shen, J.; Zhu, Y.; Zhang, K.; Zhang, W. Lithiophilic Cu-CuO-Ni Hybrid Structure: Advanced Current Collectors Toward Stable Lithium Metal Anodes. Adv. Mater. 2018, 30, 1705830. 85. Hou, Z.; Yu, Y.; Wang, W.; Zhao, X.; Di, Q.; Chen, Q.; Chen, W.; Liu, Y.; Quan, Z. Lithiophilic Ag Nanoparticle Layer on cu Current Collector Toward Stable Li Metal Anode. ACS Appl. Mater. Interfaces 2019, 11, 8148–8154. 86. Ishikawa, K.; Harada, S.; Tagawa, M.; Ujihara, T. Effect of Crystal Orientation of Cu Current Collectors on Cycling Stability of Li Metal Anodes. ACS Appl. Mater. Interfaces 2020, 12, 9341–9346. 87. Park, J. H.; Ku, J.; Lim, J.; Choi, J.; Son, I. H. Adhesive Interlayer Between Active Film and Current Collector for Improving the Performance of Silicon Anodes of Li-Ion Batteries. J. Electroanal. Chem. 2016, 778, 53–56. 88. Long, J.; Liu, H.; Xie, Y.; Tang, W.; Fu, T.; Tang, Y.; Lu, L.; Ding, X.; Tang, X. Three-Dimensional Copper Foil-Powder Sintering Current Collector for a Silicon-Based Anode Lithium-Ion Battery. Materials 2018, 11, 1338. 89. Liang, P.; Huang, Z.; Chen, L.; Shao, G.; Wang, H.; Sun, H.; Wang, C.-A. Highly Elastic and Low Resistance Deformable Current Collectors for Safe and High-Performance Silicon and Metallic Lithium Anodes. J. Power Sources 2021, 511, 230418. 90. Nayak, P. K.; Yang, L.; Brehm, W.; Adelhelm, P. From Lithium-Ion to Sodium-Ion Batteries: Advantages, Challenges, and Surprises. Angew. Chem. Int. Ed. 2018, 57, 102–120. 91. Liu, S.; Tang, S.; Zhang, X.; Wang, A.; Yang, Q.-H.; Luo, J. Porous Al Current Collector for Dendrite-Free Na Metal Anodes. Nano Lett. 2017, 17, 5862–5868. 92. Slater, M. D.; Kim, D.; Lee, E.; Johnson, C. S. Sodium-ion batteries. Adv. Funct. Mater. 2013, 23, 947–958. 93. Bhide, A.; Hofmann, J.; Dürr, A. K.; Janeka, J.; Adelhelm, P. Electrochemical Stability of Non-aqueous Electrolytes for Sodium-Ion Batteries and Their Compatibility with Na0.7CoO2. Phys. Chem. Chem. Phys. 2014, 16, 1987–1998. 94. Eshetu, G. G.; Grugeon, S.; Kim, H.; Jeong, S.; Wu, L.; Gachot, G.; Laruelle, S.; Armand, M.; Passerini, S. Comprehensive Insights into the Reactivity of Electrolytes Based on Sodium Ions. ChemSusChem 2016, 9, 462–471.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Current Collector

347

95. Saha, P.; Datta, M. K.; Velikokhatnyi, O. I.; Manivannan, A.; Alman, D.; Kumta, P. N. Rechargeable Magnesium Battery: Current Status and Key Challenges for the Future. Prog. Mater. Sci. 2014, 66, 1–86. 96. Cheng, G.; Xu, Q.; Ding, F.; Sang, L.; Liu, X.; Cao, D. Electrochemical Behavior of Aluminum in Grignard Reagents/THF Electrolytic Solutions for Rechargeable Magnesium Batteries. Electrochim. Acta 2013, 88, 790–797. 97. Wall, C.; Karger, Z. Z.; Fichtner, M. Corrosion Resistance of Current Collector Materials in Bisamide Based Electrolyte for Magnesium Batteries. ECS Electrochem. Lett. 2015, 4, C8–C10. 98. Yagi, S.; Tanaka, A.; Ichikawa, Y.; Ichitsubo, T.; Matsubara, E. Electrochemical Stability of Magnesium Battery Current Collectors in a Grignard Reagent-Based Electrolyte. J. Electrochem. Soc. 2013, 160, C83–C88. 99. Tang, H.; Chao, F.; Chen, H.; Jia, R.; Luo, H.; Xiong, F.; Yao, X.; Zhang, W.; Zuo, C.; Wang, J.; Luo, P.; An, Q. Water-Lubricated Aluminum Vanadate for Enhanced Rechargeable Magnesium Ion Storage. Small 2022, 18, 2203525. 100. Cen, Y.; Li, S.; Zhou, Y.; Cai, X.; Wang, X.; Xiang, Q.; Hu, B.; Yu, D.; Liu, Y.; Chen, C. Ultrathin VO2(B) Nanosheets as Cathode Material for High-Performance Hybrid Magnesium-Lithium Ion Batteries. J. Electrochem. Soc. 2019, 166, A1660–A1667. 101. Cheng, Y.; Liu, T.; Shao, Y.; Engelhard, M. H.; Liu, J.; Li, G. Electrochemically Stable Cathode Current Collectors for Rechargeable Magnesium Batteries. J. Mater. Chem. A 2014, 2, 2473–2477. 102. Ha, J. H.; Cho, J.-H.; Kim, J. H.; Cho, B. W.; Oh, S. H. 1-Butyl-1-Methylpyrrolidinium Chloride as an Effective Corrosion Inhibitor for Stainless Steel Current Collectors in Magnesium Chloride Complex Electrolytes. J. Power Sources 2017, 355, 90–97. 103. Ha, S.-Y.; Lee, Y.-W.; Woo, S. W.; Koo, B.; Kim, J.-S.; Cho, J.; Lee, K. T.; Choi, N.-S. Magnesium(II) Bis(Trifluoromethane Sulfonyl) Imide-Based Electrolytes with Wide Electrochemical Windows for Rechargeable Magnesium Batteries. ACS Appl. Mater. Interfaces 2014, 6, 4063–4073. 104. Lipson, A. L.; Proffit, D. L.; Pan, B.; Fister, T. T.; Liao, C.; Burrell, A. K.; Vaughey, J. T.; Ingram, B. J. Current Collector Corrosion in Ca-Ion Batteries. J. Electrochem. Soc. 2015, 162, A1574–A1578. 105. Sun, H.; Wang, W.; Yu, Z.; Yuan, Y.; Wang, S.; Jiao, S. A New Aluminum-Ion Battery With High Voltage, High Safety and Low Cost. Chem. Commun. 2015, 51, 11892. 106. Kravchyk, K. V.; Wang, S.; Piveteau, L.; Kovalenko, M. V. Efficient Aluminum Chloride−Natural Graphite Battery. Chem. Mater. 2017, 29, 4484–4492. 107. Wang, S.; Kravchyk, K. V.; Filippin, A. N.; Müller, U.; Tiwari, A. N.; Buecheler, S.; Bodnarchuk, M. I.; Kovalenko, M. V. Aluminum Chloride-Graphite Batteries With Flexible Current Collectors Prepared From Earth-Abundant Elements. Adv. Sci. 2018, 5, 1700712. 108. Torrero, D. M.; Anderson, M.; Palma, J.; Marcilla, R.; Ventosa, E. Unexpected Contribution of Current Collector to the Cost of Rechargeable Al-Ion Batteries. ChemElectroChem 2019, 6, 2766–2770. 109. Chen, Y.; Zhou, Z.; Li, N.; Jiao, S.; Chen, H.; Song, W.-L.; Fang, D. Thick Electrodes Upon Biomass-Derivative Carbon Current Collectors: High-Areal Capacity Positive Electrodes for Aluminum-Ion Batteries. Electrochim. Acta 2019, 323, 134805. 110. Torrero, D. M.; Palma, J.; Marcilla, R.; Ventosa, E. A Critical Perspective on Rechargeable Al-Ion Battery Technology. Dalton Trans. 2019, 48, 9906–9911. 111. Manthiram, A.; Fu, Y.; Chung, S.-H.; Zu, C.; Su, Y.-S. Rechargeable Lithium−Sulfur Batteries. Chem. Rev. 2014, 114, 11751–11787. 112. Yu, X.; Manthiram, A. Performance Enhancement and Mechanistic Studies of Room-Temperature Sodium−Sulfur Batteries With a Carbon-Coated Functional Nafion Separator and a Na2S/Activated Carbon Nanofiber Cathode. Chem. Mater. 2016, 28, 896–905. 113. Li, P.; Ma, L.; Wu, T.; Ye, H.; Zhou, J.; Zhao, F.; Han, N.; Wang, Y.; Wu, Y.; Li, Y.; Lu, J. Chemical Immobilization and Conversion of Active Polysulfides Directly by Copper Current Collector: A New Approach to Enabling Stable Room-Temperature Li-S and Na-S Batteries. Adv. Energy Mater. 2018, 8, 1800624. 114. Muthuraj, D.; Ghosh, A.; Kumar, A.; Mitra, S. Nitrogen and Sulfur Doped Carbon Cloth as Current Collector and Polysulfide Immobilizer for Magnesium-Sulfur Batteries. ChemElectroChem 2019, 6, 684–689. 115. Manthiram, A.; Yu, X.; Wang, S. Lithium Battery Chemistries Enabled by Solid-State Electrolytes. Nat. Rev. Mater. 2017, 2, 16103. 116. Kato, Y.; Hori, S.; Saito, T.; Suzuki, K.; Hirayama, M.; Mitsui, A.; Yonemura, M.; Iba, H.; Kanno, R. High-Power All-Solid-State Batteries Using Sulfide Superionic Conductors. Nat. Energy 2016, 1, 16030. 117. Filippin, A. N.; Rawlence, M.; Wäckerlin, A.; Feurer, T.; Zünd, T.; Kravchyk, K.; Kovalenko, M. V.; Romanyuk, Y. E.; Tiwari, A. N.; Buecheler, S. Chromium Nitride as a Stable Cathode Current Collector for All-Solid-State Thin Film Li-Ion Batteries. RSC Adv. 2017, 7, 26960. 118. Chen, X.; Sastre, J.; Aribia, A.; Gilshtein, E.; Romanyuk, Y. E. Flash Lamp Annealing Enables Thin-Film Solid-State Batteries on Aluminum Foil. ACS Appl. Energy Mater. 2021, 4, 5408–5414. 119. Choi, S.; Kim, J.; Eom, M.; Meng, X.; Shin, D. Application of a Carbon Nanotube (CNT) Sheet as a Current Collector for All-Solid-State Lithium Batteries. J. Power Sources 2015, 299, 70–75. 120. Jung, K.-N.; Shin, H.-S.; Park, M.-S.; Lee, J.-W. Solid-State Lithium Batteries: Bipolar Design, Fabrication, and Electrochemistry. ChemElectroChem 2019, 6, 3842–3859. 121. Takami, N.; Yoshima, K.; Harada, Y. 12 V-Class Bipolar Lithium-Ion Batteries Using Li4Ti5O12 Anode for Low-Voltage System Applications. J. Electrochem. Soc. 2017, 164, A6254–A6259. 122. Shin, H.-S.; Ryu, W.-G.; Park, M.-S.; Jung, K.-N.; Kim, H.; Lee, J.-W. Multilayered, Bipolar, all-Solid-State Battery Enabled by a Perovskite-Based Biphasic Solid Electrolyte. ChemSusChem 2018, 11, 3184–3190. 123. Kim, S.-H.; Kim, J.-H.; Cho, S.-J.; Lee, S.-Y. All-solid-state printed bipolar Li–S batteries. Adv. Energy Mater. 2019, 9, 1901841.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes Karim Zaghiba, MR Anil Kumara, and MV Reddyb, aDepartment of Chemical and Materials Engineering, Concordia University, Montreal, QC, Canada; bNouveau Monde Graphite, Saint-Michel-de-Saints, QC, Canada © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of J.B. Goodenough, SECONDARY BATTERIES – LITHIUM RECHARGEABLE SYSTEMS – LITHIUM-ION | Positive Electrode: Layered Metal Oxides, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 243–248, ISBN 9780444527455, https://doi.org/10.1016/ B978-044452745-5.00193-3.

1 2 3 4 4.1 4.2 4.3 5 6 7 7.1 7.2 8 9 9.1 9.2 10 References

Introduction to electrolytes Electrolytes for LIBs Electrolyte design principle and solvation Physicochemical properties Ionic conductivity Transference number Viscosity Organic electrolytes for LIBs Solid polymer organic electrolytes (SPOEs) Liquid organic electrolytes Non-flammable liquid organic electrolytes Gel polymer organic composite electrolytes (GPOEs) and its applications Inorganic electrolytes for LIBs Solid inorganic electrolytes Non-flammable inorganic liquid electrolytes Polymer composite inorganic electrolytes Conclusion

349 350 351 352 353 353 353 355 355 356 356 359 360 361 362 363 364 364

Abstract Both academia and industry have recognized the value of rechargeable batteries since the 1970s. The rising demand for e-mobility and the inclusion of renewable energy sources in the energy ecosystem have made this more noticeable. However, because certain essential elements (lithium, cobalt, etc.) are scarce, it is difficult to further expand the use of LIBs to large-volume technical applications, even though they have been very successful in the fields of energy storage and portable consumer electronics. Thus, it is anticipated that newly developed mono-valent (such as sodium and potassium) and multi-valent (such as magnesium, calcium, zinc, aluminium, etc.) batteries will be able to overcome the resource constraint and associated difficulties. The evolution of aqueous, non-aqueous, organic, and inorganic electrolytes as well as electrode–electrolyte interphases is presented here, with an emphasis on the parallels and discrepancies among lithium-based batteries. To gain a better understanding of the transport behavior in the bulk electrolyte, particular attention is paid to certain fundamental parameters related to solvents and salts, such as adequate chemical and electrochemical stability and viscosity. Also go over potential approaches to improve the electrochemical and physical (transport behavior, solvation, etc.) characteristics of electrolytes and interphases.

Abbreviations CDE CEI CIBs DFDEC DFT DMA DMMEMP DPOF EC EMC FEC GPOEs

348

Charge-discharge cycle efficiency Cathode-electrolyte interphase Cl-ion batteries Di-(2,2,2 trifluoroethyl) carbonate Density functional theory N, N-dimethylacetamide Dimethyl(2-methoxyethoxy) methyl phosphonate Diphenyl octyl phosphate Ethylene carbonate Ethyl methyl carbonate Fluoroethylene carbonate Gel polymer organic composite electrolytes

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00203-6

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

HF HFPM HIBs LIBs LiFP6 LiFSI LiTFSI LOEs MFE NLOEs PAN PEO PMMA PVC RDP SEI SPOEs TEP TMP VC

1

349

Hydrogen fluoride 1,1,1,3,3,3-Hexafluoroisopropyl methyl ether Halide ion batteries Lithium-ion batteries Lithium hexafluoro phosphate Lithium bis(fluorosulfonyl)amide Lithium bis(trifluoromethanesulfonyl)imide Liquid organic electrolytes Methyl-nonafluorobutyl ether Non-flammable liquid organic electrolytes Polyacrylonitrile Polyethylene oxide Poly(methylmethacrylate) Poly(vinyl chloride) Resorcinol bis (diphenyl phosphate) Solid-electrolyte interphase Solid polymer organic electrolytes Triethyl phosphate Trimethyl phosphate Vinylene carbonate

Introduction to electrolytes

All electrochemical current source consists of two electrodes separated by an electrolyte layer. The majority of research and development efforts have been concentrated on the electrodes because they are the important appropriate attribute of such a device that determines the storage capacity, safety and cost, in the above aspect numerous review publications have been devoted to this chapter.1 Electrolyte is the third fundamental part of a battery, has received far less attention. This chapter makes an effort to close this informational gap by providing more details on the role of electrolytes in LIBs, as a result of the inferior qualities of commercial electrolytes as well as the technical and scientific difficulties associated with developing new electrolytes for enhanced lithium-ion and post-lithium batteries. A number of conditions must be met for an electrolyte to be effective. The most basic ones are as follows: i) Adequate chemical and electrochemical stability: When a battery is being charged, the electrolyte solution must be stable at highly oxidative potentials at the cathode and at very reductive potentials at the anode. In general, the molecules found in an electrolyte are more stable when there are stronger bonds between the atoms. The highest occupied molecular orbital and the lowest unoccupied molecular orbital levels, which can be computed for a given structure and measured experimentally as oxidation and reduction potentials, respectively, can be used to quantify electrochemical stability. However, in experimental conditions, electrolyte stability is a kinetic in nature because there aren’t any polar solvents with strong enough thermodynamic properties. The presence of highly reactive electrolyte additives enhancing the durability and transport capabilities of interphase layers should be taken into account when predicting potential reactions resulting in electrolyte degradation. The particular environment in which the electrolyte components are intended to function as well as their interactions with electrode materials at extremely low and/or extremely high electrical potentials should be taken into account. The fact that electrode materials frequently have catalytic characteristics is another aspect. Catalysts for organic processes include graphite, carbon black, and substances containing cobalt, manganese, nickel, and iron. Additionally, electrolytes must not interact with other battery parts like current collectors and casing because doing so could cause passivation processes that would impede the reactions. Additionally, it is strongly advised that electrolytes avoid reacting with elements like oxygen and water that are frequently found in the environment. ii) High ionic conductivity: Electrolytes must effectively allow ions to move between electrodes through frequent cycles of charging and discharging. Therefore, throughout the entire range of operating temperatures, a strong ionic conductivity is necessary. For systems intended for high current applications and quick charging times, this is especially crucial. The molecular architectures of salts and solvents have an obvious impact on the ionic conductivity of electrolyte solutions. Viscosity measurements may be useful to identify the best compositions and concentrations because, in many systems, viscosity and molar conductivity are inversely related. The size and form of an anion have an impact on the characteristics of conductivity and viscosity (the type of cation in a particular battery system is fixed). In actuality, the processes that produce electricity are only aided by the transit of lithium cations, which are active ions, with LIBs. The concentration and dissociation energy of an electrolyte salt determine how easily it can split into simple cations, unfavorable neutral ionic pairs, and highly charged agglomerates. The transference number of a cation and its relationship to temperature and salt concentration are then recommended to be measured. Consequently, if the corresponding transference number is closer to unity, even somewhat lesser total conductivity can be accepted.

350

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

iii) Based on the economic demands: Only when the total cost of all components is financially reasonable can a particular battery design attain commercial success. The normal material cost of the battery’s electrolyte is less than 10% of the sum of the costs of all of its active and inactive materials.2 Additionally, the material costs of the required components make up a very minor portion of the overall costs of a ready-to-use battery pack, which must also take into account costs associated with setting up and running a manufacturing facility, marketing, cost of capital, etc. This is a very recognizable explanation for the seeming reluctance of industrial enterprises to adopt new components if they only offer marginally better qualities than the ones already available. As we write about electrolytes, it is important to note that significant expenses are associated with handling electrolytes, such as controlled/protective storage environments, desiccation procedures, and cell construction. iv) Acceptable operational and environmental safety: It goes without saying that a battery must be safe over the course of its whole life and even in an emergency. In particular, even when it is broken, open, overheated, or used inappropriately, a battery shouldn’t catch fire, explode, or produce hazardous or corrosive fumes. This is a very demanding requirement because a fully charged battery has a significant amount of chemical energy, and a common electrolyte is composed of volatile organic solvents that emit poisonous fumes. It is also crucial to understand what gases can be emitted during the development of the solid-electrolyte interphase (SEI), thermal degradation, and prolonged exposure to high temperatures. When a battery is thrown away or recycled, it needs to be environmentally friendly. Currently used electrolytes seldom meet this criterion. Hydrogen fluoride (HF), Fluorine, phosphorous, or arsenic levels in electrolyte salts as well as in the by-products of their thermal, chemical, or electrochemical disintegration are primarily responsible for this type of toxicity. If the salt or the products of its breakdown are toxic, other potential risk factors include the salt’s biological compatibility or ease of migration through the skin and/or cell membrane, as well as its ability to cause cancer and/or mutagenesis. A quick assessment of all the aforementioned requirements reveals that there is no such thing as an ideal electrolyte. Instead, we’ll go for a middle ground to create “the least imperfect” electrolyte that offers a tolerable balance of benefits and disadvantages. v) Good thermal stability: The construction of electrolytes with sufficient compatibility with the electrode materials and high ionic conductivity across those ranges is necessary for the development of batteries that can function over a broad operating temperature range. vi) Low cost, Easy preparation, and Minimal toxicity: These factors are considered crucial performance indicators for the commercial viability of rechargeable batteries, especially in their application for stationary electric energy storage. The cost and toxicity of these batteries significantly influence their feasibility in grid-scale applications.

2

Electrolytes for LIBs

Currently, the same electrolyte is used in practically all commercial LIBs. Electrolytes consists, liquid, solid, polymer electrolytes, for example, Lithium hexafluoro phosphate (LiFP6) is dissolved in an EC and EMC solution at 1.2 M. In order to enhance the production of stable and ionically conductive passive layers on negative (graphite) electrodes (the so-called SEI), various additives, such as vinylene carbonate (VC), are frequently utilized. This electrolyte has a conductivity slightly above 10 mS/cm at room temperature, electrochemical stability >4.8 V versus the Li/Li+ redox pair, which is sufficient for the currently used cathode based on NMC, LiFePO4 etc. and does not negatively affect other battery system components. The other electrolyte salts, which cause corrosion to aluminium connector. On the other hand, the electrolyte degrades above 70
C and is extremely sensitive to water traces, which causes hydrolysis reactions that release toxic hydrogen fluoride. The optimum Li battery electrolyte materials should have the following essential qualities: The qualities listed below stop selfdischarge: (i) a wide electrochemical potential window; (ii) Electronic conductivity below 10−8 Scm−1; (iii) Chemical stability toward electrodes throughout the operating temperature range; (iv) Transference number approximately equal to 1; (v) Thermal expansion coefficients matching those of cathode materials; (vi) High ionic conductivity of 10−3 Scm−1 at room temperature; (vii) Maintaining a stable chemical composition; Active electrode materials should undergo no crystal structure phase transformation up to or near their sintering temperatures; (viii) Sintering temperature aligning with electrode active materials; and (ix) Cost-effective and low toxicity.3 By description, an electrolyte is a material that, when dissolved in a polar solvent, creates a system with choice of ionic movements. Ionic conductivity is therefore conceivable. Such a concept is no longer appropriate because we now understand that weakly polar solvents can also provide freedom of ionic movement in addition to polar solvents. Molten salts and polymer matrices can both accomplish this. For this reason, in the entire battery system holding ions and allowing for free ion movement is referred to as an electrolyte in electrochemical or battery terminology. Fig. 1 shows the general representation illustration of the fabricated electrolyte for all solid-state lithium (Li) batteries. As a result, an electrolyte for lithium-ion (Li-ion) cells includes the entire system, including the solvent and/or the solvent combination as well as any potential additions, in addition to the lithium salt. Compared to earlier rechargeable cell technologies, an electrolyte in Li-ion cells serves a variety of additional purposes. The movement of lithium cations between electrodes is the electrolyte’s most significant function, and it has an impact on the maximum current a cell can use.4 However, the electrolyte parameters also have an indirect or direct impact on the following additional cell properties:

• •

Both in the case of solvents and salts, the thermal stability of the electrolyte serves as a direct limiter of storage and working temperatures.5 The ratio of energy input to output during a charge-discharge cycle is known as the charge-discharge cycle efficiency (CDE) and is directly influenced by the electrolyte conductivity and lithium cation transference number, as well as perhaps by the SEI layer resistance.6

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

351

Composite positive electrode Li6.6La3Zr1.4Ta0.4O12 LiCoO2

Thick supporting electrolyte

100 Pm In-Li anode Fig. 1 Schematic diagram showing the cross-sectional scanning electron micrograph of the electrolyte that was manufactured for Li batteries. Reproduced with permission of Reddy, M. V.; Julien, C. M.; Mauger A.; Zaghib, K. Sulfide and Oxide Inorganic Solid Electrolytes for All-Solid-State Li Batteries: A Review, Nanomaterials 2020, 10, 1606.

• • • 3

Energy density is indirectly impacted by restrictions brought on by the incompatibility of the electrolyte with newer or better electrodes (restraining its use), either because of a chemical incompatibility (no passivation or overreaction) or because it is not possible to achieve the high and/or low potentials necessary to maximize the electrode’s capacity.7 Acute toxicity, chronic toxicity, salt stability, solvent(s), and/or the effect of the electrolyte on the stability of the electrodes are all factors that directly affect safety and environmental impact.8 Cost (the electrolyte accounts for up to 15% of the component costs in a Li-ion cell, depending on the size of the cell and the precise technology is utilized).9

Electrolyte design principle and solvation

In this work, Zhao, X et al. present the design principles for non-aqueous electrolytes aimed at improving the performance of Halide Ion Batteries (HIBs). Based on the following reactions, the Cl-ion Batteries, a Li||FeOCl cell is chosen as a representative model to elucidate the essential criteria for electrolytes in Halide Ion Batteries10: Cathode: FeOCl + e— $ FeO + Cl— Anode: Li + Cl

—

—

$ LiCl + e

(1) (2)

Liquid electrolytes for CIBs contain three side reactions that are connected to the dissolving effect. The breakdown of transition metal is the first adverse reaction: 3FeOCl ! Fe2O3 + Fe3+ +3Cl—

(3)

Secondly, the self-discharge reaction results in the dissolution of elemental chlorine: FeOCl ! FeO + Cl

(4)

The third side reaction involves the dissolution of the discharge product (i.e., LiCl). LiCl ! Li + + Cl—

(5)

Hence, to facilitate rapid Cl− conducting kinetics and ensure a wide electrochemical stability window for preventing anodic degradation, the electrolyte for CIBs should possess not only high ionic conductivity but also low solubility in LiCl, Cl, and Fe. Through DFT calculations, solvent candidates with different dipole moments (m) were identified.11 These solvent candidates are categorized as low-polarity (HFE, a ¼ 4.373), medium-polarity (N, N-dimethylacetamide (DMA), a ¼ 3.869), and high-polarity (propylene carbonate (PC), a ¼ 5.637). The schematic molecular approach taken in the creation of binary non-aqueous electrolyte solvents with CIB with dipole moment ¼ m is shown in the Fig. 2(a).

352

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

Fig. 2 Strategies for designing electrolytes. (a) A schematic illustration of the molecular approach taken in the creation of binary non-aqueous electrolyte solvents with CIB. The dipole moment ¼ m. (b) A demonstration of the fast Li-ion transport, wide temperature range ( 60
C) stability, and soft solvation between Li ions and soft solvent. (a) Reproduced with permission of Yang, X.; Zhang,B.; Tian, Y.; Wang, Y.; Fu, Z.; Zhou, D.; Liu, H.; Kang, F.; Li, B.; Wang, C.; Wang G. Electrolyte Design Principles for Developing Quasi-Solid-State Rechargeable Halide-Ion Batteries. Nature. Comm. 2023, 14, 925. (b) Reproduced with permission of Jijian, X.; Jiaxun, Z.; Pollard, T.P. Li Q.; Tan, S.; Hou, S.; Wan, H.; Chen, F.; He, H.; Hu, E.; Xu, K.; Yang, X.-Q.; Borodin, O.; Wang, C. Electrolyte Design for Li-Ion Batteries under Extreme Operating Conditions. Nature 2023, 614, 695–702.

A large electrochemical stability window and a moderate boiling point combined with a low freezing point are the main requirements for selecting a solvent. A soft solvating capability ensuring a low Li-ion desolvation energy with minimal compromise on ionic dissociation ability. Should to be the secondary criterion (Fig. 2b). The elastic inorganic-organic interphase can endure the relatively minor volume changes that occur during lithiation and delithiation in both the LiNi0.8 Mn0.1Co0.1 O2 (NMC811) cathode and graphite anode, resulting in high cycle life (Fig. 2b).12,13

4

Physicochemical properties

Conductivity and viscosity offer details on ion transport characteristics. Super concentrated electrolytes are expected to have different physicochemical characteristics from dilute electrolytes since they contain salt concentrations that are nearly saturation levels.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes 4.1

353

Ionic conductivity

Ions move between positive and negative electrodes as charge carriers, supplying electrolytes with electrical conductivity. Ionic conductivity typically shows a maximum value with concentration fluctuations. For example, a concentration of 3.5 m represents the peak for a LiTFSI solution.14 At a certain attention, a compromise between the quantity of ionic carriers and viscosity results in maximum conductivity. Consequently, because of their decreased ionic conductivities and increased viscosity brought on by Coulombic frictions, highly concentrated solutions are not favorable. Scientists have recently attained “unusual” characteristics, including improved thermal stability, extended electrochemical stability windows, and unexpectedly higher conductivity, through elevating the salt concentration beyond conventional electrolyte levels. These remarkable outcomes provide justification for employing highly concentrated electrolytes.15,16 For example, despite the high viscosity of the electrolyte (32 mPas), a conductivity of approximately 8 mScm−1 (at 25
C) is achieved for a 21 m LiTFSI, which is comparable to that of typical non-aqueous electrolytes. KTFSI (30 m)16 and NaClO4 (17 m)17 exhibit even higher conductivities of 41 and 64 mScm−1, respectively. A conductivity of 55.6 mScm−1 is recorded for 20 m KOAc. In,18 in a mixed cation electrolyte, concentration increases to 27 m KOAc, and Li0.2K0.8OAc1.3H2O results in conductivity values of 31.4 and 5.3 mScm−1, respectively.19 A significant portion of solvent-separated Li+(H2O)4 (about 40%) with Li+ cations serving as the primary charge carriers is anticipated in the solution at more concentrations of LiTFSI (20 m). Due to the immobilization of anions in 3D anion-rich regions, Li+ ions surrounded by water exhibit a threefold increase in mobility compared to those surrounded by TFSI anions. In reality, the high conductivities source from formations involving water-solvated Li+ ions.16 This enables rapid vehicular-type ion transport, either within bulk-like water domains or through nanometric water channels (with a diameter of approximately 1.2 nm) formed by TFSI anions.20–22 As the salt concentration increases to levels where the water/salt ratio approaches 2 or lower, as observed in cases such as Li(TFSI)0.7(BETI)0.32H2O hydrate melt15 and Li 0.2K0.8 OAc1.3H2O, there is a belief that the nature of ion transport undergoes a shift toward a hopping-type process. This transformation is attributed to the insufficient presence of bulk-like water, prompting a change in the predominant mechanism.23 It is anticipated that the conductivity of bulk electrolyte will differ from that of narrow cavities when it comes to being within a nanometer’s vicinity to a charged electrode due to the alteration of nanometric domains.24 Generally, a change in temperature can alter conductivity because it can alter ion mobility. The primary issue with the majority of H2O in salt electrolytes, including 21 m LiTFSI, is that they have a freezing point of 0
C, which lowers conductivity at lower temperatures. Not all of the super concentrated aqueous electrolytes, nevertheless, behave in this way. For instance, it is stated that the tolerable ionic conductivities at −20
C for 22 m KOSO2CF3 (KOTf )25 and 51.2 m KOAc+LiOAC water-in-bisalt electrolyte26 are around 10 and 2 mScm−1, respectively. Furthermore, it has been demonstrated that the addition of acetonitrile to a 5 m LiTFSI solution27,28 enhances the ionic conductivity of the highly concentrated aqueous electrolyte, achieving significantly improved performance even at extremely low temperatures (6.25 mScm−1 at −20
C). The resulting conductivity levels are comparable to those observed in organic solvents.

4.2

Transference number

Another crucial element of ion transport is the electrolyte’s ion transference number (t), which measures the percentage of the total current like both parasitic and desirable that is attributable to a particular ionic species, primarily the ions that are the subject of the investigation.29 Because there are more charge carriers when the concentration of salt rises, the cation transference number (t+) usually increases as well.30 During rapid charge or discharge conditions, an electrochemical device being studied with a high ion transference number, crucial for ions like Li+ in the context of LIBs, exhibits minimal bulk electrolyte resistance and a further rate capability.17 While t+ in diluted electrolytes can be easily calculated by dividing cation self-diffusion by the total of cation and anion self-diffusion values, adjustments are necessary for these calculated t+ values in highly concentrated electrolytes due to strong electrostatic interionic interactions.31 Since anions and cations have different structures at the nanoscale, they diffuse differently. Since the two networks are in contact with one another, there are some interactions between them, but they are not as strong as they would be with a traditional electrolyte. Consequently, the transference number of cations in a salt in H2O system is likely to be much lower than that determined under ideal conditions.32 For 21 m LiTFSI, the uncorrelated Li+ transference number is testified to be approximately 0.7, while the expected interrelated value is 0.22.17 This value is still on par with or higher than the values found for organic electrolytes and super concentrated electrolytes.15 This is explained by the fact that Li+ moves more easily in water domains,33 which exhibit additional noticeable non-Gaussian behavior.

4.3

Viscosity

Viscosity, the inverse of fluidity and a measure of a substance’s resistance to flow, is another factor that impacts ion transport. The viscosity of an electrolyte is subject to changes due to temperature, solvent-ion interactions, and salt concentration. Among the critical factors are soft volume, solvent properties, which involves the disruption of solvent dipole orientation, ion solvation, and electrostatic ion–ion interactions.34,35 Generally, increased ionic friction and stronger electrostatic interactions result in increased viscosity as salt concentration rises.15 Upon adding Li+ cation salts such as FSI, LiFSI, and NaFSI to water at concentrations ranging

354

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

from 1 to 20 m, the viscosity elevates from approximately 2 mPas to 14 and 19 mPas at 25
C, respectively. The fact that these values are substantially lower than the 21 m LiTFSI (>30 mPas)15 shows how crucial the molecular size and nature of the anion are. Both FSI-salts have higher viscosities at concentrations of 35 m. This effect is more pronounced for NaFSI (95 mPas) than LiFSI (80 mPas). The larger radius of Li+ ions contribute to an increase in Coulombic interactions between ions, thereby accentuating the influence of cations. As anticipated, a decrease in water content is noted to considerably increase viscosity by more than two orders of magnitude compared to the 21 m LiTFSI solution. This effect is especially prominent in hydrate melts, such as Li(PTFSI)0.6(TFSI)0.41H2O. An overview of the viscosity of various H2O in bi-salt and salt solutions. Because viscosity and temperature are inversely correlated, a rise in temperature causes a decrease in viscosity because of increased mobility. As observed, at 60
C, the reported viscosity values for 35 m NaFSI and 35 m LiFSI H2O in salt electrolytes are approximately 18 mPas and 17.5 mPas, respectively.36 In contrast, water-in-salt electrolytes with 35 m NaFSI and 35 m LiFSI exhibit higher viscosity values, reaching 200 mPas and 305 mPas, respectively, result from reducing the temperature to 10
C. An overview of the viscosity of the different super concentrated electrolytes that have been studied at various concentrations and temperatures. The energy-storage devices’ (ESD) structure, type, and operating mechanism all influence the desired electrolyte properties. In a restricted sense, a liquid or solid solution that can conduct ions but not electrons are called an electrolyte. Broadly speaking, it can refer to a material that has ion-conducting elements combined with other structural or functional elements, like structural improvements, additives,37–39 and so on. Three primary types of electrolytes are employed in LIBs: liquid, solid, and gel electrolytes, each determined by their states of matter at room temperature. Irrespective of their physical state, the primary role of an electrolyte in a LIB is to facilitate the rapid and efficient movement of Li+ between two electrodes.40 Notably, as energy storage devices (ESDs) progress, the desired features and capabilities of electrolytes also evolve. Table 1 offers a comprehensive summary and comparison of significant electrolyte properties for various LIBs across different electrolyte types. Table 1 highlights a few points that should be noted. First off, the assessment of the properties in the table is purely the authors’ opinion and is not exhaustive. It primarily shows how the states of matter affect the characteristics of electrolytes. This table does not include some special electrolytes. The section dedicated to each type of electrolyte contains a thorough explanation of the properties of the electrolyte. Secondly, since some electrolyte properties cannot be quantitatively described, a qualitative comparison is made for all properties. In real-life applications, the most crucial attributes include ionic conductivity and interfacial properties. It is noted that the states of matter significantly impact the contact/interfacial properties of electrolytes, and research on interfacial properties is seldom published.41 Additionally, Table 1 exclusively includes the most critical electrolyte properties. While essential characteristics are outlined for all types of electrolytes in LIBs, certain special batteries, like lithium-sulfur and lithium-air batteries, require additional important characteristics.42 Specifically, during the charging or discharging process of lithium-sulfur batteries, it is imperative for the electrolytes employed to exhibit robust resistance against dissolving polysulfides, which are by-products of the batteries compared to lithium-ion.43–45 In contrast, electrolytes employed in lithium-air batteries necessitate excellent oxygen solubility and the ability to facilitate the repeated and highly reversible formation and decomposition of Li2O2. In simpler terms, they should exhibit high resistance to the formation and/or dissolution of undesirable by-products.25,46,47 The circumstance that liquids typically exhibit a constituent solubility that is significantly higher than solids is another factor used to evaluate the performance of each type of electrolyte. We can gather the following important conclusions from Table 1. Firstly, develop an electrolyte that can demonstrate exceptional performance across all properties. Actually, the majority of electrolyte research to date has concentrated on enhancing one or more of the characteristics indicated in Table 1. Safety enhancement for liquid electrolytes is one of these studies.48,49 enhancement of ionic conductivity in solid polymeric electrolytes. It should be noted, nevertheless, The key to a successful solid or gel electrolyte lies in consistently achieving a stable and effective contact between the electrolyte and the electrodes. Table 1 Summarizing the essential electrolyte properties sought for diverse LIBs and comparing them based on their states of matter provides crucial insights into optimizing performance.40 Electrolyte properties Essential properties Ionic conduction interfacial properties Electrochemical stability Thermal stability Dimension stability Safety Additional properties required by a specific battery Resistance to forming or dissolving by-products (e.g. polysulfides for Li–S) Ability to dissolve oxygen (e.g. Li–O2)

Liquid

Solid

Gel

high >10−3 S cm−1 good poor poor poor poor

low 10−4 S cm−1

poor good good good good

medium poor poor medium medium

low good

high poor

medium medium

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

355

Furthermore, when transitioning from liquid electrolytes to solid or gel electrolytes, meticulous attention must be devoted to the interfacial characteristics between the electrolyte and the electrodes. Third, there are various aspects that make up the stability properties of electrolytes, such as the temperature at which thermal decomposition occurs, the electrochemical window, and the modulus needed to separate the two electrodes. The intricate matter of electrolyte stability dictates a battery’s safety. In addition to the different stability characteristics listed in Table 1, other characteristics like flammability, toxicity, and viscosity (which causes leaks) are also taken into account for the electrolyte’s safety.

5

Organic electrolytes for LIBs

Because of their excellent ion conductivity, strong chemical stability, and frequently advantageous electrical properties, organic electrolytes are preferred for these uses. Based on the application’s needs and the desired performance characteristics, different types of organic electrolytes used.

6

Solid polymer organic electrolytes (SPOEs)

The solid organic polymer electrolytes for lithium batteries that are based on lithium-ion conductive poly (ethylene oxide) (PEO) and lithium salts were also thoroughly investigated as solid polymer electrolytes for lithium batteries. The concept of a solid electrolyte emerged in the 1980s, initially formulated as a solid polymer material. The range of solid electrolytes has expanded to include organic polymer materials from inorganic ceramic materials.50 In the 2000s, a growing body of research aimed to enhance the incorporation of solid electrolytes into the development of lithium metal batteries, including LIBs.51,52 The focus shifted toward solid composite electrolytes, aiming to leverage the advantages while addressing the drawbacks associated with both organic and inorganic solid electrolytes. Despite this trend, some studies persist in focusing solely on Solid Polymer Electrolytes (SPOEs). (i) Polyethylene oxide (PEO) Early in the 1980s, PEO/lithium salt was employed as a host matrix for the first time to create polymer solid electrolytes. Since then, a great deal of research on solid electrolytes based on PEO has been done. The PEO polymer chain’s fundamental chemical structure, -CH2-CH2-O-, can provide an effective Li-salt dissolvability, as shown in Fig. 3(a). This material aids in the formation of metal or lithium salt complexes and acts as a host matrix for solid electrolytes.53 Furthermore, the flexible macromolecular configuration of polyethylene oxide (PEO) chains plays a role in facilitating Li+ transport. It is important to highlight that PEO displays a semi-crystalline nature at room temperature, and the presence of crystalline domains within the PEO polymer matrix significantly hinders ion transfer. Due to this, PEO based solid electrolytes typically exhibit relatively low ionic conductivity at room temperature, typically ranging between 10−6 and 10−8 Scm−1.54 To improve the ionic conductivities of PEO based solid electrolytes by reducing PEO polymer crystallinity, three common techniques are employed: the addition of plasticizers, incorporation of cross-linked or blocked polymers, and introduction of inorganic ceramic fillers.54,55 For example, Ito et al.56 observed that introducing low-molecular-weight poly(ethylene glycol) (PEG) as a plasticizer to the PEO or Li salt composite effectively diminishes the crystalline domain, thereby enhancing ionic conductivity. Khurana et al.57 cross-linked segments of PEO with polyethylene (PE) chains to inhibit PEO’s crystallinity. (ii) Polyacrylonitrile (PAN) Excellent anti-oxidation properties make polyacrylonitrile a synthetic organic polymer resin that has been tested for use in high-quality carbon fibers and textile nanofibers, among other uses. PAN polymer emerges as an excellent choice for incorporation into solid electrolytes owing to the presence of the nitrile group within its polymer chains (Fig. 3(b)). This nitrile group, a potent electron-withdrawing entity, exhibits remarkable electrochemical stability.58 PAN-based solid electrolytes have a wide range of benefits, such as excellent compatibility with Li electrodes, high ionic conductivity, good thermal stability, and a large electrochemical stability window. Li salts and plasticizers are always mixed with PAN polymer to prepare PAN-based solid electrolytes, and at room temperature, the ionic conductivity can reach approximately 10−3 S cm−1. It is noteworthy that PAN based gel electrolytes exhibit a Li ion transference number that is comparatively higher than PEO-based electrolytes.59,60 In contrast to PEO based solid electrolytes, where the mobility of the PEO polymer chain predominantly influences ionic conductivity, the movement of Li ions along the segmental chain has minimal impact on ionic conductivity at room temperature. The two primary methods used to enhance the ionic conductivity of PAN based solid electrolytes involve

Fig. 3 (a) Chemical structures of Polyethylene oxide and (b) Polyacrylonitrile that have been used as Solid polymer organic electrolytes.

356

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

incorporating plasticizers and inorganic fillers.61 In the quest to augment the ionic conductivity within the PAN electrolyte system, Feullade et al.62 pioneered the exploration of polycarboxylate (PC) plasticizers, revealing a substantial increase in ionic conductivity to approximately 10−2 S/cm−1. (iii) Other polymers In addition, established polymers like PEO, PAN, and PVDF, various other polymer hosts, including PMMA and PVC, have been investigated for their potential in solid electrolytes. A pivotal development was introduced by Iijima63 in 1985, who pioneered the use of PMMA as a gelling agent. The choice of PMMA as the polymer host in solid electrolytes for lithium batteries brings significant advantages, notably its diminished reactivity toward the Li electrode, leading to a stable lithium-electrolyte interface. Furthermore, the presence of polar carbonyl groups (dC]Od) in PMMA chains enables coordinated bonding with lithium salts, thereby expediting rapid ion transport within solid electrolytes.64 In a study conducted by Rahman et al.,65 a PMMA-MG49LiClO4 solid polymer electrolyte was successfully fabricated using a straightforward solution casting method. Noteworthy is the achievement of the highest recorded ionic conductivity, reaching 1.47  10−8 S/cm−1 at 20 wt% LiClO4. Similar to PVDF-based electrolytes, a notable drawback of PMMA-based solid electrolytes is their relatively lower mechanical strength compared to other polymer electrolytes. Researchers have also looked into PVA as a polymer for creating solid electrolytes. Strong hydrogen binding is provided by the abundant hydroxyl groups bonded to the carbon chain backbone of PVA polymer, resulting in high melting point and superior mechanical stability. The PVA solid electrolyte system, as documented, displays a comparatively elevated ionic conductivity spanning from 10−8 to 10−4 S/cm−1.66 Study conducted by Rajendran et al., the impact of plasticizer on the ionic conductivity of PVA-LiClO4 solid electrolytes was investigated. It was observed that the inclusion of dimethyl phthalate (DMP) as a plasticizer resulted in a peak ionic conductivity of 0.149  10−3 S/cm−1 at 302 K.67

7

Liquid organic electrolytes

In the group of organic liquid electrolytes, essential components include inorganic salts and organic solvents. These electrolytes play a pivotal role as conduits for ion transport, establishing connections between cathodes and anodes while preventing detrimental electron transport within rechargeable batteries. Particularly during the initial discharge/charge cycles at low reaction potentials (3.5 V versus Li or Li+), breakdown occurs in organic liquid electrolytes, leading to the formation of passivation layers. These layers possess dual characteristics, serving as both ionic conductive and electrically insulating materials. The anode surface is enveloped by the SEI, while the cathode surface exhibits the CEI. The optimal functionality of commercially viable batteries relies on the integrity and performance of these SEI and CEI layers. These layers play an essential role in enhancing the capacity retention and Coulombic efficiency of electrodes by preventing direct contact between electrodes and electrolytes, thus minimizing ongoing side reactions. A wealth of literature underscores the crucial role of stable and robust SEI and CEI layers in achieving high cyclability and Coulombic efficiency in inorganic batteries.68,69 Similarly, within the domain of organic batteries, organic liquid electrolytes are essential for the formation of SEI and CEI, exerting a significant influence on cycle life and Coulombic efficiency. It is of paramount importance to gain a fundamental understanding of the relationship between the composition of organic liquid electrolytes and the electrochemical performance of lithium batteries.

7.1

Non-flammable liquid organic electrolytes

Numerous liquid organic electrolytes that are non-flammable (NLOEs) have been created. Conventional NLOEs may be rendered non-flammable by adding the proper flame-retardant additives to them. The flame-retardant additive content of this type of NLOEs is typically less than 40 wt%. NLOEs can also be attained by using non-flammable liquids as liquid electrolyte solvents. For flame-retardant additives or non-flammable solvents in non-flammable organic explosives to be effective, certain characteristics are essential (a) high flame retardant efficiency at low concentrations; (b) favorable solubility in conventional electrolytes (c) low volatility (d) low viscosity in conventional LEs (e) robust electrochemical stability and (f ) excellent wettability with electrodes and separators.70 Non-flammable LOEs integrate flame-retardant additives such as phosphazenes, fluorinated phosphates, fluorides, and phosphates. Subsequent sections will provide a detailed exploration of various flame-retardant additives and non-flammable solvents, elucidating their composition and characteristics in non-flammable LOEs.71 (i) Phosphates Because of their excellent flame retardance, different phosphates have been added as flame-retardant additives to conventional liquid organic electrolytes (LOEs) or used as LOE solvents. LOEs are widely used in the firefighting industry. Fig. 4 displays the chemical structures of phosphates that were utilized in non-flammable LOEs. Phosphates have been shown in numerous reports to be able to suppress the propagation of flames, resulting in electrolytes that are non-flammable.72 Certain phosphates have a high viscosity, which lowers the conductivity of ions.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

357

Fig. 4 (a)–(i) Chemical structures of phosphates and Fluorides that have been used in non-flammable liquid electrolytes.

(a) Trimethyl phosphate (TMP) Due to its cost-effectiveness, low viscosity, and commendable flame retardance, TMP has found application in non-flammable LOEs (Fig. 4a). Adding more than 12 mol% TMP can halt the burning of conventional LOEs. However, TMP possesses inherent instability and tends to disintegrate on graphite or lithium metal anodes. Wang et al. conducted a systematic investigation into the fundamental properties of LEs based on TMP.73–75 The results of their study demonstrated that, in a cosolvent or TMP binary mixed electrolyte, The enhanced flame retardance of 1 M LiPF6 was achieved through the strategic incorporation of cosolvents characterized by a high boiling point and a reduced number of hydrogen atoms, thereby contributing to the overall flame safety of the system. The minimal TMP concentrations required to establish non-flammable electrolytes, the introduction of 12 mol% trimethylolpropane (TMP) into the 1 M LiPF6/EC electrolyte, comprising PC, gamma-butyrolactone (GBL), DEC, EMC, and dimethoxy ethane (DME), resulted in a notable enhancement of non-flammability in the electrolyte system. The presence of TMP not only prevented its decomposition on graphite anodes but also contributed to the formation of robust SEIs when combined with EC/PC or EC/DEC. Both 1 M LiPF6 in EC/DEC/TMP and 1 M LiPF6 in EC/PC/TMP exhibited high conductivities (>7 mS cm−1). However, after 30 cycles, amorphous carbon (AC)/LiCoO2 (LCO) cells with EC/PC/DEC/TMP (30:30:20:20) and 1 M LiPF6 displayed relatively low-capacity retention (68%). (b) Triethyl phosphate (TEP) TEP (Fig. 4b) has low viscosity, good flame retardance, and good solvation of lithium salts; however, TEP also decomposes on the graphite or lithium metal anode. Fluoroethylene carbonate (FEC) was added by Inoue et al.76 to non-flammable LE based on TEP (1 M LiPF6 in TEP/FEC), The formation of robust SEIs on SiO2 anodes was facilitated to prevent the decomposition of triethyl phosphate (TEP) on SiO2 anodes. At room temperature, the ionic conductivity of 1 M LiPF6 in TEP/FEC was approximately 6 mScm−1. Similar to cells employing 1 M LiPF6 in EC/DEC (79%), LiNi0.8Co0.15Al0.05O2/SiO cells assembled with this non-flammable liquid electrolyte demonstrated a capacity retention of 78% after 250 cycles. The safety of the cells was significantly enhanced by the use of 1 M LiPF6 in TEP/FEC electrolyte. In short-circuit tests, cells containing 1 M LiPF6 in TEP/FEC, even with charged anodes up to 200
C, did not exhibit any exothermic reactions. (c) Dimethyl(2-methoxyethoxy) methyl phosphonate (DMMEMP) DMMEMP (Fig. 4c) was synthesized by Li et al.77,78 and utilized as a solvent for LEs. DMMEMP has a non-flammable 1 M LiTFSI, a broad electrochemical window of 5.5 V, and sufficient conductivity of 2.0 mScm−1 at 200
C. In addition, it has a high dielectric constant of 77, a moderate viscosity of 4.85 cP at 25
C, and a high boiling point of 283
C. After 10 cycles, the LiFePO4 (LFP)/Li cell showed 98.2% high-capacity retention and roughly 100% Coulombic efficiency (CE) using 1 M LiTFSI in DMMEMP electrolyte. The ionic conductivity fairly reduced to 4.7 mScm−1 and the SET of the electrolytes decreased to 15 sg−1 when 30 wt% DMMEMP was added to 1.0 M LiTFSI in EC/DMC (14:7).

358

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

(d) Diphenyl octyl phosphate (DPOF) Utilized as a flame-retardant additive, DPOF (Fig. 4d) has been employed to reduce the flammability of LEs. In a study by Zhang et al.,79 it was observed that the addition of 15% DPOF to 1.0 M LiPF6 in EC/DEC/EMC (1:1:1 vol.) resulted in a low burning speed of 6.1 mg s−1. DPOF did not impact the formation of SEI films on artificial graphite, and the capacity of the Li/graphite half-cell with 5% DPOF exceeded that of the base electrolyte, 1.0 M LiPF6 in EC/DEC/EMC (1:1:1 vol.). Kim et al.79–82 demonstrated that 1.15 M LiPF6 remained electrochemically stable in EC/EMC (4:6 vol.) electrolytes containing DPOF at concentrations ranging from 10 to 30 wt%. The stability range was extended to approximately 4.75–5.5 V, with the addition of 30 wt% DPOF elevating the electrolyte decomposition temperature to 223
C. The optimal electrolyte composition for enhancing cycle performance in LCO and MCMB cells was found to be 10% weight percentage DPOF. LCO and MCMB cells incorporating 10% DPOF exhibited an impressive 80% capacity retention and a discharge capacity of 101.8 mAhg−1 after 40 cycles. (e) Resorcinol bis (diphenyl phosphate) (RDP) Furthermore, RDP83 (Fig. 4e) serves the dual purpose of safeguarding against overcharging and providing flame retardance. The introduction of 72 vol% RDP to 1 M LiPF6 in EC/EMC (1:3 vol.) substantially reduced the burning time from 49 to 0 sg−1. Despite causing a slight decrease in the electrolyte’s conductivity, the performance of LiMn2O4/graphite cells remained unaltered with the incorporation of 10% RDP. Acting as a current absorber during overcharging, RDP undergoes polymerization at 4.4 V, preventing voltage runaway in cells. LiMn2O4/graphite cells equipped with the electrolyte containing 10% RDP stabilized at 4.4 V during overcharge up to 5 V, and even after a 100% overcharge, the voltage did not exceed 5 V. (ii) Fluorides Due to their low freezing point, superior electrochemical stability, low surface tension, and low viscosity, fluorides have been extensively investigated for their potential applications in lithium-ion batteries (LIBs). The chemical structures of the fluorides utilized in non-flammable LEs are depicted in Fig. 4. The incorporation of fluorinated solvents in electrolytes offers various advantages, including flame retardance, an extended electrochemical window, and the formation of robust SEIs on anodes and CEIs on cathodes.84 The strong electron-withdrawing effect of fluorine results in higher oxidation potentials for fluorinated molecules, which is beneficial for enhancing the cycle performance of high voltage LIBs. (f ) Fluoroethylene carbonate (FEC) FEC85 (Fig. 4f ) has a high flash point (130
C), low viscosity, high dielectric constant, and good flame retardance. In PC/FEC (90:10), for example, 1 M LiPF6 is self-quench. Another useful addition to raise LIB cycle performance is FEC.86–88 According to Stevenson et al.,89 kinetically stable SEIs primarily composed of LiF were formed as a result of the fluoride ions produced by the reduction of the FEC-based electrolytes. A totally non-flammable electrolyte comprising 1.0 M LiPF6 in FEC/F-EMC/F-EPE (3:5:2 vol.) was created by Zhang et al.84 Under high temperature cycling conditions, high-voltage LNMO and graphite LIBs demonstrated enhanced stability for both the graphite anode and the LNMO cathode when employing an electrolyte based on fluoroethylene carbonate (FEC). Comparative studies indicated superior performance of LNMO and graphite cells with the FEC based electrolyte compared to those using conventional liquid electrolytes at temperatures of both 25 and 55 degrees Celsius. After 250 cycles at 55
C, LNMO and graphite cells utilizing the FEC-based electrolyte exhibited a Coulombic efficiency (CE) exceeding 99.5% and a 50% capacity retention rate. Additionally, the FEC based electrolyte led to fewer solid breakdown products deposited on the surfaces of both the anode and cathode. (g) Methyl-nonafluorobutyl ether (MFE) Researchers have focused a lot of attention on safe electrolytes that use non-flammable hydrofluoroethers (HFEs) as flame-retardant additives because these non-flammable HFEs have several fascinating properties, including low surface tension, low viscosity, and good electrochemical stability. MFE90–92 (Fig. 4g), one of the many HFEs, has received extensive research as a solvent of non-flammable LEs. Fang et al.93 developed a non-flammable electrolyte consisting of G2E/MFE/FEC (50:45:5 wt) with 0.8 M LiTFSI. This electrolyte exhibited satisfactory lithium-ion transference number of 0.568 at 25
C and ionic conductivity of 3.8 mScm−1. Even at low temperatures, the performance of LFP and graphite cells using this electrolyte matched that of cells employing conventional liquid electrolytes. With this electrolyte, LFP and graphite cells achieved a capacity of 62 mAh g−1 at 20
C, retaining 46.3% of that capacity 134 mAh g−1 at 25
C. (h) Di-(2,2,2 trifluoroethyl) carbonate (DFDEC) The HOMO in DFDEC (Fig. 4h) is lowered from −12.59 eV to −13.11 eV by substituting fluorine atoms for the hydrogen atoms in DEC. This enhances the stability of electrochemical oxidation. In Song et al.’s report, they described a non-flammable LE with a SET of 0 sg−1 and an unmeasurable flash point that contained 1 wt% FEC + 1.0 M LiPF6 in DFDEC/PC (7:3vol.). Incorporating FEC as a film-forming additive enhanced the cycle stability of both the graphite anode and Li1.13Mn0.463Ni0.203Co0.203O2 (LNMC) cathode, especially under high charge voltage conditions (4.85 V). The cells exhibited a commendable initial capacity of 255 mAh g−1, maintained a CE of 99.8% over the course of 100 cycles and sustained good capacity retention of 80%. the electrolyte-containing cells were impressive. Capacity retention was 77% after 14 days of storage at 60
C.94,95

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

359

(i) 1,1,1,3,3,3-Hexafluoroisopropyl methyl ether (HFPM) Liu et al.96 introduced HFPM, a fluorinated ether characterized by its absence of a flash point (Fig. 4i). In distinction to EC (1.5 V) and FEC (1.7 V), HFPM displays a lower decay voltage of 1.2 V. Researchers successfully formulated a completely non-flammable LE by combining 1.0 M LiPF6 in FEC/DMC/EMC/HFPM (2:3:1:4 vol.). This innovative electrolyte, characterized by superior wettability, showcased an expanded electrochemical window of 5.5 V, excellent compatibility with graphite anodes, and an impressive high ionic conductivity of 8.57 mS cm−1. Notably, in flammability tests, the electrolyte exhibited non-flammable properties. When employed in LNMO/MCMB 18650 batteries, the system displayed remarkable performance, maintaining an impressive 82% capacity retention and an outstanding 99.6% Coulombic efficiency over 200 cycles, even at a high cut-off voltage of 4.9 V.

7.2

Gel polymer organic composite electrolytes (GPOEs) and its applications

In contrast to solid polymer electrolytes, Gel Polymer Electrolytes (GPOEs), comprising polymer matrices and plasticizers, demonstrate significantly enhanced ionic conductivity within the range of 10−4 to 10−2 S cm−1. Plasticizers play a vital role in diminishing the crystallinity of polymer matrices, improving segmental mobility, aiding lithium salt dissociation, and facilitating the transport of lithium ions. Although traditional plasticizers, frequently carbonates, are highly flammable, the utilization of flame-retardant additives as plasticizers and the integration of flame-retardant polymers as matrices in Gel Polymer Electrolytes (GPOE) can effectively reduce flammability, leading to the creation of non-flammable Gel Polymer Electrolytes. (a) PVDF and PVDF-HFP Non-flammable polymers like poly(vinylidene fluoride-hexafluoropropylene) (PVDF-HFP) and PVDF were widely employed as matrices of polymer in Gel Polymer Electrolytes (GPOEs). These polymers are characterized by robust electron withdrawing C–F groups, contributing significantly to their exceptional electrochemical stability. With a dielectric constant of approximately 8.4, Playing a pivotal role in promoting the dissolution of lithium salts and elevating the concentration of mobile lithium ions, these polymers contribute to the outstanding mechanical strength of Gel Polymer Electrolytes (GPOEs) through their high crystallinity.97–99 In the realm of GPOEs, PVDF-HFP is preferred over PVDF due to the more frequent use of PVDF-HFP as the polymer matrix. The incorporation of HFP units in PVDF-HFP introduces amorphous domains that can absorb more plasticizers, thereby enhancing ionic conductivity. The production of non-flammable GPOEs based on PVDF-HFP typically involves the casting technique, wherein a PVDF-HFP solution is cast onto a flat substrate. After the solvent has evaporated, the non-flammable GPOEs are obtained by activating the membrane by absorbing a non-flammable lithium salt solution. As flame-retardant plasticizers, phosphenes have been added to non-flammable GPOEs based on PVDF-HFP. Morita et al. created a number of non-flammables GPOEs based on PVDF-HFP that included phosphates like TMP,100–102 TEP,103,104 diphenyl phosphite (DPP),105 and Phosphates made up about 20% of the volume ratio of plasticizers, such as 0.8 M LiPF6 in EC/DEC/TMP (55,25,20).101 The non-flammable Gel Polymer Electrolytes (GPOEs) derived from PVDF-HFP showcased commendable thermal stability, exhibiting high ionic conductivity within the range of 0.9 to 6.2 mScm−1 at room temperature. These GPOEs demonstrated excellent performance in LIBs. To enhance their properties, oligomeric ionic liquid (OIL) was belonging to imidazolium, has been incorporated into PVDF-HFP-based non-flammable Gel Polymer Electrolytes as plasticizers. Examples of these include EMI-TFSI,106 PYRA1201-TFSI,59 EMIMDCA,107 BMP-TFSI,108 BMMI-TFSI,109 and 3P(MPBIm-TFSI).110 At room temperature, the ionic conductivity of these non-flammable GPOEs varied between 0.25 and 2.0 mScm−1. Chen et al.110 prepared 3-arm polymeric ionic liquids (PILs) denoted as 3P(MPBIm-X), with X counter anions, namely (CF3SO3) 2 N- (TFSI-), CF3SO3- (Tf-) or Br-. Among these, 3P(MPBIm-X) featuring the TFSI counter anion (3P(MPBIm-TFSI)) demonstrated the most favorable thermal properties and the highest ionic conductivity. To form a solid-like composite electrolyte SLCE, mixture of 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide (EMIM-TFSI) and 3P(MPBIm-TFSI), LiTFSI, were combined into electrospun PVDF-HFP films. Exhibiting notable attributes, the SLCE demonstrated high ionic conductivity (1.2 mS cm−1 at 50
C), a broad electrochemical window (4.8 V), nonflammability, and robust thermal stability (approximately 370
C). In a cycling test of 100 cycles, LFP or Li cells utilizing SLCE exhibited exceptional performance, with a high capacity of 144.1 mAh g−1, an impressive capacity retention of 95%, and a near-ideal Coulombic efficiency of approximately 100%. The applicability of SLCE extended to flexible LIBs, as demonstrated by an LFP or Li pouch battery assembled with SLCE, which maintained normal functionality even when bent at a 180
C angle. (b) Poly (ethylene glycol) (PEG) Numerous GPOEs have been developed based on poly (ethylene glycol) (PEG) and poly (ethylene oxide) (PEO), most of which are flammable, because they share the same repeat unit of -(CH2CH2O) n-. There have been reports of certain non-flammable PEG-based GPOEs.111 Non-flammable GPOEs Tetra PEG was developed by Morita et al..112 Tetra PEG-NH2 and Tetra PEG-NHS prepolymers reacted to create tetra PEG hydrogel, which was subsequently dried. To create Tetra PEG GPOEs, which had a polymer content of 6.2 wt%, the dried gels were swollen with 1.0 M LiPF6 in EC/DEC/TFEP. Tetra PEG, GPOEs were non-flammable when the ternary EC/DEC/TFEP mixture’s TFEP content exceeded 20 wt%.

360

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

Distended with 1.0 M LiPF6 in EC/DEC/TFEP (53:27:20 vol.), Tetra PEG Gel Polymer Electrolytes (GPEs) exhibited a noteworthy ionic conductivity of 2.6 mS cm−1 at 25
C. These Tetra PEG GPEs showcased improved Coulombic efficiency (CE) for graphite/Li cells during the initial cycle, resulting in a reversible capacity of 256 mAh g−1. To bolster mechanical strength and thermal stability, In their work,113 Zheng et al. implemented a sturdy supporting structure for Gel Polymer Electrolytes by employing a high-strength porous film. They created an interpenetrating Gel Polymer Electrolyte that is both rigid-flexible and non-flammable, using electro-spun poly(arylene ether ketone) (PAEKNW) as the framework for the GPEs, known for its commendable thermal and mechanical stability. The fabrication of GPEs utilizing poly(arylene ether ketone) (PAEKNW) involved the polymerization of poly(ethylene glycol) dimethacrylate (PEGDMA) and ethoxylated trimethylolpropane triacrylate (ETPTA) in a solution containing 1 M LiPF6 in EC/DMC/ DEC (1:1:1) + 4% FEC within the PAEKNW framework. These GPEs showcased a generous electrochemical window of 4.5 V, demonstrated non-flammability, and achieved a notable ionic conductivity of 1.2 mS cm−1 at room temperature. PAEKNW GPEs have the potential to mitigate lithium dendrite formation and enhance the stability of battery cycles. Upon completing 200 cycles at 0.5C, K. Deng et al. observed that LFP or Li cells with PAEKNW Gel Polymer Electrolytes (GPEs) retained a capacity of 124 mAh g−1. When subjected to heating at 80
C for one hour, LFP or Li pouch batteries assembled with PAEKNW GPEs displayed no noticeable volumetric expansion or contraction. Remarkably, these batteries continued to operate normally even after one-third of them underwent being nailed through and cut off. (c) Poly acrylate Gel Polymer Electrolytes (GPOEs) using polyacrylate as a base material are commonly produced through either the solution immersion method or in-situ polymerization.114 In the solution immersion approach, the aim is to incorporate plasticizers into the existing polymer networks. Li et al.115 innovatively developed a non-flammable Gel Polymer Electrolyte based on polyurethane acrylate (PUA). Employing UV curing, PUA was used to form a highly cross-linked film. Subsequently, the resulting PUA film underwent a solution immersion process at 50
C, being submerged in melted 1 M LiTFSI/succinonitrile (SN). This allowed the electrolyte to permeate the PUA network, resulting in the formation of GPOE-SN-IM. Succinonitrile (SN), a solid plastic crystal, exhibited excellent solubility for lithium salts and low flammability. At 25
C, GPOE-SN-IM exhibited impressive mechanical properties with a high tensile strength of 6.5 MPa and exceptional ionic conductivity of 1.63 mS cm−1. Remarkably, in ignition tests, GPOE-SN-IM displayed non-flammable characteristics. In LCO/Li cells, GPOE-SN-IM showcased superior cycle performance at 55
C compared to cells utilizing commercial electrolyte-based Gel Polymer Electrolytes (GPOE), attributed to the excellent thermal stability of GPOE-SN-IM. The durable mechanical strength of the GPOE-SN-IM film allowed film batteries based on LCO and Li4Ti5O12 to endure 100 cycles of folding without experiencing damage to the electrolyte or loss of capacity. (d) Polymeric phosphates To mitigate the reaction of minor molecule flame retardant additives with electrodes, flame-retardant elements can be chemically incorporated into polymer matrices, transforming them into flame retardant polymer matrices. Polymers enriched with phosphorus fractions have been employed to develop non-flammable Gel Polymer Electrolyte (GPOE) matrices. Cekic-Laskovic et al.116 achieved this by copolymerizing oligo(ethylene glycol) ether methacrylate (OEGEMA) and dimethyl-p-vinyl benzyl phosphonate (DMpVBnP) to create a phosphates-based copolymer (c-PPO). The subsequent creation of c-PPO/LP30 GPOE involved absorbing the c-PPO polymer films with 1 M LiPF6 in EC/DMC (1:1 wt.), resulting in a GPOE with a notable high ionic conductivity of 1.5 mS cm−1 at 25
C. The flame-retardant effect of the DMpVBnP moiety in c-PPO led to the formation of non-flammable c-PPO/LP30 GPOE. The flame retardant component and the electrodes did not have parasitic reactions because of the flame retardant moiety’s chemical bonding to the polymer. Graphite did not react with C-PPO/LP30 GPOEs, even at elevated temperatures. As a result, after 50 cycles, NCM523/graphite cells with c-PPO/LP30 GPOEs demonstrated strong capacity retentions of 94% and high CE of 99.8% at 0.2C.

8

Inorganic electrolytes for LIBs

Many new technologies are made possible by inorganic electrolytes; recent advances have significantly increased the conductivity of lithium ions. Extensive studies have been conducted on various crystal structures, such as those that are LISICON-like (a type of lithium superionic conductor), garnets, argyrodites, lithium nitrides, lithium hydrides, perovskites, and lithium halides. These structures offer the potential for both structural and compositional adjustments within a specific family of materials, aiming to attain enhanced conductivities. These lithium-ion conductors present promising opportunities for advancing solid-state lithium-ion and lithium-air batteries, particularly in the context of vehicle applications. The heightened stability and safety characteristics of inorganic electrolytes create possibilities to streamline and reconsider safety measures already integrated into battery cells. This includes revisiting features like charge-interruptible devices, overpressure vents, and addressing challenges related to complex thermal management systems and operational strategy limitations within the battery pack.117

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

9

361

Solid inorganic electrolytes

To be applicable across a broad spectrum of operating temperatures, solid electrolytes need to exhibit low activation energy (Ea) and high ionic conductivity at room temperature. Lithium nitride (Li3N), first identified in the 1970s118 and characterized by a high ionic conductivity of 6  10−3 S cm−1 at room temperature, emerges as a promising solid electrolyte.119 Unfortunately, practical applications cannot make use of it due to its low electrochemical decomposition potential. Furthermore, these solid electrolytes not only simplify battery design but also offer additional advantages, such as improved environmental stability and enhanced electrochemical stability against both the anode and cathode. In recent decades, research efforts have predominantly focused on ionically conducting oxides and sulfides. Examples include the NASICON (Na Super Ionic Conductor)-type Li1.3Al0.3Ti1.7(PO4)3,120 LISICON (Lithium Super Ionic Conductor)-type Li14ZnGe4O16,121 perovskite La0.5Li0.5TiO3,122 garnet Li7La3Zr2O12123 and glass-ceramic Li2S-P2S5.124,125 At room temperature, these conductors display ionic conductivities of approximately 10−3 S cm−1, with an Ea ranging from 0.3 to 0.6 eV.126 Amorphous LIPON (Lithium Phosphorus Oxynitride) is another system that is presently being studied as a solid electrolyte. Despite having a comparatively lower ionic conductivity of 2  10−6 S cm−1 at 25
C, LIPON can be made up for in solid-state batteries by using a very thin layer (1 mm) as the electrolyte.127,128 Additionally, it has been noted to exhibit outstanding cell performance at room temperature for thousands of cycles.129 However, thin-film batteries have drawbacks such as low cell capacity owing to low active material loading and high fabrication costs. Inorganic Li-ion conductors have been the subject of several recent overview publications, including those by Kim et al.,130 Knauth,126 Goodenough and Singh,131 and Quartarone and Mustarelli.132 Furthermore, Anantharamulu et al.133 provided a comprehensive overview of NASICON-type compositions; Teng et al.134 reviewed recent advancements in garnet solid electrolytes; Thangadurai et al.135 Performed a comparative analysis of garnet type solid state Li ion conductors for lithium batteries; additionally, Berbano et al.136 and Tatsumisago et al.137 reported advancements in the processing and fabrication of all solid state lithium batteries with a focus on the development of sulfide solid electrolytes. We discussed the development of sulfide-based solid electrolytes for lithium batteries in this chapter. However, in contrast to references,136,138 our primary emphasis is on the solid electrolytes based on sulfides from a structural perspective. Initially, this discussion provides a review of the structural advancements in Li2S-P2S5 glass and glass-ceramic. During this period, various additives, such as MxSy (M ¼ Sn, Ge, Si, Bi, etc.) and LiaXb (X ¼ Cl, Br, I, O, etc.), have been utilized to enhance the ionic conductivity of solid-state electrolytes based on sulfides. As a result, this chapter also covers the structural change and ionic conductivity of the Li2S-P2S5-LiaXb system and the Li2S-MxSy-P2S5 solid solution independently. i) Sulfide-based solid electrolytes The material design of crystalline ionic conductors relies on crucial structural criteria,139,140 which include: (a) ensuring that mobile ions have a size compatible with the lattice’s conduction pathways; (b) introducing disorder within the mobile ion sublattice; and (c) favoring highly polarizable mobile ions and anion sublattices. The substitution of S2− for O2− can significantly enlarge the size of Li+ transport bottlenecks due to the larger radius of S2−. Additionally, S2− exhibits a weaker interaction between the skeleton and Li+ ions compared to O 2−, attributed to its superior polarizability. As a result, several sulfide compounds have undergone investigation, revealing high room temperature ionic conductivity exceeding 10−5 Scm−1. For visual representation, Fig. 5 showcases various Li2S-GeS-P2S5 ternary system compounds functioning as Li-ion conductors, while Table 2 provides a summary of their respective conductivities. Sulfide glasses within the Li2S-P2S5 and Li2S-SiS2 systems, prepared through the melt-quenching method, have been identified as Li-ion conductors with room temperature conductivities exceeding 10−4 S cm−1.142,144,145 Notably, in the Li2S-P2S5 binary system, the melt-quenching method resulted in obtaining perfect amorphous structures devoid of any crystalline formation. Li2S contents of 75 mol%, and in the case of 75Li2S25P2S5, the maximum conductivity of the glassy powders was approximately 2  10−4 S cm−1 at 25
C.143,146,147,158,159 Many strategies have been put forth to increase the conductivity of glassy electrolytes. The glass precursors can be simply crystallized as an efficient method. Glass-ceramic (crystallized glass) electrolytes are obtained by precipitating thermodynamically stable Li2S-P2S5 glass electrolytes. Nonetheless, varying findings have been documented regarding the association between conductivity and crystallization. Minami and Machida,160 for example, proposed that glasses that conduct Cu + ions have advanced ionic conductivity than crystals because glasses have a bigger free volume due to their random and open structure. Li2S contents of 75 mol%, and in the case of 75Li2S25P2S5, the maximum conductivity of the glassy powders was approximately 2  10−4Scm−1 at 25
C.143,146,147,158,159 Many strategies have been put forth to increase the conductivity of glassy electrolytes. The glass precursors can be simply crystallized as an efficient method. Glass-ceramic (crystallized glass) electrolytes are obtained by precipitating thermodynamically stable Li2S-P2S5 glass electrolytes. Nonetheless, varying findings have been documented regarding the association between conductivity and crystallization. Minami and Machida,160 for example, proposed that glasses that conduct Cu+ ions have higher ionic conductivity than crystals because glasses have a larger free volume due to their random and open structure. On the other hand, a notable rise in the electronic conductivity of nano crystallized V2O5-P2O5 glasses was demonstrated by Pietrzak et al..161 Ionic conductivity in the Li2S-P2S5 system was found to be temperature-dependent.140,143

362

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

Fig. 5 A Li2S-GeS2-P2S5 ternary diagram is presented, depicting a range of sulfide compounds serving as solid electrolytes for Li-ion batteries. Produced with permission from Liu, D.; Zhu, W.; Feng, Z.; Guerfi, A.; Vijh, A.; Zaghib, K. Recent Progress in Sulfide-Based Solid Electrolytes for Li-Ion Batteries. Mater. Sci. Eng. B 2016, 213, 169–176.

Table 2

9.1

Conductivities of different sulfide compounds at 25
C.141

Composition

Conductivity at 25
C (Scm−1)

Classification

Reference

2Li2SP2S5 70Li2S30P2S5 75Li2S25P2S5 70Li2S30P2S5 80Li2S20P2S5 Li7P3S11−z Li7P3S11 Li3.25P0.95S4 -Li3PS4 ß- Li3PS4 Li3.25Ge0.25P0.75S Li10GeP2S12 Li10SnP2S12 Li11Si2PS12 80(0.7Li2S0.3 P2S)20LiI 95(0.8Li2S0.2P2S5)5LiI Li7P2S8I 56Li2S24P2S520Li2O 75Li2S21P2S54P2O5 67.5Li2S7.5Li2O25P2S5 0.33(0.7B2S30.3P2S5)-0.67L2S

1.0  10−4 5.4  10−5 2.0  10−4 3.2  10−3 7.4  10−4 5.4  10−3 1.7  10−2 1.3  10−3 3.0  10−7 1.6  10−4 2.2  10−3 1.2  10−2 4.0  10−3 >1.2  10−2 5.6  10−4 2.7  10−3 6.3  10−4 >1.0  10−4 >1.0  10−4 1.1  10−4 1.4  10−4

Glass Glass Glass Glass/ceramic Glass/ceramic Glass/ceramic Glass/ceramic Glass/ceramic Crystal-like Crystal-like Crystal-like Crystal-like Crystal-like Crystal-like Glass Glass Crystal-like Glass Glass Glass Glass

142 140 143 140 143 144 145 143 146 147 139 148 149 150 151 152 153 154 155 156 157

Non-flammable inorganic liquid electrolytes

In 1988, Foster et al. introduced a distinct liquid-IE, comprising gaseous SO2 or either liquid dissolved in molten LiAlCl4. This inorganic electrolyte demonstrated several advantageous properties like increased Li+ transference number, including a high concentration of Li ions, outstanding ionic conductivity exceeding 100 mScm−1 at room temperature, non-flammability, an enhanced di-electric constant attributed to the presence of compatibility with a lithium anode are among the desirable

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

363

characteristics and SO2 electric dipoles. The initial study by Foster et al.162 highlighted the superior performance of CuCl2 rechargeable batteries compared to Li/Li+ batteries when utilizing IEs. Subsequent research by Park et al.163 explored the electrochemical behavior of LiCoO2 vs. Li/Li+ using IE. Additionally, a study by Zinck et al.162–170 demonstrated electrochemical performance, with Li/Li+ utilizing IE exhibiting 90% capacity retention after 750 cycles vs LiCoO2. Under the leadership of John B. Goodenough et al., phosphor-olivines LiMPO4 (where M ¼ Fe/Mn) have emerged as promising cathode materials for lithium-ion batteries (LIBs). Among these phospho-olivines, despite its exceptional qualities such as safety, high thermal stability, affordability, and environmental friendliness, LFP, with a theoretical capacity of 170 mAh g−1 and a flat redox potential of 3.45 V vs Li/Li+, yielding a specific energy of 586 Wh kg−1,171–177 has encountered challenges in its performance as a cathode material with organic electrolytes (OE). Issues such as Cu dissolution from the anode and metallic redeposition on the electrode materials have been observed, leading to limited longevity (in comparison to graphite anode) and restricted over-discharge for very few cycles.178–181 The Goodenough et al. group recently demonstrated the electrochemical performance of LiFePO4 (LFP) vs. Li/Li+ with an inorganic electrolyte (IE), showcasing promising results even at low temperatures of −20
C. The study suggested the feasibility of the Fe3+/Fe2+ redox couple (3.45 V) within the potential window of IE. Subsequently, Gao et al.182,183 conducted comprehensive half-cell electrochemical studies with LFP/Li and Graphite/Li, systematically comparing the outcomes with organic electrolytes (OE). However, information regarding deep discharge or long cycle life using LFP/graphite full-cells in IE has been lacking until now. This report presents, for the first time, a commercial LFP/graphite prototype based on IE, exhibiting constant internal resistance, a coulombic efficiency of approximately 99.99%, and exceptional longevity of 50,000 cycles with 20% residual capacity. Utilizing a patented dip-coating method to coat the electrodes’ surfaces further enhances capacity retention, introducing a very stable material that contributes to a potential 5% improvement in cycling stability. While the precise cause is yet unknown, attaching an inert stable layer is believed to assist in suppressing side reactions and limiting capacity loss. Additionally, the study demonstrates the achievement of a discharge depth of up to 0 V for 150 cycles at a rate of 2C.184

9.2

Polymer composite inorganic electrolytes

The (dCH2dCH2dOd)n oligo-ether serves as the structural unit in SPEs, and the polyether structure has received the majority of attention in this field. Because poly (ethylene oxide) (PEO) has a low glass transition temperature and can dissolve Li salts, it was used in practical applications. Regretfully, in order to raise the charge rate to a useful level and prevent concentration gradients from forming in the electrolyte, a much higher ionic conductivity and transference number are required.52,185 The primary cause of PEO electrolytes’ low conductivity is their partial crystallization up to 60
C, which lowers chain mobility. Various strategies were taken into consideration to solve this issue. Grafted co-polymers are added to prevent the polyether chains from crystallizing while maintaining their mechanical characteristics. A co-polymer with grafts, consisting of poly(methoxy/hexadeca-poly(ethylene glycol) methacrylate), exhibited significant conductivity at 10−4 S cm−1 at room temperature.186 Another approach includes integrating ceramics, layered clays, and diverse mesoporous particles into the polymer matrix (for a recent overview of the influence of nanoparticles on electrolytes and the electrolyte interface, see 187). The incorporation of inorganic particles as surface plasticizers decreases the polymer’s crystalline fraction. Amorphization induced by these particles not only boosts ionic conductivity by establishing more conductive pathways but also improves overall conductivity. Furthermore, the inorganic fillers act as Lewis acid-based centers, facilitating salt dissociation and releasing additional Li+ ions, thereby augmenting the transference number. An illustrative instance of a reversible transition is evident in a PEO-based electrolyte, like PEO8-LiClO4 (8:1), which transitions into an amorphous state when heated above 60
C and reverts to a crystalline state upon cooling back to room temperature. However, the introduction of TiO2 or Al2O3 prevents the polymer chains from undergoing recrystallization, maintaining the material in an amorphous state upon cooling. As a result, the transference number increased to 0.5–0.6, and the ionic conductivity experienced a boost from 5  10−8 to −5 10 S cm−1 (refer to Fig. 6).52,188 The ceramic’s Lewis acid property plays a crucial role in elevating the lithium transference number. This property not only hinders polymer crystallization but also competes with Li+ ions, forming complexes with the polymer chains. The addition of 10 wt% S-ZrO2 to PEO20-LiBF4 resulted in a transference number increase to 0.8, attributed to the significantly higher acidity of nanosized sulfated zirconia (100% higher than sulfuric acid).189 In comparison to an 80 mAh g−1 capacity after 150 cycles at a C/7 rate under the same conditions without SZrO2, the Li/5% SdZrO2 cell containing PEO20-LiBF4/ LiFePO4 delivered a capacity of 160 mAh g−1 after the first cycle and 140 mAh g−1 after 150 cycles at a rate of C/5 (current density 0.20 mA cm−2) and 90
C. Likewise, the capacity of Li-metal//LiFePO4 cells increased to 120 mAh g−1 over 20 cycles at a 0.5C rate by incorporating 3 wt% silica aerogel powder (SAP) into PAN-LiClO4 as the electrolyte.190 Due to the Lewis acidity of the boron center, borate-based anion receptors have proven effective when combined with PEO-based electrolytes, serving either as additives or being covalently bonded to the polymer chains. This utilization enhances both ion conductivity and the transference number.191 Another strategy to prevent crystallization involves introducing non-covalent bonds.192–194 Specifically, successful electrostatic interactions between PEO+ and PEO-charged ionomers have improved mechanical strength and maintained the amorphous state. PEO and its derivatives are not the only polymer structures proposed for this purpose. For instance, a cross-linked polyphosphazene-based electrolyte with pendant ether chains, mixed with lithium LiTFSI and LiBOB salt, achieves good conductivity at room temperature and stability against Li metal. However, its practical application as an electrolyte is hindered by the low Li + transference number.195

364

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

–2

Log [V (S cm–1)]

–3

–4

–5

Ceramic-free TiO2, cooling scan

–6

TiO2, heating scan Al2O3, 1st heating scan

–7

Al2O3, cooling scan Al2O3, 2nd heating scan

–8 2.6

2.8

3.0

3.2

3.4

–1

1000/T (K ) Fig. 6 Arrhenius plots depicting the conductivity of nanocomposite systems, including PEO8LiClO (4–10 wt%) TiO2 and PEO8LiClO (4–10 wt%) Al2O3, are presented. The plot of a ceramic-free PEO8LiClO4 polymer electrolyte is provided for comparison. Produced with permission from Mauger, A.; Armand, M.; Julien, C.; Zaghib, K. Challenges and Issues Facing lithium Metal for Solid-State Rechargeable Batteries. J. Power Sources 2017, 353, 333–342.

10

Conclusion

In the past few decades, a lot of research on composite solid electrolytes which combine organic and inorganic fillers with polymer electrolyte systems (PEO, PAN, PVDF, etc.) have been presented. Improved ionic conductivity and increased mechanical strength and stability of all-solid-state electrolytes are two benefits of adding inorganic and organic fillers (active or passive fillers) to polymer electrolytes. This chapter provided an overview of the development of composite solid, liquid, gel, polymer, and electrolyte composites historically. The study assessed the impact of active Li-ion conductors, as well as inert organic and inorganic fillers, on the mechanical properties, electrochemical stability, and ionic conductivity of composite electrolytes. Conversely, a thorough discussion was held regarding potential mechanisms through which inorganic and organic fillers could enhance conductivity. A variety of composite solid electrolyte design concepts were presented and their benefits and drawbacks assessed. These included inorganic nanoparticle/polymer, inorganic nanofiber/polymer, and other inorganic/polymer composite solid electrolytes, among others. Solid electrolytes with inorganic and organic filler/polymer composites that were researched for use in different Li battery systems, including Li-ion batteries, were detailed.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

Yuan, X.; Liu, H.; Zhang, J. LIBs: Advanced Materials and Technologies; CRC Press Taylor & Francis Group, 2011. Zhu, P.; Slater, P. R.; Kendrick, E. Insights into Architecture, Design and Manufacture of Electrodes for LIBs. Mater. Design 2022, 223, 111208. Reddy, M. V.; Julien, C. M.; Mauger, A.; Zaghib, K. Sulfide and Oxide Inorganic Solid Electrolytes for All-Solid-State Li Batteries: A Review. Nanomaterials 2020, 10, 1606. Xu, K. Nonaqueous Liquid Electrolytes for Lithium-Based Rechargeable Batteries. Chem. Rev. 2004, 104, 4303–4418. Zinigrad, E.; Larush-Asraf, L.; Gnanaraj, J. S.; Gottlieb, H. E.; Sprecher, M.; Aurbach, D. Calorimetric Studies of the Thermal Stability of Electrolyte Solutions Based on Alkyl Carbonates and the Effect of the Contact with lithium. J. Power Sources 2005, 146, 176–179. Wang, Y.; Nakamura, S.; Ue, M.; Balbuena, P. B. Theoretical Studies to Understand Surface Chemistry on Carbon Anodes for LIBs: Reduction Mechanisms of Ethylene Carbonate. J. Am. Chem. Soc. 2001, 123, 11708–11718. Blomgren, G. E. Electrolytes for Advanced Batteries. J. Power Sources 1999, 81–82, 112–118. Archuleta, M. M. Toxicity of Materials Used in the Manufacture of lithium Batteries. J. Power Sources 1995, 54, 138–142. Chung, D.; Elgqvist, E.; Santhanagopalan, S. Automotive Lithium-Ion Cell Manufacturing: Regional Cost Structures and Supply Chain Considerations; NREL/TP-6A20-66086, CEMAC Tech, April 2016. Zhao, X.; Zhao-Karger, Z.; Wang, D.; Fichtner, M. Metal Oxychlorides as Cathode Materials for Chloride Ion Batteries. Angew. Chem. Int. Ed. 2013, 52, 13621–13624. Kuai, D.; Balbuena, P. B. Inorganic Solid Electrolyte Interphase Engineering Rationales Inspired by Hexafluorophosphate Decomposition Mechanisms, Dielectric Constant in Liquid Electrolytes. J. Phys. Chem. C 2023, 127, 1744–1751. Jijian, X.; Zhang, J.; Pollard, T. P.; Li, Q.; Tan, S.; Hou, S.; Hongli Wan, F.; Chen, H. H.; Enyuan, H.; Kang, X.; Yang, X.-Q.; Borodin, O.; Chunsheng, W. Electrolyte Design for Li-Ion Batteries under Extreme Operating Conditions. Nature 2023, 614, 695–702. Yang, X.; Zhang, B.; Tian, Y.; Wang, Y.; Zhiqiang, F.; Zhou, D.; Liu, H.; Kang, F.; Li, B.; Wang, C. Guoxiu Wang Electrolyte Design Principles for Developing Quasi-Solid-State Rechargeable Halide-Ion Batteries. Nature. Comm. 2023, 14, 925.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

365

14. Lannelongue, P.; Bouchal, R.; Mourad, E.; Bodin, C.; Olarte, M.; le Vot, S.; Favier, F.; Fontaine, O. Water-in-Salt for Supercapacitors: A Compromise between Voltage, Power Density, Energy Density and Stability. J. Electrochem. Soc. 2018, 165, A657–A663. 15. Borodin, O.; Suo, L.; Gobet, M.; Ren, X.; Wang, F.; Faraone, A.; Peng, J.; Olguin, M.; Schroeder, M.; Ding, M. S.; Gobrogge, E.; von Wald Cresce, A.; Munoz, S.; Dura, J. A.; Greenbaum, S.; Wang, C.; Xu, K. Liquid Structure with Nano-Heterogeneity Promotes Cationic Transport in Concentrated Electrolytes. ACS Nano 2017, 11, 10462–10471. 16. Ding, M. S.; von Cresce, A.; Xu, K. Conductivity, Viscosity, and their Correlation of a Super-Concentrated Aqueous Electrolyte. J. Phys. Chem. C 2017, 121, 2149–2153. 17. Bu, X.; Su, L.; Dou, Q.; Lei, S.; Yan, X. A Low-Cost “Water-in-Salt” Electrolyte for a 2.3 V High-Rate Carbon-Based Supercapacitor. J. Mater. Chem. A 2019, 7, 7541–7547. 18. Stigliano, P. L.; Pianta, N.; Bonizzoni, S.; Mauri, M.; Simonutti, R.; Lorenzi, R.; Vigani, B.; Berbenni, V.; Rossi, S.; Mustarelli, P.; Ruffo, R. A Physico-Chemical Investigation of Highly Concentrated Potassium Acetate Solutions towards Applications in Electrochemistry. Phys. Chem. Chem. Phys. 2021, 23, 1139–1145. 19. Lukatskaya, M. R.; Feldblyum, J. I.; Mackanic, D. G.; Lissel, F.; Michels, D. L.; Cuie, Y.; Bao, Z. Concentrated Mixed Cation Acetate “Water-in-Salt” Solutions as Green and Low-Cost High Voltage Electrolytes for Aqueous Batteries. Energ. Environ. Sci. 2018, 11, 2876–2883. 20. Zhang, M.; Hao, H.; Zhou, D.; Duan, Y.; Wang, Y.; Bian, H. Understanding the Microscopic Structure of a “Water-in-Salt” Lithium Ion Battery Electrolyte Probed with Ultrafast IR Spectroscopy. J. Phys. Chem. C 2020, 124, 8594–8604. 21. Lim, J.; Park, K.; Lee, H.; Kim, J.; Kwak, K.; Cho, M. Nanometric Water Channels in Water-in-Salt Lithium Ion Battery Electrolyte. J. Am. Chem. Soc. 2018, 140, 15661–15667. 22. Popov, I.; Sacci, R. L.; Sanders, N. C.; Matsumoto, R. A.; Thompson, M. W.; Osti, N. C.; Kobayashi, T.; Tyagi, M.; Mamontov, E.; Pruski, M.; Cummings, P. T.; Sokolov, A. P. Critical Role of Anion–Solvent Interactions for Dynamics of Solvent-in-Salt Solutions. J. Phys. Chem. C 2020, 124, 8457–8466. 23. Yamada, Y.; Usui, K.; Sodeyama, K.; Ko, S.; Tateyama, Y.; Yamada, A. Hydrate-Melt Electrolytes for High-Energy-Density Aqueous Batteries. Nat. Energy 2016, 1, 1–9. 24. Han, M.; Zhang, R.; Gewirth, A. A.; Espinosa-Marzal, R. M. Nano Heterogeneity of LiTFSI Solutions Transitions Close to a Surface and with Concentration. NanoLett. 2021, 21, 2304–2309. 25. Xu, D.; Wang, Z. L.; Xu, J. J.; Zhang, L. L.; Wang, L. M.; Zhang, X. B. A Stable Sulfone-Based Electrolyte for High Performance Rechargeable Li–O2 Batteries. Chem. Commun. 2012, 48, 11674–11676. 26. Jiang, L.; Lu, Y.; Zhao, C.; Liu, L.; Zhang, J.; Zhang, Q.; Shen, X.; Zhao, J.; Yu, X.; Li, H.; Huang, X.; Chen, L.; Hu, Y.-S. Building Aqueous K-Ion Batteries for Energy Storage. Nat. Energy 2019, 4, 495–503. 27. Deng, W.; Wang, X.; Liu, C.; Li, C.; Chen, J.; Zhu, N.; Li, R.; Xue, M. Li/K Mixed Superconcentrated Aqueous Electrolyte Enables High-Performance Hybrid Aqueous Supercapacitors. Energy Storage Mater. 2019, 20, 373–379. 28. Dou, Q.; Lei, S.; Wang, D.-W.; Zhang, Q.; Xiao, D.; Guo, H.; Wang, A.; Yang, H.; Li, Y.; Shi, S.; Yan, X. Safe and High-Rate Supercapacitors Based on an “Acetonitrile/Water in Salt” Hybrid Electrolyte. Energ. Environ. Sci. 2018, 11, 3212–3219. 29. Borodin, O.; Self, J.; Persson, K. A.; Wang, C.; Xu, K. Uncharted Waters: Super-Concentrated Electrolytes. Joule 2020, 4, 69–100. 30. Dave, A.; Gering, K. L.; Mitchell, J. M.; Whitacre, J.; Viswanathan, V. Benchmarking Conductivity Predictions of the Advanced Electrolyte Model (AEM) for Aqueous Systems. J. Electrochem. Soc. 2020, 167, 013514. 31. Yu, Z.; Curtiss, L. A.; Winans, R. E.; Zhang, Y.; Li, T.; Cheng, L. Asymmetric Composition of Ionic Aggregates and the Origin of High Correlated Transference Number in Waterin-Salt Electrolytes. J. Phys. Chem. Lett. 2020, 11, 1276–1281. 32. Horwitz, G.; Rodríguez, C. R.; Steinberg, P. Y.; Burton, G.; Corti, H. R. Mobility-Viscosity Decoupling and Cation Transport in Water-in-Salt lithium Electrolytes. Electrochim. Acta 2020, 359, 136915. 33. González, M. A.; Borodin, O.; Kofu, M.; Shibata, K.; Yamada, T.; Yamamuro, O.; Xu, K.; Price, D. L.; Saboungi, M.-L. Nanoscale Relaxation in “Water-in-Salt” and “Water-in-Bisalt” Electrolytes. J. Phys. Chem. Lett. 2020, 11, 7279–7284. 34. Gering, K. L. Prediction of Electrolyte Viscosity for Aqueous and Non-aqueous Systems: Results from a Molecular Model Based on Ion Solvation and a Chemical Physics Framework. Electrochim. Acta 2006, 51, 3125–3138. 35. Horwitz, G.; Steinberg, P. Y.; Corti, H. R. Volumetric and Viscosity Properties of Water-in-Salt lithium Electrolytes: A Comparison with Ionic Liquids and Hydrated Molten Salts. J. Chem. Thermodyn. 2021, 158, 106457. 36. Reber, D.; Figi, R.; Kühnel, R.-S.; Battaglia, C. Stability of Aqueous Electrolytes Based on LiFSI and NaFSI. Electrochim. Acta 2019, 321, 134644. 37. Zhang, S. S. A Review on Electrolyte Additives for LIBs. J. Power Sources 2006, 162, 1379–1394. 38. Wu, C. G.; Lu, M. I.; Chuang, H. J. PVdF-HFP/P123 Hybrid with Mesopores: A New Matrix for High-Conducting, Low-Leakage Porous Polymer Electrolyte. Polymer 2005, 46, 5929. 39. Zhu, Y. S.; Wang, F. X.; Liu, L. L.; Xiao, S. Y.; Yang, Y. Q.; Wu, Y. P. Cheap Glass fiber Mats as a Matrix of Gel Polymer Electrolytes for Lithium Ion Batteries. Sci. Rep. 2013, 3, 3187. 40. Park, M.; Zhang, X. C.; Chung, M. D.; Less, G. B.; Sastry, A. M. A Review of Conduction Phenomena in Li-Ion Batteries. J. Power Sources 2010, 195, 7904–7929. 41. Wu, M. S.; Liao, T. L.; Wang, Y. Y.; Wan, C. C. Assessment of the Wettability of Porous Electrodes for LIBs. J. Appl. Electrochem. 2004, 34, 797–805. 42. Bruce, P. G.; Freunberger, S. A.; Hardwick, L. J.; Tarascon, J. M. Li–O2 and Li–S Batteries with High Energy Storage. Nat. Mater. 2012, 11, 19–29. 43. Scheers, J.; Fantini, S.; Johansson, P. A Review of Electrolytes for lithium–Sulphur Batteries. J. Power Sources 2014, 255, 204–218. 44. Yin, Y. X.; Xin, S.; Guo, Y. G.; Wan, L. J. Lithium–Sulfur Batteries: Electrochemistry, Materials, and Prospects. Angew. Chem. Int. Ed. 2013, 52, 13186–13200. 45. Manthiram, A.; Fu, Y. Z.; Su, Y. S. Challenges and Prospects of Lithium–Sulfur Batteries. Acc. Chem. Res. 2013, 46, 1125–1134. 46. Xu, D.; Wang, Z. L.; Xu, J. J.; Zhang, L. L.; Zhang, X. B. Novel DMSO-Based Electrolyte for High Performance Rechargeable Li–O2 Batteries. Chem. Commun. 2012, 48, 6948–6950. 47. Black, R.; Adams, B.; Nazar, L. F. Non-Aqueous and Hybrid Li-O2 Batteries. Adv. Energy Mater. 2012, 2, 801–815. 48. Zhou, D. Y.; Li, W. S.; Tan, C. L.; Zuo, X. X.; Huang, Y. J. Cresyl Diphenyl Phosphate as Flame Retardant Additive for LIBs. J. Power Sources 2008, 184, 589–592. 49. Wang, Y.; Zhong, W.-H. Development of Electrolytes towards Achieving Safe and High-Performance Energy-Storage Devices: A Review. Chem. Electro. Chem. 2015, 2, 22–36. 50. Fenton, D. Complex of Alkali Metal Ions with Poly (Ethylene Oxide). Polymer 1973, 14, 589. 51. Li, Q.; Chen, J.; Fan, L.; Kong, X.; Lu, Y. Progress in Electrolytes for Rechargeable Li-Based Batteries and Beyond. Green Energy Environ. 2016, 1, 18–42. 52. Mauger, A.; Armand, M.; Julien, C.; Zaghib, K. Challenges and Issues Facing lithium Metal for Solid-State Rechargeable Batteries. J. Power Sources 2017, 353, 333–342. 53. Quartarone, E.; Mustarelli, P.; Magistris, A. PEO-Based Composite Polymer Electrolytes. Solid State Ion. 1998, 110, 1–14. 54. Zhu, P.; Yan, C.; Dirican, M.; Zhu, J.; Zang, J.; Selvan, R. K.; Chung, C.-C.; Jia, H.; Li, Y.; Kiyak, Y. Li0.33La0.557TiO3 Ceramic Nanofiber-Enhanced Polyethylene Oxide-Based Composite Polymer Electrolytes for all-Solid-State lithium Batteries. J. Mater. Chem. A 2018, 6, 4279–4285. 55. Bouchet, R.; Maria, S.; Meziane, R.; Aboulaich, A.; Lienafa, L.; Bonnet, J.-P.; Phan, T. N.; Bertin, D.; Gigmes, D.; Devaux, D. Single-Ion BAB Triblock Copolymers as Highly Efficient Electrolytes for Lithium-Metal Batteries. Nat. Mater. 2013, 12, 452–457. 56. Ito, Y.; Kanehori, K.; Miyauchi, K.; Kudo, T. Ionic Conductivity of Electrolytes Formed from PEO-LiCF3SO3 Complex Low Molecular Weight Poly(Ethylene Glycol). J. Mater. Sci. 1987, 22, 1845–1849. 57. Khurana, R.; Schaefer, J. L.; Archer, L. A.; Coates, G. W. Suppression of Lithium Dendrite Growth Using Cross-Linked Polyethylene/Poly(Ethylene Oxide) Electrolytes: A New Approach for Practical Lithium-Metal Polymer Batteries. J. Am. Chem. Soc. 2014, 136, 7395–7402. 58. Hu, P.; Chai, J.; Duan, Y.; Liu, Z.; Cui, G.; Chen, L. Progress in Nitrile-Based Polymer Electrolytes for High Performance Lithium Batteries. J. Mater. Chem. A 2016, 4, 10070–10083. 59. Ferrari, S.; Quartarone, E.; Mustarelli, P.; Magistris, A.; Fagnoni, M.; Protti, S.; Gerbaldi, C.; Spinella, A. Lithium ion Conducting PVdF-HFP Composite Gel Electrolytes Based on N-Methoxyethyl-N-Methylpyrrolidinium Bis(Trifluoromethanesulfonyl)-Imide Ionic Liquid. J. Power Sources 2010, 195, 559–566.

366

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

60. Wang, Z.; Huang, B.; Xue, R.; Huang, X.; Chen, L. Spectroscopic Investigation of Interactions among Components and Ion Transport Mechanism in Polyacrylonitrile Based Electrolytes. Solid State Ion. 1999, 121, 141–156. 61. Rahman, M. Y. A.; Ahmad, A.; Ismail, L.; Salleh, M. M. Fabrication and Characterization of a Solid Polymeric Electrolyte of PAN-TiO2-LiClO4. J. Appl. Polym. Sci. 2010, 115, 2144–2148. 62. Feuillade, G.; Perche, P. Ion-Conductive Macromolecular Gels and Membranes for Solid lithium Cells. J. Appl. Electrochem. 1975, 5, 63–69. 63. Iijima, T.; Toyoguchi, Y.; Eda, N. Quasi-Solid Organic Electrolytes Gelatinized with Polymethyl-Methacrylate and their Applications for Lithium Batteries. Denki Kagaku 1985, 53, 619–623. 64. Rajendran, S.; Mahendran, O.; Kannan, R. Characterisation of [(1−x)PMMA–xPVdF] Polymer Blend Electrolyte with Li+ Ion. Fuel 2002, 81, 1077–1081. 65. Su’ait, M. S.; Ahmad, A.; Hamzah, H.; Rahman, M. Y. A. Preparation and Characterization of PMMA–MG49–LiClO4 Solid Polymeric Electrolyte. J. Phys. D Appl. Phys. 2009, 42, 055410. 66. Every, H.; Zhou, F.; Forsyth, M.; MacFarlane, D. Lithium ion Mobility in Poly(Vinyl Alcohol) Based Polymer Electrolytes as Determined by 7Li NMR Spectroscopy. Electrochim. Acta 1998, 43, 1465–1469. 67. Rajendran, S.; Sivakumar, M.; Subadevi, R. Li-Ion Conduction of Plasticized PVA Solid Polymer Electrolytes Complexed with Various lithium Salts. Solid State Ion. 2004, 167, 335–339. 68. Yan, C.; Xu, R.; Xiao, Y.; Ding, J.-F.; Xu, L.; Li, B.-Q.; Huang, J.-Q. Toward Critical Electrode/Electrolyte Interfaces in Rechargeable Batteries. Adv. Funct. Mater. 2020, 30, 1909887. 69. Wang, H.; Zhai, D.; Kang, F. Solid Electrolyte Interphase (SEI) in Potassium Ion Batteries. Energ. Environ. Sci. 2020, 13, 4583–4608. 70. Deng, K.; Zeng, Q.; Wang, D.; Liu, Z.; Wang, G.; Qiu, Z.; Zhang, Y.; Xiao, M.; Meng, Y. Nonflammable Organic Electrolytes for High-Safety LIBs. Energy Storage Mater. 2020, 32, 425–447. 71. Otsuki, M.; Ogino, T. Flame-Retardant Additives for LIBs. In LIBs: Science and Technologies; Yoshio, M., Brodd, R. J., Kozawa, A., Eds.; Springer New York: New York, NY, 2009; pp. 1–15. 72. Yao, X. L.; Xie, S.; Chen, C. H.; Wang, Q. S.; Sun, J. H.; Li, Y. L.; Lu, S. X. Comparative Study of Trimethyl Phosphite and Trimethyl Phosphate as Electrolyte Additives in LIBs. J. Power Sources 2005, 144, 170–175. 73. Wang, X.; Yasukawa, E.; Kasuya, S. Nonflammable Trimethyl Phosphate Solvent-Containing Electrolytes for LIBs: I. Fundamental Properties. J. Electrochem. Soc. 2001, 148, A1058–A1065. 74. Wang, X.; Yasukawa, E.; Kasuya, S. Nonflammable Trimethyl Phosphate Solvent-Containing Electrolytes for LIBs: II. The Use of an Amorphous Carbon Anode. J. Electrochem. Soc. 2001, 148, A1066–A1071. 75. Matsumoto, K.; Nakahara, K.; Inoue, K.; Iwasa, S.; Nakano, K.; Kaneko, S.; Ishikawa, H.; Utsugi, K.; Yuge, R. Performance Improvement of Li Ion Battery with Non-Flammable TMP Mixed Electrolyte by Optimization of Lithium Salt Concentration and SEI Preformation Technique on Graphite Anode. J. Electrochem. Soc. 2014, 161, A831–A834. 76. Matsumoto, K.; Inoue, K.; Utsugi, K. A Highly Safe Battery with a Non-flammable Triethyl-Phosphate-Based Electrolyte. J. Power Sources 2015, 273, 954–958. 77. Wu, L.; Song, Z.; Liu, L.; Guo, X.; Kong, L.; Zhan, H.; Zhou, Y.; Li, Z. A New Phosphate-Based Nonflammable Electrolyte Solvent for Li-Ion Batteries. J. Power Sources 2009, 188, 570–573. 78. Jin, Z.; Wu, L.; Song, Z.; Yan, K.; Zhan, H.; Li, Z. A New Class of Phosphates as CoSolvents for Nonflammable LIBs Electrolytes. ECS Electro Chem. Lett. 2012, 1, A55–A58. 79. Lai, Y.; Ren, C.; Lu, H.; Zhang, Z.; Li, J. Compatibility of Diphenyloctyl Phosphate as Flame-Retardant Additive with LiNi1/3Co1/3Mn1/3O2/Artificial Graphite Cells. J. Electrochem. Soc. 2012, 159, A1267–A1272. 80. Shim, E.-G.; Nam, T.-H.; Kim, J.-G.; Kim, H.-S.; Moon, S.-I. Diphenyloctyl Phosphate as a Flame-Retardant Additive in Electrolyte for Li-Ion Batteries. J. Power Sources 2008, 175, 533–539. 81. Nam, T.-H.; Shim, E.-G.; Kim, J.-G.; Kim, H.-S.; Moon, S.-I. Diphenyloctyl Phosphate and Tris(2,2,2-Trifluoroethyl) Phosphite as Flame-Retardant Additives for Li-Ion Cell Electrolytes at Elevated Temperature. J. Power Sources 2008, 180, 561–567. 82. Shim, E.-G.; Nam, T.-H.; Kim, J.-G.; Kim, H.-S.; Moon, S.-I. Effect of the Concentration of Diphenyloctyl Phosphate as a Flame-Retarding Additive on the Electrochemical Performance of LIBs. Electrochim. Acta 2009, 54, 2276–2283. 83. Feng, J.; Lu, L. A Novel Bifunctional Additive for Safer LIBs. J. Power Sources 2013, 243, 29–32. 84. Hu, L.; Zhang, Z.; Amine, K. Fluorinated Electrolytes for Li-Ion Battery: An FEC-Based Electrolyte for High Voltage LiNi0.5Mn1.5O4/Graphite Couple. Electrochem. Commun. 2013, 35, 76–79. 85. McMillan, R.; Slegr, H.; Shu, Z. X.; Wang, W. Fluoroethylene Carbonate Electrolyte and its Use in lithium Ion Batteries with Graphite Anodes. J. Power Sources 1999, 81–82, 20–26. 86. Shi, P.; Fang, S.; Luo, D.; Yang, L.; Hirano, S.-I. A Safe Electrolyte Based on Propylene Carbonate and Non-Flammable Hydro Fluoro Ether for High-Performance Lithium Ion Batteries. J. Electrochem. Soc. 2017, 164, A1991–A1999. 87. Nguyen, C. C.; Lucht, B. L. Comparative Study of Fluoroethylene Carbonate and Vinylene Carbonate for Silicon Anodes in Lithium Ion Batteries. J. Electrochem. Soc. 2014, 161, A1933–A1938. 88. Zhang, X. Q.; Cheng, X. B.; Chen, X.; Yan, C.; Zhang, Q. Fluoroethylene Carbonate Additives to Render Uniform Li Deposits in Lithium Metal Batteries. Adv. Funct. Mater. 2017, 27, 1605989–1605997. 89. Schroder, K.; Alvarado, J.; Yersak, T. A.; Li, J.; Dudney, N.; Webb, L. J.; Meng, Y. S.; Stevenson, K. J. The Effect of Fluoroethylene Carbonate as an Additive on the Solid Electrolyte Interphase on Silicon Lithium-Ion Electrodes. Chem. Mater. 2015, 27, 5531–5542. 90. Arai, J. No-Flash-Point Electrolytes Applied to Amorphous Carbon/Li1+xMn2O4 Cells for EV Use. J. Power Sources 2003, 119–121, 388–392. 91. Zhang, S. S.; Read, J. Partially Fluorinated Solvent as a Co-Solvent for the Non-aqueous Electrolyte of Li/Air Battery. J. Power Sources 2011, 196, 2867–2870. 92. Liu, Y.; Fang, S.; Luo, D.; Yang, L.; Hirano, S.-I. Safe Electrolytes for LIBs Based on Ternary Mixtures of Triethylene Glycol Dimethylether, Fluoroethylene Carbonate and Non-Flammable Methyl-Nonafluorobutyl Ether. J. Electrochem. Soc. 2016, 163, A1951–A1958. 93. Fang, S.; Wang, G.; Qu, L.; Luo, D.; Yang, L.; Hirano, S.-I. A Novel Mixture of Diethylene Glycol Diethylether and Non-flammable Methyl-Nonafluorobutyl Ether as a Safe Electrolyte for lithium Ion Batteries. J. Mater. Chem. A 2015, 3, 21159–21166. 94. Pham, H. Q.; Lee, H.-Y.; Hwang, E.-H.; Kwon, Y.-G.; Song, S.-W. Non-Flammable Organic Liquid Electrolyte for High-Safety and High-Energy Density Li-Ion Batteries. J. Power Sources 2018, 404, 13–19. 95. Pham, H. Q.; Hwang, E.-H.; Kwon, Y.-G.; Song, S.-W. Approaching the Maximum Capacity of Nickel-Rich LiNi0.8Co0.1Mn0.1O2 Cathodes by Charging to High-Voltage in a Non-flammable Electrolyte of Propylene Carbonate and Fluorinated Linear Carbonates. Chem. Commun. 2019, 55, 1256–1258. 96. Xia, L.; Xia, Y.; Wang, C.; Hu, H.; Lee, S.; Yu, Q.; Chen, H.; Liu, Z. 5V-Class Electrolytes Based on Fluorinated Solvents for Li-Ion Batteries with Excellent Cyclability. Chem. Electro. Chem. 2015, 2, 1707–1712. 97. Baskakova, Y. V.; Yarmolenko, O. G. V.; Efimov, O. N. Polymer Gel Electrolytes for Lithium Batteries. Russ. Chem. Rev. 2012, 81, 367–380. 98. Song, J. Y.; Wang, Y. Y.; Wan, C. C. Review of Gel-Type Polymer Electrolytes for LIBs. J. Power Sources 1999, 77, 183–197. 99. Stephan, A. M. Review on Gel Polymer Electrolytes for Lithium Batteries. Eur. Polym. J. 2006, 42, 21–42. 100. Yoshimoto, N.; Gotoh, D.; Egashira, M.; Morita, M. Alkylphosphate-Based Nonflammable Gel Electrolyte for LiMn2O4 Positive Electrode in Lithium-Ion Battery. J. Power Sources 2008, 185, 1425–1428. 101. Yoshimoto, N.; Niida, Y.; Egashira, M.; Morita, M. Nonflammable Gel Electrolyte Containing Alkyl Phosphate for Rechargeable lithium Batteries. J. Power Sources 2006, 163, 238–242.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

367

102. Morita, M.; Niida, Y.; Yoshimoto, N.; Adachi, K. Polymeric Gel Electrolyte Containing Alkyl Phosphate for LIBs. J. Power Sources 2005, 146, 427–430. 103. Lalia, B. S.; Fujita, T.; Yoshimoto, N.; Egashira, M.; Morita, M. Electrochemical Performance of Nonflammable Polymeric Gel Electrolyte Containing Triethylphosphate. J. Power Sources 2009, 186, 211–215. 104. Lalia, B. S.; Yoshimoto, N.; Egashira, M.; Morita, M. Effects of Lewis-Acid Polymer on the Electrochemical Properties of Alkylphosphate-Based Non-flammable Gel Electrolyte. J. Power Sources 2009, 194, 531–535. 105. Lalia, B. S.; Yoshimoto, N.; Egashira, M.; Morita, M. Electrochemical Performances of Non-flammable Gel Electrolyte for Lithium Ion Battery Using LiFePO4 Positive Electrode. Electrochemistry 2010, 78, 332–335. 106. Yun, Y. S.; Song, S.-W.; Lee, S.-Y.; Kim, S.-H.; Kim, D.-W. Lithium Metal Polymer Cells Assembled with Gel Polymer Electrolytes Containing Ionic Liquid. Curr. Appl. Phys. 2010, 10, 97–100. 107. Li, Q.; Ardebili, H. Flexible Thin-Film Battery Based on Solid-like Ionic Liquid-Polymer Electrolyte. J. Power Sources 2016, 303, 17–21. 108. Yun, Y. S.; Kim, J. H.; Lee, S.-Y.; Shim, E.-G.; Kim, D.-W. Cycling Performance and Thermal Stability of lithium Polymer Cells Assembled with Ionic Liquid-Containing Gel Polymer Electrolytes. J. Power Sources 2011, 196, 6750–6755. 109. Liao, K.-S.; Sutto, T. E.; Andreoli, E.; Ajayan, P.; McGrady, K. A.; Curran, S. A. Nano-Sponge Ionic Liquid–Polymer Composite Electrolytes for Solid-State lithium Power Sources. J. Power Sources 2010, 195, 867–871. 110. Que, M.; Tong, Y.; Wei, G.; Yuan, K.; Wei, J.; Jiang, Y.; Zhu, H.; Chen, Y. Safe and Flexible Ion Gel Based Composite Electrolyte for lithium Batteries. J. Mater. Chem. A 2016, 4, 14132–14140. 111. Yun, Y. S.; Choi, J.-A.; Kim, D.-W. Lithium Polymer Batteries Assembled with In Situ Cross-Linked Gel Polymer Electrolytes Containing Ionic Liquid. Macromol. Res. 2013, 21, 49–54. 112. Hazama, T.; Fujii, K.; Sakai, T.; Aoki, M.; Mimura, H.; Eguchi, H.; Todorov, Y.; Yoshimoto, N.; Morita, M. High-Performance Gel Electrolytes with Tetra-Armed Polymer Network for Li Ion Batteries. J. Power Sources 2015, 286, 470–474. 113. Li, G.; Chen, X.; Miao, L.; Chen, J.; Zheng, J. A Hybridized Solid-Gel Nonflammable Li-Battery. J. Power Sources 2018, 394, 26–34. 114. Ha, H.-J.; Kil, E.-H.; Kwon, Y. H.; Kim, J. Y.; Lee, C. K.; Lee, S.-Y. UV-Curable Semi-Interpenetrating Polymer Network-Integrated, Highly Bendable Plastic Crystal Composite Electrolytes for Shape-Conformable all-Solid-State lithium Ion Batteries. Energ. Environ. Sci. 2012, 5, 6491–6499. 115. Lv, P.; Li, Y.; Wu, Y.; Liu, G.; Liu, H.; Li, S.; Tang, C.; Mei, J.; Li, Y. Robust Succinonitrile-Based Gel Polymer Electrolyte for LIBs Withstanding Mechanical Folding and High Temperature. ACS Appl. Mater. Interfaces 2018, 10, 25384–25392. 116. Jia, H.; Onishi, H.; Wagner, R.; Winter, M.; Cekic-Laskovic, I. Intrinsically Safe Gel Polymer Electrolyte Comprising Flame-Retarding Polymer Matrix for Lithium Ion Battery Application. ACS Appl. Mater. Interfaces 2018, 10, 42348–42355. 117. Bachman, J. C.; Muy, S.; Grimaud, A.; Chang, H.-H.; Pour, N.; Lux, S. F.; Paschos, O.; Maglia, F.; Lupart, S.; Lamp, P.; Giordano, L.; Shao-Horn, Y. Inorganic Solid-State Electrolytes for Lithium Batteries: Mechanisms and Properties Governing Ion Conduction. Chem. Rev. 2016, 116, 140–162. 118. Alpen, U. V.; Rabenau, A.; Talat, G. H. Ionic Conductivity in Li/Sub 3/N Single Crystals. Appl. Phys. Lett. 1977, 30, 621–623. 119. Lapp, T.; Skaarup, S.; Hooper, A. Ionic Conductivity of Pure and Doped Li3N. Solid State Ion. 1983, 11, 97–103. 120. Aono, H.; Sugimono, E.; Sadaoka, Y.; Imanaka, N.; Adachi, G. Ionic Conductivity of Solid Electrolytes Based on Lithium Titanium Phosphate. J. Electrochem. Soc. 1990, 137, 1023–1027. 121. Bruce, P. G.; West, A. R. The A-C Conductivity of Polycrystalline LISICON, Li2 + 2x Zn1 − x GeO4, and a Model for Intergranular Constriction Resistances. J. Electrochem. Soc. 1983, 130, 662–669. 122. Inaguma, Y.; Liquan, C.; Itoh, M.; Nakamura, T.; Uchida, T.; Ikuta, H.; Wakihara, M. High Ionic Conductivity in lithium Lanthanum Titanate. Solid State Commun. 1993, 86, 689–693. 123. Murugan, R.; Thangadurai, V.; Weppner, W.; Angew. Fast Lithium Ion Conduction in Garnet-Type Li7La3Zr2O12. Chem. Int. Ed. 2007, 46, 7778–7781. 124. Hayashi, A.; Hama, S.; Morimoto, H.; Minami, T.; Tatsumisago, M. Preparation of Li2S–P2S5 Amorphous Solid Electrolytes by Mechanical Milling. J. Am. Ceram. Soc. 2001, 84, 477–479. 125. Mizuno, F.; Hayashi, A.; Tadanaga, K.; Tatsumisago, M. New, Highly Ion-Conductive Crystals Precipitated from Li2S–P2S5 Glasses. Adv. Mater. 2005, 17, 918–921. 126. Knauth, P. Inorganic Solid Li Ion Conductors: An Overview. Solid State Ion. 2009, 180, 911–916. 127. Bates, J. B.; Gruzalski, G. R.; Dudney, N. J.; Luck, C. F.; Yu, X. Rechargeable Thin-Film Lithium Batteries. Solid State Ion. 1994, 70/71, 619–628. 128. Dudney, N. J. Solid-State Thin-Film Rechargeable Batteries. Mater. Sci. Eng. B 2005, 116, 245–249. 129. Li, J.; Ma, C.; Chi, M.; Liang, C.; Dudney, N. Solid Electrolyte: The Key for High-Voltage Lithium Batteries. Adv. Energy Mater. 2015, 5. https://doi.org/10.1002/ aenm.201401408. 130. Kim, J. G.; Son, B.; Mukherjee, S.; Schuppert, N.; Bates, A.; Kwon, O.; Choi, M. J.; Chung, H. Y.; Park, S. A Review of Lithium and Non-lithium Based Solid State Batteries. J. Power Sources 2015, 282, 299–322. 131. Goodenough, J. B.; Singh, P. Review—Solid Electrolytes in Rechargeable Electrochemical Cells. J. Electrochem. Soc. 2015, 162, A2387–A2394. 132. Quartarone, E.; Mustarelli, P. Electrolytes for Solid-State lithium Rechargeable Batteries: Recent Advances and Perspectives. Chem. Soc. Rev. 2011, 40, 2525–2540. 133. Anantharamulu, N.; Rao, K. K.; Rambabu, G.; Kumar, B. V.; Radha, V.; Vithal, M. A Wide-Ranging Review on Nasicon Type Materials. J. Mater. Sci. 2011, 46, 2821–2837. 134. Teng, S.; Tan, J.; Tiwari, A. Recent Developments in Garnet Based Solid State Electrolytes for Thin Film Batteries. Curr. Opin. Solid State Mater. Sci. 2014, 18, 29–38. 135. Thangadurai, V.; Narayanan, S.; Pinzaru, D. Garnet-Type Solid-State Fast Li Ion Conductors for Li Batteries: Critical Review. Chem. Soc. Rev. 2014, 43, 4714–4727. 136. Berbano, S. S.; Mirsaneh, M.; Lanagan, M. T.; Randall, C. A. Lithium Thiophosphate Glasses and Glass–Ceramics as Solid Electrolytes: Processing, Microstructure, and Properties. Int. J. Appl. Glass Sci. 2013, 4, 414–425. 137. Tatsumisago, M.; Nagao, M.; Hayashi, A. Recent Development of Sulfide Solid Electrolytes and Interfacial Modification for All-Solid-State Rechargeable Lithium Batteries Footnote. J. Asian Ceramic Soc. 2013, 1, 17–25. 138. Kanno, R.; Murayama, M. Lithium Ionic Conductor Thio-LISICON: The Li2 S-GeS2 - P 2 S 5 System. J. Electrochem. Soc. 2001, 148, A742–A748. 139. West, A. R. Basic Solid State Chemistry, 2nd ed.; John Wiley & Sons: Chichester, UK, 1999; p. 32. Chapter 7.5. 140. Mercier, R.; Malugani, J.-P.; Fahys, B.; Robert, G. Superionic Conduction in Li2S - P2S5 - LiI – Glasses. Solid State Ion. 1981, 5, 663–666. 141. Liu, D.; Zhu, W.; Feng, Z.; Guerfi, A.; Vijh, A.; Zaghib, K. Recent Progress in Sulfide-Based Solid Electrolytes for Li-Ion Batteries. Mater. Sci. Eng. B 2016, 213, 169–176. 142. Hayashi, A.; Minami, K.; Ujiie, S.; Tatsumisago, M. Preparation and Ionic Conductivity of Li7P3S11 − z Glass-Ceramic Electrolytes. J. Non Cryst. Solids 2010, 356, 2670–2673. 143. Seino, Y.; Ota, T.; Takada, K.; Hayashi, A.; Tatsumisago, M. A Sulphide lithium Super Ion Conductor Is Superior to Liquid Ion Conductors for Use in Rechargeable Batteries. Energ. Environ. Sci. 2014, 7, 627–631. 144. Pradel, A.; Ribes, M. Electrical Properties of lithium Conductive Silicon Sulfide Glasses Prepared by Twin Roller Quenching. Solid State Ion. 1986, 18–19, 351–355. 145. Tatsumisago, M. Glassy Materials Based on Li2S for all-Solid-State lithium Secondary Batteries. Solid State Ion. 2004, 175, 13–18. 146. Liu, Z.; Fu, W.; Payzant, E. A.; Yu, X.; Wu, Z.; Dudney, N. J.; Kiggans, J.; Hong, K.; Rondinone, A. J.; Liang, C. Anomalous High Ionic Conductivity of Nanoporous b-Li3PS4. J. Am. Chem. Soc. 2013, 135, 975–978. 147. Minami, T.; Machida, N. Preparation of New Glasses with High Ionic Conductivities. Mater. Sci. Eng. B 1992, 13, 203–208. 148. Kamaya, N.; Homma, K.; Yamakawa, Y.; Hirayama, M.; Kanno, R.; Yonemura, M.; Kamiyama, T.; Kato, Y.; Hama, S.; Kawamoto, K.; Mitsui, A. A Lithium Superionic Conductor. Nat. Mater. 2011, 10, 682–686. 149. Bron, P.; Johansson, S.; Zick, K.; Gunne, J. S.; Dehnen, S.; Roling, B. Li10SnP2S12: An Affordable Lithium Superionic Conductor. J. Am. Chem. Soc. 2013, 135, 15694–15697.

368

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

150. Kuhn, A.; Gerbig, O.; Zhu, C.; Falkenberg, F.; Maier, J.; Lotsch, B. V. A New Ultrafast Superionic Li-Conductor: Ion Dynamics in Li11Si2PS12 and Comparison with Other Tetragonal LGPS-Type Electrolytes. Phys. Chem. Chem. Phys. 2014, 16, 14669–14674. 151. Ujiie, S.; Hayashi, A.; Tatsumisago, M. Structure, Ionic Conductivity and Electrochemical Stability of Li2S–P2S5–LiI Glass and Glass–Ceramic Electrolytes. Solid State Ion. 2012, 211, 42–45. 152. Ujiie, S.; Hayashi, A.; Tatsumisago, M. Preparation and ionic conductivity of (100−x)(0.8Li2S0.2P2S5)xLiI glass–ceramic electrolytes. J. Solid State Electrochem. 2013, 17, 675–680. 153. Rangasamy, E.; Liu, Z.; Gobet, M.; Pilar, K.; Sahu, G.; Zhou, W.; Wu, H.; Greenbaum, S.; Liang, C. An Iodide-Based Li7P2S8I Superionic Conductor. J. Am. Chem. Soc. 2015, 137, 1384–1387. 154. Ohtomo, T.; Hayashi, A.; Tatsumisago, M.; Kawamoto, K. Characteristics of the Li2O–Li2S–P2S5 Glasses Synthesized by the Two-Step Mechanical Milling. J. Non Cryst. Solids 2013, 364, 57–61. 155. Hayashi, A.; Muramatsu, H.; Ohtomo, T.; Hama, S.; Tatsumisago, M. Improved Chemical Stability and Cyclability in Li2S–P2S5–P2O5–ZnO Composite Electrolytes for AllSolid-State Rechargeable Lithium Batteries. J. Alloys Compd. 2014, 591, 247–250. 156. Trevey, J. E.; Gilsdorf, J. R.; Miller, S. W.; Lee, S.-H. Li2S–Li2O–P2S5 Solid Electrolyte for all Solid-State lithium Batteries. Solid State Ion. 2012, 214, 25–30. 157. Hayashi, A.; Muramatsu, H.; Ohtomo, T.; Hama, S.; Tatsumisago, M. Improvement of Chemical Stability of Li3PS4 Glass Electrolytes by Adding MxOy (M ¼ Fe, Zn, and bi) Nanoparticles. J. Mater. Chem. A 2013, 1, 6320–6326. 158. Mizuno, F.; Hayashi, A.; Tadanaga, K.; Tatsumisago, M. High lithium Ion Conducting Glass-Ceramics in the System Li2S–P2S5. Solid State Ion. 2006, 177, 2721–2725. 159. Tachez, M.; Malugani, J.-P.; Mercier, R.; Robert, G. Ionic Conductivity of and Phase Transition in Lithium Thiophosphate Li3PS4. Solid State Ion. 1984, 14, 181–185. 160. Pietrzak, T. K.; Garbarczyk, J. E.; Gorzkowska, I.; Wasiucionek, M.; Nowinski, J. L.; Gierlotka, S.; Jozwiak, P. Correlation between Electrical Properties and Microstructure of Nanocrystallized V2O5–P2O5 Glasses. J. Power Sources 2009, 194, 73–80. 161. Kanno, R.; Hata, T.; Kawamoto, Y.; Irie, M. Synthesis of a New Lithium Ionic Conductor, Thio-LISICON–Lithium Germanium Sulfide System. Solid State Ion. 2000, 130, 97–104. 162. Foster, D. L.; Kuo, H. C.; Schlaikjer, C. R.; Dey, A. N. New Highly Conductive Inorganic Electrolytes: The Liquid Solvates of the Alkali and Alkaline Earth Metal Tetrachloroaluminates. J. Electrochem. Soc. 1988, 135, 2682–2688. 163. Park, C. W.; Oh, S. M. Performances of Li/LixCoO2 Cells in LiAlCl4  3SO2 Electrolyte. J. Power Sources 1997, 68, 338–343. 164. Zinck, L.; Borck, M.; Ripp, C.; Hambitzer, G. Purification Process for an Inorganic Rechargeable Lithium Battery and New Safety Concepts. J. Appl. Electrochem. 2006, 36, 1291–1295. 165. Fey, G. T.-K.; Liu, W.-K.; Chang, Y.-C. Temperature and Concentration Effects on the Conductivity of LiAlCl4/SOCl2 Electrolyte Solutions. J. Power Sources 2001, 97-98, 602–605. 166. Grundish, N.; Amos, C.; Goodenough, J. B. Communication—Characterization of LiAlCl4xSO2 Inorganic Liquid Li+ Electrolyte. J. Electrochem. Soc. 2018, 165, A1694–A1698. 167. Kim, A.; Jung, H.; Song, J.; Kim, H. J.; Jeong, G.; Kim, H. Lithium-Ion Intercalation into Graphite in SO2-Based Inorganic Electrolyte toward High-Rate-Capable and Safe LIBs. ACS Appli. Materi. & Interf. 2019, 9, 9054–9061. 168. Song, J.; Chun, J.; Kim, A.; Jung, H.; Kim, H. J.; Kim, Y.-J.; Jeong, G.; Kim, H. Dendrite-Free Li Metal Anode for Rechargeable Li–SO2 Batteries Employing Surface Modification with a NaAlCl4–2SO2 Electrolyte. ACS Appli. Materi. & Interf. 2018, 10, 34699–34705. 169. Hartl, R.; Fleischmann, M.; Gschwind, R.; Winter, M.; Gores, H. A Liquid Inorganic Electrolyte Showing an Unusually High Lithium Ion Transference Number: A Concentrated Solution of LiAlCl4 in Sulfur Dioxide. Energies 2013, 6, 4448–4464. 170. Ripp, C.; Hambitzer, G.; Zinck, L.; M. Borck. In Encyclopedia of Electrochemical Power Sources; Garche, J., Dyer, C. K., Eds.; Academic Press; Imprint of Elsevier: Amsterdam, Boston, 2009; p. 383. 171. Padhi, A. K.; Nanjundaswamy, K. S.; Goodenough, J. B. Phospho-Olivines as Positive-Electrode Materials for Rechargeable Lithium Batteries. J. Electrochem. Soc. 1997, 144, 1188–1196. 172. Wang, B.; Al Abdulla, W.; Wang, D.; Zhao, X. S. A Three-Dimensional Porous LiFePO4 Cathode Material Modified with a Nitrogen-Doped Graphene Aerogel for High-Power LIBs. Energ. Environ. Sci. 2015, 8, 869–875. 173. Yamada, A.; Chung, S. C.; Hinokuma, K. Optimized LiFePO4 for Lithium Battery Cathodes. J. Electrochem. Soc. 2001, 148, A224–A231. 174. Padhi, A. K.; Nanjundaswamy, K. S.; Masquelier, C.; Okada, S.; Goodenough, J. B. Effect of Structure on the Fe3+ / Fe2+ Redox Couple in Iron Phosphates. J. Electrochem. Soc. 1997, 144, 1609–1616. 175. Delacourt, C.; Poizot, P.; Levasseur, S.; Masquelier, C. Size Effects on Carbon-Free LiFePO4 Powders: The Key to Superior Energy Density. Electrochem. Solid St. 2006, 9, A352–A357. 176. Whittingham, M. S. Lithium Batteries and Cathode Materials. Chem. Rev. 2004, 104, 4271–4302. 177. Okada, S.; Sawa, S.; Egashira, M.; Yamaki, J.-I.; Tabuchi, M.; Kageyama, H.; Konishi, T.; Yoshino, A. Cathode Properties of Phospho-Olivine LiMPO4 for lithium Secondary Batteries. J. Power Sources 2001, 97-98, 430–432. 178. He, H.; Liu, Y.; Liu, Q.; Li, Z.; Xu, F.; Dun, C.; Ren, Y.; Wang, M.-X.; Xie, J. Failure Investigation of LiFePO4 Cells in Over-Discharge Conditions. J. Electrochem. Soc. 2013, 160, A793–A7106. 179. Ramar, V.; Balaya, P. Enhancing the Electrochemical Kinetics of High Voltage Olivine LiMnPO4 by Isovalent Co-Doping, Physical Chemistry Chemical Physics. Phys. Chem. Chem. Phys. 2013, 15, 17240–17249. 180. Ramar, V.; Saravanan, K.; Gajjela, S. R.; Hariharan, S.; Balaya, P. The Effect of Synthesis Parameters on the lithium Storage Performance of LiMnPO4/C. Electrochim. Acta 2013, 105, 496–505. 181. Saravanan, K.; Ramar, V.; Balaya, P.; Vittal, J. J. Li(MnxFe1−x)PO4/C (x ¼ 0.5, 0.75 and 1) Nanoplates for Lithium Storage Application. J. Mater. Chem. 2011, 21, 14925–14935. 182. Gao, T.; Wang, B.; Wang, L.; Liu, G.; Wang, F.; Luo, H.; Wang, D. LiAlCl43SO2 as a High Conductive, Non-flammable and Inorganic Non-Aqueous Liquid Electrolyte for Lithium Ion Batteries. Electrochim. Acta 2018, 286, 77–85. 183. Gao, T.; Wang, B.; Wang, F.; Li, R.; Wang, L.; Wang, D. LiAlCl43SO2: A Promising Inorganic Electrolyte for Stable Li Metal Anode at Room and Low Temperature. Ionics 2019, 25, 4137–4147. 184. Ramar, V.; Pszolla, C.; Rapp, M.; Borck, M.; Zinck, L. Non-Flammable Inorganic Liquid Electrolyte LIBs. J. Electrochem. Soc. 2020, 167, 070521. 185. Weber, R. L.; Ye, Y.; Banik, S. M.; Elabd, Y. A.; Hickner, M. A.; Mahanthappa, M. K. Thermal and io N Transport Properties of Hydrophilic and Hydrophobic Polymerized Styrenic Imidazolium Ionic Liquids. J. Polym. Sci. Part B Polym. Phys. 2011, 49, 1287–1296. 186. Zuo, X.; Liu, X.-M.; Cai, F.; Yang, H.; Shen, X.-D.; Liu, G. A Novel all-Solid Electrolyte Based on a Co-Polymer of Poly-(Methoxy/Hexadecal-Poly(Ethylene Glycol) Methacrylate) for Lithium-Ion Cell. J. Mater. Chem. 2012, 22, 22265–22271. 187. Salem, N.; Abu-Lebdeh, Y. Effect of Nanoparticles on Electrolytes and Electrode/Electrolyte Interface. In Nanotechnology for LIBs; Abu-Lebdeh, Y., Davidson, I., Eds.; Springer: New York, 2013; pp. 221–244. 188. Appetecchi, G. B.; Croce, F.; Persi, L.; Ronci, F.; Scrosati, B. Transport and Interfacial Properties of Composite Polymer Electrolytes. Electrochim. Acta 2000, 45, 1481–1490. 189. Croce, F.; Settimi, L.; Scrosati, B. Superacid ZrO2-Added, Composite Polymer Electrolytes with Improved Transport Properties. Electrochem. Commun. 2006, 8, 364–368. 190. Chen, Y. T.; Chuang, Y. C.; Su, J. H.; Yu, H. C.; Chen-Yang, Y. W. High Discharge Capacity Solid Composite Polymer Electrolyte lithium Battery. J. Power Sources 2011, 196, 2802–2809.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic and Inorganic Electrolytes

369

191. Ciosek, M.; Marcinek, M.; Z˙ ukowska, G.; Wieczorek, W. Lithium Transference Number Measurements and Complex Abilities in Anion Trapping Tri Phenyl Borane Poly (Ethylene Oxide) Dimethyl Ethere lithium Tri Fluoro Methane Sulfonate Composite Electrolyte. Electrochim. Acta 2009, 54, 4487–4493. 192. Lutkenhaus, J. L.; McEnnis, K.; Hammond, P. T. Tuning the Glass Transition of and Ion Transport within Hydrogen-Bonded Layer-by-Layer Assemblies. Macromolecules 2007, 40, 8367–8373. 193. Zhang, L.; Chaloux, B. L.; Saito, T.; Hickner, M. A.; Lutkenhaus, J. L. Ion Conduction in Poly(Ethylene Oxide) Ionically Assembled Complexes. Macromolecules 2011, 44, 9723–9730. 194. Yao, L.; Watkins, J. J. Photoinduced Disorder in Strongly Segregated Block Copolymer Composite Films for Hierarchical Pattern Formation. ACS Nano 2013, 7, 1513–1523. 195. Jankowsky, S.; Hiller, M. M.; Wiemhöfer, H. D. Preparation and Electrochemical Performance of Polyphosphazene Based Salt-in-Polymer Electrolyte Membranes for LIBs. J. Power Sources 2014, 253, 256–262.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Ionic Liquid Electrolytes Henry Adenusia and Stefano Passerinib,c,d, aDepartment of Science and Engineering of Matter, Environment and Urban Planning, Marche Polytechnic University, Ancona, Italy; bHelmholtz Institute Ulm (HIU), Ulm, Germany; cKarlsruhe Institute of Technology (KIT), Karlsruhe, Germany; dDepartment of Chemistry, Sapienza University of Rome, Rome, Italy © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of S. Passerini, W.A. Henderson, SECONDARY BATTERIES – LITHIUM RECHARGEABLE SYSTEMS | Electrolytes: Ionic Liquids, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 85–91, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5. 00835-2.

1 Introduction 2 Functional structure 3 Physicochemical properties 3.1 Viscosity 3.2 Ionic conductivity 3.3 Thermal stability 3.4 Electrochemical stability 4 Recent advances of ionic liquid electrolytes 5 Ionic liquid electrolytes for post-lithium batteries 6 Summary and outlook 6.1 Computational models 6.2 Machine learning methods 6.3 Advanced operando experiments Acknowledgments References

371 371 374 374 375 375 375 375 379 380 380 381 381 381 382

Abstract Lithium-ion battery technology poses safety issues due to the inherent incompatibility and flammability of the organic solvent-based liquid electrolytes. Ionic liquids have emerged as ideal materials that can supersede such solvents due to their unique physicochemical properties and propensity for rational design; they are integral for the development of lithium-ion batteries with superior safety, lifetime and performance. In this chapter, we present the recent advances of ionic liquid electrolytes for lithium-ion batteries outlining their functional structure, physicochemical properties and use case for postlithium batteries. Lastly, we provide an outlook for the optimization of ionic liquid-based electrolytes utilizing computational models, machine learning methods and advanced operando experiments.

Glossary Cathode electrolyte interphase A complex heterogeneous structurally disordered layer that forms in situ on the positive electrode; its formation essential for battery operation. Dendrite The formation of lithium crystals with tree-like structure during the charge of lithium batteries. Its generation is produced by faster growth along energetically favorable crystallographic directions and can cause loss of active material as well as short circuits. Electrochemical stability window The electrochemical potential window in which a given electrolyte is stable and does not decompose due to redox reactions. It is usually expressed in volts with respect to a reference redox couple. Electrolyte An electrically conductive medium consisting of free ions in solutions; composite and solid electrolytes can also be generated. Lithium-ion batteries A type of rechargeable battery that reversibly intercalates lithium ions into electronically conducting solids to store energy. Post-lithium batteries Battery technologies based on elements such as aluminum, magnesium, and sodium, that are of high abundance on earth and thus economical compared to lithium-ion batteries. Solid electrolyte interphase A complex heterogeneous structurally disordered layer that forms in situ on the negative electrode, requisite for lithium-ion batteries to reversibly charge/discharge.

370

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00176-6

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Ionic Liquid Electrolytes

371

Key points ● ● ● ● ●

1

Outline of the functional structure and the different classes of ILs. Overview of the main physicochemical properties of ILEs that influence functionality. Recent advances demonstrating the performance of ILEs in lithium batteries. Insights into the use case of ILEs for aluminum, magnesium, sodium and zinc based batteries. Outlook for the optimization of ILEs using computation, machine learning and operando experiments.

Introduction

Ionic liquids (ILs) are defined as liquids that consist entirely of ions in a liquid state below 100
C. This class of materials display excellent potential to overcome the standing problems of lithium batteries as they can function as effective modern electrolytes. ILs possess unique physicochemical and thermal properties including the ability to dissolve inorganic salts, high ionic conductivity, negligible vapor pressure thus nonflammability, along with excellent electrochemical and thermal stability.1 Additionally, chemical modulation of the ionic species permits the rational design of ILs which is of relevance for enhanced electrode/electrolyte interphases (EEIs) in lithium batteries. ILs are employed as electrolytes for both lithium-ion batteries (LIBs) and lithium metal batteries (LMBs) in two main roles: ILs are used as the sole solvent (replacing conventional solvents such as ethylene carbonate (EC), propylene carbonate (PC) and dimethyl carbonate (DMC)) dissolving a lithium salt or as additives to organic liquid electrolytes. Alternatively, ILs can be used as additives to conventional polymer-lithium salt mixtures to enhance the performance of solid polymer electrolytes (SPEs).2 Recently, the versatility of ILs has been exploited to produce novel classes of ILs with relevance for energy storage spanning concentrated ionic liquid electrolytes (ILEs) to zwitterionic variants. The substitution of organic liquid electrolytes with ILs greatly reduces the flammability of the corresponding material thus boosting the safety profile for high-energy applications.3

2

Functional structure

ILs are composed solely of ions, the main interaction force between anions and cations are Coulombic interactions which are influenced by the ionic radius. The greater the ionic radius, the smaller the force between molecules and the lower the melting point of the ionic material. The interaction energy between ions is dependent on the charge density which is related to the magnitude and distribution of the charges of anions and cations. ILs are materials which are advancing the modern electrolyte field for lithium batteries owing to their well-balanced ionic interactions and dissociation.4 The ionic conductivity of ILs are sensitive to numerous factors, namely composition of mixtures (concentration of ILs in mixtures), temperature and pressure. ILs can be modulated for specific applications with the prospect of at least 1 million binary IL combinations and 1018 ternary IL combinations.5 However, ILs are limited by several drawbacks compared to organic solvents. Due to the presence of strong Coulombic interactions and steric hindrance caused by large ions, ILs are highly viscous, their Li+ transference number (t+Li) is lower than those of organic solvents (ILs solely composed of ions migrate according to a gradient potential, reducing the Li+ conductivity). For industrial application, the synthesis process of ILs relies upon the utilization of environmentally unfriendly organic solvents and costly precursors with energy intensive as well as time-onerous steps. This includes quaternization, anion exchange reactions and purifications. Nonetheless, significant efforts are in progress to develop cost-effective, eco-friendly, and facile synthesis routes. Despite these issues, ILs are excellent materials demonstrating compatibility in various lithium battery systems; in LMBs with high-energy cathode materials, e.g., Ni-rich layered/spinel oxides, lithium rich oxides, cation disordered rock salt oxides, sulfur and even oxygen. Task specific ILs can be prepared, with tailored formulations that regulate interphase formation on electrode surfaces, those manifesting as redox active sites and producing desirable (fluorine, nitrogen, phosphorous, boron based) species with electron insulating properties, low ionic resistance and stability.6 Both anion and cation can participate in solid electrolyte interphase (SEI) formation on lithium metal anodes. The anions constituting the lithium solvation shell can access the anode, decomposing and producing inorganic species, e.g., lithium nitride (Li3N) and lithium fluoride (LiF). In contrast, cations undergo electrostatic absorption on the surface of lithium preventing the deposition of new Li+ and impeding dendritic evolution. For the implementation of ILs as electrolytes, the ILs must possess a low melting point with complete absence of crystallization (even 2 M) are utilized, though this supports uniform lithium deposition on lithium metal electrode.10 Transitioning from the conventional constituent IL ions, unique structures can be designed possessing inter-ionic chemical or physical bonds (Fig. 2). These include polymerized ILs (poly-ILs), solvated ILs (SILs) and zwitterionic ILs (ZILs), which retain the advantages of typical ILs yet attain new functionalities.11–13 Cation and anion or several cation structures can be tethered by covalent bonding yielding poly-ILs and ZILs. Polymerizing the anion structure results in the generation of Li+ single ion conductors which are similar to SPEs than poly-ILs. For ZILs, connecting anion and cation renders the material neutral therefore in an electric field the movement of ZILs is diminished. Notably, poly-ILs and ZILs exhibit higher Tm and glass transition temperatures than ILs. Conversely, SILs consist of a lithium salt whereby Li+ is coordinated by an oligomeric ether (glyme) forming [Li(glyme)]+ complexes, which lower the Tm of the salt. Another class of ILs are known as protic ILs (PILs) that can be synthesized via facile Brønsted acid-base reaction and are affordable relative to traditional ILs.14 In spite of their synthetic simplicity, PILs manifest complex structural patterns including hydrogen bonding features, nanosegregation and the presence of mobile protons (via the vehicle or Grotthuss mechanism) that can act as fast charge carriers. PILs have been successfully utilized with lithium iron phosphate (LFP) and lithium vanadium phosphate electrode materials. As a result, this class of functional materials have been proposed as alternative solvents for batteries as they display conductivities, thermal stabilities and viscosities comparable to conventional ILs. The presence of the mobile proton influences their interplay with electrode materials, electrochemical stability and reactivity with electrolyte species (solvents and salts) for SEI formation; thus they are prospective candidates to realize highly reversible and fast energy storage.14

Fig. 2 Overview of the types of ILs and the respective structuring of each class.

374

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Ionic Liquid Electrolytes

Moreover, to mitigate the ILEs insufficient Li+ transport and elevated viscosity, concentrated ILEs containing higher lithium salt mole percent 30–50% are reported to enable the operation of LMBs at higher dis-/charge rates, stemming from elevated Li+ concentrations and enhanced Li+ transference numbers due to less organic cations and free anions. Nevertheless, the increased viscosity of concentrated ILEs caused by the formation of Li+-anion complexes hinders the wettability towards thick electrodes. An effective strategy to resolve the elevated viscosity and sluggish Li+ mobility of ILEs is the addition of non-solvating co-solvents named locally concentrated ionic liquid electrolytes (LCILEs).15 These non-solvating co-solvents are usually fluorinated ethers and aromatic molecules. The strong electron-withdrawing effect of the fluorinated groups weakens the solvating ability of the co-solvent towards Li+ and this enables the dilution of concentrated ILEs, maintaining the Li+ coordination and increased Li+ transference number.15 LCILEs exhibit lower flammability and superior compatibility towards lithium metal anodes compared to locally concentrated electrolytes based on conventional organic solvents (carbonate esters, sulfones and phosphate esters). To date, specific LCILEs have been developed with superior Li+ transport and lower viscosity allowing stable cycling of LMBs with high mass loading insertion-type cathodes, lithium-oxygen, and lithium-sulfur batteries at elevated current densities.

3

Physicochemical properties

For the implementation of ILEs with electrode materials, four key physicochemical properties influence the performance of ILEs in rechargeable LIBs (Fig. 3).

3.1

Viscosity

The viscosity of the ILE dictates the ability for ion migration as well as physical contact between ILEs and porous electrode materials/ separators. In ILEs, the functional interactions such as Coulombic interactions, hydrogen bonding, van der Waals forces and p-p interactions impact the viscosity of the medium. Recent efforts have reduced the viscosity of ILEs via customization of the ionic species. Cation design has focused on ameliorating the delocalization of positive charges to lessen the Coulombic interactions between cations/anions. ILs which utilize EMI+ cation display lower viscosities (34 cP), relative to PY13+ cations (60 cP) when used with TFSI anion.16 Imidazolium cations can delocalize the positive charge reducing the Coulombic interaction between ions. On the anion side, conformational flexibility effects the viscosity of the IL whereby aliphatic ammonium cations with C2F5BF−3 anion exhibit low viscosities (68 cP, [N122.1O2C2F5BF3]) compared to BF−4 anion (426 cP, [N122.1O2BF4]), even though the former anion is considerably larger in size.17 The higher degree of conformational freedom of the C2F5BF−3 anion offers structural flexibility influencing the resultant IL viscosity. In conjunction with the ionic conductivity, the viscosity of the IL determines the ionic transport of the medium which is requisite for superior lithium batteries.

Fig. 3 (Top) Four key physicochemical properties influencing ILE performance in LIBs. (Bottom) Walden plot showcasing the degree of ionicity of select ILs.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Ionic Liquid Electrolytes 3.2

375

Ionic conductivity

The ionic conductivity of the ILE is essential for fast Li+ diffusion which impacts the overall performance of lithium batteries. ILs with lower viscosities typically exhibit higher conductivities as well as increasing the testing temperatures promotes elevated transport of ionic species. Compared to conventional liquid electrolytes, IL-based liquid electrolytes are highly conductive displaying values of 8.61 mS cm−1 for LiPF6-EMITFSI relative to 8.50 mS cm−1 for LiPF6-EC/DEC.18 ILs with TFSI anions exhibit good ionic conductivity (8.8 mS cm−1, [EMI][TFSI]), while ILs with FSI show pronounced ionic conductivity (15.6 mS cm−1, [EMI][TFSI]) owing to the reduced steric hindrance of the anion relative to TFSI. The ionic conductivity of ILEs is affected by ion aggregation and ion pairing, which reduce ion mobility and ionic concentration. The Walden plot (Fig. 3) displays the correlation of the viscous flow and conductivity on a logarithmic scale, in which the diagonal line represents the values measured with diluted aqueous solution of KCl (0.01 mol L−1). Any deviation from the “ideal” line could be ascribed to the degree of ionicity of the charged molecules in ILs. For “good” ILs, a uniform distribution of positive charge surrounds the negative charge with a Madelung energy (the larger the EMad, the lower the vapor pressure of the IL) comparable to ideal ionic crystal, e.g., EMIBF4. In “poor” ILs, a high degree of ion pairing is prevalent, e.g., EMITFESI.

3.3

Thermal stability

The thermal stability of ILEs is an excellent property enhancing the safety profile of lithium battery systems especially for high temperature utilization. The anionic and cationic moieties determine the thermal stability of ILs with numerous studies focused on the modification of cations, varying the alkyl chain lengths, functional groups, and alkyl substituents to increase the IL thermal stability. The thermal stability of ILs is reduced through alteration of the cation such as a lengthier alkyl chain, functionalization (e.g., number and position of oxygen atoms, length of O-alkyl chain) and tethering of the alkyl chain via tertiary carbon atom. Moreover, the type of anion can increase the IL stability at high temperatures as evidenced by [BF4]−, [Ntf2]− and [PF6]−. In [EMIM] [BF4] and [C2MIM][C4F9SO3], the substitution of the alkyl chain to a hydroxyl group results in facile decomposition of the hydroxyethyl group.19 [CnIMBS][HSO4] exemplifies the influence of the chain length, with the thermal decomposition temperature decreasing from 311
C to 253
C, as the chain length n increases from 1 to 16.19 Additional factors, spanning gas atmosphere, heating rates and impurities also effect the thermal stability measurements; of which the heating rate has the most significant impact on the thermal gravimetric analysis (TGA) results, with the difference in Tonset acquired at 1
C/min and 20
C/min, can be up to 100
C.

3.4

Electrochemical stability

The electrochemical stability is paramount to ensure functionality of the ILE with electrode materials along with a wide ESW and to circumvent its degradation. A relationship between the chemical structure of the IL and the ESW is evident; in imidazolium cations the ESW of the resultant IL are generally narrower due to the acidic C2 hydrogen. Whereas the use of pyrrolidinium cations instead of imidazolium enhances the ESW when such cations are paired with TFSI anion.20 [EMI][TFSI] displays lower electrode stability (Ecathodic ¼ −2.47 V and Eanodic ¼ 2.07 V vs Fc/Fc+ at 25
C) compared to pyrrolidinium cation which exhibits a superior electrochemical stability with TFSI anion (Ecathodic ¼ −3.42 V and Eanodic ¼ 2.60 V vs Fc/Fc+ at 25
C).20 In contrast, for anions this is evidenced in [EMI][TFSI], when FSI is utilized owing to the presence of SdF bonds which are less stable than SdCF3 bonds; reduced electrochemical and thermal stability are reported. In addition, the electrochemical stability of ILEs are of relevance for SEI formation as uniform interphases are demonstrated in [EMI][FSI] due to the decomposition of FSI anions producing fluorine based SEI.

4

Recent advances of ionic liquid electrolytes

Recent studies highlight the necessity of fluoride-based interphases in LIBs to realize superior battery chemistries enabling compatibility with high capacity, high voltage cathodes, particularly lithium metal electrode and post-lithium chemistries. Nanosized fluorides stemming from the solvents and anions have a role as components that can control interfacial dynamics such as electrochemical redox reactions as well as providing fast Li+ conduction pathways by connecting to semi-carbonates at nano-length scale. Generally, fluorination of the interphase is achieved using fluorinated salt anions, e.g., BF4−, PF6− or TFSI− or neutral fluorinated solvent molecules. However, such fluorinated moieties cannot populate the inner-Helmholtz layers of the electrode surface of a high Fermi energy level (e.g., Li0 anode) as evidenced in the passivation of anode surfaces in aqueous electrolytes, which would have to rely on anionic reduction. To resolve this limitation, Liu et al. developed an ILE that offers unprecedented interphasial chemistry functionalities when interacting with electrodes based on a cation possessing CdF bonds that offer enhanced interphasial fluorination relative to mutable fluorines on B, P and S heteroatoms.21 A fluorinated cation, 1-methyl-1-propyl-3-fluoropyrrolidinium was coupled with FSI− anion forming an IL (PMpyrfFSI) that exhibits zero vapor pressure, non-flammability with the ability to transport fluorine sources on both anion and cation. Notably, LiFSI can dissolve in the IL at different concentrations forming an ILE that delivers high Coulombic efficiency (CE) up to 99.9% and Li0 prevention for 900 h in LMB.

376

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Ionic Liquid Electrolytes

In pure ILEs, Li+ ion is only solvated by anions with the IL cation as spectator in Li+ ion solvation. The concentration of the ILE was optimized by altering the lithium salt concentration to ensure the electrolyte species are less absorbed on both electrode surfaces. Fig. 4 outlines the performance of the battery using (PMpyrf)0.8Li0.2FSI and (PMpyrf)0.5Li0.5FSI. (PMpyrf)0.5Li0.5FSI was tested in a NMC622/Li full cell (cell #20; NMC areal loading of 8.8 mg cm−2) between 4.6 and 3.0 V with this ILE delivering a superior initial discharge capacity of 203 mAh/g at C/3, with only 3 mAh/g lost during the first 100 cycles with an average CE of 99.9%. A high capacity retention of 89% was achieved with cycling extended to 300 cycles (Fig. 4c). These results highlight the applicability of ILEs for use on Ni-rich cathode at high voltage along with excellent Li-metal compatibility. In addition, this suggests the resultant CEI and SEI are passivating, thus mitigating additional electrolyte degradation over extended cycling. Wu et al. utilized [LiTFSI]0.2[Pyr14FSI]0.8 ILE achieving high stable cycling of Li/LiNi0.88Co0.09Mn0.03O2 cells for up to 300 cycles with a capacity retention of 88%.22 However, the pronounced performance was attained with only a low cathode mass loading (10−4 S cm−1) and is thermally stable with lithium metal electrode displaying superior electrochemical stability with high voltage cathodes with respect to other solid-state electrolytes (e.g., Li1.4Ti1.6 Al0.4(PO4)3 (LTAP), sulfides). In combination with an ILE that minimizes the interfacial resistance of ceramic LLZO particles and solid electrodes, this composite material delivers Li-ion transport through both solid-liquid (LLZO-IL) and solid-liquid-solid (LLZO-IL-solid) interfaces for high voltage pseudo-solid-state LIBs.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Ionic Liquid Electrolytes

379

The HSE displays chemical stability in the ILE along with good thermal stability even at 400
C, an electrochemical stability of 5.5 V with Li metal and an ionic conductivity of 0.4 mS cm−1. Moreover, the fabricated cells using HSE delivered initial charge-discharge capacities of 140/130 mA h g−1 (Li/HSE/LiCoO2) with 99% capacity retention at the 150th cycle. In high voltage (>8 V) pseudo-solid-state LIBs, the HSE hinders dendrite formation during cycling, emphasizing the use case of composite ILE-based electrolytes for high voltage LIBs. Furthermore, task specific liquid electrolytes namely SILs and ZILs have been developed for lithium batteries with favorable features such as non-flammability, negligible volatility, and high thermal stability. Owing to their composition, SILs can possess high t+Li due to their high concentration of Li+ and the abundance of different charged species compared to binary IL-lithium salt mixture and oxidative stability.28 SILs are designed by altering the combination and composition of lithium salts and glyme, thus tuning the physical bonds. Similarly, ZILs are under investigation for batteries with the cation and anion tethered by covalent bonds enhancing the intermolecular electrostatic interaction.29 ZILs are solid at room temperature, possessing high Tm; yet modification of the ionic constituents introducing imidazolium cations and sulfonamide anions can produce liquid ZIL derivatives. Additionally, PILs have been successfully introduced in LIBs with different electrode materials proving particularly advantageous when high current densities (elevated C-rates) are applied. This is of relevance because one of the drawbacks of ILs is the high viscosity, which causes limited power performance. Imidazolium-based PILs were synthesized and displayed good ionic conductivities as well as low viscosities in combination with LFP electrodes.30 0.5 M LiTFSI in dimethylimidazolium (1,2-DMim+) exhibited good capacity retention during prolonged charge-discharge cycles performed at 1
C and 60
C outlining PILs as a viable class of electrolyte for high performance solvent-free LIBs. As a host for PILs, solid polymeric systems based on PMMA have been utilized for PIL-polymer electrolytes with perfluorinated anions towards safe and high-performance LIBs.31 Such PIL-based electrolytes manifest elevated ion mobility within the polymeric matrix, thus the complex behavior of these solid-state electrolytes can inform on the design of advanced LIB technologies.

5

Ionic liquid electrolytes for post-lithium batteries

Alternative battery chemistries are undergoing increased research to realize durable, superior, and sustainable post-lithium solutions. A range of differing electrolytes including ILEs have been identified as viable candidates for use in post-lithium chemistries with particular focus on aluminum (Al), magnesium (Mg), sodium (Na) and zinc (Zn) batteries. Al offers high volumetric capacity (8040 mA h L−1) and high abundance (i.e., are economical), yet the standard reduction potential of Al is −1.67 V, hence they deliver lower voltage than other post-lithium chemistries. As a result of their favorable features, IL mixtures have been incorporated for Al electrodeposition based on ILs with N,N-pyrrolidinium cations and alkoxy substituents instead of alkyl chains, as changing the IL cation leads to differences in the size and morphology of the Al deposits. Al deposition in such electrolytes is reliant on the concentration and temperature, thus the equilibrium between the differing anionic species. Halides are also influential in determining the conductivity and electrochemical stability of the electrolyte, with chlorine (Cl) inducing ILEs with reduced conductivity and a lower electrochemical stability window relative to bromine (Br) and iodine (I). Chen et al. developed a feasible rechargeable Al battery comprising of inexpensive components utilizing a chloroaluminate inorganic ILE, displaying enhanced capacity (128 mAh g−1), reversibility and rate capability (63 mAh g−1 at 8000 mA g−1) at 393 K compared to the conventional AlCl3-1-ethyl-3-methylimidazolium chloride.32 The use of halide-based ILEs such as chloroaluminate ILEs necessitates handling in an inert environment due to their hygroscopic profile, hence there is a shift to substituting the halides with TFSI anions in the IL. An intricate relationship between the concentration dependent phase behavior and Al deposition in TFSI-based ILs is present even though these materials are less sensitive to air and water. Additional research is required to ascertain the role of the negatively charged Al complex in relation to electrochemical activity and to realize practical Al batteries. Similar to Al based-chemistries, Mg possesses elevated volumetric energy density (3833 mA h L−1) with the possibility to store the same energy in a smaller cell. In Mg battery chemistries, greater complexity lies at the electrode/electrolyte interface because Mg metal restricts the transport of Mg ions to the electrode, therefore the electrolyte must be reductively stable to prevent interphase layer formation. The conventional electrolytes which allow reversible Mg deposition/dissolution are Grignard reagents or Lewis acid-base pairs dissolved in ether solvents, however, the use of volatile solvents present safety risks. ILEs using TFSI anion support Mg deposition/dissolution at high temperatures (150
C). Further advancements have shown reversible Mg stripping/plating in IL solutions without the addition of a separate ether solvent, using an ether functionalized IL and a Mg(BH4)2 anion.33 Experimental results show TFSI anion is present in the first Mg solvation shell forming [Mg2+-TFSI−]+ ion pairs at the electrode/electrolyte interface, which are prone to reductive decomposition forming a passivating layer on the Mg electrode surface. Whereas ether functionalized IL reduces the likelihood of TFSI reduction and passivation formation, highlighting the relevance of functional interactions between ILE and Mg electrode. Recent works by Chellappan et al. shows the use case of composite ILEs consisting of magnesium bis(diisopropyl)amide and 1-ethyl-3-methylimidazolium tetra-chloroaluminate IL in tetrahydrofuran solvent. The electrolyte displays excellent reversibility and CE for Mg deposition/stripping process at room temperature on different working electrodes (molybdenum, graphite and stainless steel).34 Na-ion batteries (SIBs) are the most suitable alternative technology because they possess similar chemistries and can capitalize from the decades of materials research and development of LIBs. SIBs are economical, in large abundance (2.6% of the earth’s crust vs 0.0017 wt% of lithium), low geopolitical supply risks and suitable electrochemical potential (E0Na ¼ −2.71 V vs the standard

380

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Ionic Liquid Electrolytes

hydrogen electrode, SHE). Even though Na is heavier, the loss in capacity can be compensated using the lighter Al current collector. Several studies have implemented ILEs in full SIB cells displaying operation with good cycling stability at room temperature as well as with layered metal oxide cathode materials.35 Also, the inherent instability of organic electrolytes at elevated temperatures has prompted the use of ILEs in SIBs which possess excellent stability at extreme regimes, with even imide salts manifesting good performances in ILEs compared to organic solvents. Notably, ILEs display intrinsic bulk conductivity at high temperatures even without the addition of Na salt, e.g., NaFSI in N-methyl-N-propylpyrrolidinium FSI, 15.6 mS cm−2 at 353 K, where the organic electrolyte is not thermally stable.36 The limitation of ILEs in SIBs stems from the low transference of the metal cation in ILEs which limits the rate capability of the battery along with the typically higher viscosity. The coordination of the metal ions by the IL anions impacts the physical properties with larger Na-complexes, e.g., NaTFSI inducing a lower mobility and thus reduced transference numbers. “Solvent-in-salt” ILEs have been used to enhance the rate performance by increasing the salt concentration resulting in superior Na transference; this has been linked to anions coordinating to two Na cations enabling improved performance.35,37 Zn batteries based on aqueous electrolytes exhibit high theoretical capacity (820 mAh g−1), low cost and low redox potential of Zn (−0.763 V vs SHE). ILs have emerged as model electrolytes for Zn batteries because they can reduce corrosion and reversibly deposit dendrite-free Zn. Advancements have demonstrated the use of dicyanamide (DCA) anion with Zn salts display enhanced redox currents with the IL cation effecting the Zn deposition process.38,39 In Zn batteries, Zn dissolution/deposition can proceed in the presence of water when ILEs are utilized with a small amount of water employed as an additive. Water is beneficial for DCA anions, improving the overall fluidity of the system whereas water displaces triflate and TFSI anions leading to a decrease in the Zn deposition overpotential. The opposite strategy has also been tested using IL as additive in alkaline electrolytes with dendritic growth evident, yet the addition of IL altered the deposition leading to the generation of Zn particles with the absence of dendritic clusters. Lastly, the use of composite electrolyte systems mixing conventional aqueous electrolytes and ILEs offers high electrolyte conductivity and reaction rates for Zn batteries. Ultimately, the design of ILEs for post-lithium batteries will require integrating the decades of research experience on lithium batteries with a fundamental understanding of the intrinsic differences between the metal chemistries.

6

Summary and outlook

ILEs have garnered significant interest from both the academic and industrial sectors owing to their structural versatility, favorable physicochemical properties and compatibility with energy storage systems. The sheer diversity of ILs spanning poly-ILs, SILs, ZILs, PILs and LCILEs offers new avenues towards adoption in LIBs, LMBs and post-lithium chemistries. Nonetheless, ILEs are limited by their high viscosity thus lower tLi+ and complex synthesis methods. In spite of this, ILEs that achieve high ionic conductivity, excellent stability and low viscosity have been demonstrated for use in lithium battery technology. This is exemplified by the ILEs based on TFSI-IL and FSI salts which are feasible for battery applications generating uniform SEI on lithium electrode as well as mitigating dendrite formation. Also, the intrinsic safety that ILs afford in battery systems cannot be understated, thus they represent feasible alternatives to organic liquid solvent-based electrolytes which are commonplace in commercial LIBs. The propensity to design and modulate the constituent ions provides a route to tailor ILs for specific battery chemistries and to produce enhanced EEIs. Recent works outline the ILE functional structure, specific interactions in the electrolyte and with electrode materials as well as the unique properties are influential in realizing superior high energy density lithium batteries. Impressive battery performance metrics have been achieved in cells using different types of ILEs specifically in terms of CE, capacity retention and electrochemical stability even with prolonged cycling. The use of ILs with composite electrolytes consisting of traditional formulations, polymers or solid components ameliorates the compatibility at interfaces, particularly the resistance at the solid-solid interphases as well as the ability to limit side reactions. In consideration of the use of ILEs at the industrial level, the IL constituents must be economical, and this can be achieved by increasing production to leverage the economies of scale. Additionally, simplification of the purification processes using anions such as TFSI−, which is hydrophobic and water stable can streamline such procedures enabling solvent extraction and thus boosting the cost-effectiveness of ILEs towards commercialization.40 The rational development of ILEs with desirable features and ideal chemistry will rely upon the synergy between complementary techniques encompassing computational models, machine learning methods and advanced operando experiments (Fig. 8).

6.1

Computational models

Theoretical approaches that account for the complete electronic structure at the atomic scale can inform on the structure (configuration and conformation) and dynamics of ILs with the ability to predict physical properties of the system (e.g., conductivity, density, viscosity). Of which the descriptors, properties and interpretation are complementary to machine learning and operando experiments. Electronic structure modeling can describe the interactions in ILEs, reaction mechanisms and large-scale simulations can be implemented to attain time-resolved information including interfacial solvation structures, short/long range ordering and dynamics at the electrode/electrolyte interphase. As a result, such large-scale models can be utilized to parameterize these details and evolve the meso- and macroscopic understanding of diverse ILEs.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Ionic Liquid Electrolytes

381

Fig. 8 Framework for the optimization of ILEs employing computational models, machine learning methods and advanced operando experiments.

6.2

Machine learning methods

Machine learning methods can be merged with atomic and electronic structure computations to handle complex IL systems along with large data sets leveraging the information for predictive outcomes. The synergy of these approaches can bridge the gap for the study of larger ILE systems using high quality reference data to train the models with system specific descriptors for short/long range interactions, intra/intermolecular forces, and energetics. Models such as neural network potentials, graph neural networks and gaussian process regression can be employed to “learn” the correlation between structure, dynamics and energy/forces from data obtained through computational simulations and experiments.

6.3

Advanced operando experiments

In situ and operando characterization techniques present excellent advantages to characterize ILEs including in operative battery cells. Kinetic processes such as bond formation/cleavage are on the order of picoseconds, contrasted with cation-anion separation on the order of nanoseconds. These timescales and properties are amenable to characterization by modern computational and operando techniques to resolve features in real-time. Neutron scattering is non-destructive and non-ionizing offering spatio-temporal resolved information on a sample’s atomic/nano-through to meso-, to macro-scale structuring for crystalline, amorphous and liquid systems. The emanating trends from pulsed neutron source measurements are more intelligible when synergy is formed with computational modeling as well as complementary operando microscopy and spectroscopic techniques. In summary, cultivating a cross-disciplined framework will advance the optimization of ILEs to develop superior battery technologies.

Acknowledgments The authors would like to thank Stephanie Ferreira for her graphical expertise in constructing Figs. 1–3, 6 and 8.

382

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Ionic Liquid Electrolytes

References Philippi, F.; Welton, T. Targeted Modifications in Ionic Liquids—From Understanding to Design. Phys. Chem. Chem. Phys. 2021, 23 (12), 6993–7021. Balducci, A. Ionic Liquids in Lithium-Ion Batteries. Top. Curr. Chem. 2017, 375 (2), 1–27. Ma, X.; Yu, J.; Hu, Y.; Texter, J.; Yan, F. Ionic Liquid/Poly(Ionic Liquid)-Based Electrolytes for Lithium Batteries. Ind. Chem. Mater. 2023, 1 (1), 39–59. Welton, T. Ionic Liquids: A Brief History. Biophys. Rev. 2018, 10 (3), 691–706. Singh, S. K.; Savoy, A. W. Ionic Liquids Synthesis and Applications: An Overview. J. Mol. Liq. 2020, 297, 112038. Qi, S.; Liu, J.; He, J.; Wang, H.; Wu, M.; Wu, D.; Huang, J.; Li, F.; Li, X.; Ren, Y.; Ma, J. Structurally Tunable Characteristics of Ionic Liquids for Optimizing Lithium Plating/Stripping via Electrolyte Engineering. J. Energy Chem. 2021, 63, 270–277. 7. Brutti, S.; Simonetti, E.; De Francesco, M.; Sarra, A.; Paolone, A.; Palumbo, O.; Fantini, S.; Lin, R.; Falgayrat, A.; Choi, H.; Kuenzel, M.; Passerini, S.; Appetecchi, G. B. Ionic Liquid Electrolytes for High-Voltage, Lithium-Ion Batteries. J. Power Sources 2020, 479, 228791. 8. Pringle, J. M.; Golding, J.; Baranyai, K.; Forsyth, C. M.; Deacon, G. B.; Scott, J. L.; MacFarlane, D. R. The Effect of Anion Fluorination in Ionic Liquids—Physical Properties of a Range of Bis(Methanesulfonyl)Amide Salts. New J. Chem. 2003, 27 (10), 1504–1510. 9. Tsurumaki, A.; Branchi, M.; Rigano, A.; Poiana, R.; Panero, S.; Navarra, M. A. Bis(Oxalato)Borate and Difluoro(Oxalato)Borate-Based Ionic Liquids as Electrolyte Additives to Improve the Capacity Retention in High Voltage Lithium Batteries. Electrochim. Acta 2019, 315, 17–23. 10. Tong, J.; Wu, S.; von Solms, N.; Liang, X.; Huo, F.; Zhou, Q.; He, H.; Zhang, S. The Effect of Concentration of Lithium Salt on the Structural and Transport Properties of Ionic Liquid-Based Electrolytes. Front. Chem. 2020, 7, 1–10. 11. Ohno, H. Design of Ion Conductive Polymers Based on Ionic Liquids. Macromol. Symp. 2007, 249–250, 551–556. 12. Yoshizawa-Fujita, M.; Ohno, H. Applications of Zwitterions and Zwitterionic Polymers for Li-Ion Batteries. Chem. Rec. 2023, 23 (8), 1–11. 13. Mandai, T.; Dokko, K.; Watanabe, M. Solvate Ionic Liquids for Li, Na, K, and Mg Batteries. Chem. Rec. 2019, 19 (4), 708–722. 14. Stettner, T.; Balducci, A. Protic Ionic Liquids in Energy Storage Devices: Past, Present and Future Perspective. Energy Storage Mater. 2021, 40, 402–414. 15. Liu, X.; Mariani, A.; Adenusi, H.; Passerini, S. Locally Concentrated Ionic Liquid Electrolytes for Lithium-Metal Batteries. Angew. Chem. Int. Ed. 2023, 62, e202219318. 16. Seki, S.; Kobayashi, T.; Kobayashi, Y.; Takei, K.; Miyashiro, H.; Hayamizu, K.; Tsuzuki, S.; Mitsugi, T.; Umebayashi, Y. Effects of Cation and Anion on Physical Properties of Room-Temperature Ionic Liquids. J. Mol. Liq. 2010, 152 (1–3), 9–13. 17. Bin Zhou, Z.; Matsumoto, H.; Tatsumi, K. Low-Melting, Low-Viscous, Hydrophobic Ionic Liquids: Aliphatic Quaternary Ammonium Salts With Perfluoroalkyltrifluoroborates. Chem. - A Eur. J. 2005, 11 (2), 752–766. 18. Guerfi, A.; Dontigny, M.; Charest, P.; Petitclerc, M.; Lagacé, M.; Vijh, A.; Zaghib, K. Improved Electrolytes for Li-Ion Batteries: Mixtures of Ionic Liquid and Organic Electrolyte With Enhanced Safety and Electrochemical Performance. J. Power Sources 2010, 195 (3), 845–852. 19. Xu, C.; Cheng, Z. Thermal Stability of Ionic Liquids: Current Status and Prospects for Future Development. Processes 2021, 9, 337. 20. Liu, K.; Zhou, Y. X.; Han, H. B.; Zhou, S. S.; Feng, W. F.; Nie, J.; Li, H.; Huang, X. J.; Armand, M.; Bin Zhou, Z. Ionic Liquids Based on (Fluorosulfonyl)(Pentafluoroethanesulfonyl) Imide With Various Oniums. Electrochim. Acta 2010, 55 (23), 7145–7151. 21. Liu, Q.; Jiang, W.; Xu, J.; Xu, Y.; Yang, Z.; Yoo, D. J.; Pupek, K. Z.; Wang, C.; Liu, C.; Xu, K.; Zhang, Z. A Fluorinated Cation Introduces New Interphasial Chemistries to Enable High-Voltage Lithium Metal Batteries. Nat. Commun. 2023, 14 (1), 1–15. 22. Wu, F.; Fang, S.; Kuenzel, M.; Mullaliu, A.; Kim, J. K.; Gao, X.; Diemant, T.; Kim, G. T.; Passerini, S. Dual-Anion Ionic Liquid Electrolyte Enables Stable Ni-Rich Cathodes in Lithium-Metal Batteries. Joule 2021, 5 (8), 2177–2194. 23 Liu, X.; Mariani, A.; Diemant, T.; Di Pietro, M. E.; Dong, X.; Kuenzel, M.; Mele, A.; Passerini, S. Difluorobenzene-Based Locally Concentrated Ionic Liquid Electrolyte Enabling Stable Cycling of Lithium Metal Batteries with Nickel-Rich Cathode. Adv. Energy Mater. 2022, 12 (25), 2200862. 24. Atik, J.; Diddens, D.; Thienenkamp, J. H.; Brunklaus, G.; Winter, M.; Paillard, E. Cation-Assisted Lithium-Ion Transport for High-Performance PEO-Based Ternary Solid Polymer Electrolytes. Angew. Chem. Int. Ed. 2021, 60 (21), 11919–11927. 25. Zhang, W.; Wang, X.; Zhang, Q.; Wang, L.; Xu, Z.; Li, Y.; Huang, S. Ceramic Nanofiber-Incorporated Solid Polymer Electrolytes for Flexible Lithium Batteries. ACS Appl. Energy Mater. 2020, 3 (6), 5238–5246. 26. Zhai, W.; Zhu, H. J.; Wang, L.; Liu, X. M.; Yang, H. Study of PVDF-HFP/PMMA Blended Micro-Porous Gel Polymer Electrolyte Incorporating Ionic Liquid [BMIM]BF4 for Lithium Ion Batteries. Electrochim. Acta 2014, 133, 623–630. 27. Kim, H. W.; Manikandan, P.; Lim, Y. J.; Kim, J. H.; Nam, S. C.; Kim, Y. Hybrid Solid Electrolyte With the Combination of Li7La3Zr2O12 Ceramic and Ionic Liquid for High Voltage Pseudo-Solid-State Li-Ion Batteries. J. Mater. Chem. A 2016, 4 (43), 17025–17032. 28. Shigenobu, K.; Shibata, M.; Dokko, K.; Watanabe, M.; Fujii, K.; Ueno, K. Anion Effects on Li Ion Transference Number and Dynamic Ion Correlations in Glyme-Li Salt Equimolar Mixtures. Phys. Chem. Chem. Phys. 2021, 23 (4), 2622–2629. 29. Quan Nguyen, D.; Loi Nguyen, T.; Loan Phung Le, M.; Phong Mai, T.; Sik Kim, H. A Zwitterionic Salt With One Sulfonate and Two Ether Functional Groups as an Additive for Lithium-Ion Battery Electrolyte. Electrochem. Commun. 2022, 137, 107269. 30. Stettner, T.; Walter, F. C.; Balducci, A. Imidazolium-Based Protic Ionic Liquids as Electrolytes for Lithium-Ion Batteries. Batter. Supercaps 2019, 2 (1), 55–59. 31. De Araujo, G.; Enrica, M.; Castiglione, F.; Fazzio, P.; Martinez, M.; Fraenza, C. C.; Stallworth, P.; Greenbaum, S.; Triolo, A.; Appetecchi, G. B.; Mele, A.; Enrica, M.; Castiglione, F.; Fazzio, P.; Martinez, M.; Fraenza, C. C.; Stallworth, P. Polymer Electrolytes Based on Protic Ionic Liquids With Perfluorinated Anions for Safe Lithium-Ion Batteries. Electrochim. Acta 2023, 143466. 32. Chen, C. Y.; Tsuda, T.; Kuwabata, S.; Hussey, C. L. Rechargeable Aluminum Batteries Utilizing a Chloroaluminate Inorganic Ionic Liquid Electrolyte. Chem. Commun. 2018, 54 (33), 4164–4167. 33. Watkins, T.; Kumar, A.; Buttry, D. A. Designer Ionic Liquids for Reversible Electrochemical Deposition/Dissolution of Magnesium. J. Am. Chem. Soc. 2016, 138 (2), 641–650. 34. Chellappan, L. K.; Jayasayee, K.; Kvello, J.; Tolchard, J. R.; Dahl, P. I.; Hanetho, S. M.; Berthelot, R.; Fiksdahl, A. Non-Nucleophilic Electrolyte Based on Ionic Liquid and Magnesium Bis(Diisopropyl)Amide for Rechargeable Magnesium-Ion Batteries. ACS Appl. Energy Mater. 2020, 3 (10), 9585–9593. 35. Domingues, L. S.; de Melo, H. G.; Martins, V. L. Ionic Liquids as Potential Electrolytes for Sodium-Ion Batteries: An Overview. Phys. Chem. Chem. Phys. 2023, 25 (18), 12650–12667. 36. Ding, C.; Nohira, T.; Hagiwara, R.; Fukunaga, A.; Sakai, S.; Nitta, K. Electrochemical Performance of Hard Carbon Negative Electrodes for Ionic Liquid-Based Sodium Ion Batteries over a Wide Temperature Range. Electrochim. Acta 2015, 176, 344–349. 37. Forsyth, M.; Yoon, H.; Chen, F.; Zhu, H.; MacFarlane, D. R.; Armand, M.; Howlett, P. C. Novel Na+ Ion Diffusion Mechanism in Mixed Organic-Inorganic Ionic Liquid Electrolyte Leading to High Na+ Transference Number and Stable, High Rate Electrochemical Cycling of Sodium Cells. J. Phys. Chem. C 2016, 120 (8), 4276–4286. 38. Xu, M.; Ivey, D. G.; Xie, Z.; Qu, W. Rechargeable Zn-Air Batteries: Progress in Electrolyte Development and Cell Configuration Advancement. J. Power Sources 2015, 283, 358–371. 39. Ilyas, F.; Ishaq, M.; Jabeen, M.; Saeed, M.; Ihsan, A.; Ahmed, M. Recent Trends in the Benign-by-Design Electrolytes for Zinc Batteries. J. Mol. Liq. 2021, 343, 117606. 40. Greer, A.; Jacquemin, J.; Hardacre, C. Industrial Applications of Ionic Liquids. Molecules 2020, 25, 5207. 1. 2. 3. 4. 5. 6.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives Ujwal Shreenag Medaa,b, Nidhi Bhata,b, Om Madan Raikara,b, Tribikram Guptac, and Kalpana Sharmad, aDepartment of Chemical Engineering, RV College of Engineering, Bengaluru, India; bCentre for Hydrogen and Green Technology, RV College of Engineering, Bengaluru, India; cDepartment of Physics, RV College of Engineering, Bengaluru, India; dDepartment of Physics, Ramaiah Institute of Technology, Bengaluru, India © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 1.1 1.2 1.2.1 1.2.2 1.2.3 1.3 2 2.1 2.2 2.2.1 2.2.2 2.2.3 2.2.4 2.2.5 2.2.6 2.2.7 2.2.8 2.2.9 2.2.10 2.2.11 2.2.12 3 4 5 6 References

Introduction What is SEI? Issues associated with SEI Thermal failure Chemical failure Mechanical failure Need for additives Additives Working principle of additives Different additives Fluoroethylene carbonate 2–Fluoropyridine (C5H4FN) Trimethylsilyl isothiocyanate 3-Sulfolene Lithium hexafluorophosphate Bisfluoroacetamide (C2H3F2ON) Diphenyl disulfide (C12H10S2) Lithium bis(oxalato)borate (C4O8B− Li+) Dimethoxydimethylsilane (C4H12SiO2) Bis (4-fluorophenyl) sulfone (C12H8F2SO2) Heptafluorobutyric anhydride Lithium hexamethyldisilazide Disadvantages of additives Alternatives to conventional additives Future of additives Conclusion

384 384 384 384 385 385 385 385 385 386 386 386 387 387 388 388 390 390 391 391 392 393 393 394 395 395 395

Abstract Solid Electrolyte Interphase (SEI) can prolong battery life but has issues with stability and electrochemical performance, which can be overcome by additives. Electrolyte additives can stabilize SEI and Cathode Electrolyte Interphase (CEI) layers, prevent dendrite formations, increase capacity retention, and enhance the cycling performance of batteries. Drawbacks such as lithium plating should be addressed before the development of electrolyte additives for commercial use. The functional synthetic additives developed through retrosynthesis may create a thermodynamically stable interphase and add spatial versatility. Rigorous theoretical calculations in conjunction with cutting-edge experimental methodology and SEI film characterization are necessary to comprehend the proper reaction route.

Glossary Accretion Growth or increase by the gradual accumulation of additional layers or matter. Capacity retention A measure of the ability of a battery to retain stored energy during an extended open-circuit rest period. Density functional theory (DFT) A computational quantum mechanical modelling method used to calculate the electronic structure of atoms, molecules, and solids. Graphite exfoliation The exfoliation of graphite is a phase transition involving the vaporization of the intercalate in the graphite. Exfoliated graphite is an expanded graphite with a low density. Intercalation The addition of lithium ions into a host material without significantly changing the host’s structure. Kinetic Monte Carlo Method (KFC) A computer simulation intended to simulate the time evolution of some processes occurring in nature.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00119-5

383

384

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

Overpotential The potential difference (voltage) between a half-reaction’s thermodynamically determined reduction potential and the potential at which the redox event is experimentally observed. Pulverization The process of pressing or crushing something until it becomes powder or a soft mass. Retrosynthesis An analytical method of chemical synthesis which involves “deconstructing” a target molecule into its readily available, simple starting materials in order to assess the best synthetic route. Surfactants Surfactants are chemical compounds that decrease the surface tension or interfacial tension between two liquids, a liquid and a gas, or a liquid and a solid.

Key points

• • • • •

1 1.1

SEI enhances battery performance by preventing further electrolyte decomposition Additives facilitate the morphological growth of SEI and improve battery operation Some additives reduce a battery’s lifetime and degrade the sustainability of the SEI Retro-synthesis of synthetic additives can overcome issues of conventional additives Theoretical calculations, studies and characterization of SEI can improve additives

Introduction What is SEI?

The anode and cathode materials of a Lithium Ion Battery (LIB) are in a non-equilibrium condition at the initial stage of the activated operation. This leads to the sacrificial decomposition of the electrolyte and the electrode. The decomposed material chemically adsorbs onto the electrodes, leading to the formation of passivation components on electrode surfaces. These passivation components are known as solid-electrolyte-interphase (SEI) on the anode and cathode-electrolyte-interphase (CEI) on the cathode. The reaction of the alkali metal with the electrolyte generates the SEI almost instantaneously. The SEI blocks the flow of electrons through it but allows Li ions to pass through it. As the SEI grows in thickness, the tunneling barrier faced by the electrons grows, thus stopping the chemical reaction between the electrode and the electrolyte and thereby the growth of the SEI. The SEI acts as a protective layer, preventing undesired chemical degradation of the electrode. A very high degree of passivation reduces the power output of the cell and is not desirable. The SEI layer on the negative cathode particle surface is formed within just the first five charge-discharge cycles of carbonaceous electrodes. The SEI layer formed on the negative cathode is thick while the SEI layer formed on the positive anode is not thick enough to provide a complete passivation barrier between the electrolyte and oxidizing environment. Anode Oxidation of the electrolyte forms a nanometer-thick SEI layer only primarily composed of organic compounds. A successful SEI layer separates the active electrode material from the electrolyte, by allowing only the intercalation and deintercalation of the Li + ions. The type of electrolyte, active materials, binder, and conductive substances used, are the chemical factors that affect SEI formation. Temperature and battery cycling conditions are some of the other physical factors.

1.2

Issues associated with SEI

The formation of the SEI when the electrodes are immersed in the electrolyte is far from chemical equilibrium and is in quasi-chemical equilibrium. Ideally, the SEI should be thin and lightweight, lead to structural stabilization of the electrodes, be insensitive to the presence of impurities, and have low impedance, low polarization, and high discharge plateau. However, in real life, every SEI film is saddled with one or many of the issues that lead to non-ideal behavior, as discussed further.1 A thin and light SEI layer is desirable since it reduces the degradation in output power and boosts the capacity and cyclability of the battery. However, the size of the SEI layer grows until the electrons are unable to tunnel through them. The depth of the potential well is dictated by the choice of electrolyte and electrode. Thus, tuning the thickness of the SEI layer is a complex issue that has evaded a proper understanding both theoretically and experimentally. The stability of the electrodes can be achieved only if the protective SEI layer is formed so fast that it can prevent the sacrificial degradation of the electrode efficiently. However ultrafast formation of SEI causes inhomogeneous accretion on the electrode resulting in irregular pathways for ionic conductivity. This is a big issue with SEI formation and the electrode stability may suffer in the following three ways.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

385

1.2.1 Thermal failure For high C-rate operation, heat is generated due to both irreversible Joule dissipation and reversible heat generated due to entropic effects and exothermic chemical reactions. The SEI decomposes at high temperatures, resulting in an inhomogeneous and thicker SEI layer as revealed by XPS, SEM, and ToF-SIMS techniques. The SEI becomes unstable at high temperatures due to enhanced electrochemical activity caused by thermal activation. Due to the fast rate of dissolution of organic components in organic electrolytes, the SEI starts to come off. Besides, at high temperatures, electrochemical surface tension at the interphase increases causing mechanical instability.

1.2.2 Chemical failure Cycling of the battery renders it chemically unstable for certain ranges of parameters.

1.2.3 Mechanical failure Cycling of the batteries dramatically alters the elastic properties of the electrode causing mechanical failure of batteries. The formation of the SEI film is strongly influenced by the presence of impurities, which modify the energy landscape of the electrode and thereby affect the composition and morphology of the SEI, affecting the transport of Li ions. Low battery resistance is desirable as it leads to low Joule dissipation. The resistance is proportional to the thickness and regularity of the SEI film, both of which are hard to control. Polarization refers to the difference between the measured terminal voltage and the internal open circuit voltage, due to the passage of electrical current through that cell. It can be ohmic polarization, concentration polarization, and electrochemical polarization. It reduces the performance of batteries and should be controlled. Strong cross-linking of polymeric structures formed in the SEI makes the SEI dense and insoluble, another common SEI problem (Presence of vinoXyl radicals and sp2-hybridized species in fluoroethylene carbonate (FEC)).

1.3

Need for additives

SEI layer enhances the performance of lithium-ion batteries by preventing further electrolyte decomposition, thereby maintaining cycling stability. Additives facilitate the regular morphological growth of the SEI by playing a crucial role in every stage of SEI formation, stability, and optimal performance. Additives ensure that the SEI layer affixes stably to the electrode and blocks the flow of electrons while allowing the passage of Lithium ions. Additives influence the porosity and thickness of the SEI layer on both electrodes by altering the rate of the irreversible interfacial reactions, that is the cathodic reduction reactions and anodic oxidation reactions. This affects the diffusion rate of lithium ions through the SEI.2 A high rate of SEI formation results in cracks due to which there is an inhomogeneous transport rate of Li-ions through it, causing the emergence of irregular patterns. Adding certain additives to the electrolyte can prevent mechanical degradation, stop dendrite formation, and help in the formation of robust SEI. Certain electrolyte additives can enhance the mechanical performance at the interphase. Additives stabilize lithium-ion batteries, which are prone to irreversible capacity degradation due to SEI formation by helping in the formation of an engineered SEI layer at the negative electrode. This is because the additive has a lower/higher reduction potential than the electrolyte. Density Function Theory predicts that a few additives considerably enhance cyclic stability. LIBs based on graphite electrode due to possessing a lower reduction potential compared to the electrolyte forms a protective SEI layer on the graphite electrode. Certain additives enhance cyclic stability due to a higher reduction potential than electrolytes. As per cyclic voltammetry (CV) measurements, a combination of electrolyte and the additive demonstrates a reduction peak at a specific voltage, thereby inhibiting the electrolyte reduction. Raman spectroscopy and electrochemical impedance spectroscopy (EIS) show that SEI formation in the presence of such additives can reduce structural disorder in the graphite electrode as well as lower interfacial and charge transfer resistance. Capacity retention increases too after a few cycles.

2 2.1

Additives Working principle of additives

A proper understanding of how additives work has eluded the researchers so far. The limited understanding of how additives work is completely based on the insights offered by theoretical methods like Quantum Chemistry (QC), Density Functional Theory (DFT), First Principle Molecular Dynamics (FPMD), DFT-MD, Kinetic Monte Carlo (KMC), and Red Moon Method (KMC + MD). The additives undergo a complex chain of reactions with both the electrolyte and the electrode. “Trans-esterification” (An alcohol R group is swapped with an ester R’ group), enables these additives to suppress a host of undesirable satellite reactions between the bulk electrolyte solvents. Electrolyte additives modify lithium deposition and thus influence the battery operation at two crucial levels. It regulates the rate of aggregation of SEI on the electrode and then influences both the onset and growth of lithium metal deposition. SEI layer gets formed in a certain sequence each being influenced by the presence of additives.

386

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

Reduction of the electrolyte: In the absence of any meaningful experiment at this stage, all inferences like reduction pathways and electrochemical reactions are achieved with the help of theoretical simulations. The methods used range from first-principles molecular dynamics (FPMD), density functional theory, tight binding, kinetic Monte Carlo, and reaction force field (ReaxFF). Aggregation of reduced products into the SEI: This stage is poorly understood both experimentally and theoretically. What initiates the aggregation process, what drives it, and what selects the reaction pathways is not clear. Formation of a heterogeneous bilayer structure within the SEI: An inorganic salt layer starts aggregating on the electrode side. An organic salt layer starts aggregating on top of the inner inorganic layer. The solvent molecules between the electrode(anode) and the inorganic layer get reduced to form the inner SEI components. Additives alter the rate of the Irreversible interfacial reactions on the surface of both electrodes (cathodic reduction reactions and anodic oxidation reactions), thereby affecting the overall conduction of lithium ions through the SEI layer.3 When water is used as an additive, even in trace ppm levels, Lithium metal grows columnar on metallic electrodes made of copper, brass, nickel, and stainless steel. Water reacts with PF6 anions to form HF. If the potential of the working electrode is reduced, electrocatalytic reduction of HF occurs on the surface to form LiF. This LiF layer controls the shape and morphology of the lithium deposition. Similar results are seen using other electrolyte additives like CsPF6, LiAsF6, or ex-situ deposited LiF layers. The connection between the presence of LiF in the initial SEI and the resulting shape of the growth has not been found so far. The optimum quantity of additive concentration cannot be estimated.

2.2

Different additives

2.2.1 Fluoroethylene carbonate Fluoroethylene Carbonate (FEC) is a popular film-forming additive that works well in Lithium Ion Batteries (LIB). Research shows that FEC could promote the production of protective surface films on anodes and cathodes. Defluorination and ring-opening are the two most likely degradation pathways for FEC, as seen in Fig. 1. Three types of FEC decomposition products can be distinguished: LiF and non-fluoride species; LiF and organic fluorides; organic fluorides (unstable). FEC experiences more fluoride polymerization at the cathode than at the anode. On the cathode surface, fluoride takes part in interphase formation through polymerization. Solvent, as well as salt decomposition, could be excellently suppressed and LiCoO2 could be shielded from being destroyed by the stable cathode-electrolyte interphase. Hydrogen Fluoride (HF) and polycarbonate species on the anode are mostly the decomposition products of FEC, which may then be further reduced to produce species with higher stability. Lithium-ion salt decomposition could be reduced by the interface during cycling, leaving a smaller number of inorganic species, LiF and LiCO3, on the anode surface, encouraging the transport of charges at the electrode active particle-liquid electrolyte interphase. As a consequence, the capacity cycle is enhanced over subsequent cycles at a higher voltage. LiCoO2/graphite pouch cells display enhanced cycling performance in 5% FEC-containing electrolytes in the voltage range of 3.0–4.4 V. A 64% increase in the discharge capacity retention after 100 cycles is observed.4

2.2.2 2–Fluoropyridine (C5H4FN) 2-Fluoropyridine (2-FP) is an effective and inexpensive electrolyte additive that has been utilized in carbonate-based and ether-based electrolytes for Lithium Metal Batteries (LMBs) with favorable results.

Fig. 1 Scheme depicting possible decomposition reaction patterns and products involving FEC. Reprinted with permission Liu, D.; Qian, K.; He, Y.-B.; Luo, D.; Li, H.; Wu, M.; Kang, F.; Li, B. Positive Film-Forming Effect of Fluoroethylene Carbonate (FEC) on High-Voltage Cycling with Three-Electrode LiCoO2/Graphite Pouch Cell. Electrochim. Acta 2018, 269, 378–387. https://doi.org/10.1016/j.electacta.2018.02.151.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

387

2-FP impacts the distribution and transference number of Li+ ions. The lone-pair electrons of nitrogen in the pyridine molecule can behave as Lewis bases. Through acid-base interactions, Li ions (Lewis acids) are adsorbed strongly into the electrolyte. This interaction facilitates the distribution of metallic Li nuclei uniformly on the surface of the anode. Additionally, the pyridine ring’s F atoms are separated with ease during the charging/discharging cycle and participate in SEI formation. The reduction of 2-FP leads the C-F bond to be broken, releasing the F atom, which combines with Li ions to develop a stable SEI layer composed of LiF throughout the preferential reduction process. Fig. 2 depicts the evolution of SEI layer compositions with the electrolyte in the presence and absence of 2-FP as an electrolyte additive. 2-FP helps reduce nucleation, and deposition overpotentials and promotes the development of a robust and dense SEI layer rich in LiF and Li3N. This SEI layer successfully suppresses parasitic reactions between electrolyte and Li metal, allowing for rapid Li+ ion transportation. LMBs with 2-FP having a high areal capacity (1.5 mAh cm−2) and as an additive to LiNi1/3Mn1/3Co1/3O2 cathode, it shows significant improvement in high voltage cyclability and energy density. Even when a controlled amount of 2-FP was added to the LiFePO4|Li metal cell’s electrolyte, a capacity retention of 68.04% after a 200-cycle test at a high current density of 1.5 mA cm−2 was observed.5

2.2.3 Trimethylsilyl isothiocyanate (Trimethylsilyl)isothiocyanate (TMSNCS) is a compound containing both amino silane and isothiocyanate functional groups. It effectively scavenges Phosphorus Pentafluoride (PF5) and hydrogen fluoride (HF), which can negatively affect the thermal stability of fluoroethylene carbonate (FEC) and bring about considerable compositional and structural changes in the electrode interphases. The nitrogen in TMSCNS has a high nucleophile strength, and forms a complex with PF5. PO3F, which is highly reactive, can create H(PO2F2) and H2(PO3F) through hydrolysis. However, the production of PO3F is not driven by simply deactivating PF5. TMS-F is generated through beneficial complexation with HF and the amino silane (Si-N) group present in TMSCNS is responsible for the creation of TMS-F. The mechanism for HF removal and PF5 stabilization is given in Fig. 3. In LiPF6-containing electrolytes, TMSNCS addition also aids in maintaining the interfacial layers and improves thermal stability at high temperatures. TMSNCS develops interfacial layers, that can transmit ions, at the surfaces of the electrodes and reduces the disintegration of the CEI and SEI layers after recurrent cycles. Therefore, the NCM622/graphite full cells’ superior fast-charging ability is also made possible by its addition. For storage at elevated temperatures, the TMSNCS addition further enhances the performance of full cells based on graphite anodes and NCM622 cathodes by successfully mitigating the removal of Lewis acid-promoted F from FEC in electrolytes having LiPF6.6

2.2.4 3-Sulfolene 3-sulfolene (3SF) is a novel additive for graphite anodes that exhibits a strong reductive activity. Vinylene Carbonate (VC) is a well-known electrolyte additive that gets reduced to Propylene Carbonate (PC) while forming SEI film on the graphite surface. Under storage, VC tends to polymerize resulting in a loss of film-forming capacity. PC-based electrolytes provide good safety and a

Fig. 2 Illustration of the orientation of SEI layer with the electrolyte in the absence and presence of the electrolyte additive 2-FP. Reprinted with permission Xie, Z.; Wu, Z.; An, X.; Yue, X.; Yoshida, A.; Du, X.; Hao, X.; Abudula, A.; Guan, G. 2-Fluoropyridine: A Novel Electrolyte Additive for Lithium Metal Batteries with High Areal Capacity as Well as High Cycling Stability. Chem. Eng. J. 2020, 393, 124789. https://doi.org/10.1016/j.cej.2020.124789.

388

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

Fig. 3 Schematic representation of the mechanism for removal of Hydrogen Fluoride removal and stabilization of PF5 by the Trimethylsilyl isothiocyanate additive. Reprinted with permission Han, J. G.; Jeong, M. Y.; Kim, K.; Park, C.; Sung, C. H.; Bak, D. W.; Kim, K. H.; Jeong, K. M.; Choi, N. S. An Electrolyte Additive Capable of Scavenging HF and PF5 Enables Fast Charging of Lithium-Ion Batteries in LiPF6-Based Electrolytes. J. Power Sources 2020, 446(November 2019), 227366. https:// doi.org/10.1016/j.jpowsour.2019.227366.

wide operating range for lithium-ion batteries due to high flash points and low melting points. However, at voltages above 0.8 V, PC-based electrolytes begin to get co-inserted and reduced leading to severe exfoliation of the graphite electrode. Characterization studies such as charge/discharge cycle, SEM, TEM, and elemental mapping revealed that an addition of 3% 3SF by weight produces a uniform and thin SEI film on the graphite surface effectively suppressing the continuous co-insertion and reduction of PC-based electrolyte. This led to the high cyclic stability of Li/graphite cells in PC-based electrolytes. Noticeably, 3SF is superior to the other additives such as prop-1-ene-1 and 3-sultone, to replace vinylene carbonate and to increase the interfacial stability of graphite/PC-based electrolyte.7 Exfoliation of graphite in the absence of 3SF and prevention in its presence is visualized in Fig. 4.

2.2.5 Lithium hexafluorophosphate Lithium hexafluorophosphate (LiPF6), dissolved in mixtures of cyclic carbonates, particularly ethylene carbonate (EC), and linear carbonates, such as ethyl methyl carbonate, make up the majority of the electrolytes in today’s commercial LIBs. A stable SEI is formed when carbonate/LiPF6 electrolytes combine with graphite-negative electrodes. It inhibits ongoing electrolyte deterioration while permitting reversible charging and discharging. LiPF6 predominantly and quickly reacts with Li2CO3 generating POF3 where hydrolysis should be kinetically inhibited at moderate temperatures. These POF3s preferentially react with highly anionic oxygens thereby demonstrating autocatalysis. By altering the distribution of inorganic carbonate species or by limiting the transport of PF6 through the SEI, the reactivity of LiPF6 can be controlled. Fig. 58 illustrates the formatting of SEI film in presence of LiPF6. LiPF6 a functional material acts as a material scavenger by eliminating or deactivating reactive substances produced by the degradation of the electrolytes. It effectively provisions fine-tuning of interfacial structures of electrodes. Electrolyte instability which leads to lifespan reduction and performance decline, is effectively mitigated by LiPF6.9

2.2.6 Bisfluoroacetamide (C2H3F2ON) Bisfluoroacetamide (BFA) can be used to create a surface rich in C-F bonds and a bottom rich in LiF in a gradient SEI design. The SEI enables the even and dense deposition of Li+ ions owing to the strong polarity of the C-F bonds, which can behave as Lewis base sites for Li+ adsorption. The LiF layer at the bottom has lower Li + surface diffusion potential and higher interfacial energy, facilitating the prompt transportation of Li+ ions and a smooth interphase creation. BFA lowers the nucleation and plateau overpotentials for the deposition of Li+ ions. Additionally, the LiF layer effectively blocks the transport of electrons across the interfacial phase, preventing dendrite formations and side reactions with the electrolyte. The pseudo capacitance generated on and inside the electrode surface causes the Li+ transport rate to increase and provides effective protection to the electrode material. Fig. 6 visualizes the CEI formation and the underlying mechanism. BFA decreases the activation energy of the electrolyte’s Li+ ions, creating more favorable conditions for their prompt transport. This leads to stable cycling performance in Li||Li symmetric cells comprising 1.0 wt% BFA for over 700 h at a current density of 1.0 mA cm−2. In Li||LiNi0.6Co0.2Mn0.2O2 (Li||NCM622) full cells, the addition of BFA helps to uphold the cathode structure’s integrity by generating pseudo capacitance. This reduces the pulverization of the anode and results in capacity retention of 65.1% and better cyclic stability after 200 cycles.10

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

389

Fig. 4 Schematic representation illustrating the benefits of 3SF as an electrolyte additive. Reprinted with permission Wang, K.; Xing, L.; Zhi, H.; Cai, Y.; Yan, Z.; Cai, D.; Zhou, H.; Li, W. High Stability Graphite/Electrolyte Interface Created by a Novel Electrolyte Additive: A Theoretical and Experimental Study. Electrochim. Acta 2018, 262, 226–232. https://doi.org/10.1016/j.electacta.2018.01.018.

Fig. 5 Schematic representation of SEI film forming in presence of LiPF6. Reprinted with permission van Ree, T. Electrolyte Additives for Improved Lithium-Ion Battery Performance and Overcharge Protection. Curr. Opin. Electrochem. 2020, 21, 22–30. https://doi.org/10.1016/j.coelec.2020.01.001.

390

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

Fig. 6 Illustration of BFA’s working principle. The impact of adding BFA on the electrolyte, cathode, and anode, along with CEI layer formation on the surface of NCM622 is visualized. Reprinted with permission Sun, Z.; Wen, Z.; Chen, Y.; Ma, Y.; Zhang, J.; Li, Y.; Li, L.; Chen, R. A Gradient Solid Electrolyte Interphase with High Li+ Conductivity Induced by Bisfluoroacetamide Additive for Stable Lithium Metal Batteries. Nano Res. 2023. https://doi.org/10.1007/s12274-022-5363-6.

2.2.7 Diphenyl disulfide (C12H10S2) Diphenyl Disulfide (DPDS) additive belongs to a group of cathode additives that form passivation layers on the surface of the cathode. This functionality comes from its aromatic nature which facilitates polymerization. The passivation layers help in stabilizing the CEI and enable electron conduction owing to the conjugated structure of the products formed due to polymerization. DPDS can also take part in the generation of an SEI film on the surface of the anode and the decomposition of the electrolyte due to reduction is also simultaneously suppressed. The SEI films derived from DPDS protect the Li1.2Mn0.54Ni0.13Co0.13O2 electrode and graphite structures from deterioration during high-temperature cycling and enhance the performance at elevated temperatures. Radical cations are formed due to preferential oxidation which reacts with other additive molecules leading to the formation of the conducting film on the cathode as shown in Fig. 7. Capacity retention, resistance after 100 cycles, and cell expansion showed around 22% increase, 76% decrease, and 27% decrease respectively, due to the addition of DPDS.11

2.2.8 Lithium bis(oxalato)borate (C4O8B− Li+) Lithium bis(oxalato)borate (LiBOB) prevents unintended electrolyte decomposition on the surface of cathodes such as Li1.17Ni0.17Mn0.5Co0.17O2. In the absence of LiBOB, the electrolytes decompose continuously at high voltages and a resistive deposit is formed on the cathode due to unstoppable side reactions. LiBOB additive efficiently prevents severe oxidative degradation of LiPF6-based electrolytes by forming a protective layer on the cathode surface as depicted in Fig. 8.

Fig. 7 The reaction mechanism illustrating the formation of a conducting film on cathode through oxidative polymerization of DPDS. Reprinted with permission Zuo, X.; Zhao, M.; Ma, X.; Xiao, X.; Liu, J.; Nan, J. Effect of Diphenyl Disulfide as an Additive on the Electrochemical Performance of Li1.2Mn0.54Ni0.13Co0.13O2/ Graphite Batteries at Elevated Temperature. Electrochim. Acta 2017, 245, 705–714. https://doi.org/10.1016/j.electacta.2017.05.155.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

391

Fig. 8 Visualization of LiBOB derived Cathode-Electrolyte Interphase. Reprinted with permission Kazzazi, A.; Bresser, D.; Kuenzel, M.; Hekmatfar, M.; Schnaidt, J.; Jusys, Z.; Diemant, T.; Behm, R. J.; Copley, M.; Maranski, K.; Cookson, J.; de Meatza, I.; Axmann, P.; Wohlfahrt-Mehrens, M.; Passerini, S. Synergistic Electrolyte Additives for Enhancing the Performance of High-Voltage Lithium-Ion Cathodes in Half-Cells and Full-Cells. J. Power Sources 2021, 482 (June 2020), 228975. https://doi.org/10.1016/j.jpowsour.2020.228975.

Cycle stability and rate capability of cathodes such as Li1.17Ni0.17Mn0.5Co0.17O2 are enhanced due to the addition of LiBOB as an additive. In addition, a better discharge capacity of 115 mAh g−1 was observed. A full cell with LiBOB retains its Open Circuit Voltage (OCV) better than that of a full cell without the additive during storage at 60
C. LiBOB-added electrolyte effectively mitigates electrolyte decomposition resulting in a relatively thin protective film on the cathode. The cathode cycled in the reference electrolyte without LiBOB formed a thick and unstable surface film on the lithium-rich cathode surface, which is likely the result of the significant electrolyte decomposition. This difference was confirmed by SEM and XPS studies. Also, the completely charged graphite/Li1.17Ni0.17Mn0.5Co0.17O2 whole cells’ storage performance at 60
C was significantly enhanced by the protective layer produced from LiBOB.12

2.2.9 Dimethoxydimethylsilane (C4H12SiO2) Silyl-functionalized dimethoxydimethylsilane (DODSi) is suggested to provide cyclic stability to layered nickel-rich cathode material at high voltage. Because of its stability under anodic polarization, DODSi does not produce artificial cathode-electrolyte interphases through electrochemical oxidation. Decomposition of LiPF6 (LiPF6 $ LiF+PF5) under atmospheric conditions generates F− species in the electrolyte. Ni-rich NCM cathode materials’ performance is impacted by the generated F− species through a chemical reaction at the surface. For the Ni-rich cathode materials, the nucleophilic F− readily binds the electrophilic transition-metal-ions (Fig. 1A), resulting in the formation of MFx species that are highly soluble in the electrolyte. The silyl-ether (Si-O) chemical moiety is present in DODSi. Due to its ease in donating non-bonding electrons and the strong affinity of the Si element toward the F− species, the O element traps positively charged species excellently. Si-O is therefore seen to scavenge F− species in the cell very effectively. The mechanism of DODSi is visualized in Fig. 9. DODSi possesses a broad electrochemical window up to 5.0 V (vs. Li/Li+), demonstrating that it does not impact the electrochemical stability of traditional carbonate-based electrolytes. At high potential, DODSi can improve cycling retention: a cell cycled with 0.25 wt% DODSi shows a retention ratio of 60.8% compared to a cell cycled with only standard electrolyte (41.9%). Using too much DODSi (2.0 wt%) does not improve the cell’s ability to cycle because it reduces the ionic conductivity of the electrolyte.13

2.2.10

Bis (4-fluorophenyl) sulfone (C12H8F2SO2)

Bis (4-fluorophenyl) sulfone (BFS) forms CEI that can significantly enhance the cycling stability and prevent the cathode from acute degradation through the formation of a smooth and sturdy film. It is suitable for Nickel rich lithium-ion batteries and is effective in high-voltage applications. Experiments revealed that the capacity retention was enhanced by around 8% after 100 cycles at the rate of 1C. The sulfone group of BFS is efficient in forming a protective layer on the electrode, protecting the electrolyte from decomposition, and reducing the degradation of the cathode at high voltage. The aromatic ring enhances the oxidation stability of the electrolytes at high voltages and also enhances the overcharge tolerance. A uniform CEI film is formed in presence of BFS and the same is shown in Fig. 10. The electrochemical performance of BFS containing half-cells (0.5% BFS) is much higher. In the presence of BFS, the discharge capacities of the half-cells when evaluated after 100 cycles at the rate of 1C increased by around 16 mAh g−1 to 173.76 mAh g−1.

392

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

Fig. 9 Scheme depicting the effect of hydrogen fluoride at the top and DODSi at the bottom. The surface stability of anode and cathode are visualized. Reprinted with permission Jang, S. H.; Jung, K.; Yim, T. Silyl-Group Functionalized Organic Additive for High Voltage Ni-Rich Cathode Material. Curr. Appl. Phys. 2018, 18(11), 1345–1351. https://doi.org/10.1016/j.cap.2018.07.016.

Fig. 10 Depiction of uniform cathode-electrolyte interphase formation due to the addition of BFS in comparison with blank electrolyte without BFS. Reprinted with permission Liu, L.; Gao, W.; Cui, Y.; Chen, S. A Bifunctional Additive Bi(4-Flurorophenyl) Sulfone for Enhancing the Stability and Safety of Nickel-Rich Cathode Based Cells. J. Alloys Compd. 2020, 820, 153069. https://doi.org/10.1016/j.jallcom.2019.153069.

Overcharge tolerance of the half-cells also witnessed an improvement due to the addition of 2% BFS. Overcharging tests were conducted on the half-cells which proved that BFS contributes to making the handling of these batteries safer, as it alleviates the overcharge behavior mechanism.14

2.2.11

Heptafluorobutyric anhydride

Heptafluorobutyric Anhydride (HFA) can concurrently enhance electrode-electrolyte interphases as well as Li-ion flux/solvation. HFA exhibits a more favored reduction trend at the anode than the base electrolyte, enabling the creation of a dense and even SEI rich in inorganic elements. Lithium Fluoride (LiF) is produced by HFA’s high fluoride content. This is favorable for Li-ion

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

393

conduction in inner SEI and a C-F-rich surface of SEI is developed, by which the adsorption of Li-ions can be promoted. HFA at the cathode can aid in the creation of a CEI that is more compact and thinner, protecting the active material’s structure and reducing the dissolution of transition metal ions. The rising amount of LiF in CEI can promote prompt Li-ion transport. HFA can be used as a surfactant in the separator to lower the electrolyte surface tension and improve the separator’s wettability, which increases the Li-ion flux. The even deposition of Li benefits from the uniform Li-ion flux. The introduction of HFA reduces the cumulative coordination number of the solvent molecules in the bulk electrolyte, aiding in Li-ion desolation. Fig. 1115 provides the schematic representation of the mechanism of SEI formation in the presence of HFA. HFA prevents needle-like dendrite formations on the anode surface. This leads to improved cycling stability in Li||Li symmetric cells containing 1.0 wt% HFA, which can cycle for over 340 h with only mild polarization, whereas the overpotential rapidly rises in blank electrolytes after 100 h. In Li||NCM622 full cells, HFA addition significantly enhances the cycling performance, and the capacity retention rises after 250 cycles.15

2.2.12

Lithium hexamethyldisilazide

Lithium hexamethyldisilazide (LiHMDS) is an additive suitable for Nickel-rich cathodes like LiNi0.8Co0.1Mn0.1O2 (NCM811). NCM811 shows tremendous potential to assist lithium-ion batteries (LIBs) in achieving high specific energy (>400 W h kg−1). Li|| NCM811 batteries can experience improved cycling stability when exposed to high-stress circumstances like high voltages (4.5 V) and temperatures (60
C), owing to LiHMDS, which has a low oxidation potential. F NMR measurements on several electrolyte solutions can be conducted to understand the working mechanism of LiHMDS. These measurements show that LiHMDS reacts with H2O more vigorously than with LiPF6 and can remove HF from the electrolyte. The reaction mechanism of LiHMDS with HF and H2O is shown in Fig. 12. The capacity retention of the Li||NCM811 batteries with LiHMDS is very good. It effectively suppresses electrolyte side reactions along with the removal of HF and H2O from the electrolyte. It also provides NCM811 cathode with a good thermal shock resistance.16

3

Disadvantages of additives

There are no broad guiding principles in the field of additives, making progress in this field tardy. One of the common undesirable fallouts of additives is the deposition of non-intercalated Li on the anode surface, also called plating. Plating should be addressed while developing electrolyte additives for commercial applications. A standard additive Vinylene carbonate with Graphite negative electrode blocks the intercalation channels by forming highly resistive surface films and triggers plating.17 A study of negative electrode SEI with different electrolyte additives like triallyl phosphate (TAP), tris(−trimethyl silyl)-phosphite, Vinylene Carbonate (VC), trimethylsilyl methane sulfonate (TMSMS), and Ethylene sulfite (ES) along with 1M LiPF6 as electrolyte indicates that though these additives induce low negative electrode charge transfer resistance thereby allowing for low-temperature operation and high rate of charging, they can dramatically reduce the lifetime of the battery. Sulfur-containing electrolyte additives have poor compatibility with negative graphite electrodes and easily get oxidized on the positive electrode at high temperatures. Some of them are limited in their usage in commercial battery systems due to their harmful effect on human health. Table 1 shows the disadvantages of some sulfur-containing electrolyte additives.18 Fluorinated additives are seen to improve the electrochemical performance of LIBs, but degrade the sustainability of the SEI and have adverse effects on the chemical reaction pathways. The use of FEC additive at the high potential difference is untenable in graphite cells because it causes an abrupt degradation in performance after about 1000 cycles (rollover failure). The problem gets aggravated if the upper cut-off potential is sought to be increased since that helps in enhancing the energy density of the LIB. The conductive additives block the active sites on the electrode causing degradation in the transport of ions and the overall performance of the electrolyte.

Fig. 11 Schematic diagram illustrating the effect of qua-functional HFA. Reprinted with permission Su D Powerful Qua-Functional Electrolyte Additive for Lithium Metal Batteries. Green Energy Environ. 2022, 7(3), 361–364. https://doi.org/10.1016/j.gee.2021.10.005.

394

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

Fig. 12 Schematic representation of LiHMDS in Li||NCM811 cells. (A) Reaction mechanism of LiHMDS with HF and H2O. (B–D) Working mechanism of Li|| NCM811 cells. Reprinted with permission Zhang, D.; Liu, M.; Ma, J.; Yang, K.; Chen, Z.; Li, K.; Zhang, C.; Wei, Y.; Zhou, M.; Wang, P.; He, Y.; Lv, W.; Yang, Q.-H.; Kang, F.; He, Y.-B. Lithium Hexamethyldisilazide as Electrolyte Additive for Efficient Cycling of High-Voltage Non-Aqueous Lithium Metal Batteries. Nat. Commun. 2022, 13(1), 6966. https://doi.org/10.1038/s41467-022-34717-4.

Table 1

Disadvantages of some sulfur containing electrolyte additives.

Additives

Disadvantages

1, 3-propane sultone (PS) prop-1-ene-1,3-sultone (PES) 3,2-dioxathiolane-2,2-dioxide (DTD) Ethylene Sulfite (ES) Ethylene Carbonate (EC)

Harms the Environment and degrades on exposure to moisture. SEI resistance is high when subjected to high voltage cycling. Storing problem and copious liberation of gas when formed. Storage is below par, large SEI resistance, unstable on oxidation copious liberation of gas when formed and stored. Inflexible at room temperature

From Tong, B.; Song, Z.; Wan, H.; et al. Sulfur-Containing Compounds as Electrolyte Additives for Lithium-Ion Batteries. InfoMat. 2021; 3(12): 1364–1392. https://doi.org/10.1002/inf2.12235.

Electrolytes can cause exfoliation of graphite electrodes, leading to a chemically unstable SEI, as seen in Propylene Carbonate (PC). It solvates Li ions faster than its other additive peers. The flip side is that PC does not allow Li+ to de-solvate from it, and Li-ion to intercalate into graphite along with PC. This causes the graphite to strip off, causing mechanical degradation of the electrode. This harmful effect of co-intercalation has been controlled by covering the graphite edge, for example, by covalently linking polyethylene oxide (PEO). A study of the Vinyl Carbonate (VC) additive-derived SEI with nano-silicon anodes found a very flexible surface film of SEI with high resistance for Li+ migration making VC unfavorable for high-power applications.

4

Alternatives to conventional additives

The present set of additives in LIBs is unable to satisfy the gold standard of durability and high-speed charging-discharging process. The use of non-aqueous liquid organic electrolytes in LIBs is a matter of concern due to their inflammable nature, restricting their use in big-scale energy storage devices. Effective and thermally stable flame retardants like fluorinated phosphazine and fluoro ester

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

395

additives need to be carefully designed since the position and the number of fluorine substitution affects their performance. Additives with active functional groups that contain flame retarding elements are expected to perform better. The design of synthetic additives allows the formation of a thermodynamically stable interphase by capping the electrolyte with fluorine and silane. This treatment makes it robust enough to withstand volume expansion of Si-embedded anodes during lithiation and at the same time, these synthetic additives optimize ion channels for the fast transport of Lithium ions. The design of synthetic additives follows the path of retro-synthesis. Retro-synthesis is a strategy in which the footprint of organic synthesis is implemented by transforming a target molecule into simpler precursor structures. It installs highly desirable features like spatial flexibility.19 Retro-synthesis of functional synthetic VC derivatives has proved to be challenging. Si/P/B-centered compounds undergo oxidation at the Ni-rich cathodes before electrolyte decomposition, thereby creating a stable CEI. Hence, these molecules are targeted to alleviate the degradation of those Nickel laden cathodes. The electrochemical reversibility of the Nickel loaded cathodes is a concern that can be improved by introducing electron-donating moieties such as phosphite, amine, amino silane, and silyl ether. These electron-donating moieties capture Hydrogen Fluoride, which would otherwise corrode the Cathode cations (from the Transition metal group). This prevents the structural damage of the interphase at both the Cathode and Anode, thus ensuring better performance. Similarly, additives working on dimethyl vinylene carbonate (DMVC) scaffold centers around the use of the Trifluoromethoxy group as a fluorine source generates Lithium Fluoride and the utilization of the Trimethylsilyl group as a Hydrogen Fluoride scavenger. The Solid Electrolyte Interphase is constructed on the Silicon Carbide anode by a reductive copolymerization of Dioxolone derivatives.

5

Future of additives

Additives can modify the behavior of both electrolytes and electrodes, working alone or in tandem with other additives. They can therefore impact the SEI formed and thereby affect the stability of the battery to repeated cycling apart from the energy and power delivered by the device. Some of the important issues that would govern the future discourse on additives are as follows. The behavior of the additive is strongly influenced by the salt and the solvent used in the electrolyte. One, therefore, has to be careful when drawing inferences from theory and trying to connect with actual experiments. Capturing the science of SEI and how trace quantities of additives influence it would need the use of multiscale computational approaches, that cover the full range from the nanoscale to macroscopic scale. With the emergence of machine learning and neural networks, the marriage of quantum chemistry with machine learning will usher in the next revolution in the selection of potential materials and further in the design of new materials with desirable properties.20 The appropriateness of using a half-cell configuration for simulations and experiments, with Li metal (not Graphite) as a negative electrode for investigating the properties of electrolyte additives designed for LIBs needs to be questioned. Dendritic growth of Li(plating) and high reactivity toward electrolyte need to be understood. The interdiffusion of soluble SEI/CEI species to the surface of the counter electrode needs to be understood since this cross-talk between the SEI and the CEI mediated by the electrolyte would be one of the issues driving additive research in the future.

6

Conclusion

Additives stabilize lithium-ion batteries, which are prone to irreversible capacity degradation due to irregular SEI formation, by helping in the formation of an engineered SEI layer at the negative electrode. Additives enhance lithium-ion flux, prevent dendrite formation, improve cyclic stability, increase capacity retention, and aid in the formation of efficient SEI and CEI layers. However, there are a few challenges. Understanding the physical working mechanism of the additives will be a must to develop functional electrolytes with improved effectiveness. The complexity of the SEI layer, its stability, and the sequence of chemical reactions have defied thorough understanding so far. The use of advanced techniques such as liquid and solid-state Nuclear Magnetic Resonance studies would be needed. The SEI film needs to be studied experimentally and characterized, to develop better additives. Robust theoretical calculations coupled with the above characterization may prove to be a game changer in identifying the right reaction pathway.

References 1. Peled, E.; Menkin, S. Review—SEI: Past, Present and Future. J. Electrochem. Soc. 2017, 164 (7), A1703. https://doi.org/10.1149/2.1441707jes. 2. Mosallanejad, B.; Javanbakht, M.; Shariatinia, Z.; Akrami, M. Improvement of Cycle Stability for Graphite-Based Lithium-Ion Batteries Via Usage of Phenyl Methanesulfonate as an Electrolyte Additive. Batteries 2022. https://doi.org/10.3390/batteries8100152. 3. Kasse, R. M.; Geise, N. R.; Ko, J. S.; Nelson Weker, J.; Steinrück, H.-G.; Toney, M. F. Understanding Additive Controlled Lithium Morphology in Lithium Metal Batteries. J. Mater. Chem. A 2020, 8 (33), 16960–16972. https://doi.org/10.1039/D0TA06020H. 4. Liu, D.; Qian, K.; He, Y.-B.; Luo, D.; Li, H.; Wu, M.; Kang, F.; Li, B. Positive Film-Forming Effect of Fluoroethylene Carbonate (FEC) on High-Voltage Cycling with Three-Electrode LiCoO2/Graphite Pouch Cell. Electrochim. Acta 2018, 269, 378–387. https://doi.org/10.1016/j.electacta.2018.02.151.

396

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Electrolyte Additives

5. Xie, Z.; Wu, Z.; An, X.; Yue, X.; Yoshida, A.; Du, X.; Hao, X.; Abudula, A.; Guan, G. 2-Fluoropyridine: A Novel Electrolyte Additive for Lithium Metal Batteries with High Areal Capacity As Well as High Cycling Stability. Chem. Eng. J. 2020, 393, 124789. https://doi.org/10.1016/j.cej.2020.124789. 6. Han, J. G.; Jeong, M. Y.; Kim, K.; Park, C.; Sung, C. H.; Bak, D. W.; Kim, K. H.; Jeong, K. M.; Choi, N. S. An Electrolyte Additive Capable of Scavenging HF and PF5 Enables Fast Charging of Lithium-Ion Batteries in LiPF6-Based Electrolytes. J. Power Sources 2020, 446 (November 2019), 227366. https://doi.org/10.1016/j.jpowsour.2019.227366. 7. Wang, K.; Xing, L.; Zhi, H.; Cai, Y.; Yan, Z.; Cai, D.; Zhou, H.; Li, W. High Stability Graphite/Electrolyte Interface Created by a Novel Electrolyte Additive: A Theoretical and Experimental Study. Electrochim. Acta 2018, 262, 226–232. https://doi.org/10.1016/j.electacta.2018.01.018. 8. van Ree, T. Electrolyte Additives for Improved Lithium-Ion Battery Performance and Overcharge Protection. Curr. Opin. Electrochem. 2020, 21, 22–30. https://doi.org/10.1016/ j.coelec.2020.01.001. 9. Spotte-Smith, E. W. C.; Petrocelli, T. B.; Patel, H. D.; Blau, S. M.; Persson, K. A. Elementary Decomposition Mechanisms of Lithium Hexafluorophosphate in Battery Electrolytes and Interphases. ACS Energy Lett. 2023, 8 (1), 347–355. https://doi.org/10.1021/acsenergylett.2c02351. 10. Sun, Z.; Wen, Z.; Chen, Y.; Ma, Y.; Zhang, J.; Li, Y.; Li, L.; Chen, R. A Gradient Solid Electrolyte Interphase with High Li+ Conductivity Induced by Bisfluoroacetamide Additive for Stable Lithium Metal Batteries. Nano Res. 2023. https://doi.org/10.1007/s12274-022-5363-6. 11. Zuo, X.; Zhao, M.; Ma, X.; Xiao, X.; Liu, J.; Nan, J. Effect of Diphenyl Disulfide as an Additive on the Electrochemical Performance of Li1.2Mn0.54Ni0.13Co0.13O2/Graphite Batteries at Elevated Temperature. Electrochim. Acta 2017, 245, 705–714. https://doi.org/10.1016/j.electacta.2017.05.155. 12. Kazzazi, A.; Bresser, D.; Kuenzel, M.; Hekmatfar, M.; Schnaidt, J.; Jusys, Z.; Diemant, T.; Behm, R. J.; Copley, M.; Maranski, K.; Cookson, J.; de Meatza, I.; Axmann, P.; Wohlfahrt-Mehrens, M.; Passerini, S. Synergistic Electrolyte Additives for Enhancing the Performance of High-Voltage Lithium-Ion Cathodes in Half-Cells and Full-Cells. J. Power Sources 2021, 482 (June 2020), 228975. https://doi.org/10.1016/j.jpowsour.2020.228975. 13. Jang, S. H.; Jung, K.; Yim, T. Silyl-Group Functionalized Organic Additive for High Voltage Ni-Rich Cathode Material. Curr. Appl. Phys. 2018, 18 (11), 1345–1351. https://doi.org/ 10.1016/j.cap.2018.07.016. 14. Liu, L.; Gao, W.; Cui, Y.; Chen, S. A Bifunctional Additive bi(4-Flurorophenyl) Sulfone for Enhancing the Stability and Safety of Nickel-Rich Cathode Based Cells. J. Alloys Compd. 2020, 820, 153069. https://doi.org/10.1016/j.jallcom.2019.153069. 15. Su, D. Powerful Qua-Functional Electrolyte Additive for Lithium Metal Batteries. Green Energy Environ. 2022, 7 (3), 361–364. https://doi.org/10.1016/j.gee.2021.10.005. 16. Zhang, D.; Liu, M.; Ma, J.; Yang, K.; Chen, Z.; Li, K.; Zhang, C.; Wei, Y.; Zhou, M.; Wang, P.; He, Y.; Lv, W.; Yang, Q.-H.; Kang, F.; He, Y.-B. Lithium Hexamethyldisilazide as Electrolyte Additive for Efficient Cycling of High-Voltage Non-Aqueous Lithium Metal Batteries. Nat. Commun. 2022, 13 (1), 6966. https://doi.org/10.1038/s41467-022-34717-4. 17. Wang, A.; Kadam, S.; Li, H.; Shi, S.; Qi, Y. Review on Modeling of the Anode Solid Electrolyte Interphase (SEI) for Lithium-Ion Batteries. npj Comput. Mater. 2018, 4 (1), 15. https://doi.org/10.1038/s41524-018-0064-0. 18. Tong, B.; Song, Z.; Wan, H.; Feng, W.; Armand, M.; Liu, J.; Zhang, H.; Zhou, Z. Sulfur-Containing Compounds as Electrolyte Additives for Lithium-Ion Batteries. InfoMat 2021, 3 (12), 1364–1392. https://doi.org/10.1002/inf2.12235. 19. Park, S.; Jeong, S. Y.; Lee, T. K.; Park, M. W.; Lim, H. Y.; Sung, J.; Cho, J.; Kwak, S. K.; Hong, S. Y.; Choi, N.-S. Replacing Conventional Battery Electrolyte Additives with Dioxolone Derivatives for High-Energy-Density Lithium-Ion Batteries. Nat. Commun. 2021, 12 (1), 838. https://doi.org/10.1038/s41467-021-21106-6. 20. Xu, N.; Shi, J.; Liu, G.; Yang, X.; Zheng, J.; Zhang, Z.; Yang, Y. Research Progress of Fluorine-Containing Electrolyte Additives for Lithium Ion Batteries. J. Power Sources Adv. 2021, 7 (November 2020), 100043. https://doi.org/10.1016/j.powera.2020.100043.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Separators Takashi Ikemoto, Asahi-Kasei Corporation, Tokyo, Japan © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of H. Tanji, SECONDARY BATTERIES – LITHIUM RECHARGEABLE SYSTEMS - LITHIUM-ION | Separators, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 356–367, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00209-4.

1 2 3 4 5 6 7 8 9 10 10.1 10.2 10.3 10.4 10.5 11 12 12.1 12.2 12.3 13 14 14.1 14.2 14.3 14.4 14.5 14.6 15 15.1 15.2 15.3 16 17 18 19 20 References Further reading

Introduction Application of microporous membranes to battery separators Functions and requirements of separators for lithium-ion secondary batteries Chemical stability Mechanical property Thinness of the membrane Current shutdown capability Electrolyte holding capacity Others Technology and process of manufacturing microporous membrane Materials and their composition Polyethylene monolayer membrane Polypropylene monolayer membrane Polyethylene–polypropylene multilayer membrane Polyethylene–polypropylene microphase separation membrane Coating separators Manufacturing process and technology of microporous membrane Wet process (microphase separation method) Dry process (drawing method) Coating process Cell performance and separator design Characterization of properties of separator Separator thickness Micropore size Porosity Morphology of microporous structure Mechanical properties Thermal properties Separator design based on cell requirements Cell assembling processability Cell performance (charge and discharge performance) Safety Subject for a further study Higher strength and lower thickness Ion permeability Thermal properties (shutdown properties, heat shrinkage) Conclusion

398 399 399 400 401 401 401 401 402 402 402 402 402 404 404 404 404 404 406 406 407 407 407 407 407 409 409 409 410 410 410 410 410 410 410 411 411 411 411

Abstract Polyethylene-based microporous membrane separators and coated separators are a key device for lithium-ion secondary batteries (LIBs). In addition to their general functions and requirements, that is, separation and insulation of each electrode and holding of electrolyte, they are characterized by their thinness, chemical stability, and shutdown properties, which are required particularly for higher cell capacity and safety characteristics of LIBs. There are two major methods for manufacturing microporous membranes: wet and dry process. In addition, separators with functional layers coated on polyolefin microporous membranes are widely used. The characteristics of microporous membranes are defined by geometrical parameters, microporous structural properties, mechanical properties, and thermal properties.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00268-1

397

398

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Separators

Key points

• • • • • • • •

The separator, a component of the LIB, is an important material for battery safety. The separator is mainly made of a microporous polyolefin membrane. Recently, coated separators with fillers and polymers are widely used to add functions such as heat resistance. The basic functions required of separators are electrical insulation and ion conductivity. In addition, chemical stability against electrolyte and electrochemical stability during battery charging and discharging are also required. Wet process and dry process are the main production methods for microporous membranes used in separators. An important function of separators is the shutdown property, in which the separator melts at high temperatures. This is a function in which the melting of the polymer closes the micropores and shuts off the flow of electric current. The performance of a battery is highly dependent on the cell design and other components, so it is very difficult to satisfy all requirements by separator design alone. Coated separators have been adopted as one means of resolving the trade-off characteristics.

Abbreviations ABS ASTM BET DIPS DMC DSC EC HDPE JIS LDPE LIB LIP PAN PE PET PIPS PMMA POE POM PP PS PTFE PVC PVDF SEM TEM TIPS TMA UHPE

1

Acrylonitrile butadiene styrene American Society for Testing and Materials Brunauer–Emmett–Teller Diffusion-induced phase separation Dimethyl carbonate Differential scanning calorimetry Ethylene carbonate High-density polyethylene Japanese Industrial Standards Low-density polyethylene Lithium-ion secondary battery Lithium-ion polymer battery Poly(acrylonitrile) Polyethylene Poly(ethylene terephthalate) Pressure-induced phase separation Poly(methyl methacrylate) Poly(oxyethylene) Poly(oxymethylene) Polypropylene Polystyrene Polytetrafluoroethylene Polyvinyl chloride Polyvinylidine difluoride Scanning electron microscopy Transmission electron microscopy Thermally induced phase separation Thermal mechanical analysis Ultrahigh molecular weight polyethylene

Introduction

Lithium-ion secondary batteries (LIBs) now play a very significant role as power sources in various portable IT equipment such as laptop computers and smartphone, in the mobility field such as electric vehicles, and in power storage systems. These batteries are characterized by the fact that separators are the key material particularly for ensuring safety.1 In this article, the fundamental aspects of polyethylene(PE)-based microporous membrane separators and coated separators in LIBs and lithium-ion polymer batteries (LIPs), that is, the choice of raw materials, battery performance characteristics, and manufacturing technology, are described.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Separators

2

399

Application of microporous membranes to battery separators

Separator materials for various batteries and particularly component materials for major portable secondary batteries are listed in Tables 1 and 2, respectively. Materials for separators are selected based on the requirements and characteristics of batteries. Polyethylene microporous membranes were originally used as separators for lead–acid batteries; then, by making the full use of their characteristics, their application was extended to lithium-based batteries. The market of microporous membrane separators has made a marked growth since LIBs were commercialized and they rapidly took a major place in powering portable electronic devices. As LIBs use electrolytes composed of nonaqueous organic solution, separator materials with specific properties such as solvent resistance, thinness, and current shutdown properties at a certain temperature range are required. The PE microporous membrane is the preferred choice that can meet these requirements.2 Recently, coated separators are commonly used to increase the heat resistance of these PE microporous membranes for LIBs.

3

Functions and requirements of separators for lithium-ion secondary batteries

General functions and requirements of separators are summarized in Table 3. Fundamental and common functions required of all battery separators are as follows: 1) separation and electrical insulation between positive and negative electrodes, that is, preventing contact of active materials of each electrode (i.e., short-circuit prevention) and. 2) holding of electrolyte, that is, forming passages for ion transfer (ion conductivity). Table 1

Separators for various batteries.

Material

Secondary battery

Form

Raw material

LIB

LIP

Microporous membrane

Polyethylene Polypropylene Others Polypropylene Polyamide Others Polyethylene Regenerated cellulose Polyethylene Craft paper





Nonwoven fabric Film Paper

⃝

Ni–MH

Ni–Cd

Primary battery Lead–acid

Lithium





Manganese

Alkaline manganese

⃝



⃝ 



Silver oxide















LIB, lithium-ion secondary battery; LIP, lithium-ion polymer battery.  Mainly used; ⃝ possibly used.

Table 2

Component materials of portable secondary batteries.

Batteries

Active materials

Electrolyte

Positive electrode

Negative electrode

Nickel–cadmium

NiOOH

Cd

Nickel–metal hydride LIB (liquid electrolyte)

NiOOH

MH

LiCoO2, etc.

Carbon

LIB (gel electrolyte)

LiCoO2, etc.

Carbon

Separator Materials

KOH aqueous Polyamide nonwoven fabric solution KOH aqueous Hydrophilically augmented polypropylene solution nonwoven fabric Lithium salt Polyethylene microporous membrane nonaqueous solution (EC, DMC) Gel sheet (PVDF–POE–acrylate–PAN) (polymer + electrolyte solution + microporous membrane)

Requirements Electrolyte retainability Prevention of self-discharge Organic solvent durability Shutdown property Thinner membrane Same as above + electrolyte retainability

LIB, lithium-ion secondary battery; EC, ethylene carbonate; DMC, dimethyl carbonate; PVDF, polyvinylidine difluoride; PoE, poly(oxyethylene); PAN, poly(acrylo)nitrile.

400

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Separators

Table 3

Functions and requirements of separators.

Functions and performance

Objectives and functions

Requirements

Basic functions and performance

Separation of cathode and anode (preventing the short circuit) Holding electrolyte (forming passage for ion transfer)

Sheet or plate material

Intrinsic properties for LIB and LIP

Durability in the cell

High capacity

Mechanical stability (mechanical strength, mechanical durability) Mechanical strength (thin membrane, assembling processability, prevention of short circuit) Affinity to electrolyte (shortening electrolyte immersing time) Thinner membrane (large electrode surface area compensating low ion conductivity)

Charge–discharge Overcharge High temperature

Capacity for holding electrolyte (micro-porosity, pore size, pore structure) Shutdown (low shutdown temperature, high meltdown temperature) Durability for membrane rupture at high temperature

Productivity and cost of cell assembling Cell performance Cell safety

Electron nonconductivity Microporous material Ion conductivity (capacity of holding electrolyte) Chemical stability (electrochemical stability, organic solvent durability)

LIB, lithium-ion secondary battery; LIP, lithium-ion polymer battery.

Characteristics of LIB

Characteristics of cell structure

Requirements for separator

High voltage High energy density

Existence of lithium ion

High safety level

Nonaqueous electrolyte

Solvent durability

Low ion transfer

High porosity

Large electrode surface area

Thin and dense jelly roll

Thin membrane Appropriate mechanical properties

Fig. 1 Characteristics of separator for lithium-ion batteries (LIBs).

In conclusion, essential performance-related features required of separator materials are (1) electrical insulation, (2) microporous structure, and (3) affinity to electrolyte. In addition, there are some specific requirements of separators for LIBs as shown in Fig. 1. These requirements come from the fact that the LIB contains an active lithium compound as an electrode and a flammable organic solvent as an electrolyte solution. Moreover, the LIB has considerably high energy density, so requirements concerning safety are also very important. Details of these requirements are described in the following.

4

Chemical stability

Separator materials should be chemically stable to nonaqueous organic solvent in an electrolyte solution and should be electrochemically inert during charge–discharge reaction in a battery. Moreover, they should be unreactive to the products of side reactions such as decomposition of other components.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Separators

Can

Negative electrode

401

Positive electrode

Electrolyte

Separator

Fig. 2 Schematic of a cylindrical lithium-ion battery.

5

Mechanical property

The schematic structure of a cylindrical LIB is shown in Fig. 2. In the cell assembling process, a separator is inserted between the positive and negative electrodes and is wound together to make a jelly roll. Then, the jelly roll is placed into a cell case made of materials such as metallic cans and laminated film packs. In this process, various types of mechanical and internal stresses are applied to the separator; therefore, it should have enough strength and elastic modulus to endure various stresses generated in the process. Recent increasing needs for higher cell capacity also require separators to be thinner without compromising high strength.

6

Thinness of the membrane

The simplest way to increase the capacity of a cell is by increasing the quantity of active materials contained in the cell. Moreover, a separator itself does not contribute positively to electrochemical charge and discharge reaction. As a result, the volume and the space allowed for the separator are preferably minimized. In addition, in the case of LIBs, electrolytes containing a nonaqueous organic solvent have electroconductivity two-digit order lower than those containing an aqueous solution. Consequently, separators should be much thinner to provide electrodes enough surface area. Such thinner separators are also required to achieve higher energy density. However, to prevent the mechanical and electrical breakdown of the separators, the thickness should be more than adequate. In general, the thickness should range between 10 and 30 mm for cylindrical-type and prismatic-type LIBs and around 10 mm for laminated-type LIPs.

7

Current shutdown capability

Separators are one of the key safety devices of lithium-based batteries. In the case of overcharge, for example, a temperature rise in the battery causes thermal runaway, which may result in an unsafe mode. To prevent such a critical state, the separator is designed to cut current flow by closing the micropores when the temperature reaches a particular value. In addition, the separator should maintain its dimension and strength to sustain current interruption even though the temperature exceeds the critical temperature. A typical example of a shutdown curve is illustrated in Fig. 3.

8

Electrolyte holding capacity

To achieve higher performance of the battery, a sufficient amount of electrolyte should be retained in the microporous structure of the separator, in order to avoid electrolyte exhaustion caused by repeated inflation and deflation of the electrode during cyclic charge and discharge.

402

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Separators Shutdown mechanism

By heating a model cell at constant heating rate, pores in the separator close at a certain temperature range around polymer material melting point, resulting in an abrupt increase in impedance (shutdown). On further heating, rupture, shrinkage, and/or fusion of separator occur and impedance drops again (meltdown) 6

Key issue: Pores close completely in order to maintain electrical separation between electrodes

log impedance (: cm2)

5

4

3

2

1

70

90

110

130

150

170

Temperature (qC)

Fig. 3 Schematic of a shutdown curve and shutdown mechanism.

Therefore, separator material and its microporous structure are designed to have enough porosity and certain affinity to electrolyte solution.

9

Others

Dimensional parameters such as thickness and width should be uniform and consistent to obtain satisfactory processability in the cell assembling process.

10

Technology and process of manufacturing microporous membrane

10.1 Materials and their composition Fundamental properties of various synthetic resins are summarized in Table 4. Among them, PE has the most balanced performances. Particularly, organic solvent durability and low melting point are important for LIB applications.

10.2 Polyethylene monolayer membrane Polyethylene monolayer membranes are most preferably used as separators in LIBs. High molecular weight (viscosity average molecular weight of 0.2–0.3 million) high-density PE (HDPE), ultrahigh molecular weight PE (UHPE) (viscosity average molecular weight of 1–2 million), and a blended mixture of them are generally used. In some cases, low-density PE (LDPE) and copolymerized PE are added to adjust the shutdown performance of separators.

10.3 Polypropylene monolayer membrane In terms of shutdown properties, the melting point of polypropylene (PP) is so high that it cannot be used as a single agent for LIB separator. Therefore, the use of PP monolayer membranes for LIB separators is restricted to certain types of manganese oxide–lithium primary batteries.

Table 4

Properties of synthetic resins.

Property

Izod notch (kg cm cm−2)

Strong acidic

LDPE

HDPE

UHPE

0.091–0.925 105–110

0.941–0.965 135 < − 110 220–390 20–1000 10–85 2.3 Lix TiS2

(IV)

as illustrated in Fig. 3. Insertion into the space between strongly bonded layers is commonly referred to as intercalation. With lithium metal as the anode and the LixTiS2 solid-solution range 0 < x < 1, the TiS2/LiClO4/lithium cell was shown to have a capacity of one lithium per titanium disulfide molecule and to have an open-circuit voltage described by the Nernst equation: V OC ¼ f2:2 + ln½ð1 − x Þ=x gV:

(3)

Lithium perchlorate was the salt introduced into the dry organic electrolyte. However, a passivation layer on the lithium anode resulted in dendrite formation at the anode on repeated recharging of the cell; growth of dendrites across the separator of the cell gave an explosive short circuit. If another insertion compound of lower EFA was to be used as the anode, the VOC would be reduced and the battery would not be competitive with an MH/KOH/NiOOH secondary battery. Because the Ti4+/Ti3+ redox couple lies at the top of the S–3p bands, a significant increase in the voltage of a sulfide is not an option; the sulfides cannot take advantage of the larger window offered by the nonaqueous electrolytes. On the contrary, higher oxidation states can be realized in first-row transition metal oxides. Therefore, oxide-insertion compounds are used as cathodes in rechargeable lithium batteries.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic Electrolyte Cells

441

Fig. 3 Schematic of the Li/LiClO4/TiS2 cell during discharge.

6

Layered LiMO2 cathodes

Layered MO2 compounds analogous to titanium disulfide are not found; layered oxides are stable only where vanadyl (V ¼ O)3+ or molybdyl (Mo ¼ O)4+ ions are formed. On the contrary, LiMO2 oxides having a cation radius ratio RM/RLi < 0.83 may have Li+ and M3+ ions ordered into alternate (111) octahedral-site planes of a face-centered cubic array of oxide ions (Fig. 4). This structure, like that of lithium titanium disulfide (LiTiS2), has strongly bound layers of edge-shared (MO2)− octahedra that are bonded to one another by Li+ ions in the intervening octahedral-site planes. From a structural viewpoint, extraction of lithium from between the layers should proceed as in the case of lithium titanium disulfide as long as the Li+ and M3+ ions of the parent phase are well ordered; however, the strong bonding within the MO2 layers is too ionic to allow weaker van der Waals forces to stabilize a layered MO2 phase. On removal of more than a critical fraction of lithium from ordered LiMO2, the Li1–xMO2 phase becomes metastable; electrostatic forces would drive M cations into the lithium planes. However, the M cations must pass through a tetrahedral site to reach the lithium planes, and if they have a strong octahedral-site preference, metastable MO2 layers may be retained on removal of nearly all the lithium. The requirement of a strong octahedral-site preference as well as the need to have the M4+/M3+ redox couple at a low enough energy to give a high VOC versus a lithium anode limits the eligible M3+ cations for a LiMO2 cathode. The lightest transition metal atoms M belong to the 3d block, and the M4+/M3+ redox couples of this block having a sufficiently low energy are M ¼ Cr, Mn, Fe, Co, and Ni. The Fe3+ ion does not have a strong octahedral-site preference and two Mn3+ ions readily disproportionate into Mn2+ and Mn4+ to yield a Mn2+ ion with no octahedral-site preference. Moreover, Cr4+ ions may undergo the disproportionation reaction. 3 Cr4+ ! Cr6+ + 2 Cr3+ 6+

+

(V) 3+

in which the Cr ion occupies a tetrahedral site to block Li -ion diffusion. Therefore, the only eligible 3d-block M ions are Co3+ and Ni3+. Although the cation ratio RNi/RLi ratio is marginally small enough for ordering of the Li+ and Ni3+ ions, partial reduction of Ni3+ to Ni2+ in a high-temperature synthesis of lithium nickel oxide (LiNiO2) increases this ratio and stabilizes Ni2+ ions in the Li+-ion planes. It is therefore difficult to synthesize well-ordered lithium nickel oxide. However, the small size and strong octahedral-site preference of low-spin Co3+ and Co4+ make straightforward fabrication of well-ordered lithium cobalt oxide (LiCoO2). Unfortunately, Li1–xCoO2 is only stable over the solid-solution range 0  x  0.5, but it gives a VOC  4.0 V versus a lithium anode.4 The traditional battery manufacturers could not imagine fabricating a battery with a discharged cathode. However, electrical engineers of the SONY Corporation of Japan recognized that they could extract lithium from lithium cobalt oxide and insert it into a carbon anode, the lithium intercalating into the graphite sheets to form LiC6. Because the LiCoO2/Li+/C cell is chargeable to Li1–xCoO2/Li+/LixC6, starting with a discharged cathode allows charging the carbon anode. The charged ‘rocking-chair’ battery had a high enough energy density to permit the company to manufacture the cell telephone and launch the wireless revolution. The capacity of the Li1–xCoO2 cathode is limited to 0.5 lithium per cobalt because the Co4+/Co3+ redox couple is pinned at the top of the O-2p bands. On removal of more than 0.5 lithium per cobalt, the O2− ions at the surface become oxidized to peroxide (O2)2− ions rather than additional Co3+ to Co4+, and oxygen is evolved as oxygen in the reaction.

442

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic Electrolyte Cells

Fig. 4 Structure of layered LiCoO2.

2 ðO2 Þ2− > 2O2− + O2 "

(VI)

Gas evolution introduces safety problems that are not easily controlled in large batteries containing multiple cells. However, for single-cell batteries used in electronic portable devices, they have been proven to be safe. The high price and toxicity of cobalt has motivated searches for alternative LiMO2 cathodes, and partial substitution of nickel for cobalt has given Li1–xCo0.2Ni0.8O2 with a capacity of 0 < x < 0.6 and only a slightly reduced VOC.5 The smaller size of the low-spin Co3+ ion keeps the mean hRMi/RLi ratio small enough with only 0.2 cobalt per formula unit to allow the synthesis of well-ordered M3+ and Li+ ions. Good ordering of the cations is essential for fast Li+-ion diffusion in the oxide, that is, for a high power output and a fast recharge. Alternatively, Ni2+ is stable in the presence of Mn4+, and both of these ions have a strong octahedral-site preference. The Mn4+ ion is not further oxidized on charging and there is no voltage step on passing from the Ni3+/Ni2+ to the Ni4+/Ni3+ redox couple because both couples are pinned at the top of the O-2p bands. However, the hRMi/RLi ratio of LiNi0.5Mn0.5O2 is only marginally small enough to allow synthesis of well-ordered materials. Nevertheless, a high degree of order has been achieved by a low-temperature preparation of LiNi0.5+dMn0.5–dO2 or by synthesizing NaNi0.5Mn0.5O2 and subsequently exchanging lithium for sodium. These cathodic oxides contain no cobalt, and charge/discharge rates of 5C (12 min) have been achieved in laboratory half-cells (lithium anode) with well-ordered oxides. Some reduction in capacity occurs at fast charge/discharge rates, but full capacity is recovered on lowering the rate of charge/discharge. A simpler synthetic route to a well-ordered LiNi0.5Mn0.5O2 compound is made possible by adding a small amount of the low-spin Co3+ ion in LiN0.5–dMn0.5–dCo2dO2 with 0 < d  1/6. Nevertheless, the problem of either oxygen evolution from the surface or oxidation of the electrolyte remains and limit the capacity to less than 0.6 lithium per formula unit. In order to increase the capacity and safety by suppression of oxygen evolution from the surface and any reaction with the electrolyte, the oxide particles have been coated with a main-group oxide so as to reduce contact of the transition metal oxide with the electrolyte, but without limiting Li+-ion migration between the electrolyte and the oxide. A capacity as high as 280 mAhg−1 (about 0.9 lithium per formula unit) has been obtained with LiNi0.5–dMn0.5–dCo2dO2 coated with aluminum oxide (Al2O3).6–10 In summary, the layered LiMO2 cathode materials offer the best volume as well as specific energy density with a large capacity. Therefore, they promise to continue to power the wireless electronics revolution. However, they probably will not be competitive

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic Electrolyte Cells

443

with three-dimensional (3D) framework oxide hosts for high-power applications where multiple-cell batteries are required and internal temperatures may disorder these metastable hosts. The 3D framework oxide hosts have been shown to be capable of powering power tools and small or hybrid electric vehicles.

7 7.1

Three-dimensional framework oxide hosts Spinel framework

The A[M2]O4 cubic spinel structure of Fig. 5 contains tetrahedral-site A cations on the 8a sites and octahedral-site M cations on the 16d sites; the oxygen atoms form a face-centered cubic array. The [M2]O4 array forms a strongly bonded 3D framework in which the 8a tetrahedral sites and 16c octahedral sites form a 3D interconnected interstitial space. Insertion of additional lithium into the spinel Li[Mn2]O4 displaces all the lithium into the 16c sites, and Li1+z[Mn2]O4 has a VOC ¼ 3.0 V versus a lithium anode. Removal of lithium from the 8a sites in Li1–y[Mn2]O4 gives a VOC  4.0 V. In this structure, the disproportionation reaction 2Mn3+ ! Mn2++Mn4+ is confined to the surface of the oxide; the bulk framework is thermodynamically stable for the entire solid-solution range 0 < x < 2 of Lix[Mn2]O4, but a cooperative orbital ordering at the Mn3+ ions at room temperature for x > 1 induces a distortion of the structure to tetragonal symmetry and a separation into lithium-rich and lithium-poor phases at room temperature in the interval 1 < x < 1.8. The 1 V step at x ¼ 1 means that the practical capacity of the spinel framework is limited to 0.5 lithium per manganese atom. The VOC  4.0 V of Li1–y[Mn2]O4 is of greater interest, but problems with manganese dissolution to the electrolyte on cycling have forced modification of the composition. Substitution of 0.1 lithium and 0.1 nickel for manganese per formula unit has been shown to reduce strongly manganese dissolution, but at the expense of capacity. Up to 0.2 fluorine atoms per formula unit can be substituted for oxygen to improve the capacity, and in a laboratory half-cell (lithium anode) Li[Mn1.8Li0.1Ni0.1]O3.8F0.2 has given stable cycling with a specific capacity of 110mAhg−1 at a C/10 (10 h) discharge rate or 100mAhg−1 at 4
C (15 min) rate. Although the voltage is attractive, the specific capacity is disappointingly low.

7.2

Olivine framework

The olivine structure of LiMPO4, shown in Fig. 6, has a 3D MPO4 framework that is stable at high temperatures. Strong covalent bonding in the (PO4)3− polyanion lowers the energies of the M3+/M2+ redox couples, and Li1–xFePO4 gives a VOC ¼ 3.45 V versus a lithium anode, which is suitable for a polymer electrolyte. A subtle structural change between lithium iron phosphate (LiFePO4) and iron phosphate (FePO4) provides a flat voltage versus the state of charge in accordance with the Gibbs phase rule. However, the structure confines the Li+ ions to move in 1D channels, which imposes two constraints: (1) the lithium channels must not be blocked by either disorder of the lithium and iron atoms or by the presence of a foreign phase; (2) the cathode particle must be small without stacking faults that block the channels. The small lithium iron phosphate particles are platelets with the lithium

Fig. 5 Two quadrants of cubic spinel structure.

444

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic Electrolyte Cells

Fig. 6 Structure of LiFePO4.

Fig. 7 Cell voltage vs. specific capacity of a Li/LiPF6/FePO4 cell at different discharge rates after charging at C/5 rate. The solid lines represent carbon-coated LiFePO4 attached to polypyrrole and the dotted lines represent carbon-coated LiFePO4 with carbon and Teflon additives for electronic conduction collector and Teflon as binder of the cathode mass.

channels parallel to the short dimension; this morphology allows the Li+ ions to move cooperatively within the channels in a successive wave to give a moving phase boundary between the two phases. Therefore, despite the 1D channel for Li+-ion motion and a phase separation that limits the electronic conductivity, 10C (6 min) charge and discharge rates can be achieved with lithium iron phosphate cathodes. The specific capacity approaches 160 mAhg−1 for thousands of cycles at a C/5 discharge rate, and the reversible capacity loss at higher discharge rates is small (see Fig. 7). Rechargeable multicell batteries with lithium iron phosphate cathodes have been proved to be safe and to be capable of powering cordless electric tools and small electric vehicles.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Organic Electrolyte Cells

445

The olivine lithium manganese phosphate (LiMnPO4) offers an open-circuit voltage of 4.0 V versus a lithium anode, but the preparation of small particles that allow extraction of lithium has proved to be a challenge.

7.3

Increasing Capacity

In order to increase capacity to more than one Li+ ion per transition metal cation in a framework host, investigations are being made of structures that can accommodate the required number of Li+ ions and also a transition metal atom having two redox energies of nearly the same energy. Such frameworks appear to contain polyanions as illustrated by Li2MSiO4 and Li3M2(PO4)3, where M is a transition metal ion; however, there are few suitable M cations.

8

Outlook

In consumer electronic applications, LCO has been widely adopted because of its reliability and safety since the 90s. LCO has been used in smartphones, laptop and notebook PCs and other portable devices for over 30 years since the first industrial production. Later, a ternary NidCodMn system (NCM) has been developed and applied particularly to xEV. At present, LFP has been adopted for specific uses such as low-cost EV applications and stationary use. All of these cathode active materials were originally proposed in the early ‘80s to late ‘90s. We are reminded of Professor Goodenough’s great achievements and his foresight through his article.

•

Layered LiMO2 Cathodes

At present, the stability of LCO has improved by heteroatom doping and surface treatment. This technical advance allows a high charge operation potential above 4.4 V and a specific capacity greater than 170 mAg−1. The doping and surface coating techniques prevent dissolution of Co4+, especially due to the transition from monoclinic to hexagonal phase when x > 0.5. The ternary transition metal oxide, LiNixCoyMnzO2 (x + y + z ¼ 1, so called NCM) or LiNi1-x-yCoxAlyO2 (NCA) have been mainly adopted in e.g., electric power tools, and xEV applications. Development has now shifted from NCM111 to NCM811 or higher nickel content to improve capacity. However, safety risks increase with the nickel content as noted by Prof. Goodenough. The thermal run away (oxygen release) temperature decreases with the nickel content. Also, the development of a manganese rich phase is of increasing interest on account of material costs in 2022–24.

•

Spinel Framework

The spinel manganese LiMn2O4 is well known and is used in LIB. Also, other manganese spinels, Li4Mn5O12 (lithium rich spinel) and Li2Mn4O9 (defect-spinel) have attracted attention since they are more stable than LiMn2O4. Remarkably, when they co-exist in a solid solution with Li2MnO3 (lithium rich layered rock salt) these spinels increase stability. To increase the specific capacity of cathodes, a promising approach is to use Li2MnO3 based solid solutions to stabilize the framework structure. The biggest operational issue in the use of manganese based active materials has been dissolution of the manganese ions into the electrolyte, especially at high temperatures. Whereas for lithium rich cathodes, O2 evolution and/or formation of Li2O has been problematic. However, the stability of these cathode materials has been increasing year by year. Also, recently, high voltage spinel metal oxides have attracted attention, such as the LiNi0.5Mn1.5O4 (5 V system). However, organic electrolytes tend to oxidize above 4.2–4.3 V. The development of stable electrolytes is required for the general applicability of 5 V- spinel system.

•

Olivine Framework

Nowadays, Lithium Iron Phosphate (LFP) is been widely used in low budget EVs, stationary energy storage systems, and portable applications due to its low cost and safety. LiMnxFe(1-x)PO4 has attracted increasing interest since mid-2020 for such applications. These materials were suggested by Prof. Goodenough’s team in 1996. Also, it should be mentioned that the successful commercialization of these materials required great effort for the stabilization of the active material frameworks, purification, and various additives in the background of the development of LIBs. Also, the reduction of cobalt and nickel has been studied in light of increasing mineral cost due to increased competition in 2022–23. The high voltage spinel, LiNi0.5Mn1.5O4 has been of increasing interest, but a new generation of highly stable electrolytes may be needed to enable the transition to high potential systems. Because natural resources are limited, the proportion of xEV batteries being recycled must be increased in the next decades. Although material development has been essential in the development of batteries, it will also be important to develop batteries that are easy to disassemble and separate to realize a sustainable society.

References 1. Harris, W. S. Electrochemical Studies in Cyclic Esters; Thesis, University of California: Berkeley, CA, 1958. 2. Mizushima, K.; Jones, P. C.; Wiseman, P. J.; Goodenough, J. B. LixCoO2 (050% of the lithium stored in the positive electrode is cycled. It has been shown that up to 70% of the lithium could be deintercalated out of the lithium cobalt dioxide in an inorganic system using purified electrodes.

Fig. 5 Schematic of a lithium inorganic cell.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Inorganic Electrolyte Cells 7.2

455

Purification of the positive electrode

It is assumed that freshly manufactured lithium cobalt oxide (LiCoO2) is highly hygroscopic and reacts with water described in reaction [VI]. Therefore, lithium cobalt dioxide, in its delivered state, has a thin layer of lithium hydroxide (LiOH) on the surface: LiCoO2+x H2O !x LiOH + Li1−xHx CoO2

[VI]

This process results in lithium ions from the lithium cobalt oxide being replaced by protons. For a material with a mean grain size of 5 mm, the proportion of lithium ions in the lithium cobalt lattice replaced by protons is 6 mol%. A purification method was developed to remove lithium hydroxide from the electrode.25 The process involves a Lewis acid that reacts with the lithium hydroxide. Following the removal of the reaction products of the process, the electrodes are ready for operation in the cell. The purification leads to a much better cycle stability and a higher capacity. While cycling positive electrodes with a capacity of 41 mAh cm−2 and a thickness of 0.6 mm in a potential range 3.5–4.5 V, the measured capacity without purification was 58% of the theoretical capacity. If the electrodes are purified, the capacity increases to 71% of the theoretical capacity, because all hydrogen ions are replaced with lithium ions. Figs. 6A and B compare 40 cycles of cyclic voltammograms for untreated (A) and purified (B) LiCoO2 electrodes. The cyclic voltammetry experiments comprise a three-electrode system (working electrode: LiCoO2; counter electrode: Li; and reference electrode: Li). The voltammograms were recorded across the potential range from 3.5 to 4.5 V with a sweep rate of 0.2 mV s−1. During cycling of the untreated electrode, the peak potential increases from 3.95 to 4.20 V. This means that the internal resistance of the cell is increasing. Fig. 6B shows the cyclic voltammograms for a cell with purified electrode. There is no drift of the peak voltage and no change in the internal resistance. Within an inorganic system using purified electrodes, it has been shown that up to 70% of the lithium was deintercalated out of the lithium cobalt dioxide, without any negative impact. These measurements show that, in spite of the higher charging level, the LiCoO2 lattice is not destroyed. The capacity over many cycles remains constant. Fig. 7 shows the capacity development over 750 cycles measured with cyclic voltammetry.

100 80

(A)

Z=40

60 current / mA

40 20

Z=1

0 Z=40

–20 –40

Z=1

–60 –80 –100 3,4

3,5

3,6

3,7

3,8 3,9 4,0 4,1 4,2 potential vs. Li/Li+/V

4,3

4,4

4,5 4,6

100 80

(B)

60

Z=1 to Z=40

current / mA

40 20 0 –20 –40 –60 –80 –100 3,4

3,5

3,6

3,7

3,8

3,9

4,0

4,1

4,2

4,3

4,4

4,5 4,6

+

potential vs. Li/Li V Fig. 6 (A) Cyclic voltammograms of a contaminated LiCoO2 electrode (40 cycles). (B) Cyclic voltammograms of a purified LiCoO2 electrode (40 cycles).

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Inorganic Electrolyte Cells 4,4 70

4,35

60

4,3

50

4,25

40

4,2 4,15

30

4,1 20

peak potential / V

discharge capacity / % of theoretical capacity

456

4,05

10

4

0 0

250

500

3,95 1000

750

number of cycles Fig. 7 Discharge capacity and peak potential of an LiCoO2 electrode; beginning with a deintercalation state >70% (about 750 cycles).

High power discharge, mean power 120 W, 1475 Whkg-1 140

80

120

70

power / W

50

80

40 60

power / W current / A

40

30 20

20

10

0 -5

current / A

60

100

0 0

5

10

15 time / s

20

25

30

35

Fig. 8 Power and current of a 30-s high-power discharge with a specific power of 1475 W kg−1.

7.3

Cell data

The specific energy of the inorganic LiC6/LiCoO2 system is up to 250 Wh kg−1, and the energy density can exceed 750 Wh L−1.24 In cells optimized for high power, it is possible to achieve very high constant power levels. Fig. 8 shows power and current of a 30-s high-power discharge with a specific power of 1475 W kg−1. The usable capacity of cells is up to 70% of the theoretical capacity. In purified cells, 200 cycles without capacity loss or increase in the internal resistance were reached in laboratory full cells. After 200 cycles, the water contamination of the non hermetically sealed cells leads to an increase in the internal resistance and to a loss of capacity. While cycling up to 4.2 Volt the not sufficient reversibility of the overcharge reactions leads to a decrease of the cycle efficiency and in addition to a capacity decrease. The cell has a very good deep discharge characteristic. Starting with a cell discharged to 1.5 V, it is possible to make several deep discharge cycles without observing a change in the behavior of the cell. No significant changes in capacity or internal resistance were observed.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Inorganic Electrolyte Cells

8

457

Battery cells with LiFePO4 positive electrode

A change from lithium cobalt oxide to lithium iron phosphate (LFP) as the active material of the positive electrode in a cell with an inorganic SO2-based electrolyte increased the usable capacity and stabilized the capacity decrease. The overcharge reaction is reduced due to the lower upper potential of lithium iron phosphate compared to lithium cobalt oxide.

8.1

Cell data

Battery cells with lithium iron phosphate/graphite and inorganic SO2-based electrolyte show high longevity of 50,000 cycles with a charge/discharge rate of 2C reaching 20% of residual capacity with coulombic efficiency of 99.99% and a low and constant ohmic resistance at room temperature (Fig. 9).18 Up to 90% of the theoretical capacity of lithium iron phosphate can be used while cycling the cells.26

8.2

Deep discharge

The deep discharge of lithium iron phosphate/LiAlCl4 xSO2/graphite cells down to 0 V shows comparable cycling stability of cells which are discharged to 2.5 V (Fig. 10). The cells were cycled between the voltage window of 0–3.6 V at 2C rate and the time interval between charge and discharge was 10 min. After 150 deep discharge cycles to 0 V, no change in the behavior of the cell was observed. No significant changes in capacity and internal resistance were detected. This phenomenon may encourage researchers for the development of future energy storage systems which can be capable of tolerating the extreme discharge and charge conditions including the nonflammable inorganic SO2-based electrolyte in the electrochemical cells.

8.3

Surface coating

Zinck et al. proposed a dip-coating process for lithium iron phosphate or graphite electrodes.18,26 Preferably coating refers to a thin layer of Al2O3 or SiO2. Either the insertion active material intended for processing into the electrode or the whole electrode is brought into contact with a reaction solution that contains starting materials suitable for the formation of the layer. A temperature treatment is then performed to form and harden the layer. Cells were assembled with dip-coated lithium iron phosphate electrodes and standard graphite electrodes and results were compared with uncoated LFP and standard graphite electrode cells with SO2-based inorganic electrolyte. The capacity retention is improved by 5% at 1C. Also, the negative electrode can be coated. The capacity loss in the first cycle for a coated graphite electrode is significantly less than for an untreated electrode. The stability of cycle life of a lithium iron phosphate/LiAlCl4 xSO2/graphite cell can be improved by coating processes which shows significant improvements compared to cells with uncoated electrodes.

Fig. 9 Cycling of prismatic cells: voltage window 2.5–3.6 V; 2C cycling rate.

458

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Inorganic Electrolyte Cells

Fig. 10 Potential profile of deep discharged cells and normally discharged prismatic cells at 2C rate cycling.

9

Battery cells with inorganic electrolyte stable at high potentials

As discussed above the known issue of the LiAlCl4 xSO2 electrolyte is the oxidative building of chlorine at voltages above 4.0 Volt.28 The challenge is to expand the advantages of an inorganic SO2-based electrolyte with a stable high voltage behavior. The use of a proprietary conducting salt (details may not be disclosed at this stage), which is stable at higher potentials, shows promising results in a wide voltage range and open the possibility to develop new high voltage cathode materials.

9.1

Lithium iron phosphate battery cells

Fig. 11 shows three charge/discharge cycles of lithium iron phosphate (LFP)/high voltage salt  xSO2/graphite cells with different charge potentials of 4.6 V, 4.8 V and 5.0 V.

Fig. 11 Three charge/discharge cycles of lithium iron phosphate/high voltage salt  xSO2/graphite cells with different charge potentials of 4.6 V, 4.8 V and 5.0 V.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Inorganic Electrolyte Cells

459

The achieved cycle efficiencies and discharge capacities are almost identical for each charge potential and are >99,6%. This means that the discharge capacities obtained are independent of the charge potential. A higher charge potential does not cause any undesired reactions, such as decomposition of the electrolyte or irreversible destruction of the active material LFP.

9.2

Nickel manganese cobalt oxide (NMC) battery cells

The use of NMC as an active material in battery cells with organic electrolyte is limited to voltages not exceeding 4.2–4.3 V depending on the proportion of the metals nickel, manganese, and cobalt.27 High voltage leads to reactions between the NMC cathode active material and the organic electrolyte solvent resulting in a decomposition of electrolyte with formation of hydrogen fluoride (HF) and aggressive Ni4+. HF passivates the current collector and Ni4+ reacts with solvent to form a cathode electrolyte interface layer. This behavior leads to lower stability resulting in low cycle life and early cell failure. The use of a SO2-based electrolyte formulation, which is stable at high voltages prevents reactions with the NMC cathode active material at high potentials. No decomposition of electrolyte or formation of HF may be detected. A stable high voltage cycling of NMC/Graphite full battery cells are possible. Fig. 12 shows the discharge capacities and the resistances of two identical NMC/ graphite battery cells, one filled with organic (1 M LiPF6 in EC:EMC 3:7d with 2 wt% VC) and the other with the high voltage stable SO2-based electrolyte. Both cells were cycled in a voltage range of 2.8 to 4.5 V at C/2 discharge and 1C charge rate. While the organic cell shows turnover after only 10 cycles, accompanied by a significant increase of the internal resistance, the inorganic system is able to achieve stable discharge capacities and resistances under the exact same cycling conditions.

10

Conclusions

Sulfur dioxide solvates of lithium tetrachloroaluminate have low sulfur dioxide vapor pressure and high conductivity. Lithium grows in a needle-like shape and builds whiskers, instead of branched dendrites as in the inorganic electrolyte environment. A thin lithium dithionite surface layer on the negative electrode is present. The electrolyte is stable at potentials above 4 V and highly suitable for low voltage lithium cells. SO2-based electrolytes with more advanced conducting salts, which are stable at higher voltages show promising results. Different lithium-ion rechargeable battery cells have been presented, where lithium cobalt oxide or lithium iron phosphate are used as the active material for the positive electrode. Graphite serves as the active material of the negative electrode. The electrolytes are sulfur dioxide solvates of lithium tetra chloroaluminate or of a proprietary conducting salt stable at high voltages. Lithium cobalt oxide cells with inorganic LiAlCl4 xSO2 electrolyte demonstrated good electrochemical properties. The overcharge reaction must have a reversibility of 99.9% to get a stable capacity retention. The change of the active material from LiCoO2 to LiFePO4 increases the usable capacity and stabilized the capacity decrease. The overcharge reaction is reduced due to the lower upper potential of lithium iron phosphate compared to lithium cobalt oxide. The use of the high voltage electrolyte formulation shows promising results. The discharge capacities are independent of the charge potential. A high voltage charge potential does not cause any undesired reactions, such as decomposition of the electrolyte or irreversible destruction of the active material LFP. Battery cells with NMC as active material show an outstanding high voltage performance.

Fig. 12 Discharge capacity and resistance in the voltage range 2.8 V to 4.5 V of battery cells with NMC active material and high voltage SO2-based electrolyte respectively organic electrolyte.

460

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Inorganic Electrolyte Cells

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

Xu, K. Nonaqueous Liquid Electrolytes For Lithium-Based Rechargeable Batteries. Chem. Rev. 2004, 104, 4303–4417. Thomas, B. Reddy Linden’s Handbook of Batteries, 4th edn; McGraw-Hill, 2011. (Chapter 14.5–14.7). Kühnl H (1971) Germ. Pat. DT 2,140,146. Kühnl, H.; Bredemeyer, G.; Hartjenstein, D.; Jostak, J. Dichte, Leitfähigkeit und Elektrolyse flüssiger Phasen in wasserfreien Systemen vom Typ MCl/AlCl3/SO2 (M¼Li, Na). Z. Anorg. Allg. Chem. 1984, 515, 187–198. Kühnl, H.; Strumpf, A.; Gladwiza, M. Die Systeme des Typs MCl/AlCl3/SO2 (M¼Li, Na, K, NH4). Zeitschrift für Anorganische und Allgemeine Chemie 1979, 449, 145–156. Finke H-D (1986) Germ. Pat. DE: 3,604,541. Dey, A. N.; Kuo, H. C.; Foster, D.; Schlaikjer, C. R.; Kallianidis, M. Inorganic electrolyte rechargeable Li/SO2 system. Prog. Batteries Sol. Cells 1987, 6, 73–80. JEC Press, Inc, Cleveland, OH. Foster, D. L.; Kuo, H. C.; Schlaikjer, C. R.; Dey, A. N. New Highly Conductive Inorganic Electrolyte. J. Electrochem. Soc. 1988, 135, 2682–2686. Dey, A. N.; Bowden, W. L.; Kuo, H. C.; Gopikanth, M. L.; Schlaikjer, C. R.; Foster, D. Inorganic Electrolyte Li/LiCuCl2 Rechargeable Cell. J. Electrochem. Soc. 1989, 136, 1618–1621. Hambitzer, G.; Döge, V.; Ripp, C.; Pinkwart, K. Abscheidung und Wachstum von Lithium in anorganischer Elektrolytlösung. GDCh-Monographien 1999, 18, 82–88. Ripp, C.; Döge, V.; Pinkwart, K.; Hambitzer, G. Wiederaufladbare Lithiumbatterie mit anorganischer Elektrolytlösung. GDCh-Monographien 1997, 12, 311–318. Hambitzer, G.; Hefer, B.; Lutz, C.; Heitbaum, J. Zyklisches Verhalten von Lithium-Metallelektroden. GDCh-Monographien 1996, 3, 136–149. Hambitzer, G.; Hefer, B.; Lutz, C. Rechargeable lithium battery with inorganic electrolyte; In Proceedings of the 37th Power Sources Conference, Cherry Hill: New Jersey, USA, 1996. Stassen, I.; Hambitzer, G. Metallic lithium Batteries for High Power Applications. J. Power Sources 2002, 105, 145–150. Jeong, G.; Kim, H.; Sug Lee, H.; et al. A room-temperature sodium rechargeable battery using an SO2-based nonflammable inorganic liquid catholyte. Sci. Rep. 2015, 5, 12827. Kim, A.; Jung, J.; Song, J.; Kim, H. J.; Jeong, G.; Kim, H. Lithium-Ion Intercalation into Graphite in SO2-Based Inorganic Electrolyte toward High-Rate-Capable and Safe Lithium-Ion Batteries. ACS Appl. Mater. Interfaces 2019, 11, 9054–9061. Ripp, C.; Hambitzer, G.; Zinck, L.; M., Borck. In Encyclopedia of Electrochemical Power Sources; Garche, J., Dyer, C. K., Eds.; Academic Press; Imprint of Elsevier: Amsterdam, Boston, 2009; p 383. Ramar, V.; Pszolla, C.; Rapp, M.; Borck, M.; Zinck, L. Non-flammable Inorganic Liquid Electrolyte Lithium-Ion Batteries. J. Electrochem. Soc. 2020, 167, 70521. Logan, E. R.; Tonita, E. M.; Gering, K. L.; Li, J.; Xiaowei Ma, L. Y.; Beaulieu and J. R., ; Dahn, A. Study of the Physical Properties of Li-Ion Battery Electrolytes Containing Esters. J. Electrochem. Soc. 2018, 165 (2), A21–A30. Peled, E. The Forming Reaction at the lithium/Electrolyte Interface. J. Power Sources 1983, 9, 253–266. Aubay, M.; Lojou, E. Film Formation on lithium Electrode in LiAlCl4/SO2Cl2 and LiAlCl4:3SO2 Based Electrolytes. J. Electrochem. Soc. 1994, 141, 865–872. Prochazka, P.; Cervinka, D.; Martis, J.; Cipin, R.; Vorel, P. Li-ion Battery Deep Discharge Degradation. ECS Trans. 2016, 74 (1), 31–36. Bard, A. J.; Faulkner, L. R. Electrochemical Methods – Fundamentals and Applications; John Wiley & Sons, Inc., 1980 Wang, H.; Jang, Y.-I.; Huang, B.; Sadoway, D. R.; Chiang, Y.-M. TEM Study of Electrochemical Cycling-Induced Damage and Disorder in LiCoO2 Cathodes for Rechargeable lithium Batteries. J. Electrochem. Soc. 1999, 146, 473–480. Zinck, L.; Borck, M.; Ripp, C.; Hambitzer, G. Purification Process for an Inorganic Rechargeable Lithium Battery and New Safety Concept. J. Appl. Electrochem. 2006, 36, 1291–1295. Zinck et al. (n.d.) US 2017/047612 A1. Jung, R.; Metzger, M.; Maglia, F.; Stinner, C.; Hubert, A. Gasteiger Oxygen Release and Its Effect on the Cycling Stability of LiNixMnyCozO2 (NMC) Cathode Materials for Li-Ion Batteries. J. Electrochem. Soc. 2017, 164 (7), A1361–A1377. Grundish, N.; Amos, C.; John, B. Goodenough Communication—Characterization of LiAlCl4  x SO2 Inorganic Liquid Li+ Electrolyte. J. Electrochem. Soc. 2018, 165 (9), A1694–A1696.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production Paul R Shearinga,b, Denis Cummingb,c, and Emma Kendrickb,d, aThe ZERO Institute, University of Oxford, Oxford, United Kingdom; b The Faraday Institution, Harwell Science and Innovation Campus, Quad One Becquerel Avenue, Harwell, Didcot, United Kingdom; cDepartment of Chemical and Biological Engineering, University of Sheffield, Sheffield, United Kingdom; dMetallurgy and Materials, University of Birmingham, Edgbaston, United Kingdom © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 Introduction 2 Powder handling and electrode formulation 3 Mixing 4 Ink and slurry characterization 5 Coating 6 Coating metrology 7 Drying 8 Calendering 9 Slitting and notching 10 Cell assembly 11 Formation aging and testing 12 Conclusions Acknowledgement References

461 462 463 464 466 466 467 468 469 470 470 471 471 471

Abstract Li-ion batteries are manufactured globally at scale by a large number of companies and in a diverse range of cell formats. The applications for these cells are also diverse, ranging from micro-batteries to grid-scale storage, although in the past decade, the Li-ion battery manufacturing industry has been heavily dominated by cells destined for electric vehicle operation. While this breadth of cell type and application, and the intellectual property protection unique to each manufacturer, inevitably leads to nuances in individual manufacturing facilities, there are generic unit operations that are common to most. This chapter summarizes those unit operations for generic Li-ion electrode and cell manufacture, and introduces key scientific concepts to illustrate the process.

Key points

• • • •

1

To introduce the common unit operations of Li-ion battery electrode manufacture and cell assembly, To explain the integration of these unit operation in electrode and cell manufacture in a typical manufacturing environment, To introduce key under-pinning scientific concepts germane to electrode manufacturing, To explore the role of metrology tools in electrode and cell manufacturing.

Introduction

The manufacture of Li-ion batteries has accelerated globally over the past decade, particularly to service the needs of the electric vehicle industry. So called ‘giga-factories’ (those manufacturing facilities capable of producing over a GWh of energy storage units per year) are increasingly geographically diverse, however, while manufacturing capacity in European and North American facilities is growing rapidly, Asia remains the focal point for much of global Li-ion battery production. Modern Li-ion batteries are commodity products, and the facilities that make them are notable for their scale and the high degree of automation. With the rapid increase in demand for high energy density batteries, innovation is growing in both the battery chemistries to serve a range of applications, the design of cells and batteries, and the manufacturing processes to supply them. In this chapter, we focus on the most widely used processes for Li-ion battery manufacture beginning with electrode processing and concluding with cell Formation, Aging and Testing (FA&T). Cell manufacture at scale is an energy-intensive process, and we note that innovations are constantly being developed to improve product quality and manufacturing yield and to reduce energy consumption and ‘embedded carbon’. Alongside an improved understanding of cell performance and end-of-life recycling, these innovations are helping the industry to move toward a circular

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00316-9

461

462

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production

Fig. 1 A schematic diagram of a typical cell manufacturing process highlighting key unit operations.

economy. This chapter is focused on Li-ion battery manufacture, but there are parallels with other ‘post Li-ion’ battery manufacture. For example, sodium-ion batteries are generally considered to be a ‘drop-in’ technology that can leverage the established and mature Li-ion manufacturing base. Here our focus is on manufacturing techniques for porous, non-metallic electrodes assembled in cells with non-aqueous liquid electrolytes. While alternative next generation battery chemistries (including solid state batteries and lithium-sulfur batteries) will diverge in some aspects of their manufacture (for example in handling metallic foil electrodes and solid ceramic electrolytes), they will also share some of the unit operations described herein. A comprehensive review of each of the unit operations involved in the manufacturing processes for Li-ion electrodes and cell assembly is beyond the scope of this chapter, and throughout, the reader is signposted to further reading.1 Fig. 1 illustrates the process flow diagram for the generic unit operations in Li-ion battery manufacture which defines the scope of this chapter. For illustrative purposes we will consider a generic Li-ion battery chemistry to describe the manufacturing operation: a predominantly graphite negative electrode deposited onto a copper current collector, a metal oxide (e.g. NMC811) positive electrode deposited on an aluminium current collector, assembled into a cell using an electrically insulating polymer separator and filled with liquid electrolyte (generally LiPF6 in EC/DMC solvent). While the electrode materials are not, compared with metallic electrodes, highly air or moisture sensitive, cathode (positive electrode) materials will generally be manufactured in an atmospherically controlled environment with reduced moisture content. These positive electrode materials generally also utilize PVDF binders which are soluble in NMP, and therefore a significant infrastructure in electrode manufacture is also devoted to solvent handling and recovery

2

Powder handling and electrode formulation

The balance of materials in a typical lithium-ion cell contains common components such as housing or cans, current collector foils, separator and various electrode tabs and integral safety devices. These components are highly varied depending on the cell format and proprietary cell design. They are typically manufactured using standard manufacturing processes for metals or polymers and use traditional joining techniques such as welding.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production

463

The remaining components that are used to make the electrodes are composed of a cathode and anode active material, conductive additives to aid the electronic conduction of the electrode composite and a polymeric binder to hold the other components together and adhere the electrode to the current collection foil. Materials in a typical Li-ion cell generally use graphite-based anodes and metal oxide cathodes. Graphite can be mined from natural deposits or produce synthetically using very high temperature (in excess of 3000
C) processing. Cathode active materials are synthesized from lithium, and transition metal hydroxides and sulfates and these are produced via standard mineral extraction and chemical processing routes. Conductive additives are typically carbon black, a finely divided form of carbon which is produced from the incomplete combustion of natural gas. Finally, the polymer component is made using well establish petrochemical based routes and the cathode binders use fluorinated compounds which are more specialist but routinely manufactured by industry. Downstream processes such as mixing use the starting components in powdered form. For all of the subsequent mixing and coating processes the quality of the product is correlated to the powder specification and properties. Particle size, shape, and distribution are three essential parameters that manufacturers strive to control within and between batches of powder. As discussed below, most mixing processes work best when the main component of the mix in the micron range as very small particles can cause unpredictable rheology and very large particles tend to settle under the force of gravity and so are not stable over time. To move powders around, it is best if they flow easily and do not clump so that storage and transport to and from storage hoppers is easy. For more advanced processing (such as solvent free dry processing) powders can flow directly into processing unit operations without the aid of a liquid or solvent. Size and shape are the two parameters which control slurry properties and powder flow and size distribution in batch can influence both properties and so also need careful control. For widely deployed poly-crystalline cathode materials, most manufacturing systems aim for a cathode active material with mean particle size of ca. 10-15 mm. Fortunately, state of the art synthesis processes currently produces most layered metal oxide cathodes with near spherical particles in desired size ranges. Advanced synthesis routes are being developed and scaled to compositionally grade the inside of particles so that the lithium storage and transport can be tailored to improve performance and lifetime.2 Graphitic anode materials generally possess irregularly shaped particles in the 10-20 mm range due to the crystal structure of graphite and to the limitations of how graphite can be processed. However, spherical graphite has a lower specific surface area (BET), which minimizes capacity losses; furthermore, the tap density and specific volumetric capacity are increased. Conductive additives typically have very small particle size (nanosized primary and submicron secondary aggregates), and so great care needs to be taken when processing the conductive additive to ensure adequate distribution in the slurry and the final electrode structure. Polymer binders, while supplied in powder form, are dissolved in the slurry solvent and so the size and shape are less critical. Solvent-free, dry processing is emerging as an alternative, but this area is evolving very rapidly, and it is unclear what the best structure for the polymer will be in this scenario. Additional new materials that are being used in small quantities to improve energy storage or performance include new active materials such as nano-sized silicon on the anode side and carbon nanotubes as conductivity modifier. There will continue to be new advanced materials added as the field develops, however, for the purpose of this article we wish to emphasize that there are core rules for formulation which remain constant despite large variation in material types, cell chemistry or cell architecture. Formulation encompasses two different sets of choices that need to be simultaneously satisfied. The first is the formulation of the final electrode: this fundamentally is the ratio of the active, conductive and polymeric components. Depending on the application of the cell, energy or power density can be prioritized. For energy dense cells, compositions can contain greater than 96% by weight of active material, with the balance consisting of conductive and polymer materials, typically in a 50:50 ratio (i.e. 2 wt% carbon black and 2 wt% polymer). As mentioned above, particle size and shape, while important for processing, also control the final electrode microstructure and therefore cell performance, which is why the choice of size and shape plays a critical role. Cell manufacturers will have numerous proprietary variations to manipulate the cell performance to their desired product specification, and most of these compositions are trade secrets; consequently, it is not possible to describe them in more detail even if the breadth of the composition could be analyzed and categorized. The second, simultaneous, formulation is for the slurry to enable the mixing and formation of the electrode shape during coating. The particle size and shape and the relative ratios of the small and large sized components, as well as the surface chemistry and compatibility with the solvent all dictate how well the slurry can be processed. This area of formulation is still highly empirical and requires great skill and experience as well and a deep understanding of colloidal science. In the following section the mixing and coating process with be covered in detail.

3

Mixing

The electrode components: active material, binder, and conductive additive are combined into an ink, slurry or paste for coating or extrusion onto a metallic foil current collector. The mixing step ensures that the conductive additives, usually carbon, are well dispersed throughout the ink, which produces the 3D electronically conductive network within the electrode coating once dried. The binder is dispersed throughout such that after drying a cohesive and adhesive film is produced. Several binder-solvent systems can be utilized; to date, electrode manufacturing has been dominated by N-methyl pyrrolidone (NMP) and Poly Vinylidene polymer (PVDF) binders, which have limited solubilities in solvents. NMP is a harmful solvent and is used because of PVDF’s solubility, however, a search for more green and less harmful solvents which are compatible with PVDF or offer alternative binder solutions is required. Water-based systems can be used for air-stable materials such graphite and lithium

464

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production

iron phosphate, LiFePO4, (LFP). Aqueous processing has more challenges due to the high surface tension and wettability of active material components and foils. However, the low toxicity and cost of water-systems means that solvent recovery is not needed unlike for organic solvents used in cathode/NMP/PVDF systems; consequently, the cost of production and the manufacturing footprint can be substantially reduced. The surface tension and the components’ dispersibility can be improved by adding surfactants. Surfactants are comprised of mixed hydrophobic and hydrophilic character which can adhere to the surface of the active material components, while also interacting with the solvent molecules, keeping the materials more easily suspended and hence providing a more stable ink. More recently dry processing has become preferable due to the reduced cost in manufacturing from the removal of the drying process. Dry processing requires polymers that can fibrillate to produce a web that attaches all the components and adheres the components to the current collector. Different mixing methods, such as jet-milling or twin-screw extruders which can disperse the carbon black and fibrillate the polymers are required and are currently an active field of development.3 Different mixer types are used for ink production and range from small lab-scale mixers to industrial continuous mixing. For a given formulation, the method of mixing will heavily influence the resulting ink properties and coating characteristics. The key variables are formulation, order of addition, rotation speed and energy, mixing time and temperature. An energy profile of the mix gives some information about the key stages of the mixing process, rheology is measured to ascertain the viscosity and coat-ability of the slurries, and conductivity measurements are performed to note the dispersion of the conductive additives. Four mixing types are discussed here, which are used from Lab scale to Giga factory scale; Dual Asymmetric Centrifugal, High torque, High Energy and Continuous. A Dual Asymmetric Centrifuge (DAC) is a device that combines centrifugal and rotational forces to quickly mix the contents of a sample holder without entraining any air. The sample holders are designed to balance the load during mixing. Once mixed, the material is transferred to the dispensing vehicle. This method is highly effective in eliminating air bubbles in small quantities of materials and is commonly used in laboratories. One point to note is the importance of properly fill the mixing vessels to ensure optimal mixing. A high torque slurry mixing method, is where the ingredients are mixed at high viscosities or as pastes, this requires a force mixing device. A standard procedure is as follows; initially, the dry ingredients are combined, followed by the addition of a solvent or binder solution which is then kneaded to form a dough. The viscosity of the mixture is adjusted by adding the appropriate amount of solvent for coating purposes. It is important to be cautious during the mixing process to avoid over-processing the conductive carbons, which can result in the formation of carbon coatings on active materials. This can lead to electrode coatings with high resistance. To counter this issue, graphite, or other long range conductive carbons, such as carbon fibers or carbon nanotubes can be added to establish long-range conductive pathways that connect the carbon-coated particles. To improve the mixing, scrapers can be incorporated into the mixer design which prevent the need for manual scraping of the sides and the paddles periodically during the mix. High energy slurry mixing is typically performed by pre-dispersing a conductive additive in the binder solution or solvent, before the addition of the active material component. This process helps to break down agglomerates before adding the active material, particularly with additives such as carbon black. The lower viscosity of the mix aids in the dispersion of carbon black through the ink. The active material is then added in aliquots to ensure that the viscosity does not get too high during the mixing process, ensuring good homogeneity of the ink. By using this method, the solid content of the inks can be higher than the high torque method for the same achieved slurry viscosity. This method also encourages a 3D conductive network of carbon black, and additional carbon types are not required to form the conduction bridges. Continuous slurry mixing is a widely used technique in large-scale manufacturing that reduces the time and energy required for the mixing process. This method typically involves a co-rotating screw mixer that can continuously dose raw materials, pre-mix, knead, disperse, and degas them. The precise and continuous dosing of solid and liquid components ensures fast mixing, and the residence time is usually less than one minute. This allows for quick processing and the mixing of materials that are sensitive to air or moisture, making it suitable for direct application onto a coating line. However, this method is not ideal for smaller research and development processes, as it requires changes to optimize the process for different material types and larger material quantities per run. If the material set is fixed, optimization can be achieved by manipulating material flow rates into the mixer, generating multiple different compositions from a single run.

4

Ink and slurry characterization

The slurry viscosity can be modified through formulation and mixing, changing the solid loading, conductive additive type and content, or binder and solvent ratio. Specifically, the solid and solvent ratio in the slurry is the key to achieving good homogeneous coatings. In all cases for printing the electrode, the slurry must flow and deposit onto the collector homogeneously and retain its structure into the drying process. The slurry properties are governed by the viscosity, viscoelastic and yield stress properties of the inks and slurries. A high viscosity at low shear rates is required to keep the slurry dispersion stable and maintain the structure of the coating. It also prevents the wet coating from spreading or ‘slumping,’ which results in thin and uneven coatings. At high shear, a low viscosity is required so that the slurry will flow and coat smoothly onto the current collector, without excessive pressure build-up which limits coating speed. To enable high throughput and less waste, the slurry needs to have a fast response to each shear

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production

465

rate applied and allow any slumping to occur before the slurry returns to a semi-solid state once the shear is removed. The viscoelastic properties also determine the coat-ability of the mixtures, for example if in an elastic region at the coating shear rate, coating instabilities results in uneven coating. Therefore, the inks and slurries are characterized by their viscosity, rheology, and stability before coating so that the coat-ability and period of stability of the inks can be assessed. The exact rheological properties required depends on the specific printing or application process. However, viscosity is mostly used as the standard measure, and the apparent viscosity is reported at a specific shear rate. The apparent viscosity for a non-Newtonian fluid is given in Eq. (1). The required viscosity must be within the correct shear parameters or defects such as streaking or pooling can occur during the application. An apparent viscosity sweep against shear illustrates the shear thickening or thinning behavior of these composite slurries, and is used to ascertain the viscosity at the shear applied in the specific printing process. Eq. (1)—Non-Newtonian flow equation where g is the shear rate, h is the apparent viscosity and k is the shear stress. h ¼ k:g−1

(1)

A shear thinning non-Newtonian fluid can be described by the cross model (Eq. 2) with two limiting Newtonian viscosities at low shear, h0 and high shear, h¥. m ¼ 0 describes a Newtonian fluid, whereas m ¼ 1 a shear-thinning liquid. Eq. (2)—Cross equation which describes the full viscosity vs. shear of a non Newtonian fluid. K is time dependent and m is dimensionless. h − h¥ 1 ¼ h0 − h¥ 1 + ðKgm Þ

(2)

Apparent yield stress describes the point in time when the composite ink particle network changes from an elastic solid type of behavior to a liquid. This is important to understand, as it gives an idea of what minimum shear is required for the slurry to flow onto the current collector. The thixotropy of the ink describes the restructuring of particles in the liquid after the slurry has been deposited. It describes the ‘slumping’ or surface levelling after application, the correct thixotropy will mean clean coating edges and homogeneous electrode coatings. Thixotropy, or material dispersion properties, can be determined from the hysteresis between the up and down sweeps in a viscosity-shear rate curve. A second measurement type is a Three Interval Thixotropy Test (3ITT) wherein a low shear measurement is applied followed by a high shear and then a low shear. This simulates the structure at rest then structure breakdown and regeneration. Fast semi-solid particulate structure regeneration is required to retain the shape of the deposition but not too fast that the coatings can level after deposition.4 Ink stability is important for coat-ability and ensuring that the same coating is achieved at the start and end of the coating. Particle agglomeration, phase separation or sedimentation can occur when the slurry is not optimized. Stabilization can be achieved by electrostatic, steric, or electro-steric methods. Polymers provide steric hindrance between particles while surface-modifying dispersants achieve electrostatic repulsion. An ink thickener such as carboxy methyl cellulose (CMC) can be used to increase ink viscosity and stabilize the ink. If the ink is not stable, sedimentation of the particles will occur, and this can be measured by the Stokes equation. Eq. (3)—Stokes equation which describes the rate of sedimentation (V cm s−1) of a particulate dispersion in a liquid. Spherical particles of diameter, d are assumed and are of a sufficient distance apart to not interact. r1 is the density of the particles and r2 of the solvent-binder, g gravity and h0 viscosity of the ink. V¼

d2 ðr1 − r2 Þg 18h0

(3)

Routes to improve ink stability include reduction in particle size, increase in binder-solvent density and increase in viscosity, although these parameters have to be considered as part of a wider multi-objective optimization. For many systems, the particles are not completely isolated, and agglomeration or flocculation of the particles can occur which increases the sedimentation rate. This is especially true for water-based ink systems. Physical particulate interactions in the slurry cause the above stabilities; however, chemical processes can also occur. One example is for inks with high nickel content layered oxides such as lithium-containing NMC8115 or Sodium nickel oxides.6 As shown for the sodium-based cathode inks, the surface residuals, carbonates, and hydroxides react with the binder, causing a chemical gel that is uncoatable. Gelation is caused by defluorination of the PVDF, which results in crosslinking of the polymer over time, this can be controlled through pH, water content and heat reduction. In addition, these basic inks can corrode the aluminium current collector, and acids are sometimes used to help neutralize the pH and prevent corrosion. However, care must be taken, as water is produced which can inadvertently also affect the degradation of the materials during long term operation of the battery. The properties of ink or slurry are influenced by a variety of factors, such as particle size and shape, surface area, surface properties of the material, molecular weight, binder concentration, solvent nature, mixing method, container size, and propeller. Each of these factors is systematically adjusted in order to optimize electrochemical performance. If a new active material with different particle size is introduced, the optimization process is repeated to ensure optimal results. High-energy coatings with high coat weights require high-viscosity, high-solid content inks, while lower viscosity inks are preferred for low loadings generally associated with high power cells. Throughout the process, the solvent content is kept to a minimum to facilitate drying and minimize solvent recovery needs. The mixing sequences and methods used can have a significant impact on electrode performance.

466

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production

The surface properties of a slurry determine its wettability and tendency to slump after coating, providing additional information about the slurry’s structure. To gauge these properties, measurements of surface tension and interfacial tension with the current collector are performed, such as contact angle measurements. High contact angles hinder spreading, which may additionally result in poor adhesion of the coatings. Homogeneity of the electrode slurry can be determined through density measurements, predicted density is based on component ratios and is compared to measured density to ensure consistency. Poor mixing can lead to density variations and component interactions can alter slurry density. Agglomerates in the slurry can significantly impact coating quality. To ensure proper dispersion of the active material agglomerates are broken down into individual particles, which also improved the stability. Measuring particle size in opaque slurries with high solid weights can be challenging, as light scattering measurements may require dilution of the inks, which destroys the slurry structure. However, particle size distributions from light scattering can provide valuable information to determine if the active material is dispersed properly. Standard industrial methods include the use of a Hegman gauge which can determine agglomerate size of an ink.

5

Coating

The slurry is coated onto a current collector using a tape casting method: Slot die, comma bar or doctor blade techniques are commonly used. Small laboratory coatings are generally performed by doctor blade, while industry typically uses a slot die coater which has superior control and tolerances. More recently dry coating processes are being commercialized, removing the need for drying and hence reducing costs significantly. Doctor blade coating, also known as knife coating or blade coating, is a popular technique for fabricating thin films. It involves either running a blade over the substrate or moving a substrate underneath the blade. This ink and slurry-processing technique produces high-quality films at a low cost. The process begins with coating the substrate with a thin layer of solution, which is then spread uniformly across the substrate using a blade. The thickness of the resulting film depends on the gap between the blade and the substrate. In a comma bar process, the electrode coating is created by passing the substrate through a gap between the comma roll and the back roll. This technique is best suited for medium to high-viscosity materials and produces medium-thickness film coatings. The variation of the comma direct (box type) method is particularly useful for producing thick coatings. It allows for a wider gap, with a structure that ensures even coating fluid supply and prevents fluid from flowing. Reverse comma bar coating is a type of contouring coating method that is similar to the comma direct method. It involves adding a rubber transfer roll to the comma bar, resulting in a shorter deposition length. This method is used when it is not desirable to make the substrate run through a narrow gap. It also allows for intermittent or patch coating, which is used for cylindrical cells electrodes, and has higher precision deposition with cleaner edges. The slot die process was first developed in the 1950s for the industrial production of photographic papers and is highly effective for production of battery electrodes due to the versatility and ability to print thicker and higher mass coatings. The slurry is delivered onto the surface of the substrate through a slot-die coating head, these can be of various designs, they all however have a high aspect ratio outlet, which carefully controls the final delivery of the coating liquid onto the substrate. This results in the continuous production of a wide layer of coated material on the substrate, with adjustable width depending on the dimensions of the slot-die outlet. By closely controlling the rate of solution deposition and the relative speed of the substrate, coatings with easily controllable thicknesses up to hundreds of micrometers are achieved.

6

Coating metrology

During the coating stage, the electrode slurry is applied onto the current collector, which is then dried and calendered to produce the final electrode. Quality control (QC) measures can be taken during both the wet and dry coating processes, as well as after calendering (Table 1). The purpose of QC during the coating stage is to assess the properties of the electrode and enable the production of electrodes that are free of defects and have controlled energy density, which can be balanced against a counter electrode. Table 1

Metrology measurement summary for electrode coatings.

Electrode properties

Measurement techniques

Thickness Coat weight Porosity Adhesion Conductivity

Micrometer, light interferometer, laser sensors, laser Ultrasound, Beta sensors, x-ray Mercury porosimeter, BET, density and volume measurements Scratch testing, bend test, peel test, pull off-test 4-point probe

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production

467

Coat weight and thickness are interlinked properties that determine the capacity of an electrode. Thickness allows for density and porosity estimation which is needed for calendering the coating to a fixed porosity, enhancing conductivity and volumetric capacity without compromising ion transport. Thickness is often measured using laser triangulation, which accurately determines the distance of the coating from the sensor. However, the coating needs to be stable and passed over a fixed roller, and the bare foil needs to be measured for reference. Laser calipers produce highly accurate single-point measurements, but they are difficult to traverse. X-ray or beta sensors can be used for coat weight measurement, but they require radiation safety controls. Alternatively, ultrasound is a radiation-free alternative. By measuring both thickness and coat weight in-line, density maps can be created for in-line control of calendering pressure or gap height. Porosity is frequently estimated from the volume of the electrode components and known mass contents, this relies upon the known density of the components, which is often difficult to ascertain for commercial electrodes. Off-line techniques such as electron microscopy and X-ray tomography methods can be used to reconstruct the electrode and analyze the carbon binder domain and porosity. Mercury porosimetry and BET can analyze different pore size volumes and give some information into the range of the porosity. Coating conductivity can be measured in-plane or through-plane. In-plane techniques are useful to examine the coating, such as with 4-point probes, but the through-plane is more representative of final cell conductivity. For the (in-plane) 4-point probe, the coating is typically on a dielectric such as mylar, to remove the electronic conductivity contribution from the current collector. Good adhesion is needed for coatings to handle calendering and cell assembly without delamination. Volume expansion and contraction during cycling can cause delamination, leading to premature cell failure. Therefore, ensuring strong adhesion at this stage can prevent cell failures. Adhesion can be measured using peel testing, which involves peeling the coating at a 180-degree angle and measuring the force required. Pull-off testing involves removing the coating vertically upwards. Peel testing is useful for predicting adhesion during handling and calendering, while pull-off testing gives a more direct adhesion measure that is easier to use in models. Other methods include bend testing and scratch testing; however, these are very qualitative and difficult to reproduce.

7

Drying

After coating the slurry onto the current collector, the electrode is dried to remove the solvent and create the 3D porous structure (see Fig. 2). This affects the adhesion, conductivity, and pore distribution in the final electrode coating. The drying process follows three stages: film shrinkage, solvent evaporation, and bonding to the substrate. Factors such as drying rate, binder types, temperature, and drying procedures can significantly impact the quality of the electrode and, consequently, the battery performance.7 Various drying methodologies, such as air-drying, inert gas flow, laser drying, and spray drying, can be utilized – all of which affect the microstructure and cohesion of the electrode film. The drying processes is a large contribution to the manufacturing costs because of the heat required to remove the solvent, and the energy needed to condense the solvent and prevent release to the atmosphere. Therefore, the choice of binder, such as organic-based or water-based binders, can greatly influence the cost and environmental impact of the drying process. Typically, in a laboratory setting, hot plates are used to heat the electrode either from the bottom or in some cases from the top and the bottom. This transfers heat into the electrode and encourages the solvent to evaporate. This process is not scalable, or suitable for double sided coatings, and in industrial ‘reel-to-reel’ settings, hot air drying is typically used, as this can be incorporated after the coating deposition process. If evaporated too quickly the binder and carbon black can migrate toward the electrode surface causing delamination from the current collector. The temperature required to control the evaporation depends upon the solvent and binder type. Lower solvent ratios require less energy to remove and is also less likely to have binder migration occurring as it dries. Therefore, precise control of the temperature and solvent evaporation profile affects both the microstructure and homogeneity of the final electrode coating, in addition to the cohesion and adhesion to the current collector.

Fig. 2 Drying process for a slurry coated film showing the (a) the coated slurry (b) the consolidation and (c) the final solvent evaporation via capillary action to form the final electrode (d). Reproduced from Zhang Y. S. et al. A Review of Lithium-Ion Battery Electrode Drying: Mechanisms and Metrology. Advanced Energy Materials 2022, 12(2), 2102233, https://doi.org/10.1002/AENM.202102233.

468

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production

Convection drying is where hot air streams are blown onto the surface of the coating, this can occur from the top and the bottom. The benefit of drying integration into a reel-to-reel system is that there can be different temperatures and flows set for different stages of the process, and large drying units of several tens of meters, incorporating different drying chambers are employed to control the extraction of the solvent. Typical temperatures range from 50
C to 180
C. For some chemistries, wet air can cause some degradation of the material properties, so the airflow can be modified for dry air or inert gas, which is common for high nickel-based cathodes. Defects during the drying process of coatings include delamination, cracks, and blistering. Delamination is caused by poor contact or fast drying and is more common in thicker coatings. Cracks are caused by internal stresses and can be reduced with slower drying rates. Blistering is similar to pinholes caused by evolving gases and can be prevented by optimizing the slurry or surface.

8

Calendering

Calendering is a thickness reduction operation whereby the electrode is passed through the gap between two large rollers. It is the final step in electrode manufacturing and serves to set the final thickness and porosity, and provide a suitable surface finish, flattening any imperfections formed during coating or drying. It is the last step in the electrode manufacturing process that substantively changes the electrode microstructure and so has a critical impact on electrode and cell performance (see Fig. 3). Calendering is a very cost-effective process used to enhance the volumetric energy density, by bringing active material (usually greater than 95% by weight) closer together to occupy a smaller volume.8 This is achieved by reducing the porosity (e) from circa 50% down to 20–40%, depending on the electrode application. Compression improves the internal electrode cohesion and adhesion to the current collector, however, over compression can close off the pores making the electrode difficult to wet with electrolyte solution and limit the Li-ion diffusion during cell use. The process itself is relatively simple in application: an electrode is passed between a gap set by two large, mechanically stiff and hard rollers. This imparts a compression force on the electrode resulting in thickness reduction. In practice, the force is normalized by the coating width to calculate the line load, Qline. Qline ¼ Fn =Wc Where Fn is the force applied through the rollers and Wc is the width of the coating. The only other practical parameters that can be controlled are roller temperature and speed. Calendering parameters (line load, temperature and speed) must be adjusted according to the electrode properties such as formulation, coating properties from the preceding operations such as drying and slurry solids content, which mayrequire adjustment of the calendering parameters due to variations in binder location and local variations in porosity and electrode thickness. Historically, this optimization has been heavily empirical. During compression the mechanical properties of the electrode will determine the critical ‘spring back’ where the elastic component of the compression recovers after leaving the roll gap. Composition of the electrode, which is already determined during the formulation stage, will control the mechanical properties of the electrode.9,10 Roll temperature can significantly influence compaction behavior particularly when the binder is a thermoplastic such as PVDF used is most state of the art cathode formulations.11 New active materials such as nickel rich materials, NMC811 and above, and additions such as carbon nanotubes and silicon additives can change the mechanical response of electrode and therefore change the way the electrode will behave during calendering. Broad interactions are shown in Fig. 4. But the specific relationships are not fundamentally understood and need more scientific and applied research, both experimental and through modelling, which is advancing rapidly.12–14

Fig. 3 A schematic of the calendering process showing the compression of the high porosity (e) dried electrode (yellow). Formulation and drying procedure will control the mechanics of the film and therefore the amount of elastic recovery after exiting the roller (green) as well as the final microstructure of the electrode (purple).

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production

469

Fig. 4 Structural relationships in a calendered electrode where one property must be traded off against another. The thickness is controlled entirely by the process and the resulting porosity/density controls the final microstructure and ultimate electrochemical performance of the cell.

In addition to optimization of the microstructure, the calendering operation must avoid introduction of defects. These can include variation is local porosity due to inhomogeneity in the coating, delamination at the current collector interface or internally in the electrode or wrinkling of the current collector (typically seen at the edges). Determination of the correct parameters is often highly empirical. Development of new insights into the link between process parameters, microstructural development and electrochemical performance, combined with new types of metrology and process sensors will open the possibilities of high throughput, lean manufacture.

9

Slitting and notching

Prior to cell assembly, dried, calendered electrodes must be unwound from their reels and cut into a suitable geometry for the desired cell format. This cutting process involves two key steps: slitting, whereby the electrode is cut to the desired size; and notching where the tab geometry (on the bare foil) is formed for subsequent welding. The slitting and notching processes are both generally considered to be low-energy and high-throughput operations in the overall cell manufacture, and are generally achieved using mechanical cutting. Both slitting and notching operations must ensure high dimensional tolerance and should provide ‘clean’ cuts of the electrodes without burring or the production of swarf. Large burrs can lead to a risk of short circuiting in the assembled cell, while swarf that may result from the cutting process needs to be eliminated from the cut electrode, as metallic foreign matter within an operating cell can pose performance and safety risks. The ‘blade cutting’ processes generally have lower capital outlay, but periodic maintenance is required; alternative approaches using laser slitting and notching are therefore gaining increasing attention. A photographic illustration of some of the manufacturing unit operations described in this chapter is shown in Fig. 5.

Fig. 5 Unit operations in battery manufacture (a) mixing (b) printing cathode (c) printing anode (d) drying (e) calendering (f ) cell winding and assembly. All images courtesy UKBIC.

470

10

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production

Cell assembly

There are three main Li-ion cell formats that are widely manufactured at scale; these are pouch cells, prismatic cells and cylindrical cells. Cylindrical cells are generally categorized according to their diameter and height in mm (where for example an 18,650 is a cylinder of 18 mm in diameter and 65.0 mm in height). Double sided electrodes (anodes and cathodes) are separated with a polymeric separator material which can be tightly wound in a ‘jelly roll’ configuration before insertion into a metal (generally steel) can. The spiral wound nature of cylindrical cells lends itself well to roll to roll processes readily enabling automated cell assembly off the roll. Careful monitoring of the winding tension for the spiral wound electrode/separator assembly must be ensured to guarantee product quality. The position and number of current collector tabs are determine, in part, the cell’s power capability; generally printed electrodes will contain periodic ‘gaps’ in the active material coating, allowing direct welding of the tab to the current collector. More recent innovation in so-called ‘tab-less’ designs circumvent the need for this additional tab welding step. The spiral assembly is placed inside an empty steel can, generally with the negative terminal at the bottom, and the positive terminal at the top; electrolyte filling follows under environmentally controlled (dry) conditions. The top of the cell also contains safety hardware which may include current interrupt devices (CIDs), positive temperature co-efficient devices (PTCs) and pressure burst disks. The cell cap is joined to the cell body with sufficient electrical insulation to prevent shorting the terminals. While cylindrical cell formats were initially developed for consumer electronics applications, their performance, widespread availability and comparatively low-cost has also enabled their integration into EV applications. In addition, the lower energy of the small cells with a higher specific surface area (m2/kWh) compared to larger prismatic cells provides greater fault tolerance. Over the course of the past decade there has been a substantial increase in the volumetric capacity of cylindrical cell formats associated with improvements to cell engineering as well as constituent materials. More recently, there has been a significant motivation to increase the size of cylindrical cells, and cells with diameters >40 mm are now routinely available. Pouch and prismatic cells have similar ‘mono-lithic’ designs and are generally differentiated by the casing materials; pouch cells have a ‘soft’ external shell, generally comprising aluminized plastic, while prismatic cells are enclosed inside a solid metal casing. Both variants can be manufactured in a wide range of capacities and footprints servicing an array of sectors. The constituent electrodes can be assembled in a flattened wound format, or in a stacked format (which may be a laminated ‘brick’ design of the electrode/separator stack, or a ‘Z-fold’ of separators woven between the electrode layers). In the flattened winding, the ‘jelly rolls’ are assembled using a similar winding assembly to cylindrical cells, while in the Z-stack and brick design, the electrodes are stacked robotically layer by layer. The soft casing of pouch cell formats can be heat sealed, but do not accommodate any significant gas generation; after electrolyte filling an adjacent gas reservoir is required to capture gases that are evolved during the formation process (see Section 11). After de-gassing, this residual gas reservoir is removed and the pouch cell is hermetically sealed. During long term operation, pouch cell formats are also prone to expansion in the event of gas evolution. By contrast, prismatic cells formats generally have sufficient mechanical integrity to accommodate gas generated, and include pressure relief vents in case of over pressure. In both prismatic and pouch cell formats there is a general trend toward larger formats for EV markets, with cells >100 Ah capacity now routine. Managing current and temperature distribution across these large format cells requires robust manufacturing processes and stringent quality control.

11

Formation aging and testing

Formation is performed after the cell is assembled and the electrolyte is added. During this process the cell undergoes one or more slow charge and discharge steps, often with various rests at different temperatures, this allows the electrolyte to wet into the electrode pores and form an interface with the active materials. This wetting, formation and aging procedure can take days or weeks to complete, making it one of the most expensive parts of the manufacturing process.15 During the formation processes, the solid electrolyte interphase (SEI) layers are formed as electrolyte components decompose at the surface of the graphite electrode. With increasing cycle and calendar lifetime the SEI continues to grow. Leakage currents are monitored during aging to ensure the final cell has the correct rated capacity and monitor the interface stability. Formation protocols for each cell format and chemistry are different due to the changes in the electrode materials and architecture. The solid electrolyte interfaces need to be thin, Li+ conductive, and stable at a range of temperatures. The quality of these interfaces is essential to achieving good capacities and cycle life. The optimization of the formation protocol is performed using an empirical matrix of experiments, and due to the high cost in time, and energy, protocols for faster formation are being actively investigated. A typical formation and aging process has two stages. Stage 1 – Electrolyte filling and wetting. In this step, the electrolyte is injected into the cell, and the cell is then held in a partial or full vacuum to remove the gas in the pores of the electrode and encourage the wetting of the electrolyte into the electrode and separator pores. Further wetting can be performed, where the cell is sealed and then left at rest at a specific temperature for hours or days. However, rest periods must be managed as the copper current collector may start to oxidize, depending upon the cell’s positive to negative capacity balance and the material type.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Production

471

Stage 2 – Formation and Aging. The formation step is where the cell is cycled at a low rate to a specific state of charge. After reaching the designated state of charge the cell is rested either at room temperature or elevated temperature, this further encourages wetting of the electrolyte into the electrode and separator. After formation, often gases are produced, these are then removed under a vacuum before a final seal. After wetting and formation, an aging process is applied where the cell is left at rest for up to 3 weeks at a designated temperature, this chemically rearranges the interface composition and stabilizes it. Aging is also sometimes called conditioning and should not be confused with calendar or cycle life aging. The cells can undergo stage 2 multiple times, at different rates and temperatures. However, after each step, an open circuit cell voltage (OCV) is recorded which ascertains any current leakages. Once the cells are formed and aged, they are electrically tested assessed for any defects and sorted into grades based on their capacity. Physical checks include mass, thickness and dimension measurements, visual checks for any tears, dents, protrusions, and tab seal defects. Electrical testing includes capacity check, DCIR, impedance check for resistance. QC-rejected cells are disposed of at zero state of charge where possible, however in some cases, they are discharged down to zero volts, this oxidizes the copper and short circuits the cell and prevents any entry of these cells into the marketplace.

12

Conclusions

This chapter has provided an overview of the key unit operation in Li-ion battery electrode and cell manufacture. Over the past decade, the battery industry has witnessed significant growth which is predicted to persist in the coming decade particularly to service electrification of vehicles and large scale energy storage. Predictions of annual installed output continue to rise significantly, and it is estimated that annual production in 2023 has exceeded 1TWh for the first time. With this scale of manufacture, there is an imperative to continue to optimize established processes, and to develop new ones, particularly to improve manufacturing yield and sustainability. With increasingly demanding performance criteria across a range of applications, there is a parallel need to improved cell and electrode metrology to reduce scrap rates and provide robust quality control.

Acknowledgement The authors gratefully acknowledge their participation in the Faraday Institution Nextrode project. P.R. Shearing was supported by the Department of Science, Innovation and Technology (DSIT) and the Royal Academy of Engineering under the Chair in Emerging Technologies programme (CiET1718/59).

References 1. Grant, P. S.; et al. Roadmap on Li-Ion Battery Manufacturing Research. J. Phys. Energy 2022, 4, 042006. https://doi.org/10.1088/2515-7655/ac8e30. 2. Pardikar, K.; Entwistle, J.; Ge, R. H.; Cumming, D.; Smith, R. Status and Outlook for Lithium-Ion Battery Cathode Material Synthesis and the Application of Mechanistic Modeling. J. Phys. Energy 2023, 5 (2). https://doi.org/10.1088/2515-7655/acc139. 3. Fernandez-Diaz; et al. Mixing Methods for Solid State Electrodes: Techniques, Fundamentals, Recent Advances, and Perspectives. Chem. Eng. J. 2023, 464, 142469. https:// doi.org/10.1016/j.cej.2023.142469. 4. Barnes, H. A. Thixotropy - A review. J. Non-Newtonian Fluid Mech. 1997, 70 (1–2), 1–33. https://doi.org/10.1016/S0377-0257(97)00004-9. Elsevier. 5. Kim, Y.; Park, H.; Warner, J. H.; Manthiram, A. Unraveling the Intricacies of Residual Lithium in High-Ni Cathodes for Lithium-Ion Batteries. ACS Energy Lett. 2021, 6 (3), 941–948. https://doi.org/10.1021/acsenergylett.1c00086. 6. Roberts, S.; Chen, L.; Kishore, B.; Dancer, C. E. J.; Simmons, M. J. H.; Kendrick, E. Mechanism of Gelation in High Nickel Content Cathode Slurries for Sodium-Ion Batteries. J. Colloid Interface Sci. 2022, 627, 427–437. https://doi.org/10.1016/J.JCIS.2022.07.033. 7. Zhang, Y. S.; et al. A Review of Lithium-Ion Battery Electrode Drying: Mechanisms and Metrology. Adv. Energy Mater. 2022, 12 (2), 2102233. https://doi.org/10.1002/ AENM.202102233. 8. Abdollahifar, M.; Cavers, H.; Scheffler, S.; Diener, A.; Lippke, M.; Kwade, A. Insights into Influencing Electrode Calendering on the Battery Performance. Adv. Energy Mater. 2023, 13 (40). https://doi.org/10.1002/aenm.202300973. 9. Diener, A.; Ivanov, S.; Haselrieder, W.; Kwade, A. Evaluation of Deformation Behavior and Fast Elastic Recovery of Lithium-Ion Battery Cathodes Via Direct Roll-Gap Detection during Calendering. Energy Technology 2022, 10 (4). https://doi.org/10.1002/ente.202101033. 10. Haselrieder, W.; Ivanov, S.; Christen, D. K.; Bockholt, H.; Kwade, A. Impact of the Calendering Process on the Interfacial Structure and the Related Electrochemical Performance of Secondary Lithium-Ion Batteries. Ecs Trans. 2013, 50 (26), 59–70. https://doi.org/10.1149/05026.0059ecst. 11. Meyer, C.; Weyhe, M.; Haselrieder, W.; Kwade, A. Heated Calendering of Cathodes for Lithium-Ion Batteries with Varied Carbon Black and Binder Contents. Energy Technology 2020, 8 (2). https://doi.org/10.1002/ente.201900175. 12. Ge, R. H.; Cumming, D. J.; Smith, R. M. Discrete Element Method (DEM) Analysis of lithium Ion Battery Electrode Structures from X-Ray Tomography-the Effect of Calendering Conditions. Powder Technol. 2022, 403. https://doi.org/10.1016/j.powtec.2022.117366. 13. Ngandjong, A. C.; Lombardo, T.; Primo, E. N.; Chouchane, M.; Shodiev, A.; Arcelus, O.; Franco, A. A. Investigating Electrode Calendering and Its Impact on Electrochemical Performance by Means of a New Discrete Element Method Model: Towards a Digital Twin of Li-Ion Battery Manufacturing. J. Power Sources 2021, 485. https://doi.org/10.1016/j. jpowsour.2020.229320. 14. Giménez, C. S.; Schilde, C.; Froböse, L.; Ivanov, S.; Kwade, A. Mechanical, Electrical, and Ionic Behavior of Lithium-Ion Battery Electrodes via Discrete Element Method Simulations. Energy Technology 2020, 8 (2). https://doi.org/10.1002/ente.201900180. 15. Weng, A.; et al. Predicting the Impact of Formation Protocols on Battery Lifetime Immediately after Manufacturing. Joule 2021, 5 (11), 2971–2992. https://doi.org/10.1016/ J.JOULE.2021.09.015.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling Takehiko Okui, Nissan Motor Co., Ltd., Yokosuka, Kanagawa, Japan © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 2.1 2.2 2.3 2.3.1 2.3.2 3 3.1 3.2 4 References Further reading

Introduction Three types of recycling technologies Pyrometallurgical recycling Hydrometallurgical recycling Direct recycling Separation process Healing process Impurities on recycling Influence of impurity Impurity removal Conclusion

473 473 473 474 475 475 478 479 479 479 482 482 483

Abstract Recycling of lithium-ion batteries (LiB) is attracting attention in recent years due to concerns over environmental protection, regulations and resource security resulting from the rapidly expanding market for electric vehicles (EV) and LiB. Recycling of cathode active materials (CAM) containing highly valuable metals such as Li, Ni and Co is particularly important. Recycling technologies for CAM are generally classified into three types: pyrometallurgical, hydrometallurgical, and direct recycling. Although direct recycling has not been industrialized yet, it is considered to be the most efficient technology because CAM is recovered without being decomposed and further CAM production is not necessary. Separation of CAM from LiB components and healing of deteriorated CAM are required in direct recycling. Removal of impurities such as Cu, Al, polyvinylidene difluoride (PVDF) and carbon black is especially important because such impurities can deteriorate CAMs during the healing process.

Glossary Black mass fine powder containing mixtures of valuable metals, which is produced from LiB by mechanical processes. Relithiation lithium insertion into cathode active material recovered from LiB. Cathode electrolyte interphase coating layer formed on the surface of CAM by decomposition of electrolyte. Cation mixing antisite defects in layered cathode active materials caused by site exchange between lithium and transition metal ions.

Key points

• • • •

Overview of major three types of cathode recycling technologies for lithium-ion batteries; pyrometallurgical, hydrometallurgical and direct recycling, are described. Direct recycling is a promising technology, although it has not been industrialized yet and its application is limited. Impurities such as Cu, Al, PVDF and carbon can deteriorate the cathode active material. Several methods for impurity removal without deteriorating CAM are described.

Abbreviations CAM CEI Cyanex272 D2EHPA

472

Cathode active materials Cathode electrolyte interface Bis(2,4,4-trimethylpentyl)phosphinic acid Di(2-ethylhexyl)phosphoric acid

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00161-4

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling

EG EV LCO LiB NMC NMP PC-88A PVDF TEP

1

473

Ethylene glycol Electric vehicles Lithium cobalt(III) oxide Lithium-ion batteries Lithium nickel manganese cobalt oxides N-methyl-2-pyrrolidone (2-ethylhexyl)phosphonic acid mono-2-ethylhexyl ester Polyvinylidene difluoride Triethyl phosphate

Introduction

In recent years – with the rapid expansion of the market for electric vehicles (EV) and the lithium-ion batteries (LiB) used therein – LiB recycling is attracting attention from the perspective of environmental protection, regulations and resource security. In terms of environmental protection, LIBs contain hazardous and toxic substances such as organic electrolytes and heavy metals that must be disposed of carefully. When landfilling LiBs, mechanical damage to the LiB can cause these substances to leak out and pollute the environment. Furthermore, mechanical damage can cause the LiB to short circuit, resulting in an explosion. Therefore, appropriate handling of spent LiBs is essential. In terms of regulations, the new EU battery regulation proposed in 2020 includes articles describing the minimum recycled content of metals in industrial and automotive batteries and minimum recycling efficiencies of recycling processes. If the articles go into effect, recycling of battery grade metals becomes essential. In terms of resources, there are risks of supply deficiency due to the rapid expansion of LiB demand and supply disruption due to the unequal distribution of resources by country. Such risks promote recycling for resource security and have resulted in the recent trend of closed-loop recycling which can create battery grade materials from waste LiB. There are various challenges in realizing a LiB recycling system, such as constructing a collection scheme for spent LiBs, developing an efficient LiB recycling process and building a traceable supply chain for recycled materials.1 This chapter focuses on the recycling process for cathode active materials (CAM) which contain high value metals such as Li, Ni, and Co. There are three major recycling technologies for CAM: pyrometallurgical recycling, hydrometallurgical recycling, and direct recycling.1,2 Pyrometallurgical recycling is the process of recovering the metals contained in LiB as an alloy by heating at high temperatures over 1000
C. Hydrometallurgical recycling is the process of leaching the metals from crushed LiB scrap and separating each metal by chemical methods such as precipitation, solvent extraction, and electrolysis. Most of commercialized LiB recycling processes apply pyrometallurgical recycling, hydrometallurgical recycling or a combination of these two technologies. However, it is currently difficult for these two technologies to achieve efficient recycling processes due to the high energy consumption of high temperature heating or high chemical consumption during leaching and solvent extraction. The profitability of these recycling processes depends on the CAM chemistry and the market conditions of raw materials such as Li, Co, and Ni. The third technology, direct recycling is the process of recovering CAM without decomposing it into an alloy or separate metal compounds, but instead healing the CAM to enable its re-use as a component of a new LiB. Although direct recycling has not been industrialized yet, it is attracting attention as an efficient technology compared to conventional pyrometallurgical and hydrometallurgical recycling because it does not require a CAM manufacturing process. Various studies are underway in order to realize direct recycling.1,3–17 In particular, separating CAM from LiB components, removing impurities such as polyvinylidene difluoride (PVDF), carbon, Al and Cu, and healing deteriorated CAM are key technologies for direct recycling because these technologies greatly affect the characteristics of recycled CAM. This chapter provides an overview of these three major recycling technologies with a particular focus on direct recycling. Impurities can significantly affect the characteristics of recycled CAM. This chapter also discusses the influences of impurities on recycled CAM and technologies of removing impurities in direct recycling process.

2

Three types of recycling technologies

Many industrial cases of pyrometallurgical and hydrometallurgical recycling processes are described in books, patent, corporate websites and other literature.1,3,18 On the other hand, direct recycling is mostly documented as descriptions of specific processing technologies because there are no industrial examples. In this section, an overview of three types of recycling technologies, industrial cases of pyrometallurgical and hydrometallurgical recycling, and importantly, several processing technologies used for direct recycling are described.

474 2.1

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling Pyrometallurgical recycling

Pyrometallurgical recycling is the process of recovering the metals contained in LiB as an alloy by heating LiB scrap at a high temperature. This process has already been industrialized by Umicore, Sumitomo Metal Mining Co., Ltd. and several other companies. The Cu, Ni and Co in LiBs are recoverable metals as an alloy after the high temperature process, while Li is contained in the slag together with Mn, Al and the other metals. Organic components such as electrolyte and separator are utilized as a fuel source in the pyrometallurgical process. Carbon components such as anode active materials and conductive additives are also utilized as carbothermal reducing agents to produce the alloy. The alloy can be further separated into its constituent metals by subsequent hydrometallurgical processes. There are several technologies begin developed to recover Li from the slag.1 Although these burgeoning technologies are currently costly processes, soaring Li prices in recent years and technological advances can boost profitability of Li recovery from the slag. As an example of industrialized pyrometallurgical recycling, Sumitomo Metal Mining Co., Ltd. – a Japanese company developing and producing non-ferrous metals and advanced materials – is operating the recovery of the Cu and Ni from spent LiB through a combination of Cu smelting/refining and the Ni smelting/refining processes. The recovered Ni is recycled into CAM. Recently, Sumitomo Metal Mining Co., Ltd. and Kanto Denka Kogyo Co., Ltd. have jointly developed a new recycling technology capable of recovering Cu, Ni, Co and Li and recycling them into new LiB components. Fig. 1 shows an overview of conventional and improved recycling process flows for LiBs. In the conventional process (a), LiB scraps are put into the converter and Cu is recovered by electrowinning. Ni in the residue is recovered as NiSO4 by additional hydrometallurgical process and used for the CAM production. In contrast, the improved process (b) can recover Cu, Ni, and Co as an alloy with less impurities by using an independent pyrometallurgical refining plant. Cu is separated from the alloy by leaching and electrolysis and recycled to LiB or other components. Ni and Co in the leachate are recycled to raw materials for CAM after the additional hydrometallurgical refining. Furthermore, the Li in the slag is recovered as high purity Li compounds through another hydrometallurgical refining process and used as a raw material for CAM and electrolyte. The major advantage of pyrometallurgical recycling is its simplicity and speed. It does not require the separation of LiB components.1 The process is adaptable to wide variety of batteries.1,3 Furthermore, it can treat large amount of materials in a short time by utilizing the existing smelting equipment.3 On the other hand, the major disadvantage of pyrometallurgical recycling is that the recoverable materials are limited to metals which are easily reduced. Furthermore, the recovered metal is an alloy, which requires additional hydrometallurgical processes in order to separate the individual metals.1,3 Considering the recent trend of closed-loop recycling, the combined process of pyrometallurgical followed by hydrometallurgical methods could be significant in the future.

2.2

Hydrometallurgical recycling

Hydrometallurgical recycling utilizes a series of processes to leach metals from crushed LiBs and separate them using technologies such as precipitation, solvent extraction and electrolysis. The electrolyte is removed by heat treatment, in some cases, before crushing the LiBs. An inert atmosphere is often used during crushing or heat treatment in order to reduce the risk of explosion. Inorganic acids such as H2SO4 and HCl are widely applied for leaching because of their high efficiency and low cost compared to organic acids and

Fig. 1 Schematic flowchart of (A) conventional and (B) improved recycling process by SMM.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling

475

bases.1,3 Each metal ion can be separated by solvent extraction using acidic agents, such as D2EHPA and PC-88A.19 Each separated metal ion can be recovered as either a pure metal by electrowinning, or a metal compound by precipitation. In some cases, the metal solution is used directly for CAM production. Hydrometallurgical recycling can recover Li and Mn as well as Cu, Ni and Co. As an example of industrialized hydrometallurgical recycling, TES-AMM (Singapore) Pte Ltd. – a major recycler of E-Waste and batteries – is operating the recovery of the Cu, Al, Co and Li in spent LiB through serial mechanical separation, leaching and precipitation processes. Fig. 2 shows an overview of the recycling flows for LiB. Spent LiB can go through a series of shredding and hammer milling. The electrolyte evaporates during this process and goes through further treatment. The shredded LiB components are separated by a series of physical separation processes. Plastics and papers are removed by air flow, and the steel is separated by magnetic separation. Cu and Al are separated by a vibrating screen and a powder known as “black mass” is extracted. The black mass goes through a series of chemical processes. After leaching the metals from the black mass and separating the graphite as a solid, Co(OH)2 and Li2CO3 are precipitated sequentially. The aqueous leaching solution is reused after precipitation to help ensure that no wastewater is discharged. Currently, a total recovery rate of 90% with 99% purity of the recovered materials has been achieved. The major advantages of hydrometallurgical recycling are the high recovery-rate and high purity. Based on hydrometallurgical technologies developed for refining of mineral resources, many kinds of metals including Li are recoverable as battery grade materials.1,18 On the other hand, the major disadvantage of hydrometallurgical recycling is the relatively complicated nature of the process. Crushing and mechanical separation are required before hydrometallurgical processes in order to make the black mass. It is necessary to optimize the process conditions depending on the composition of the black mass and the types of impurities contained.18 Nevertheless, hydrometallurgical methods are fundamental to closed-loop recycling and are therefore considered to become the mainstream of LiB recycling.

2.3

Direct recycling

Direct recycling is the process of separating CAM from LiB without decomposing it into an alloy, metal ions or individual metal compounds and healing the CAM for re-use as new LiB components. Although this recycling process has not been industrialized yet, it has the potential to reduce the cost of manufacturing CAM compared to current industrialized recycling processes because it does not require additional CAM synthesis from the recycled materials.4 Fig. 3 shows an overview of the direct recycling process flow. Firstly, spent LiB are crushed by some equipment and broken to fragments. Disassembly and electrode separation are necessary in some cases. Then, the cathode powder is peeled off from the Al foil and separated from the foil and other LiB components. The cathode powder contains binders and conductive additives, such as PVDF and carbon black, which are regarded as impurities for CAM. Fine pieces of foils may also be mixed in the cathode powder. It is necessary to remove such impurities before the healing process. Since the CAM recovered from spent LiB is often degraded to some extent, it is necessary to replenish lost lithium and repair damage to the crystal structure. Several studies on the separation and healing processes for direct recycling are described below.

2.3.1 Separation process Acidic chemicals cannot be used in the direct recycling process because the CAM dissolves in acidic solutions. Furthermore, heat treatment prior to separating PVDF from CAM is not feasible because PVDF decomposes into hydrogen fluoride (HF) above 350
C and HF can decompose the CAM. Therefore, non-acidic solvents and mechanical methods are explored for the separation process.

Fig. 2 Schematic flowchart of recycling process by TES.

476

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling

Fig. 3 Schematic flowchart of direct recycling process.

Any method on direct recycling needs shredding or crushing process as described in the hydrometallurgical recycling process, or electrode separation after disassembling LiB. Table 1 shows several examples of different separation methods in direct recycling. In the mechanical separation method, a series of processes of agitation and froth flotation are studied. CAM can be delaminated from the Al foil and de-agglomerated by agitating the cathode sheets with water in a high-shear blender. R. Zhan et al. studied the separation of CAM by agitation and presumed three mechanisms of its effect: (1) a breakage of intermolecular bonds between PVDF and CAM, (2) a breakage of metallic bonds within CAM, and (3) a breakage of covalent bonds within PVDF molecules.5 After sieving fine particles and foils, the fine particles including CAM are separated from PVDF and carbon by taking advantage of the differences in surface hydrophobicity in the subsequent froth flotation process. Hydrophobic PVDF and carbon attach to air bubbles and float in froth layers, while hydrophilic CAM leaves in tailing. H. Shin et al. demonstrated CAM separation from the mixture of cathode and anode materials by water-based froth flotation, achieving a CAM recovery yield of more than 80%.6 Another example of mechanical separation involves ultrasonication in water. C. Lei et al. demonstrated a 99.5% recovery of CAM by using high-power ultrasound in a 0.1 mol L−1 NaOH aqueous solution. The cavitation generated by the ultrasound can break the adhesive bonds between the CAM and the current collector and delaminate the cathode from the Al foil. Surface wettability, binder solubility, and the presence of interfacial voids affect the efficiency of the delamination process.7 The mechanical methods described above can separate the CAM without deterioration because heat treatment and acid chemicals are not applied. Furthermore, the scale up is relatively easy and could potentially results in low-cost processes because such methods are commonly used in various industries such as chemical and mining fields. Dissolving the binder using an organic solvent is an alternative method for CAM separation. N-methyl-2-pirrolidone (NMP) dissolves PVDF and is therefore widely used for cathode electrode production. However, it is said to be generally toxic and harmful to human. Y. Bai et al. demonstrated an efficient separation process by using triethyl phosphate (TEP) solvent which presents less risk to human and environment. Spent cathode sheets were placed into the TEP at 150
C with subsequent stirring and bath sonication for 30 min each; the cathode materials were then separated from Al foils through sieving, filtration and centrifugation. The mass fraction of PVDF in the recovered cathode powder was 0.41%, which is less than one tenth of the initial content (5%). Additionally, the mass fraction of carbon decreased from 5% to 1.79% because some of the carbon particles remained dispersed in the TEP after centrifugation.8 Another solvent-based method involves deactivating the bonding between PVDF and Al foil without dissolving the PVDF. PVDF can form hydrogen bonds with the Al foil, which typically exhibits a thin passivation layer of Al2O3. Although ethylene glycol (EG) does not dissolve PVDF, it can displace the PVDF from the Al foil, as it forms stronger hydrogen bonds with Al2O3 than PVDF does.9 The PVDF and the carbon are not removed from the cathode composite by this method and the recovered electrode does not exhibit any changes in crystal structure, morphology or electrochemical characteristics. Therefore, this method may be suitable for the direct recycling from manufacturing scraps in cases where the impurity removal and the healing of CAM is not required. If the CAM is not degraded, it may be possible to re-make a slurry from the recovered electrode without removing the remaining PVDF and carbon. The solvent-based methods described above can also easily recover the Al foil current collectors because they do not require any crushing process. On the other hand, solvent costs are relatively expensive compared to mechanical process costs. It is necessary to establish a solvent recovery and reuse system in order to industrialize these methods.

Table 1

An overview of separation methods in direct recycling. Feature

Test sample

Test condition

Test result

References

Agitation and froth flotation

Water-based mechanical method. Agitation breaks the bond between CAM and PVDF. Froth flotation separates CAM from PVDF and carbon by using the difference in surface hydrophobicity. Water-based mechanical method. Cavitation breaks the bond between CAM and current collector within 0.5 s.

Cathode composite from End-of-life LiB.

Agitating 50–100 g cathode composite with 1 L water for 16 min by commercial blender followed by froth flotation using kerosene as collector. 20 kHz and 70 W cm−2 ultrasound in 0.1 M NaOH solution.

94% recovery yield and 98.3% CAM purity.

5

99.5% recovery yield.

7

Residual PVDF and carbon were 0.41 wt% and 1.79%. Separated CAM exhibited no morphological and crystalline change. Completely peeled off from Al foil.

8

Ultrasonication PVDF dissolution by solvent

Using organic solvent such as NMP and TEP. TEP has less risk to human and environment.

Electrode delamination by EG

EG forms stronger hydrogen bonding with Al2O3 at the surface of Al foil than PVDF.

Cathode (LMO/NMC) sheet from Nissan LEAF spent LIB. Cathode (NMC622) sheet from 1000 charge/ discharge cycled cell. Cathode (NMC532) sheet from 1000 charge/ discharge cycled cell.

Stirring into TEP at 150
C for 30 min followed by bath sonication for 30 min. Stirring into EG at 160
C for several seconds.

9

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling

Method

477

478

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling

2.3.2 Healing process Recovered CAM from spent LiB are degraded to some extent and need to be healed before re-use. There are various healing methods being studied such as solid state,10 hydrothermal,11,12 molten salt,13 electrochemical,14 and microwave plasma methods. Several studies have shown that chemically delithiated CAM can be regenerated to almost the same electrochemical characteristics as pristine CAM after the healing process, indicating that relithiation is complete.10,12 However, the CAM from spent LiB exhibits some microcracks, phase transformations, and cathode electrolyte interphase (CEI) which deteriorate the electrochemical characteristics.20 As each spent LiB cell can differ in the extent and mode of degradation, there have only been a few studies on healing such CAM from spent LiBs. Instead, current healing technologies are only applicable to direct recycling of manufacturing scraps where there are no cracks or crystal structure changes in the CAM. Table 2 shows several examples of different healing methods in direct recycling. In the solid state method, a mixed powder of CAM and Li precursor is calcined at a high temperature (c.a. 500
C or higher). The solid state method is similar to the calcination step in the current CAM manufacturing process. Li2CO3 or lithium hydroxide monohydrate (LiOH.H2O) is often used as the Li precursor. The Li precursor melts during calcination and accelerates the reaction. B. J. Ross et al. studied the healing of the chemically delithiated NMC111 by the solid state method and demonstrated that the NMC111 without PVDF binder is relithiated successfully by mixing LiOH.H2O with a mass fraction of 10% and annealing at 500
C. This healed NMC111 showed almost the same electrochemical performance as pristine NMC111. On the other hand, the chemically delithiated NMC111 containing 3% and 5% mass fraction of PVDF and carbon black was not fully healed by heating at 500
C or 925
C and showed slightly poorer electrochemical performance.10 Although the excess Li could be used to react with PVDF, forming LiF and preventing the formation of HF which decomposes NMC, some fluorine (F) is potentially doped into the NMC, changing its characteristics. In the hydrothermal method, CAM powder is placed in LiOH aqueous solution and heated for several hours. Hydrothermal treatment at 220
C is necessary in order to recover the same Li composition ratio as pristine NMC. Short annealing is also necessary after hydrothermal treatment in order to recover the crystallinity of the NMC. Y. Shi et al. demonstrated that spent NMC after charge/discharge cycle test was regenerated under these conditions. The electrochemical performances of healed NMC111 and NMC532 were almost same as those of pristine samples.11 In the molten salt method, CAM powder is placed in Li+ molten salt solution and treated for several hours. Among various Li+ molten salts, the mixture of LiNO3 and LiOH in a molar ratio of 3:2 has an especially low melting point of around 175
C. Y. Shi et al. studied the healing of NMC532 by using this eutectic molten salt. The spent NMC532 after charge/discharge cycle test were mixed with an excess amount of Li salt composed of LiNO3 and LiOH in a molar ratio of 3:2, and then heated to 300
C for four hours. After washing with deionized water to remove the Li salt, the NMC532 was mixed with 5% excess amount of Li2CO3 and sintered at 850
C in O2 atmosphere for four hours. The authors demonstrated that this regenerated NMC532 exhibited almost the same electrochemical performance as pristine NMC532 as well as the recovery of the same chemical composition and crystal structure vs. pristine one.13 In the electrochemical method, the cathode sheet is directly treated by an electrochemical device. Li. Zhang et al. studied this healing method using lithium cobalt(III) oxide (LCO) cathode sheets recovered after 350 cycles of charge/discharge test. (The capacity of the LiB decreased by approximately 20%.) In a three-electrode configuration, the LCO sheet served as the cathode and Pt plate as the anode in addition to an Ag/AgCl reference electrode; tests were conducted in a Li2SO4 solution with a concentration

Table 2

An overview of healing methods in direct recycling.

Method

Feature

Test sample

Test condition

Test result

References

Solid state

Mixing CAM and Li compound as powder. Same method as pristine CAM manufacturing. Stoichiometric ratio of Li addition should be considered. Residual PVDF and carbon are decomposed. Using LiOH solution as Li source. No need to consider the stoichiometric ratio of Li addition. Solid state annealing is necessary after hydrothermal. Using molten salt as Li source. No need to consider the stoichiometric ratio of Li addition.

Chemically delithiated NMC111

Mixed with 10 wt% LiOH.H2O followed by calcination at 500
C.

Same capacity at 2C as pristine NMC111

10

NMC532 from 200 charge/ discharge cycled cell. NMC532 from 400 charge/ discharge cycled cell.

LiOH solution, 220
C for 4 h followed by annealing with a small amount of excess Li at 850
C for 4 h. Mixed with LiOH-LiNO3 molten salt (molar ratio of 2:3), 300
C for 4 h followed by annealing with an excess 5% Li2CO3 at 850
C for 4 h. Electrochemical insertion of Li with 1 M Li2SO4 solution at 0.42 mAcm−2, followed by annealing at 700
C for 2 h.

Same capacity and cyclability at 5C as pristine NMC532 Same capacity and cyclability at 5C as pristine NMC532

11

Hydrothermal

Molten salt

Electrochemical

Electrochemical insertion of Li+ in the form of electrode sheet. Post annealing is necessary after electrochemical process.

Cathode (LCO) sheet from 350 charge/ discharge cycled cell.

97% capacity and almost same cyclability at C/5 as pristine NMC532

13

14

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling

479

ranging from 0.1 to 1.0 mol L−. A constant cathodic current density between −0.12 to −0.42 mA cm−2 was applied. In the case of a 1 mol L−1 Li2SO4 solution and a current density of −0.42 mA cm−2, the potential of LCO electrode gradually decreased and eventually stabilized after one hundred minutes, indicating that the electrochemical Li+ insertion had completed. Finally, the LCO cathode sheet was annealed at 700
C for two hours to produce LCO powder. The electrochemical performance of the regenerated LCO showed around 97% initial capacity compared to the pristine LCO.14

3

Impurities on recycling

Battery grade raw materials are required for LiB manufacturing in order to prevent LiB failure related to short circuits and deterioration of cycling stability. The overall purity level of each compound and the limit of each impurity contained in the compound are different depending on the compounds, the producer and the type of impurity.21 In order to achieve closed-loop recycling, it is important to reduce impurities to a battery grade level. In this section, influences of impurities on CAM and technologies of removing impurities are described.

3.1

Influence of impurity

Residual impurities derived from spent LiB or recycling processes can have various influences on LiB characteristics. Cu foils are generally used for the anode current collectors and steel is used for cell casings and recycling equipment. When such metal particles contaminate cathode, they dissolve in the electrolyte, migrate from the cathode through the separator and deposit onto the anode. This can lead to dendrite formation and cause an internal short circuit.22 Cu or Al impurities can also change the characteristics of the CAM when they are incorporated into the crystal structure of the CAM. For example, when a 1% mol fraction of Cu ions are introduced into NMC622, Cu ions can partially substitute the Ni2+ site and decrease the cation mixing between Li+ and Ni2+, improving the electrochemical characteristics of the cathode. In contrast, excess amount of Cu ions deteriorates the cathode characteristics due to the increase of the cation mixing between Li+ and Cu2+.22 The similar phenomenon is observed in the case of the Al impurity. When 0.2% mol fraction of Al ions introduced into NMC622 during precursor synthesis, Al ions substitute the Ni2+ and reduce the ratio of Ni2+ and Ni3+. Consequently, the cation mixing between Li+ and Ni2+ decreases and the electrochemical characteristics are improved. In contrast, an excess mol fraction of Al ions (5%) hinders the Li+ diffusion and deteriorates the electrochemical characteristics.23 Although these examples do not necessarily indicate that Cu and Al impurities need to be completely removed, the amounts of Cu and Al impurities need to be controlled. PVDF and carbon black are generally used for the cathode components as the binder and conductive additive. Such materials do not remain as impurities in the case of pyrometallurgical recycling because PVDF decompose to HF gas and carbon is oxidized to CO2 at high temperatures. Similarly, PVDF and carbon black are able to be easily removed by filtration in hydrometallurgical recycling processes during solid-liquid separations. In contrast, direct recycling recovers CAM as a solid powder and therefore impurities need to be separated as liquid or gas phase in order to purify the recovered CAM to battery grade. However, PVDF and carbon black may deteriorate the CAM during heat treatment. For example, HF generated by decomposition of PVDF can react with NMC as shown in Eqs. (1) and (2). F atoms are incorporated into the NMC by partially substituting for O atoms in Eq. (1), and F atoms pull Li from the NMC and form a LiF coating in Eq. (2). The surface structure of the NMC can change to a spinel or rock salt phase as a result of the latter reaction.10 This phase transition causes capacity loss and increases the resistance of NMC. 2LiMO2 + 2HF ! 2LiMO1.5F + H2O

(1)

4LiMO2 + 4HF + O2 ! 4MO2 + 4LiF + 2H2O

(2)

(M; Transition metals such as Ni, Mn, Co) Carbon impurities in NMC precursor can suppress the oxidation of transition metals on the NMC particle surface during thermal synthesis and increase the cation mixing and lattice distortion in NMC, which deteriorates the electrochemical characteristics.24 As such, the valence of transition metals in NMC may decrease during heating in the direct recycling process if thermal synthesis method is applied. These facts explain the necessity of impurity removal originating from LiB components.

3.2

Impurity removal

In current hydrometallurgical recycling, metal impurities are removed through the separation of each valuable metal from the leachate by several means, such as chemical precipitation, solvent extraction and electrolytic deposition.3 Chemical precipitation utilizes the differences in solubility of each metal compound in order to separate them. Cu, Fe and Al ions are generally removed by this method. Cu ions can be reduced and precipitated by Fe, which is known as cementation. Fe and Al ions can be precipitated by regulating pH of the solution using some neutralizer such as NaOH and Na2CO3. When NaOH is used, 99.75% of Fe and 93.1% of Al are precipitated as hydroxides at pH 4.8. In the case of Na2CO3, 99.6% of Fe and 96.7% of Al are precipitated as carbonates at pH 5.7. However, around 10% of valuable metals such as Ni, Co and Li are co-precipitated and lost from the solution in either case.25 Achieving both high purity and high yield is an ongoing issue.

480

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling

Solvent extraction is a method for selectively separating specific metal ions in an aqueous solution into an organic phase by bringing an immiscible organic solvent containing extractants into contact with the aqueous solution and controlling the pH. The target metal ions migrate to the organic phase due to the extractant, while other metal ions containing impurities remain in the aqueous phase. Acid extractants such as D2EHPA, PC-88A and Cyanex272 are widely used.3,19 Extraction rates of Mn2+, Co2+, Ni2+ and Li+ with D2EHPA and PC-88A from aqueous solution are shown in Fig. 4. D2EHPA is suitable for extracting Mn and PC-88A is useful for extracting Co after Mn separation.19 A countercurrent multistage mixer-settler is widely used for solvent extraction. Although it can achieve bulk processing, a large footprint and high cost are required. As an alternative to the mixer-settler, the emulsion flow method has been attracting attention in recent years because it has a smaller footprint and lower cost compared to the mixer-settler. Acidic solutions are generally used in hydrometallurgical recycling as mentioned above. By contrast, acid cannot be used in direct recycling because CAM dissolves to acid. Direct recycling requires the removal of impurities in liquid or gas phase while maintaining CAM as a solid under neutral or alkaline conditions. F. Tsang and P. Hailey invented the method for removing Cu and Al from the CAM by using an alkaline aqueous solution with a pH value higher than 10 in conjunction with an oxidizing agent and complexing agent.15 In this example, Al in LiOH aqueous solution dissolves as following reactions: anodic reaction Al(s) ! Al3+ + 3e−

(3)

3H2O + 3e- ! 3OH- + 3/2H2(g)

(4)

Al (aq) + 3OH (aq) ! Al(OH)3(s)

(5)

Al(OH)3(s) + OH-(aq) ! Al(OH)4-(aq)

(6)

cathodic reaction

3+

-

The solubility of Al(OH)3 depends on the pH of the solution and pH 10 or higher is preferable. Cu dissolves as following reactions: anodic reaction. Cu(s) ! Cu2+(aq) + 2e-

(7)

1/2O2 + 2H2O + 2e- ! 2OH-

(8)

Cu2+(aq) + 4NH3(aq) ! Cu(NH3)42+(aq)

(9)

cathodic reaction.

O2 is consumed for the dissolution of Cu and Cu(OH)2 precipitates without a complexing agent. Therefore, O2 gas and NH4OH are added to the LiOH aqueous solution as oxidizing agent and complexing agent for the dissolution of Cu.

Fig. 4 Extraction rates of Mn2+, Co2+, Ni2+ and Li+ with D2EHPA and PC-88A from aqueous solution. From Horai, K.; Shibata, J.; Murayama, N.; Koyanaka, S.; Niinae, M. Recycling Technology for Lithium Ion Battery by Crushing and Classification, and Hydrometallurgical Process. J. Japan Inst. Met. Mater. 2014, 78: 250–257.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling

481

It is necessary to remove PVDF and carbon by means other than heat treatment in order to prevent the generation of HF or reduction of CAM. PVDF removal can be performed by dissolving in the organic solvent such as NMP and TEP, which is mentioned in Section 2.8 Another method for PVDF removal is to use molten salt. Y. Ji et al. demonstrated that LiOH-LiNO3 eutectic system in the molar ratio of 2:3 decomposes PVDF into alkylamines, LiF and H2O at 260
C as shown in Fig. 5.16 The melting point of PVDF and LiOH-LiNO3 are around 160
C and 175
C respectively, resulting in liquid-liquid contact between PVDF and LiOH-LiNO3 and facilitating the decomposition of PVDF. Although HF forms during PVDF decomposition, it is captured by LiOH to form LiF and does not deteriorate CAM. Molten salt is also applicable to carbon removal. H. Yang et al. demonstrated that the molten salt system of LiOH-KOH (molar ratio of 3:7) with LiNO3 and O2 as oxidants can remove carbon as well as PVDF. They showed the possible reactions and the corresponding Gibbs free energies of carbon oxidation as following equations and Fig. 617: C + O2 ¼ CO2

(10)

C + 4LiNO3 ¼ 2Li2O +4NO2 + CO2

(11)

CH2CF2 + 2O2 ¼ 2CO2 + 2HF

(12)

CH2CF2 + 8LiNO3 ¼ 4Li2O + 8NO2 + 2HF + 2CO2

(13)

The Gibbs free energy of Eqs. (10) and (11) are below zero when the temperature is above 300
C. This indicates that adding LiNO3 and air blow into the molten salt can facilitate the carbon oxidation reaction above 300
C. Similarly, the PVDF decomposition is accelerated at the temperature where the Gibbs free energy of Eqs. (12) and (13) are negative.

LiOH-LiNO3 eutectic system Physical change (175°C): Melting

Chemical change (260°C): Decomposition

Li+ PVDF PVDF OH–

NO3–

NMC Li+

NMC

LMO

OH–

H

F

C

C

NO3–

Li+ PVDF LMO Li+

NO3– + H

F

C

C

H2O

LiF

Alkylamines

–

H+

Al foil

F OH–

+

Li

Fig. 5 Decomposition mechanism of PVDF in LiOH-LiNO3 eutectic system. From Ji, Y.; Jafvert, C. T.; Zhao, F. Recovery of Cathode Materials from Spent Lithium-Ion Batteries Using Eutectic System of Lithium Compounds. Resour. Conserv. Recycl. 2021, 170, 105551.

100

DrGpT (ë4.18kJ/mol)

Eqs. (11)

0

'rGTT= 0

–100 Eqs. (10) Eqs. (13)

–200 Eqs. (12)

–300 0

200

400 600 Temperature (°C)

800

1000

Fig. 6 Gibbs free energy of reaction in Eqs. (10) to (13). Modified from Yang, H.; Deng, B.; Jing, X.; Li, W.; Wang, D. Direct Recovery of Degraded LiCoO2 Cathode Material from Spent Lithium-Ion Batteries: Efficient Impurity Removal Toward Practical Applications. Waste Manag. 2021, 85–94.

482

4

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling

Conclusion

LiB recycling is in an incipient stage where different processes are being proposed and industrialized. Some current recycling technologies require a series of complex processes to separate valuable metals, purify them to battery grade, and apply the recovered metals to CAM manufacturing in order to achieve a closed-loop recycling system. Direct recycling, which is the process of recovering as CAMs and reusing them as LiB components, is one of the solutions for efficient LiB recycling. The two key processes in direct recycling, separating CAM from LiB components and healing the degraded CAM, require the unique technologies which are significantly different from conventional pyrometallurgical and hydrometallurgical recycling processes. In order to separate the CAM without deteriorating its characteristics, various methods for separation and impurity removal have been proposed instead of heat treatment and acid leaching. Several methods for healing CAM such as solid state, hydrothermal and molten salt have been proposed; however, case studies on healing CAMs of various chemistries and degradation states are still insufficient. Based on these circumstances, direct recycling from manufacturing scraps is expected to be industrialized first. The LiB market is expanding rapidly, and the importance of the LiB recycling industry is increasing accordingly. It is hoped that this chapter will be of some help in realizing a circular economy.

References 1. Chen, M.; Ma, X.; Chen, B.; Arsenault, R.; Karlson, P.; Simon, N.; Wang, Y. Recycling End-of-Life Electric Vehicle Lithium-Ion Batteries. Joule 2019, 3, 2622–2646. 2. Dunn, J. B.; Gaines, L.; Barnes, M.; Sullivan, J. L.; Wang, M. Material and Energy Flows in the Materials Production, Assembly, and End-Of-Life Stages of the Automotive Lithium-Ion Battery Life Cycle; OSTI.GOV: United States, 2014 (ANL/ESD/12-3 Rev.). 3. Lin, X.; Wang, X.; Liu, G.; Zhang, G. Recycling of Power Lithium-Ion Batteries: Technology, Equipment, and Policies; John Wiley & Sons, 2022. 4. Xu, P.; Dai, Q.; Gao, H.; Liu, H.; Zhang, M.; Li, M.; Chen, Y.; An, K.; Meng, Y. S.; Liu, P.; Li, Y.; Spangenberger, J. S.; Gaines, L.; Lu, J.; Chen, Z. Efficient Direct Recycling of Lithium-Ion Battery Cathodes by Targeted Healing. Joule 2020, 4, 2609–2626. 5. Zhan, R.; Payne, T.; Leftwich, T.; Perrine, K.; Pan, L. De-Agglomeration of Cathode Composites for Direct Recycling of Li-Ion Batteries. Waste Manag. 2020, 105, 39–48. 6. Shin, H.; Zhan, R.; Dhindsa, K. S.; Pan, L.; Han, T. Electrochemical Performance of Recycled Cathode Active Materials Using Froth Flotation-Based Separation Process. J. Electrochem. Soc. 2020, 167, 020504. 7. Lei, C.; Aldous, I.; Hartley, J. M.; Thompson, D. L.; Scott, S.; Hanson, R.; Anderson, P. A.; Kendrick, E.; Sommerville, R.; Ryder, K. S.; Abbott, A. P. Lithium ion Battery Recycling Using High-Intensity Ultrasonication. Green Chem. 2021, 23, 4710–4715. 8. Bai, Y.; Essehli, R.; Jafta, C. J.; Livingston, K. M.; Belharouak, I. Recovery of Cathode Materials and Aluminum Foil Using a Green Solvent. ACS Sustainable Chem. Eng. 2021, 9, 6048–6055. 9. Bai, Y.; Muralidharan, N.; Li, J.; Essehli, R.; Belharouak, I. Sustainable Direct Recycling of Lithium-Ion Batteries Via Solvent Recovery of Electrode Materials. ChemSusChem 2020, 13, 5664–5670. 10. Ross, B. J.; LeResche, M.; Liu, D.; Durham, J. L.; Dahl, E. U.; Lipson, A. L. Mitigating the Impact of Thermal Binder Removal for Direct Li-Ion Battery Recycling. ACS Sustainable Chem. Eng. 2020, 8, 12511–12515. 11. Shi, Y.; Chen, G.; Liu, F.; Yue, X.; Chen, Z. Resolving the Compositional and Structural Defects of Degraded LiNixCoyMnzO2 Particles to Directly Regenerate High-Performance Lithium-Ion Battery Cathodes. ACS Energy Lett. 2018, 3, 1683–1692. 12. Xu, P.; Yang, Z.; Yu, X.; Holoubek, J.; Gao, H.; Li, M.; Cai, G.; Bloom, I.; Liu, H.; Chen, Y.; An, K.; Pupek, K. Z.; Liu, P.; Chen, Z. Design and Optimization of the Direct Recycling of Spent Li-Ion Battery Cathode Materials. ACS Sustainable Chem. Eng. 2021, 9, 4543–4553. 13. Shi, Y.; Zhang, M.; Meng, Y. S.; Chen, Z. Ambient-Pressure Relithiation of Degraded LixNi0.5Co0.2Mn0.3O2 (0 < x < 1) via Eutectic Solutions for Direct Regeneration of Lithium-Ion Battery Cathodes. Adv. Energy Mater. 2019, 1900454. 14. Zhang, L.; Xu, Z.; He, Z. Electrochemical Relithiation for Direct Regeneration of LiCoO2 Materials from Spent Lithium-Ion Battery Electrodes. ACS Sustainable Chem. Eng. 2020, 8, 11596–11605. 15. Tsang, F.; Hailey, P. Method for Removing Copper and Aluminum from an Electrode Material, and Process for Recycling Electrode Material from Waste lithium-Ion Batteries; US 10 103–413 B2, 2018. 16. Ji, Y.; Jafvert, C. T.; Zhao, F. Recovery of Cathode Materials from Spent lithium-Ion Batteries Using Eutectic System of Lithium Compounds. Resour. Conserv. Recycl. 2021, 170, 105551. 17. Yang, H.; Deng, B.; Jing, X.; Li, W.; Wang, D. Direct Recovery of Degraded LiCoO2 Cathode Material from Spent lithium-Ion Batteries: Efficient Impurity Removal toward Practical Applications. Waste Manag. 2021, 85-94. 18. Velázquez-Martínez, O.; Valio, J.; Santasalo-Aarnio, A.; Reuter, M.; Serna-Guerrero, R. A Critical Review of Lithium-Ion Battery Recycling Processes from a Circular Economy Perspective. Batteries 2019, 5, 68. 19. Horai, K.; Shibata, J.; Murayama, N.; Koyanaka, S.; Niinae, M. Recycling Technology for Lithium ion Battery by Crushing and Classification, and Hydrometallurgical Process. J. Japan Inst. Met. Mater. 2014, 78, 250–257. 20. Geldasa, F. T.; Kebede, M. A.; Shura, M. W.; Hone, F. G. Identifying Surface Degradation, Mechanical Failure, and Thermal Instability Phenomena of High Energy Density Ni-Rich NCM Cathode Materials for Lithium ion Batteries: A Review. RSC Adv. 2022, 12, 5891. 21. Nasser, O. A.; Petranikova, M. Review of Achieved Purities after Li-Ion Batteries Hydrometallurgical Treatment and Impurities Effects on the Cathode Performance. Batteries 2021, 7, 60. 22. Zhang, R.; Meng, Z.; Ma, X.; Chen, M.; Chen, B.; Zheng, Y.; Yao, Z.; Vanaphuti, P.; Bong, S.; Yang, Z.; Wang, Y. Understanding Fundamental Effects of Cu Impurity in Different Forms for Recovered LiNi0.6Co0.2Mn0.2O2 Cathode Materials. Nano Energy 2020, 78, 105214. 23. Zhang, R.; Zheng, Y.; Yao, Z.; Vanaphuti, P.; Ma, X.; Bong, S.; Chen, M.; Liu, Y.; Cheng, F.; Yang, Z.; Wang, Y. Systematic Study of Al Impurity for NCM622 Cathode Materials. ACS Sustainable Chem. Eng. 2020, 8, 9875–9884. 24. Zheng, Y.; Zhang, R.; Vanaphuti, P.; Fu, J.; Yang, Z.; Wang, Y. Unveiling the Influence of Carbon Impurity on Recovered NCM622 Cathode Material. ACS Sustainable Chem. Eng. 2021, 9, 6087–6096. 25. Wang, H.; Friedrich, B. Development of a Highly Efficient Hydrometallurgical Recycling Process for Automotive Li-Ion Batteries. J. Sustain. Metall. 2015, 1, 168–178.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Recycling

483

Further reading 1. Tawonezvi, T.; Nomnqa, M.; Petrik, L.; Bladergroen, B. J. Recovery and Recycling of Valuable Metals from Spent Lithium-Ion Batteries: A Comprehensive Review and Analysis. Energies 2023, 16, 1365. 2. Wang, S.; Tian, Y.; Zhang, X.; Yang, B.; Wang, F.; Xu, B.; Liang, D.; Wang, L. A Review of Processes and Technologies for the Recycling of Spent Lithium-ion Batteries. Mater. Sci. Eng. A 2020, 782, 022025. 3. Pinegar, H.; Smith, Y. R. Recycling of End-of-Life Lithium Ion Batteries, Part I: Commercial Processes. J. Sustain. Metall. 2019, 5, 402–416. 4. European commission. Proposal for a REGULATION OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL concerning batteries and waste batteries, repealing directive 2006/66/ec and amending regulation (EU) No 2019/1020; 2020. 5. Kotaich, K.; Sloop, S. E. Recycling | Lithium and Nickel–Metal Hydride Batteries. In Encyclopedia of Electrochemical Power Sources; 2009; pp. 188–198. 6. Qu, G.; Li, B.; Wei, Y. A Novel Approach for the Recovery and Cyclic Utilization of Valuable Metals by Co-Smelting Spent Lithium-Ion Batteries with Copper Slag. Chem. Eng. J. 2023, 451, 138897. 7. Gaines, L.; Dai, Q.; Vaughey, J. T.; Gillard, S. Direct Recycling R&D at the ReCell Center. Recycling 2021, 6, 31. 8. Peng, C.; Chang, C.; Wang, Z.; Wilson, B. P.; Liu, F.; Lundström, M. Recovery of High-Purity MnO2 from the Acid Leaching Solution of Spent Li-Ion Batteries. JOM 2020, 72, 790–799. 9. Cheng, L.; Tang, X.; Zhang, Y.; Li, L.; Zeng, Z.; Zhang, Y. Process for the Recovery of Cobalt Oxalate from Spent Lithium-Ion Batteries. Hydrometallurgy 2011, 108, 80–86. 10. Sloop, S.; Crandon, L.; Allen, M.; Koetje, K.; Reed, L.; Gaines, L.; Sirisaksoontorn, W.; Lerner, M. A Direct Recycling Case Study from a lithium-Ion Battery Recall. Sustain. Mater. Technol. 2020, 25, e00152. 11. Sloop, S. E.; Crandon, L.; Allen, M.; Lerner, M. M.; Zhang, H.; Sirisaksoontorn, W.; Gaines, L.; Kim, J.; Lee, M. Cathode Healing Methods for Recycling of Lithium-Ion Batteries. Sustain. Mater. Technol. 2019, 22, e00113. 12. Chang, Z. R.; Yu, X.; Tang, H. W.; Yuan, X. Z.; Wang, H. Synthesis of LiNi1/3Co1/3Al1/3O2 Cathode Material with Eutectic Molten Salt LiOH-LiNO3. Powder Technol. 2011, 207, 396–400. 13. Ardia, P.; Stallone, S.; Cericola, D. A Quantification Method for Fe Based Particle Contaminants in High Purity Materials for Lithium-Ion Batteries. Talanta 2021, 224, 121827. 14. Marshall, J. E.; Zhenova, A.; Roberts, S.; Petchey, T.; Zhu, P.; Dancer, C. E. J.; McElroy, C. R.; Kendrick, E.; Goodship, V. On the Solubility and Stability of Polyvinylidene Fluoride. Polymers 2021, 13, 1354.

Relevant websites 1. 2. 3. 4.

www.smm.co.jp/en—Sumitomo Metal Mining Co., Ltd. www.tes-amm.com—TES-AMM (Singapore) Pte Ltd. https://www.6kinc.com—6K, Inc. https://emulsion-flow.tech/—Emulsion Flow Technologies Ltd.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market Tsuyoshi Nishimura, Masamichi Yamaguchi, and Kai Shimozono, Fuji Keizai Co., Ltd., Osaka, Japan © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 3 3.1 3.1.1 3.1.2 3.1.3 3.1.4 3.1.5 4 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 5 5.1 5.1.1 5.1.2 5.1.3 6 6.1 6.2 6.3 7 7.1 7.1.1 7.1.2 7.1.3 8 9 9.1 9.1.1 9.1.2 9.1.3 10 10.1 10.1.1 10.1.2 10.1.3 11 11.1 11.1.1 11.1.2 11.1.3 12 12.1 12.1.1 12.1.2 12.1.3

484

Introduction Primary batteries and secondary batteries LIB application market - Consumer electronics market Trends from around 2010 to the early 2020s Until 2011 2012–14 (Significant changes in the smartphone market) 2013–15 (Wearable devices announced and launched consecutively) Since 2016 (true wireless stereo (TWS) market expansion) Late 2010s Impact of COVID-19 TWS Smartphone Tablet PC Notebook PC Wearable device Power tool and garden tool Vacuum cleaner Electric motorcycle LIB application market - xEV market Trends from around 2010 to the early 2020s Since 2009 Mid-2010s Around 2020 Impact of COVID-19 BEV HEV PHEV LIB application markets - ESS, UPS, BTS markets Trends from around 2010 to the early 2020s Since 2011 Around 2018 Around 2020 Impact of COVID-19 LIB market Trends from around 2010 to the early 2020s Around 2010 Mid-2010s Around 2020 Battery material market trends - Cathode active material market Trends from around 2010 to the early 2020s Early 2010s Late 2010s Around 2020 Battery material market trends - Anode active material market Trends from around 2010 to the early 2020s Early 2010s Late 2010s Around 2020 Battery material market trends – Electrolyte solutions market Trends from around 2010 to the early 2020s Early 2010s Late 2010s Around 2020

Encyclopedia of Electrochemical Power Sources, 2nd Edition

486 487 487 487 487 488 488 488 488 488 488 488 488 488 489 489 489 489 489 489 489 489 490 490 490 490 490 490 491 491 491 491 491 491 491 491 491 492 492 493 493 493 494 495 495 495 495 495 495 495 495 495 496

https://doi.org/10.1016/B978-0-323-96022-9.00152-3

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market 13 13.1 13.1.1 13.1.2 13.1.3 14 14.1 15 16 17 18 19 20 21 22 References

Battery material market trends – Separator market Trends from around 2010 to the early 2020s Early 2010s Late 2010s Since 2020 Significance of ASSB development and market trends Significance and advantages of developing ASSB Types of ASSBs Market trend of ASSBs Recent development and commercialization trends of sulfide-based ASSBs Market outlook for sulfide-based ASSBs in the 2020s Long-term forecast for sulfide-based ASSBs after 2030 Market outlook for oxide-based ASSBs in the 2020s Long-term forecast for oxide-based ASSBs after 2030 Conclusion

485 496 496 496 496 496 496 496 496 498 499 499 500 500 501 502 502

Abstract This paper provides an overview of the past, present, and future of the lithium-ion secondary battery (LIB) market from approximately 2010–25 (2035 for the next-generation battery market). LIBs have served as versatile energy devices for consumer electronics, power applications, xEVs (general terms for electrified vehicles), and energy storage systems (ESSs) since their introduction in the early 1990s. The period around 2010 saw the global economic recession triggered by the financial crisis in September 2008, the Greek financial crisis in October 2009, and the Great East Japan Earthquake in March 2011. The first practical electric vehicles (EVs) using the LIB were also introduced during this period. In the mid-2010s, the battery electric vehicle (BEV) market began to take off again, and consumer electronics such as wearable devices and true wireless stereos (TWSs) began to be announced and marketed. Around 2020, while the BEV boom proved to be a significant tailwind for the LIB market, it was a time when many industries experienced significant supply chain disruptions due to COVID-19. In addition to existing LIBs, the all-solid-state battery (ASSB) market is expected to expand after 2025.

Glossary All-solid-state-battery A battery in which the electrolyte and separator of a conventional lithium-ion battery are replaced by a solid electrolyte; it is being developed as an evolution of the lithium-ion battery. It is expected to outperform lithium-ion batteries in terms of safety, high-temperature characteristics, capacity, input/output and cycle life. Anode active material A material with an electrochemical potential close to lithium is used, primarily carbon materials such as graphite and silicon-based materials. Cathode active material Lithium ions move from the cathode to the anode during charging and from the anode to the cathode during discharging. Lithium-containing metal oxides such as LiNiMnCoO2, LiNiCoAlO2, and LiFePO4 are commonly used. Complex hydride-based solid electrolyte A complex hydride solid electrolyte in which alkali metal or alkaline earth metal ions, such as lithium, are combined with hydrogen-containing ions, such as BH4. Electrolyte solution A medium through which lithium ions move during charging and discharging; electrolyte salts such as LiPF6 (lithium hexafluorophosphate) mixed with organic solvents and additives. Oxide-based solid electrolyte Solid electrolyte made of inorganic oxide materials. Safer than sulfide-based solid electrolyte. Polymer-based solid electrolyte A solid electrolyte that is a polymer complex consisting of a polymer material such as polyethylene oxide and an inorganic lithium salt. Separator A microporous membrane that insulates the cathode and anode, holds the electrolyte solution, and allows lithium ions to move. Polyethylene and polypropylene are the most commonly used materials. Sulfide-based solid electrolyte A sulfur-containing solid electrolyte. Among solid electrolytes, sulfide-based solid electrolyte has particularly high lithium ion conductivity.

Key points

• • •

LIB application market trend Past, present, and future of LIB and LIB materials market All-solid-state battery market trend

486

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

Abbreviations ASSB BEV BTS ESS EV HEV ICE LCO LFP LIB LME LMO LTO NCA NMC PHEV TWS UPS xEV

1

All-solid-state battery Battery electric vehicle Base transceiver station Energy storage system Electric vehicle Hybrid electric vehicle Internal combustion engine LiCoO2 LiFePO4 Lithium ion battery London metal exchange LiMn2O4 Li4Ti5O12 LiNi1-x-yCoxAlyO2 LiNi1-x-yMnxCoyO2 Plug-in hybrid electric vehicle True wireless stereo Uninterruptible power supply General terms for electrified vehicles, e.g., HEV (hybrid electric vehicle), BEV (battery electric vehicle), PHEV (plug-in hybrid electric vehicle), FCEV (fuel cell electric vehicle)

Introduction

This paper reports on the past, present, and future prospects of LIBs and their evolving forms, which are important technologies for solving climate change issues and have been used in a variety of applications, including consumer electronics, and have greatly improved our quality of life. This article represents the viewpoint of a market research company, and the content of this paper is based on Fuji Keizai’s market research materials on the battery market and edited by adding the authors’ findings. Low-carbon and decarbonization efforts will be crucial to solving the problem of climate change. According to the International Energy Agency, transportation, along with coal power and industry, account for the largest contributions to global CO2 emissions.1 Replacing the internal combustion engine (ICE) vehicles with lower CO2 emitting xEVs is important in the fight against climate change. The BEVs were commercialized before ICE vehicles around 1900. However, they had low battery energy density and short range.2 In contrast, LIBs have low weight, high energy density, and high power density compared to other batteries, making them an excellent power source for BEVs and an essential technology for the replacement of ICE vehicles with xEVs.2 Originally developed for consumer electronics, LIBs are now being used in applications as diverse as xEVs, ESSs, medical devices, and drones, resulting in significant price reductions and power density increases since their introduction.3–5 These lower prices and higher energy densities are essential for the expansion of the xEV market, and together with xEV adoption incentives from national governments, the xEV market has expanded significantly since 2020. In particular, global sales of BEVs and plug-in hybrid electric vehicles (PHEVs) exceeded 10 million units in 2022, with the number of different BEV and PHEV models in the market reaching 500.6 Examples of government incentives for purchasing xEVs and strengthening regulations for xEVs in the U.S. include the Inflation Reduction Act, which provides tax incentives, promotes Made in America, and reduces dependence on specific countries, and California’s Advanced Clean Cars II regulations, which impose an annual zero-emission vehicle requirement on automakers and provide incentives for consumers.7,8 Examples in Europe include the Fit for 55 proposals, which tighten CO2 emission standards for cars and vans by expanding the EU emissions trading system that encourages the reduction of greenhouse gas emissions, and the Net Zero Industry Act, which aims to increase Europe’s competitiveness in the LIB industry and other clean technologies by creating the conditions necessary to encourage investment in net-zero technology manufacturing projects and improving the skills of the workforce.6,9,10 In addition to existing LIBs, the development of post-LIBs is also progressing. The development of sodium-ion batteries, similar to LIBs but use sodium instead of lithium due to resource and materials cost constraints, is being driven primarily by Chinese battery manufacturers.6,11 In addition, many companies are developing ASSBs that utilize a lithium-metal anode for higher energy density (enabling longer EV range) and a nonflammable and flame-retardant solid electrolyte for higher safety than existing LIBs.2 Since their commercialization in the early 1990s, LIBs have been used in a wide variety of products, including consumer electronics, xEVs, and ESSs. Research and development has significantly improved their performance, while mass production and economies of scale have significantly reduced costs. However, as further performance and safety improvements are needed, battery development has entered a new phase, including the development of next-generation batteries embodied by ASSBs.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

2

487

Primary batteries and secondary batteries

Batteries are broadly divided into primary and secondary batteries. Typical primary batteries include zinc carbon batteries, alkaline batteries, silver oxide batteries, lithium manganese dioxide batteries, and zinc air batteries. Typical secondary batteries include lead-acid batteries, nickel-cadmium batteries, nickel-metal hydride batteries, and LIBs. The market trends for primary and secondary batteries are shown in Fig. 1. Lead-acid batteries had the largest market size among secondary batteries until 2018, but the market size of LIBs surpassed that of lead-acid batteries starting in 2019. As of 2022 (forecast), the global market for major secondary batteries was valued at more than US$ 115.8 billion (15,233.3 billion yen), with the market for lead-acid batteries valued at US$ 29.5 billion (3886 billion yen) (25%) and that of LIB valued at US$ 83.1 billion (10,924.8 billion yen) (72%). The market for all other secondary batteries, such as nickel-cadmium and nickel-metal hydride batteries, was valued at only US$ 3.2 billion (422.5 billion yen) (3%).

3

LIB application market - Consumer electronics market

LIB applications can be broadly divided into four categories: (1) consumer electronics, (2) motor drive applications (power applications) such as power tools, garden tools, vacuum cleaners, and electric bicycles, (3) xEVs, and (4) ESSs, uninterruptible power supplies (UPSs) and base transceiver stations (BTSs).

3.1

Trends from around 2010 to the early 2020s

3.1.1 Until 2011 The market for tablet PCs, which have less battery capacity per unit than notebook PCs, expanded as a replacement for notebook PCs. The saturation of PC terminals themselves, the European debt crisis, the slowdown in the U.S. and other economies, and natural disasters such as the 2011 Great East Japan Earthquake and flooding in Thailand disrupted supply chains and reduced consumption, resulting in a slowdown in the growth of the consumer electronics market. In spite of these disruptions, the market for new, promising applications such as smartphones and tablet PCs expanded remarkably.

(billion US$) 160 140 120

Original data is based on Japanese Yen and converted to US$ for this paper. Exchange rate assumptions for 2024 and beyond are the same as for 2023. The decrease in the 2022 market in US$ is due to currency fluctuations.

100 80 60 40 20 0 2008

2009

2010

2011

2012

2013

2014

2015

2016

Actual Major Primary Batteries Total

Fig. 1 Market trends and forecasts for primary and secondary batteries.12

2017

2018

2019

2020

2021 Estimate

Major Secondary Batteries Total

2022

2023

2024

Forecast

2025

488

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

3.1.2 2012–14 (Significant changes in the smartphone market) The overall market for consumer electronics deteriorated, and the situation was very difficult, especially for Japanese LIB manufacturers. The global market for notebook PCs in 2012 was slightly lower than in 2011, and the market for digital still cameras also declined significantly. On the other hand, the global LIB market was sustained due to the expansion of market size for smartphones and tablet PCs, as well as the full-scale production of cylindrical 18650 LIB cells for Tesla, Inc.’s (Tesla) BEVs. In terms of the market share of LIB by country, the share of manufacturers in South Korea and China increased, while that of Japanese companies decreased. Samsung Electronics Co. Ltd., Apple Inc., Nokia Corporation, HTC Corporation, and the former Research In Motion Limited (currently BlackBerry Limited) were the top companies by market share in 2012; however, by 2014, Nokia Corporation’s device business was acquired by Microsoft Corporation, and HTC Corporation and BlackBerry Limited had fallen to the bottom of the rankings. During this same period, Chinese smartphone manufacturers made rapid progress so that by 2014, 6 of the top 10 companies in the smartphone market were Chinese. The leading Chinese smartphone manufacturers at that time included Huawei Technologies Co., Ltd., ZTE Corporation, Lenovo Group Limited, Yulong Computer Telecommunication Scientific Shenzhen Co. Ltd. (currently Coolpad Group Limited), Xiaomi Corporation, TCL Technology Group Corporation, and Guangdong Oppo Mobile Telecommunications Corp., Ltd. Xiaomi’s rapid progress can be attributed not only to its lineup of low-cost handsets sold at less than RMB 1000 but also to its ability to meet consumer needs, such as slim handsets, accurately.

3.1.3 2013–15 (Wearable devices announced and launched consecutively) Wearable devices were announced and launched between 2013 and 2015 (Samsung Galaxy Gear: Announced in 2013; Google Glass: Announced in 2013; Apple Watch: Launched in 2015), and LIB manufacturers began to focus on developing LIBs for wearable devices. Although wearable devices did not contribute significantly to the LIB market at the time, they led to promising applications for LIBs in the future.

3.1.4 Since 2016 (true wireless stereo (TWS) market expansion) In September 2016, AirPods were announced alongside the iPhone 7. In addition, the iPhone 7 eliminated the earphone jack, which significantly expanded the TWS market.

3.1.5 Late 2010s The smartphone market in regions such as Japan, North America, China, and Western Europe was saturated. Although the introduction of 5G communication services contributed to the expansion of the market, the impact on the smartphone market was minimal.

4 4.1

Impact of COVID-19 TWS

Although there were temporary production stoppages and cutbacks at some sites, the impact on the market was minimal. Demand for TWS in online conferencing systems increased.

4.2

Smartphone

Consumption was down due to fewer outings, and the replacement cycle lengthened worldwide. In addition, the closure of smartphone factories and retailers slowed the production and sales of smartphones.

4.3

Tablet PC

The tablet PC market was benefited from the rise of telecommuting and online education.

4.4

Notebook PC

Although demand for notebook PCs was expected to decline in 2020 as the end of Windows 7 support led to a pause in replacement demand, demand for PCs grew significantly due to the increase in working from home and online education. On the other hand, the supply of CPUs, driver ICs, display panels, and power management ICs could not keep up with the rapid growth in demand.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market 4.5

489

Wearable device

Demand for smartwatches and health bands increased due to growing health awareness.

4.6

Power tool and garden tool

The market expanded due to the increased demand for DIY and gardening tools, which stemmed from mandates and requests to stay at home because of COVID-19 (The market for LIB-equipped power tools and garden tools has performed well amid the trend of “from nickel-cadmium batteries to LIBs” and “from engines to LIBs” even before COVID-19).

4.7

Vacuum cleaner

Stay-at-home demand drove market growth.

4.8

Electric motorcycle

Market growth was significant prior to COVID-19 due to China’s “Safety Technical Specification for Electric Bicycle (GB 177612018)” policy. This policy led to the replacement of substandard electric bicycles. During COVID-19, the electric motorcycle market grew steadily due to widespread avoidance of public transportation.

5 5.1

LIB application market - xEV market Trends from around 2010 to the early 2020s

5.1.1 Since 2009 Improvements in the performance and safety of LIBs led to the Full-fledged installation of LIBs in xEVs, beginning in 2009. The financial collapse of Greece, Italy, Spain, Ireland, Portugal, and other countries spread to major countries such as Germany and France, slowing the economic growth rates of these major European economies. In response to deteriorating public finances, countries enacted one austerity measure after another, resulting in sluggish sales of the public/official BEVs that automakers had counted on. The European debt crisis was also a negative factor for the expansion of the LIB market, as the worsening employment situation led to weak consumer spending and sluggish exports from China.

5.1.2 Mid-2010s Unlike the more mature LIBs used in consumer electronics, LIBs used for xEVs at the time had low yields at the start of mass production, triggering the recycling of in-process scrap from xEV LIBs to begin in earnest (recycling of end-of-life LIBs did not begin in earnest until around 2020). LIBs designed for early xEV models were high in low-cost manganese. They contained less cobalt and nickel compared to consumer electronics LIBs, making their scrap significantly less valuable. Therefore, unlike end-of-life consumer electronics LIBs, which were sold to recyclers as valuable resources, processing fees would have to be paid to recycle xEV batteries. The xEV market began to show positive signs in 2013. Sales of Tesla’s Model S were robust. The relatively high-priced car, priced at approximately US$70,000, sold as well as the NISSAN MOTOR CO., LTD. (NISSAN)’s LEAF, priced at approximately US$30,000. This was encouraging for many industry stakeholders concerned with the BEV market, which did not expand as expected. In a new step, Samsung SDI Co., Ltd. (Samsung SDI) began in-house production of several hundred tons of LiNi1-x-yCoxAlyO2 (NCA) per year in 2013, likely with plans to supply LIBs to Tesla. Tesla’s robust performance provided a tailwind for Panasonic Corporation’s (currently Panasonic Energy Co., Ltd.) LIB business. They signed a contract with Tesla to supply approximately 2 billion cylindrical LIB cells from 2014 to 2017. This volume was equivalent to the global market for cylindrical LIBs in 2013. Until 2014, only Panasonic had supplied Tesla, but the possibility of sourcing from 2 companies sparked industry interest. The possibility of sourcing from Samsung SDI or LG Chem Ltd. (currently LG Energy Solution Ltd.) was suggested because Tesla was exploring the possibility of in-house LIB production. Tesla’s planned production exceeded Panasonic’s production capacity. In the wake of a German automaker’s clean diesel cheating scandal, the automaker decided the direction of its future strategy on October 12, 2015. The company announced that its green car development initiative, previously focused on diesel engines, would shift to PHEVs, BEVs, and 48 V mild hybrid powertrains and that its future top-of-the-line model would be electric. The clean diesel cheating later led to a BEV boom in Europe, boosting the LIB and LIB materials markets from 2015. In 2015, China’s production of BEV passenger cars, BEV buses, and BEV specialty vehicles exploded. BEV passenger car production exceeded 155,000 units, with China accounting for one-third of the world’s BEV passenger car production. BEV bus production exceeded 96,000 units, with China accounting for most of the world’s production. BEV specialty vehicle production reached 49,000 units.

490

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

BEV and PHEV passenger cars remained robust in Japan, the United States, and Europe, with more than 115,000 units sold in the United States, 43,000 in the Netherlands, 34,000 in Norway, 28,000 in the United Kingdom, 27,000 in France, and 25,000 in Japan in 2015. As a result, xEV LIB’s production capacity reached 23 GWh in 2015. In anticipation of future xEV production expansion in Europe, there has been a growing trend since around 2017 to build LIB plants in the U.S. and Europe. Asian LIB manufacturers pushed to build plants mainly in European countries with low labor costs, such as Poland and Hungary. Swedish battery manufacturer Northvolt AB announced the construction of a large-scale plant with a production capacity of 32 GWh/year. In addition, Bayerische Motoren Werke AG (BMW) Group formed a consortium with Umicore S.A., a LIB cathode and battery recycling company, to develop a European xEV LIB value chain. Swiss battery manufacturer Leclanché S.A. has partnered with Indian battery giant EXIDE INDUSTRIES LIMITED to build LIB plants in India.

5.1.3 Around 2020 As the Chinese government reduced subsidies, BEV and PHEV sales in the world’s largest LIB-producing country fell below the previous year’s level (2019) for the first time since the statistics began being tracked in 2010. However, Tesla’s Model 3 has been performing well since 2020, and European automakers have made further progress in their shift to xEVs. At the United Nations Climate Change Conference (COP26) in the United Kingdom in the fall of 2021, a joint statement was announced to end the sale of new ICE vehicles by 2040 as one of the measures to combat global warming. The statement was not signed by Japan, the United States, Germany, France, China, South Korea, or other countries with large auto industries. This led to a belief that the larger the auto industry in a country, the more that country saw a 100% transition to zero-emission vehicles as unrealistic. However, U.S. automakers General Motors Company and Ford Motor Company, Germany’s Mercedes-Benz AG, and China’s BYD Auto Co., Ltd. signed the statement. In addition, in December 2021, TOYOTA MOTOR CORPORATION (TOYOTA) announced a target of 3.5 million BEV sales by 2030, an upward revision from its previous target of 2 million units. Moreover, in January 2022, Renault S.A., NISSAN, and MITSUBISHI MOTORS CORPORATION announced a plan to invest more than €20 billion in xEV development over five years and to develop more than 30 models by 2030. While many barriers to 100% zero-emission vehicles still exist – including rising prices for resources such as cobalt and lithium, the persistence of high LIB costs due to supply and demand imbalances, underdeveloped charging infrastructure, and safety concerns over LIB ignition – the transition to BEVs is steadily progressing.

6 6.1

Impact of COVID-19 BEV

BEV sales increased in many regions in 2019–20 as many automakers reduced production and sales of ICE vehicles to prioritize BEVs, and governments around the world increased subsidies for BEV purchases.

6.2

HEV

The global HEV market was impacted by COVID-19 in 2020. However, the market expanded compared to 2019. In Europe, where HEV sales have been stable in recent years, the full-year performance in 2020 exceeded the previous year’s results despite a sharp decline in demand since March 2020.

6.3

PHEV

The global PHEV market declined year-on-year in 2020 due to the impact of market stagnation caused by COVID-19 in Japan and North America. In contrast, the market expanded in Europe and China, where fuel efficiency regulations are more stringent, as major manufacturers expanded their lineups and expanded policies to support product penetration. Europe is the largest demand region in the world, with sales of more than 600,000 units, nearly tripling compared to 2019.

7

LIB application markets - ESS, UPS, BTS markets

The use of LIBs in ESS, UPS/Data Center, and BTS enables longer life and space savings compared to competing lead-acid batteries. In addition, LIB is being adopted for ESS due to its excellent input/output characteristics.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market 7.1

491

Trends from around 2010 to the early 2020s

7.1.1 Since 2011 After the 2011 Great East Japan Earthquake, the ESS market began to attract attention, with secondary alkaline batteries and LIBs coming into the spotlight alongside the sodium-sulfur and lead-acid batteries that had been the dominant storage batteries.

7.1.2 Around 2018 Unlike ESS, lead-acid batteries are dominant in the UPS/BTS sector, but a major telecommunications infrastructure service provider, China Tower Corporation Limited, decided in March 2018 to stop procuring lead-acid batteries for both new construction and renewal of BTS and to use spent LIBs for re-use. The ESS market expanded significantly in 2018, supported by the Renewable Energy 3020 Plan announced by the South Korean government in late 2017. Despite the rapid expansion of the market, ESS fires occurred in South Korea due to installation issues. After an investigation into the cause, it was concluded that the fires were not caused by the LIB cells, and installations resumed in the second half of 2019.

7.1.3 Around 2020 The expansion of renewable energy generation was a global theme and a tailwind for the expansion of the ESS market. Although the ESS market in South Korea stagnated in the first half of 2019, the ESS market grew globally. In addition, the expansion of mobile communications, including the introduction of 5G services, has increased the demand for backup power in BTSs, and this increased demand, along with the shift from lead-acid batteries to LIBs, has led to a significant increase in the LIB market.

8

Impact of COVID-19

Although the market for residential ESSs in Japan was expected to continue to expand steadily in 2020, the market stagnated in the first half of 2020 due to restrictions on sales activities as a result of COVID-19. On the other hand, stay-at-home demand, demand charges, self-consumption of solar power generation, peak power shifting, system stabilization, and ancillary services expanded the ESS market.

9

LIB market

LIB market trends and forecasts by type are shown in Fig. 2.

9.1

Trends from around 2010 to the early 2020s

9.1.1 Around 2010 The LIB market declined in 2009 from the previous year due to the global economic recession that began in the second half of 2008. At that time, Japanese LIB manufacturers had a high market share. The majority of LIBs used in consumer electronics were produced in Asia, and the majority of batteries for HEVs used in the U.S. were produced in Japan. To break the Japanese and Asian oligopoly in LIB production, the U.S. Department of Energy awarded a total of US$2.4 billion in grants under the American Recovery and Reinvestment Act to 48 projects related to the production of advanced batteries and BEV parts and the popularization of BEVs.13 This was the largest single investment in advanced battery technology in the world at that time, and it marked the beginning of serious battery technology development. There was a high probability that non-Japanese batteries would become the de facto standard for LIBs for PHEVs and BEVs, unlike LIBs for consumer electronics and nickel-metal hydride batteries for HEVs, where Japanese batteries had been the de facto standard. As such, Japanese producers feared that they would not be able to maintain their leading position. At that time, Japan was ahead of the world in LIB and xEV and had established a material-LIB-xEV supply chain in Japan (except for mineral resources for LIB materials). In contrast, Europe and the U.S. had low production volumes of LIBs and LIB materials, and they had to establish supply chains across national borders. At that time, China produced many LIBs and LIB materials but few xEVs.

9.1.2 Mid-2010s As the xEV market expanded, the LIB market for xEVs also expanded significantly. The xEV LIB market reached 23 GWh in 2015. The increase from 2014 to 2015 was notable due to the significant growth of BEV buses and BEV special vehicles in China in 2015.

492

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

(billion US$) 120

100

80

Original data is based on Japanese Yen and converted to US$ for this paper. Exchange rate assumptions for 2024 and beyond are the same as for 2023. The decrease in the 2022 market in US$ is due to currency fluctuations.

60

40

20

0 2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

Actual LIB (small, Cylindrical, Except for xEV and ESS) LIB (small, Pouch) LIB (for ESS, UPS and BTS)

2020

2021 Estimate

2022

2023

2024

2025

Forecast

LIB (small, Prismatic) LIB (for xEV)

Fig. 2 LIB market trends and forecasts by type.12

In consumer electronics, notebook PCs, tablet PCs, and smartphones did not grow significantly. Wearable devices and TWSs were expected to be market drivers in this area. Power supplies for motor drives, such as power and garden tools, vacuum cleaners, cleaning robots, and drones, were expected to be promising applications, and motor drive applications are still promising for LIBs today. When the market for LIBs for ESS started to form, many were skeptical about the expansion of this market because the ESS market would not be viable without a further new energy generation and political support, but ESS LIBs have established a certain market with the increase in political support for ESS and new energy generation coming online.

9.1.3 Around 2020 The xEV LIB market grew steadily due to the expansion of BEV production led by Tesla and the expansion of the xEV market due to the tightening of environmental regulations in Europe. In China, the country with the highest production and sales of xEV in 2020, the installed LIB for xEV exceeded 150 GWh in 2021, and the production volume of LIB for xEV reached 210 GWh. The global production volume of LIB for xEV was 359 GWh in 2021, more than double that of 2020 (166.4 GWh). The ESS, UPS, and BTS markets are expanding for the self-consumption of solar power generation, power peak shift, grid stabilization, and ancillary services in Europe, North America, and China, and the LIB market is expanding accordingly. As for LIBs for consumer electronics, the market for cylindrical LIBs is expected to expand due to demand for power tools and garden tools, power banks, TWSs, etc., and the market for pouch LIBs is expected to expand due to their adoption in smartphones, notebook PCs, and wearable devices.

10

Battery material market trends - Cathode active material market

The market for the LIB cathode, anode, electrolyte solutions, and separator is forecasted in Fig. 3.

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

493

(billion US$) 80 70 60

Original data is based on Japanese Yen and converted to US$ for this paper. Exchange rate assumptions for 2024 and beyond are the same as for 2023.

50 40 30 20 10 0

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

Actual LIB cathode active material

LIB anode active material

2019

2020

2021

2022

2023

2024

Estimate

Forecast

LIB electrolyte solution

LIB separator

2025

Fig. 3 Market trends and forecasts for LIB’s four main materials.12

10.1 Trends from around 2010 to the early 2020s 10.1.1

Early 2010s

NICHIA CORPORATION, Umicore S.A., and China’s Beijing Easpring Material Technology Co., Ltd. were the top three manufacturers of cathode active materials in 2011; the fourth and fifth largest manufacturers in the world were also Chinese companies. Although South Korean manufacturers L&F CO., LTD. and COSMO ADVANCED MATERIALS & TECHNOLOGY CO., LTD. had a strong presence, the in-house production volume of South Korean LIB manufacturers was so high that it accounted for about 45% of cathode active material production in South Korea in 2012. At that time, high-end smartphones used LIBs charged to high voltages (4.4 V) to increase the degree of LiCoO2 (LCO) utilization, and this development direction continues today. Although cobalt prices fluctuate widely, this technology has become an important technology that can contribute to higher capacity while reducing the amount of cobalt used (while controlling cost increases).

10.1.2

Late 2010s

The direction of LIB development for xEVs since the late 2010s has been to adopt NCA, LiNi0.8Mn0.1Co0.1O2 (NMC811), and LiNi1x-yMnxCoyO2 (NMC) with high capacity per unit cathode, and to increase the mixing ratio of these materials (this trend continues after 2022). In China, where the development and adoption of NCA and NMC811 had been delayed compared to Japan and South Korea, adoption increased, led by Contemporary Amperex Technology Co., Limited (CATL), a major global LIB manufacturer for xEV. Cobalt and lithium salt prices rose sharply in 2017. Concerns about tight future supply and demand led mining companies to launch a number of new projects and expand existing lithium and cobalt projects. There are concerns that supply and demand for mineral resources will be more imbalanced after 2022 than in 2017. In 2017, LiFePO4 (LFP) made up less than 50% of the cathode active material used for xEVs produced in China, while NMC made up less than 45%. In 2018, the share of LFP decreased to just under 40%, and that of NMC exceeded 55%. The ratio of NMC and LFP to the total cathode in China reflects the xEV policy. The proportion of LFP tends to increase again after 2020 (Fig. 4). The total production volume of NMC811 and NCA in China exceeded 13,000 tons in 2018 (almost 10 times more than in 2017), more than 22,000 tons in 2019, more than 54,000 tons in 2020, and more than 130,000 tons in 2021 (estimated).

494

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Other types LFP NMC

2014

2015

2016

2017

2018

2019

2020

2021

1.9% 69.2% 28.8%

2.8% 67.6% 29.6%

3.6% 73.6% 22.7%

6.2% 49.1% 44.7%

2.6% 39.1% 58.3%

1.4% 33.4% 65.1%

0.5% 41.4% 58.1%

0.2% 51.7% 48.1%

January to May 2022 0.2% 60.8% 39.1%

Fig. 4 Trends in xEV sales in China by LIB type (based on number of xEVs installed).12

10.1.3

Around 2020

China’s market for LiMn2O4 (LMO) and LFP expanded. In the past, Japan and South Korea had large LMO markets, as early BEV models, such as the first-generation NISSAN LEAF, used a cathode of LMO mixed with NCA or NMC811. While LCO, NMC, and NCA have a layered rock salt structure, LMO has a more stable spinel structure and relies only on manganese, which is less expensive than cobalt and nickel; this makes LMO a good LIB material for BEVs in terms of stability and low cost. However, LMO has a lower energy density than NMC, and the problem of leaching manganese ions into the electrolyte under high-temperature conditions leads to battery degradation. In addition, the design concept of LIBs for BEVs is shifting to high capacity due to the robust performance of Tesla, which uses NCA, and the market share of LMO in the total cathode continues to shrink. Despite these circumstances, the production of LMO has increased in China because it is highly valued for its stability and low cost and because of its technological maturity, as demonstrated by its use in the early NISSAN LEAF. The energy density of LMO-based batteries can be improved by blending a certain percentage of NMC. LFP production begins to increase again due to several factors. (1) LFP LIBs are suitable for BTS of 5G communication. (2) The new industry standard for electric bicycles in China – Safety Technical Specification for Electric Bicycle (GB 17761–2018) – requires a vehicle weight of 55 kg or less, including the battery, resulting in a shift from heavy lead-acid batteries to lightweight LIBs, leading to a growing market for LFP LIBs as well as other chemistries. (3) The reduction and elimination of xEV subsidies in China by the end of 2022 requires further cost reductions, and low-cost LFP LIBs are suitable for applications where over-specification is not required. (4) LFP LIBs are expected to be used in ESS due to their high safety, low cost, and long life. (5) Tesla has started to use CATL’s LFP cells. From 2022, the expansion of the LFP market will be more pronounced. Manufacturers of cathode-active materials are accelerating their efforts to meet the growing demand for LFPs. Hunan Yuneng New Energy Battery Material Co., Ltd., and Guangdong Guanghua Sci-Tech Co. Ltd. have started LFP production. The draft EU Battery Regulation, published in December 2020, aims to make the battery value chain sustainable. It will require a declaration of the carbon footprint of the battery and its life cycle, including the total CO2 emissions at each stage of the life cycle, as well as the disclosure of the amount of recycled raw materials used in batteries, including cobalt, lithium, and nickel. It will also introduce a minimum requirement for the percentage of recycled raw material used in the battery. As Russia is a major producer of nickel and aluminum, nickel trading on the London Metal Exchange (LME) was temporarily suspended in March 2022 due to supply disruptions and price spikes caused by sanctions against Russia. There are concerns about a future imbalance between supply and demand for mineral resources such as nickel due to a further boom in BEVs and an uneven

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

495

distribution of resources. Indonesia has imposed restrictions on the export of raw nickel ore, and South American countries, including Chile, which has rich brine-type lithium deposits and produces a large amount of lithium salts, have restricted the production and export of such minerals as strategic resources. Since the civil war in the Democratic Republic of Congo, international standards such as the US Dodd-Frank Act and the EU Conflict Minerals Regulation have been established for the responsible sourcing of minerals. They aim to prevent armed groups from financing their activities through illegal mining and from violating the human rights of miners, including children. They also impose due diligence requirements on U.S.-listed producers and companies importing ores and raw materials into the EU.

11

Battery material market trends - Anode active material market

11.1 Trends from around 2010 to the early 2020s 11.1.1

Early 2010s

Since anode active materials are directly related to battery performance, high-performance materials from Japanese companies were widely used. As of 2010, no South Korean companies had entered the market (later, the current POSCO FUTURE M CO., LTD. entered the market and became a leading global company). As a result, Japanese LIB manufacturers sourced mainly from Japanese anode active material suppliers, Korean LIB manufacturers sourced from Japanese and Chinese suppliers, and Chinese LIB manufacturers sourced mainly from Chinese anode active material suppliers. Driven by the expansion of the LIB market for xEVs, the market for anode active materials also expanded significantly during this period. Many believed that natural graphite-based anodes would be the dominant anode in xEV LIBs due to their low cost and high capacity compared to artificial graphite-based anodes. However, natural graphite-based anodes tend to expand in volume, especially during fast charging (which reduces cycle life). Therefore, the artificial graphite-based anode, which has less volume expansion, has continued to perform well in the xEV market.

11.1.2

Late 2010s

There has also been a noticeable movement toward alloy-based anodes, such as silicon-based anodes, due to the demand for higher capacity. The production of LIBs using nonalloy Li4Ti5O12 (LTO) by TOSHIBA CORPORATION, Kokam Co., Ltd., Altair Nanotechnologies, Inc. (cell production by Kokam Co., Ltd.), Microvast Holdings, Inc., Leclanché S.A., and others also increased. The production of LIBs using LTO by manufacturers other than Toshiba Corporation later decreased due to the low energy density and high cost of LTO. In 2016, production by anode type was as follows: Hard carbons 1000 tons, soft carbons 500 to 1000 tons, LTO several hundred tons, silicon-based products 500 to 1000 tons, and graphite-based products more than 160,000 tons.

11.1.3

Around 2020

This market has also grown significantly, driven by the expansion of the xEV market. The majority of the market has been graphitebased. With the proliferation of fast-charging and high-rate LIBs, the proportion of artificial graphite-based products has increased, and the silicon-based market has expanded due to the demand for higher capacity.

12

Battery material market trends – Electrolyte solutions market

12.1 Trends from around 2010 to the early 2020s 12.1.1

Early 2010s

In the early 2010s, three Japanese companies – Ube Industries, Ltd. (currently UBE Corporation), Mitsubishi Chemical Corporation, and Tomiyama Pure Chemical Industries, Ltd. – and one South Korean company, Panax Etec Co., Ltd. (currently DONGWHA ELECTROLYTE CO., LTD.), were the main electrolyte solution manufacturers. Ube Industries, Ltd. added features to its electrolyte solutions so that they were relatively expensive but had a high market share. Due to the accident at the Fukushima Daiichi Nuclear Power Plant caused by the 2011 Great East Japan Earthquake, production at Tomiyama Pure Chemical Industries, Ltd.’s Okuma Plant in Futaba-gun, Fukushima Prefecture, was suspended. Among the four main materials of LIBs, electrolyte solutions are hazardous materials subject to import and export restrictions, and local production for local consumption is more active than other materials due to transportation costs.

12.1.2

Late 2010s

(1) Importing electrolyte solutions into Europe was difficult due to chemical regulations. (2) Electrolyte solutions are often shipped in stainless steel canister drums that are thicker and larger than typical steel drums, and because of the high cost of these canister drums, they are returnable (returned to the original shipper), incurring freight and other costs. (3) Transporting electrolyte solutions from Asia to Europe takes approximately 1 month, and there are quality assurance concerns about placing electrolyte solutions in a high-temperature environment.

496

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

The factors listed above favored the local production and consumption of battery electrolyte solutions, and as such, the production of battery electrolytes in Europe has increased with increasing xEV production.

12.1.3

Around 2020

China’s xEV market slowed in 2019 and did not expand as much as originally expected in 2020 due to COVID-19, and the price of electrolyte solutions fell due to a corresponding oversupply of the raw material LiPF6. Leading manufacturers have established and expanded production facilities in Europe due to the rapid growth in demand for LIBs for xEVs expected in Europe from 2021.

13

Battery material market trends – Separator market

13.1 Trends from around 2010 to the early 2020s 13.1.1

Early 2010s

Initially, the supply of separators for xEV LIBs exceeded actual demand due to low LIB production yields and other factors. However, as the market for consumer electronics LIBs continued to grow, the market for separators began to expand significantly in 2011. In 2012, the supply of LIB separators for xEVs remained weak due to improved yields, a certain number of batteries and battery materials in inventory, and the market expansion of xEVs falling short of initial expectations. However, the market for separators continued to grow from 2011 due to the expansion of the market for smartphones, which use large capacity batteries compared to feature phones. From 2013 to 2014, the market for LIB separators expanded significantly as the xEV market began to take shape, and the LIB market for consumer electronics expanded significantly. In addition, the demand for coated separators increased more than the actual demand due to increased distribution inventory in the coating process.

13.1.2

Late 2010s

The market for LIB separators continued to expand as the market for LIBs for consumer electronics and xEVs grew significantly.

13.1.3

Since 2020

Although demand for LIB separators stagnated in 2020 due to the suspension of production by xEV LIB manufacturers, demand recovered in the second half of the year. In particular, top separator manufacturers with good supply relationships with the world’s leading LIB manufacturers, such as Panasonic, LG Energy Solution Ltd., Samsung SDI, and CATL, had a high market share. In China, large manufacturers have increased production capacity to take advantage of economies of scale. This has made it difficult for medium and smaller separator manufacturers to keep pace, and the consolidation of separator manufacturers has continued. In 2021, the separator market was on a significant uptrend, driven by the continued expansion of the LIB market for xEVs. In addition, many separator manufacturers have expanded production capacity, and both production and shipments have increased.

14

Significance of ASSB development and market trends

14.1 Significance and advantages of developing ASSB ASSBs are secondary batteries in which a solid electrolyte replaces the liquid electrolyte and separator. It is well known that replacing the liquid electrolyte with a nonflammable and flame-retardant solid electrolyte provides a high level of safety. The increase in safety may also improve the volumetric energy density of the battery pack by reducing the cooling system and safety space. In addition, the use of the solid electrolyte allows the use of new cathode and anode active materials, resulting in higher capacity. Due to their potential for high energy density and safety, ASSBs are being actively developed as a major technology for the next generation of batteries. By using these technologies, xEVs will benefit from increased safety, allowing for higher density battery cell layouts; higher rates allowing for xEVs to charge quickly; higher capacities enabling extended EV range (Fig. 5). Japan has been a world leader in the development of ASSBs. In particular, Japan has contributed to a number of breakthroughs in the research and development of sulfide-based ASSBs, including the discovery of solid electrolytes with ionic conductivity comparable to that of liquid electrolytes. Private companies, universities, and research institutes are working together under the “Development of Material Evaluation Techniques for Advanced and Innovative Batteries (Phase 2)” (SOLiD EV) to address issues such as the establishment of mass production processes and good interface formation.

15

Types of ASSBs

ASSBs are classified into four main types according to the type of solid electrolyte (Figs. 6 and 7).

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

annotation Na: Sodium-ion secondary battery K : Potassium ion secondary battery Mg: Magnesium secondary battery Li-S battery: Lithium-Sulfur battery

(Wh kg-1) 600

500

Improved performance through structural changes such as the use of high-capacity materials, ionic liquids, high concentrated electrolytes, and thicker electrode films

400

497

High capacity oriented Practical application of Post-LiBs Metal-air secondary battery, Mg Improved performance through the use of new materials such as 5V class cathode, Li-rich cathode, metallic Li anode, and the thinner electrolyte layer

Li-S battery All-solid-state battery (New material adopted)

New LIB 300

Performance limits of current LIBs

200

All-solid-state battery (Current materials used)

Non-Li secondary batteries for rare metal-free applications

Na, K

Current LIB Na

100 Polymer-based all-solid-state battery

Product commercialization progresses targeting ESS and xEV

Market launch of sulfide-based all-solid-state batteries, although significant performance improvement is unlikely due to the use of current materials such as NMC (cathode) and graphite (anode) in the initial stage, the volume density per pack will be improved by simplifying the cooling structure due to improved safety

0 2020

2025

2030

2035

14

Fig. 5 ASSB and next-generation battery technology roadmap.

Overview

Type

All-solid-state batteries using sulfide solid electrolytes containing sulfur in the composition of the solid electrolyte Compared with the other solid electrolyte, it has high ionic conductivity and is developed for practical applications in xEV.

Sulfide-based

Sulfide-based solid electrolytes have a wide potential window, and a high potential cathode, such as a 5V-class cathode, can be used to increase the working voltage, resulting in high energy density. Manufactured by lamination, pressurizing and sintering, etc., using solid oxide-based electrolytes. Bulk type

Oxide-based

Development is progressing with large area and large capacity as target, but low plasticity and dense formation of electrode-solid electrolyte interface is a problem. Development is also progressing as a semi-solid with trace amounts of liquid

Thin-film type

Semiconductor Manufacturing Technology is applied and battery with electrode and electrolyte layers is prepared by vacuum deposition High capacity is difficult to achieve but commercialization is possible due to compactness and flexibility

Laminated type

Small-capacity chip-shaped batteries using a green sheet batch sintering process, etc., is used in passive components such as multilayer ceramic capacitors and chip inductors, etc. A series of product announcements by Japanese manufacturers has successively accelerated commercialization All solid-state battery using polymeric materials as solid electrolyte

Polymer-based

Due to its superior reduction resistance, lithium metal can be used as an anode A market was formed from earlier to focus on overseas companies

Complex hydride based

A novel solid electrolyte with research reports in the works since 2007 Good moldability due to excellent flexibility, adhesion is good between anode active material and electrolyte interface

Fig. 6 Major types of solid electrolytes.14

498

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

Type

Typical composition, materials

Ionic conductivity (Scm-1)

Sulfide-based

Li2S-P2S5, LGPS, LSiPSCl, LSiSnPS, etc.

-4 -2 10 ~ 10

Oxide based

LLTO, LATP, LLZO, LATP, LAGP, LIPON, etc.

10 ~ 10

Potential window

Airborne Stability

Plasticity

Thermal resistance

~ 300qC

-6

-3

~ 600qC

-5

Polymer-based PEO-LiTFSI, etc.

10 ~ 10

-4

~ 160qC

LiBH4, LiBH4-LiI, Complex LiBH4-LiNH2, LiBH4Hydride based P2S5, etc.

10-7 ~ 10-3

~ 150qC

Excellent in the order of

>

> -2

* Ion conductivity: Higher the value, higher the conductivity 10 indicates a high order of magnitude performance than 10 * Potential window: A range of electrically stable potentials that are not decomposed by oxidation or reduction.

-3

Fig. 7 Classification of solid electrolytes by performance.14

(MWh)

1,00,000 80,000 60,000 40,000 20,000 0

Sulfide-based Oxide-based Polymer-based Complex hydride-based

2020 0 20 34 0

2025 80 1,250 300 0

2030 5,900 3,803 1,240 10

2035 79,000 20,514 3,860 150

Fig. 8 Market trend of ASSBs.14

16

Market trend of ASSBs

Until 2020, the market was dominated by polymer-based ASSBs, which were mass-produced. However, the commercialization of oxide-based ASSBs has been progressing since 2021. Sample shipments of small sulfide-based ASSB cells began in 2019. But the market is not expected to really take off until after 2025 when mass production of xEVs is expected to begin. And materials and equipment manufacturers are gearing up for production (Fig. 8).

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

499

(billion US$) 14

12

10

Original data is based on Japanese Yen and converted to US$ for this paper. Exchange rate assumptions for 2025 and beyond are the same as for 2023.

8

6

4

2

0

Sulfide-based

2020 0.00

2025 0.02

2030 0.94

2035 11.51

Fig. 9 Market trend of sulfide-based ASSBs.14

17

Recent development and commercialization trends of sulfide-based ASSBs (Fig. 9)

Most of the recent activity has been in small and specialty applications because xEVs require larger batteries, which is a tall order. xEV companies are expected to become more active in the mid-2020s and beyond. In terms of materials, Idemitsu Kosan Co., Ltd. and Mitsui Mining and Smelting Company, Limited have begun mass production trials of solid electrolytes, and it is expected that expanding demand will accelerate mass production. Maxell, Ltd. is supplying samples of compact ASSBs (approximately 10 mAh, coin type and ceramic package type) mainly for factory automation and infrastructure and is beginning to show full-scale demand acquisition. Production started in 2023. Hitachi Zosen Corporation, together with the Japan Aerospace Exploration Agency (JAXA), is conducting the charging and discharging demonstrations of “AS-LiB” (140 mAh) and has confirmed that it can be used in the space environment by attaching it to the Small Payload Support Equipment (SPySE). Ongoing evaluation of the effects of charge and discharge characteristics and the space environment on capacity degradation is expected to lead to use in other space environment devices. For xEVs, development schedules and commercialization targets have been announced primarily by Japanese, Korean, and European automakers, and xEVs equipped with ASSBs are expected to be available in the market in earnest from the mid-2020s to 2030.

18

Market outlook for sulfide-based ASSBs in the 2020s (Fig. 9)

In the late 2020s, Japanese automakers such as TOYOTA, NISSAN, HONDA MOTOR CO., LTD., and other companies such as Hyundai Motor Company and BMW are also expected to launch BEVs with ASSB. For example, TOYOTA plans to use ASSB in BEVs by 2027–28. The company originally planned to use ASSB in HEVs in the early 2020s but has changed course. In the 2020s, ASSB is expected to be installed in high-end vehicles due to cost, and the number of vehicle models is expected to be limited. Although Chinese battery and automobile manufacturers are promoting research and development, in the Chinese market, from the perspective of cost and mass production, oxide-based semi-solid-state batteries will be used in practical applications earlier (mass production from 2022), and many people believe that the market for sulfide-based ASSBs will not be established until the 2030s or later. For applications other than xEVs, it is expected to be used in extreme environments, such as lunar and volcanic region observers and small lunar rovers. If the battery capacity can be increased, it is expected to be used in large lunar exploration vehicles. Small batteries are expected to be used in factory automation equipment and infrastructure monitoring devices (such as sensors and memory backup) that require heat resistance and long life; automotive equipment (such as sensors installed around the engine, on the roof, and on the tires); and medical devices (such as surgical instruments) that require high-temperature sterilization. It is also expected to be used in industrial wearable devices.

500

19

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

Long-term forecast for sulfide-based ASSBs after 2030 (Fig. 9)

After the 2030s, second-generation ASSBs with new cathode and anode active materials are expected to enter the market and contribute to improved mobility performance, such as the extended EV range. In addition, ASSBs are expected to have high safety, high energy density, and high input/output characteristics and thus expand the possibilities for mobility electrification. Different types of batteries will be used, including ABBSs. This will depend on the requirements and use cases of electric mobility. For xEVs, the number of vehicles equipped with ASSBs will increase. The use of ASSBs in automated vehicles will also increase. In addition to passenger cars, ASSBs are expected to be used in shared vehicles, taxis, and commercial vehicles such as buses and trucks. It is also expected to be used in motorcycles, drones, electric aircraft, ships, industrial vehicles, etc., contributing to the electrification of mobility to promote carbon neutrality. For mobility applications such as xEVs, oxide-based ABBSs have many technical challenges, and semi-solid-state batteries are expected to be commercialized sooner. Semi-solid-state batteries are being developed primarily by Chinese manufacturers. Taiwanese and U.S. startups are also involved. For multi-layer thin-film compact ASSBs, several manufacturers have started or plan to start mass production. It has been adopted in multiple applications, mainly for sample supply, such as consumer devices and IoT sensor devices that require heat resistance.

20

Market outlook for oxide-based ASSBs in the 2020s (Fig. 10)

For mobility applications, the development of semi-solid-state batteries is progressing, mainly by Chinese manufacturers, and is expected to be used in a limited number of electric motorcycles and BEVs. Compared to existing LIBs, oxide-based ABBSs are expected to provide added value, such as safety and high energy density per battery pack, and to be an option for the electrification of mobility. For consumer devices, the multilayer ASSBs have limited capacity, but their added value, such as heat resistance and long operating life, is expected to expand their adoption for IoT sensor devices and backup power supplies. The use of the thin-film type is expected to expand in implantable medical devices in the mid-to-late 2020s following clinical trials.

(billion US$) 4

3

Original data is based on Japanese Yen and converted to US$ for this paper. Exchange rate assumptions for 2025 and beyond are the same as for 2023. 2

1

0

Stacked type Thin-film type Bulk all-solid Bulk semi-solid

2020 0.01 0.00 0.00 0.01

Fig. 10 Market trends for oxide-based ASSBs.14

2025 0.07 0.01 0.00 0.24

2030 0.17 0.01 0.02 0.60

2035 0.41 0.03 0.77 2.04

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

21

501

Long-term forecast for oxide-based ASSBs after 2030 (Fig. 10)

After 2030, oxide-based ASSBs are expected to be developed and mass-produced, becoming the favorite candidate battery for xEVs. For consumer devices, the combination of increased energy capacity, cost reduction, and energy harvesting is expected to be realized, and the adoption of ASSBs will expand to wearable and hearable devices and new applications. Because the demand for batteries for xEVs is the largest of the major applications and the demand for ASSBs is expected to grow, the LIB market for xEVs was compared to the ASSB market for xEVs (Fig. 11). The LIB market continues to expand as a result of policies to popularize environmentally friendly vehicles and environmental regulations in various countries. In 2020, the overall vehicle market shrank significantly due to COVID-19, but the xEV market expanded. The forecast for 2035 is 1329 GWh. As of 2021, only polymer-based ASSBs were commercialized for xEV. Bolloré SE’s BEV Bluecar, STELLANTIS N.V.’s E-MEHARI, and DAIMLER TRUCK HOLDING AG’s BEV bus used Blue Solutions’ polymer-based ASSB LMP. Solid electrolytes other than polymer-based solid electrolytes face high technical difficulties due to interface formation problems and lack of mass production technology. Therefore, by using an oxide-based solid electrolyte as the base and adding liquid electrolyte or gel polymer to make semi-solid batteries, the technical difficulty has been reduced, leading to the commercialization of the product. As of 2022, QingTao (Kunshan) Energy Development Co., Ltd.’s semi-solid-state batteries have been used and tested in the prototype of BAIC Motor Corporation Limited’s EX3 model and Hozon New Energy Automobile Co., Ltd.’s Neta U BEV. It is also planned to be used in SAIC Motor Corporation Limited’s BEVs in 2023. Taiwan’s ProLogium Technology Co., Ltd. also plans to use it in BEVs from NIO Inc. of China and VinFast LLC of Vietnam (scheduled for 2024) and has developed a battery swapping system (2.4 kWh) for electric motorcycles with Taiwan’s Gogoro Inc. and Mercedes-Benz AG. In addition, QuantumScape Corporation is leading the development of mass-production technology with the goal of commercialization in 2025. By 2022, polymer-based ASSBs and oxide-based semi-solid-state batteries dominated solid-state batteries for xEVs. TOYOTA will develop sulfide-based ASSBs and install them in BEVs in the second half of the 2020s. With the introduction of sulfide-based ASSB, the market for ASSB for xEVs is expected to expand in earnest. Other global manufacturers are also actively developing ASSB and investing in ASSB startups, and the ASSB market is expected to expand in the future. In xEVs, ASSBs are expected to extend EV range and improve safety, and the number of vehicles equipped with ASSBs is expected to increase as mass production technology is established. Although the share of ASSBs in the total demand for LIBs will be around 1% in 2030, it is expected to increase to 7.6% in 2035, mainly in high-end vehicles.

Capacity (MWh) 14,00,000 1,329,365MW LIB market forecast for xEVs

12,00,000

All-Solid-State Battery Market Forecast for xEVs

10,00,000

950,755MWh

8,00,000 6,00,000 4,00,000 101,660 MWh

2,00,000 44 MWh

1,570 MWh

10,620 MWh

2025

2030

0 2020 Fig. 11 ASSB market trends for xEV.

14

2035

502

22

Lithium Batteries – Lithium Secondary Batteries – Li-ion Battery | Market

Conclusion

Since their commercialization in the early 1990s, LIBs have served as a versatile power source for various applications, including consumer electronics, xEVs, and ESSs. Initially, the majority of LIB applications were in consumer electronics. BEVs were introduced around 2010, and although the BEV market did not initially expand as automakers and LIB manufacturers expected, BEV sales began to show signs of increasing in 2013 due to the success of Tesla. In 2015, a clean diesel scandal at a European automaker came to light, and Europe turned its attention to BEVs. In addition to Europe, China’s xEV strategy has led to a BEV boom, with global sales of BEV passenger cars reaching around 4.4 million units in 2021. In 2021, 76% of LIBs were for xEVs, 8% for ESSs (including BTSs and UPSs), and 16% for consumer electronics and non-xEV power applications, with xEV applications accounting for three-quarters of the total. The cost of LIBs has dropped significantly as companies around the world have researched, developed, and mass-produced LIBs. However, energy density is approaching performance limits, and safety concerns have not been addressed. In addition, energy device needs are changing amid trends such as the proliferation of different types of mobility, diversification of usage scenarios, and decarbonization. The demand for higher performance and lower cost is also high for secondary batteries. Competition to develop and commercialize next-generation batteries, particularly ASSBs, is also intensifying, and LIBs are entering a new phase.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Global Energy-Related CO2 Emissions by Sector – Charts – Data & Statistics. www.iea.org. Corrigan, D. A. Electric Vehicle Batteries: Past, Present, and Future. Electrochem. Soc. Interface 2022, 31 (3) 63–68. https://doi.org/10.1149/2.f09223if. Grey, C. P.; Hall, D. S. Prospects for Lithium-Ion Batteries and Beyond—A 2030 Vision. Nat. Commun. 2020, 11 (1). https://doi.org/10.1038/s41467-020-19991-4. Ziegler, M. S.; Song, J.; Trancik, J. E. Determinants of Lithium-Ion Battery Technology Cost Decline. Energy & Environ. Sci. 2021, 14 (12) 6074–6098. https://doi.org/10.1039/ d1ee01313k. Masias, A. Lithium-Ion Battery Design for Transportation. In Behaviour of Lithium-Ion Batteries in Electric Vehicles; Springer International Publishing: Cham, 2018; pp. 1–33. https://doi.org/10.1007/978-3-319-69950-9_1. Global EV Outlook., 2023. www.iea.org. Slowik, P.; Searle, S.; Basma, H.; Miller, J.; Zhou, Y.; Rodríguez, F.; Buysse, C.; Kelly, S.; Minjares, R.; Pierce, L.; Orvis, R.; Baldwin, S. Analyzing The Impact of the Inflation Reduction Act on Electric Vehicle Uptake in the United States; 2023. www.theicct.org. Advanced Clean Cars II | California Air Resources Board. https://ww2.arb.ca.gov. Fit for 55 - The EU’s plan for a green transition – Consilium. www.consilium.europa.eu/. Net Zero Industry Act. https://single-market-economy.ec.europa.eu. Walter, M.; Kovalenko, M. V.; Kravchyk, K. V. Challenges and Benefits of Post-lithium-Ion Batteries. New J. Chem. 2020, 44 (5) 1677–1683. https://doi.org/10.1039/ c9nj05682c$. Total Survey of Battery-Related Markets; Fuji Keizai Co., LTD.: Tokyo, Japan, 2010–2022. 2.4 Billion in Grants to Accelerate the Manufacturing and Deployment of the Next Generation of US Batteries and Electric Vehicles. whitehouse.gov. obamawhitehouse.archives.gov. Complete Overview of Next-Generation Battery-Related Technologies and Markets; Fuji Keizai Co., LTD.: Tokyo, Japan, 2020.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Overview Till Fuchsa,b, Burak Aktekina,b , Felix Hartmanna,b , Felix H Richtera,b , and Jürgen Janeka,b,c , aInstitute of Physical Chemistry, Justus Liebig University Giessen, Giessen, Germany; bCenter for Materials Research (ZfM/LaMa), Justus Liebig University Giessen, Giessen, Germany; cBattery and Electrochemistry Laboratory (BELLA), Karlsruhe Institute of Technology (KIT), Eggenstein-Leopoldshafen, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of P. Kurzweil, K. Brandt, SECONDARY BATTERIES – LITHIUM RECHARGEABLE SYSTEMS | Overview, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 1–26, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00185-4.

1 Introduction 2 General considerations of different SSB cell concepts 3 (Electro)chemical and morphological challenges of interfaces in SSBs 3.1 Separator: General requirements of solid electrolytes 3.2 Lithium metal anode: Challenges and ideas 3.3 Composite cathodes: Challenges and ideas 4 Summary and outlook Acknowledgments References

504 505 506 506 507 508 509 509 510

Abstract Solid-state batteries (SSBs) are promising energy storage technologies due to their potential for higher energy density, enhanced safety, and longer lifespan compared to lithium-ion batteries. This overview provides a comprehensive insight into the current promises and challenges of SSBs. We further discuss the most used cell architectures regarding their structural configurations and design principles. Further, the typology commonly used in battery research is critically looked upon, as it is not intuitive, has developed historically, and is often based on descriptive language. Additionally, the chapter addresses common electrochemical and mechanical challenges encountered during SSB operation, present mainly at interfaces between electrodes and the electrolyte. After discussing the key aspects of SSBs, this overview serves as a comprehensive introduction for researchers and engineers seeking to understand the current landscape and prospects of SSB technology.

Glossary Anode active material (AAM) Electrochemically active material in the negative electrode. Cathode active material (CAM) Electrochemically active material in the positive electrode. Cathode composite Powder mixture of active material and solid electrolyte used to produce a composite for the positive electrode. Cathode electrolyte interphase (CEI) Electronically insulating and ionically conducting passivation layer (interphase) that can form at the positive electrode of a battery. Composite cathode Positive electrode typically used in thick planar configuration composed of active material and solid electrolyte. Dry polymer electrolyte (DPE) A form of solid polymer electrolyte that does not contain any low-molecular-weight plasticizers or liquids. Electric vehicle (EV) Vehicle powered by a battery instead fossil fuel. Gel electrolyte (GE) Polymer that has taken up solvent or liquid electrolyte by swelling to form a gel electrolyte. Halide electrolyte Type of inorganic solid electrolyte, for example Li3YCl6. Hybrid electrolyte (HE) A combination of at least two different types of electrolytes in particle-in-matrix or multilayer assembly. Inorganic solid electrolyte (ISE) Ceramic solid electrolytes excluding carbon-based solid polymer electrolytes. Liquid electrolyte Electronically insulating and ionically conducting salt solution used in batteries to enable the transport of ions between the electrodes. Lithium-ion battery (LIB) State-of-the-art battery based on lithium ions as the charge carriers and employing liquid electrolytes. Lithium phosphorus oxynitride (LIPON) Solid electrolyte typically used in thin film applications. Mixed-conducting interphase (MCI) Interphase that can form in a battery allowing both electron and ion transport. Oxide/phosphate electrolyte Typically harder, brittle types of inorganic solid electrolyte, for example garnets (oxides) or NASICON materials (phosphates).

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00281-4

503

504

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Overview

Plasticized polymer electrolyte (PPE) A form of solid polymer electrolyte that contains low-molecular-weight plasticizers. Separator Ion-conducting interlayer to electronically separate the electrodes in a battery. Solid electrolyte (SE) Electronically insulating and ionically conducting solid used in solid-state batteries to enable the transport of ions between the electrodes. Solid electrolyte interphase (SEI) Electronically insulating and ionically conducting passivation layer (interphase) that can form in a battery; the term was empirically established for such interphases at the negative electrode. Solid polymer electrolyte (SPE) Electronically insulating and ionically conducting polymer used in solid-state batteries to enable the transport of ions between the electrodes. Solid-state battery (SSB) Battery employing predominantly solids as the ionically conducting components. Stack pressure Uniaxial pressure required in a cell for operation. Sulfide/thiophosphate electrolyte Typically soft, sulfur-based inorganic solid electrolytes, a representative being Li6PS5Cl.

Key points

• • • •

1

An overview of the current promises and challenges of SSBs is reported The typology of SSBs employing different types of electrolytes is discussed An overview over the possible cell architectures is given The most common challenges present during operation of an SSB are presented and discussed

Introduction

The transition to electric transportation through electrochemical energy storage is a key step toward reducing CO2 emissions, alongside other pollutants, to help mitigate climate change and related threats to the environment. Electrochemical energy storage is also useful for stabilizing the electric grid, which will increasingly rely on renewable energy sources. These technological advances have spurred rapid market growth and substantial investments in battery research and development, resulting in continuous performance enhancement such as increased energy and power density, longer cycle life, and cost reductions.1 Still, current battery technologies face physicochemical limitations, particularly in meeting the demand for electric vehicles (EVs) with extended driving range. This necessitates battery packs with higher energy density, achieved by a more compact battery pack design, such as through innovative cell-to-pack configurations, and enhanced energy density on cell level.2,3 However, the push to higher energy density on cell level also raises safety concerns associated with flammable liquid electrolytes in conventional lithium-ion batteries (LIBs), especially when using high-capacity anodes such as lithium metal or silicon.4,5 Lithium solid-state batteries (SSBs) are expected to overcome certain limitations inherent to LIBs based on liquid electrolytes. A range of cell designs with different solid electrolytes (SEs) appears to progress toward commercialization, yet significant uncertainties persist on the way regarding production methods cost and safety. While high expectations strongly push industrial and scientific efforts, fundamental questions for competitive SSB development remain - concerning the required stack pressure, the accessible operational temperature window, and processability, which must be solved for large-scale implementation.6,7 Simultaneously, the pursuit of high-performance LIBs continues, presenting “moving targets” for SSB development. The past decade has witnessed accelerated research and development at the material and component levels for SSBs and SEs to overcome these challenges. Key research topics hereby involve the dendrite-free and stable operation of lithium metal (and silicon) anodes, employing single-ion-conducting SEs to mitigate concentration polarization, thus reducing the proportion of flammable electrolytes, and utilizing thermally stable SEs. The most promising SEs can be divided in organic compounds, mostly polymers, and inorganic solids such as sulfides/thiophosphates, oxides/phosphates, halides, and oxyhalides. It may well be that not all challenges on the way to competitive SSBs are solved solely by one single electrolyte, and thus, hybrid cell designs, incorporating more than one SE, may be necessary to achieve the targets. Either way, interface design is paramount and the necessity of implementing practical as well as reasonable processing and production steps for scalable development is evident. At the same time, the demand for SEs is rising sharply and is surpassing the growth of current global material supply chains. SE production must be ramped up in parallel to expand the tests and findings on R&D level by feeding larger SSB pilot plants that are already under construction. In addition, the paradigm of inherently safe SSBs needs to be checked, as thermodynamic considerations and reactivity tests have shed some doubts on too optimistic expectations. In fact, aside from techno-economic and sustainability aspects along the entire value chain for SSBs,6,8–10 safety considerations for (single) battery cells as well as full battery packs will be decisive and a prerequisite for the commercialization of a specific SSB concept. The safety of lithium SSBs is frequently simply presumed because of the nonflammability and durability of most SEs. Nevertheless, recent findings revealed the emission of hazardous gases and strong heat dissipation caused by damaged cells and the formation of lithium dendrites.11,12 This suggests that the safety of lithium SSBs might have been overestimated and in-depth studies are essential on component level, first and foremost considering thermal

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Overview

505

properties and material interactions in electrodes, and further at the cell and battery pack level with systematic abuse testing on a large scale. This chapter “Overview – Solid-state batteries” addresses a few general considerations on SSB cell concepts and key challenges for positive and negative electrodes, separators, and interfaces. The various properties of all promising SEs, their advantages and disadvantages on material, component, and cell level, concepts for negative electrodes, thin and bulk SSB cell designs, and production routes are not discussed in depth in this overview, but thoroughly covered in the subsequent chapters.

2

General considerations of different SSB cell concepts

This chapter gives a brief overview of the different cell types that together make up SSB research. For a more detailed perspective on each, we refer the reader to their respective focus chapters. SSBs are rechargeable batteries that use only solid components.13 As any battery, they are essentially made up of three layers (plus current collectors): two electrode layers and a separating electrolyte layer. They are particularly distinguished by using solid electrolytes (SEs) instead of a combination of liquid electrolyte (LE) and separator. The SEs serve both as ion conductor and as separator because their mechanical properties are such that they can physically maintain separation of the two electrodes. They encompass a wide range of materials such as inorganic solid electrolytes (ISEs), solid polymer electrolytes (SPEs) and hybrid electrolytes (HEs). As such, the term SSB is rather general, and a closer look at which SE is used is usually required to evaluate and interpret the data presented.14 To avoid ambiguity about the type of electrolyte used, a systematic cell typology was proposed that classifies electrolytes by their conduction mechanism as LE, gel electrolyte (GE), plasticized polymer electrolyte (PPE), dry polymer electrolyte (DPE), SE (here we use ISE instead), and HE.15 Fig. 1a schematically demonstrates the proposed typology and shows how commonly used battery terms fit into the picture. The term all-solid-state battery (ASSB) is typically used simply as a synonym for SSB and covers ISEs, SPEs and HEs. The term almost-solid-state battery describes cells in which a small quantity of LE is used in addition to SE to help with ion transfer by wetting the interface with the cathode active material (CAM).13 The terms quasi-solid-state batteries and semi-solid-state batteries are typically used to describe cells that use electrolytes within the continuous range between GEs and PPEs, which contain a considerable amount of solvent or liquid plasticizer.16,17

Fig. 1 (a) Battery classification as roughly indicated by solid lines that group electrolytes into different battery types (inorganic solid electrolyte (ISE), dry polymer electrolyte (DPE), plasticized polymer electrolyte (PPE), gel electrolyte (GE), liquid electrolyte (LE), hybrid electrolyte (HE)). (b) Particle-in-matrix and multilayer assemblies of HE batteries. (c) Different cell architectures and electrode designs: thin-film battery, thick planar battery, and 3D-interdigitated battery. (d) Schematic of the bipolar stacking cell architecture, here containing three stacked mono cells. (a, b) Adapted from reference Sen, S.; Richter, F. H. Typology of Battery Cells – From Liquid to Solid Electrolytes. Adv. Sci. 2023, 10(33), 1–11. https://doi.org/10.1002/advs.202303985. (c, d) Based on references Xu, K.; Zhao, N.; Li, Y.; Wang, P.; Liu, Z.; Chen, Z.; Shen, J.; Liu, C. Design and 3D Printing of Interdigitated Electrode Structures for High-Performance Full Lithium-Ion Battery. Chin. J. Mech. Eng. Addit. Manuf. Front. 2022, 1(4), 100053. doi: 10.1016/j.cjmeam.2022.100053; Gambe, Y.; Sun, Y.; Honma, I. Development of Bipolar All-Solid-State Lithium Battery Based on Quasi-Solid-State Electrolyte Containing Tetraglyme-LiTFSA Equimolar Complex. Sci. Rep. 2015, 5, 10–13. https://doi.org/10.1038/ srep08869.

506

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Overview

In contrast, ISEs do not contain liquid additives or plasticizers. Most ISEs are crystalline inorganic materials that rely on ion hopping via vacant lattice sites within a rigid sublattice formed by the counterion. Typically, cations such as Li+, Na+, Ag+, Mg2+, etc. move through an anion sublattice, as for example in Li6PS5Cl, Li10GeP2S12 (LGPS), Li1.4Al0.4Ti1.6(PO4)3 (LATP), Li7La3Zr2O12 (LLZO), Li3InCl6, Na3SbS4, sodium b-alumina, AgI or MgSc2Se4. The anion mobility is typically very low, and the cation transference number is very close to one. Depending on their band structure, ISEs have a non-zero electronic partial conductivity that is mostly very low which leads to a negligible electronic transference number. The properties of different ISEs vary widely and each requires an individually targeted research depending on how they can be best prepared and incorporated into a cell; how sensitive they are when exposed to air; and how compatible they are electrochemically in combination with the other cell components. In addition, their ionic conductivity varies over many orders of magnitude. Notably, the best conducting ISEs known today have an ionic conductivity that rivals, and in some cases even exceeds, that of LEs at ambient temperature.18,19 There is also a wide range of glassy or amorphous ISEs, which mostly show lower conductivity than their crystalline counterparts with equal chemical composition. Often, polycrystalline ISEs contain a fraction of amorphous phase that is hardly recognized by the usual XRD characterization and may influence the electrochemical properties. In the following, the term ISE includes both crystalline and amorphous inorganic solid electrolytes. In comparison, the currently best SPEs have much lower ionic conductivity at ambient temperature than LEs. Transport of ions in SPEs typically occurs in coordination with the chain segmental motion of polymers above their glass transition temperature. In most SPEs a conducting salt is dissolved in the polymer matrix, but there are also examples of single-ion-conducting SPEs in which the moving ion is ionically bound to the polymer directly.20 The transference number of Li+ ions in SPEs varies strongly dependent on the composition and is usually smaller than 0.5. SSBs with SPEs typically operate at temperatures of 40–100  C to lower the viscosity of the SPE, which increases the mobility of the polymer chains, leading to higher ionic conductivity. Additionally, use of low-molecular-weight plasticizers often significantly increases the ionic conductivity, which is why making the distinction between plasticizer-free dry polymer electrolytes (DPEs) and plasticized polymer electrolytes (PPEs) was proposed.15 Hybrid electrolytes (HEs), as the name suggests, use a combination of different electrolyte types. Many such combinations have been investigated. The combination of SPE-ISE is the most common one, but also combinations with liquid electrolytes like LE-ISE, SPE-LE, and LE-ISE-SPE, are explored. The unifying feature of these combinations is that there is a change of conduction mechanism for lithium ion transport within the battery.15 There are interfaces present at which the moving ion must transfer from one electrolyte type and conduction mechanism to another. Two possible assemblies are commonly investigated for HEs: the particle-in-matrix assembly and multilayer assembly (Fig. 1b). The particle-in-matrix assembly relies on the dispersion of particles of one electrolyte type within the matrix phase of another electrolyte type. In the multilayer assembly, two or more consecutive layers of different electrolytes are applied between the electrodes. As HEs require ion transfer from one electrolyte type to another, the interface resistance for ion transfer has become a focus area. Often, this “hetero-ionic” interface resistance is relatively high, posing a significant hurdle for the development of competitive HE batteries.21 Finally, the electrode and electrolyte layers can be arranged in several different cell architectures (Fig. 1c and d). An early concept of a layered SSB is the thin-film battery.22 It relies on the deposition of thin planar layers via the gas phase, forming an anodeelectrolyte-cathode planar stack with small layer thicknesses (commonly in the range of a few mm) and relatively low area capacity (approx. 1 mAh cm−2). Thin-film batteries have been commercialized, yet mostly rely on lithium phosphorus oxynitride (LIPON) SE with low ionic conductivity. In contrast, the development of SEs with higher ionic conductivity, and the ease of processing when using SPEs, offer the opportunity for SSBs to act as a drop-in-technology to existing LIB manufacturing processes. Such cells are produced using tape casting, which results in SSBs with thick planar (composite) electrode layers (typically 50–100 mm thickness) and higher area capacity (approx. 5 mAh cm−2) comparable with LIBs.7,23 Finally, the solid nature of SSBs also lends itself to more complex battery architectures with 3D interdigitated electrodes24 and the possibility of bipolar stacking of cells.25,26 Further development is still needed in scaling up materials synthesis and SSB cell manufacturing.

3 3.1

(Electro)chemical and morphological challenges of interfaces in SSBs Separator: General requirements of solid electrolytes

The primary function of an electrolyte is to permit fast ion transport while strictly blocking the electron transport between two electrodes. Therefore, a high ionic conductivity of lithium ions is desirable for any SE, ideally exceeding 10 mS cm−1.7 The separator thickness is typically less than 25 mm in commercial lithium ion batteries with liquid electrolytes.27 For a 25 mm SE separator thickness in SSBs, an ionic conductivity of 10 mS cm−1 corresponds to 0.25 O cm2 resistance, which makes only a minor contribution to the overall internal cell resistance. The internal cell resistance should ideally be less than 40 O cm2,14 here, SEs with ionic conductivities less than 0.1 mS cm−1 could result in resistances more than 25 O cm2 alone (i.e., for the same thickness of 25 mm) showing the critical role of electrolyte ionic conductivity. Additionally, SEs must be mechanically stable and (electro)chemically compatible with other battery components during operation and storage. Lastly, cost is another important consideration for SEs as they should preferably consist of widely available and affordable elements, and their synthesis should be scalable. Considering these general requirements, the most promising electrolytes for SSBs can be classified into four main classes: oxide-type ISEs, sulfide-type ISEs, halide-type ISEs and SPEs. These materials exhibit diverse mechanical and chemical properties, making each material type well-suited for specific applications, as well as presenting unique challenges.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Overview

507

For instance, some sulfide-type ISEs have very high ionic conductivity comparable or even exceeding that of liquid electrolytes, e.g., 12 mS cm−1 for Li10GeP2S1228 and 32 mS cm−1 for Li9.54(Si0.6Ge0.4)1.74P1.44S11.1Br0.3O0.6.18 Another important advantage of this material class is that they can be sintered at room temperature, which renders them chemically more compatible with a wider range of electrode active materials and reduces battery assembly costs.29 Halide- and oxyhalide type ISEs also attract significant attention due to their high ionic conductivity (>1 mS cm−1), e.g., Li3YCl6, Li3YBr6, Li4YI7, etc., which can similarly be sintered at room temperature.30–32 In this sense, SPEs similarly do not require high temperature sintering to ensure an intimate contact with the electrode active materials. However, the ionic conductivity at room temperature of SPEs, e.g., lithium salt in poly(ethylene oxide) (PEO), is lower as compared to other SE types. Conductivities of about 0.1 mS cm−1 are generally considered high for SPEs.33 On the other hand, oxide-type ISEs require high temperatures during synthesis and post-synthesis sintering which increases the production costs.34 High temperature sintering is necessary to ensure a good interfacial contact between electrode active materials (e.g., CAMs) and oxide-type ISEs, but such a high temperature co-sintering process can result in chemical mixing reactions,35 and thus make some promising CAMs incompatible with specific oxide-type ISEs, e.g., LiNi0.6Co0.2Mn0.2O2 (NCM622) with Li1.3Al0.3Ti1.7(PO4)3 (LATP).36 Their ionic conductivities are typically less than 1 mS cm−1.37 but they can have superior chemical stability in ambient air and better (electro)chemical stability with electrode active materials than other ISEs, for example Li7La3Zr2O12 (LLZO), which shows an excellent stability with lithium metal.38 Most SEs have a limited thermodynamic stability window, and are unstable at low electrode potentials (e.g., when used with lithium metal anode).38,39 As a result, undesired electrolyte reduction reactions occur at the Li|SE interface. For instance, LGPS ISE are subject to continuous reduction reactions upon lithium metal contact since reaction products form a decomposition layer consisting of electronically conductive species such as Ge or Li-Ge alloys.40 Such decomposition layers, i.e., mixed conducting interphase (MCI), can continuously grow at the interface and eventually cause cell failure due to increasing impedance and lithium inventory depletion. On the other hand, reactions resulting in kinetic stabilization are also possible as in the case of sulfide-type Li6PS5Cl41–43 and oxide-type lithium phosphorus oxynitride (LiPON) ISEs,44 where a self-limiting solid electrolyte interphase (SEI) is formed at the anode|SE interface. At high potentials, electrolyte oxidation reactions similarly occur when high-voltage CAMs such as LiCoO2, nickel-rich NCM or spinel LiNi0.5Mn1.5O4 oxides are used in the composite cathode, particularly for sulfide-type ISEs45–48 and polymer-type SPEs.49,50 Therefore, the compatibility of electrolytes with commonly used anode and cathode active materials are crucial for practical SSB applications demanding high energy density and stable cycling.

3.2

Lithium metal anode: Challenges and ideas

The ultimate motivation to develop SSBs is the prospect of utilizing lithium metal as negative electrode material due to its unmatched specific capacity of 3861 mAh g−1. There are several reports both by academia and industry on different cell types including a lithium metal anode (LMA), surpassing either the energy or power density of state-of-the-art LIBs.51,52 However, several electrochemical and morphological challenges still exist when realizing an LMA without consensus on how to overcome said problems to date. Fig. 2 schematically displays the issues correlated with the use of LMAs during cycling in combination with an SE separator (Issue #1 - #4). The first challenge (Issue #1) is the ever-present lithium degradation upon contact with electrolytes, which is not limited to LEs. However, in the case of SEs, degradation products remain at the interface and can form resistive SEIs or fast growing MCIs.53,54 This is detrimental for cell operation and can even facilitate the undesired growth of lithium metal dendrites. There has yet to be found an SE with high conductivity (>10 mS cm−1) combined with thermodynamic stability when contacted to lithium metal.

Fig. 2 Schematic graph of an SSB employing a composite cathode and a planar lithium anode. The eight most common morphological and electrochemical issues of frequently used electrodes are depicted with numbers and classified in the graph at the right side regarding their sequence of occurrence during cycling and phenomenological nature.

508

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Overview

One of the most pressing challenges during cycling is the formation of pores at the interface between the lithium metal anode and an SE separator (Issue #2) during the discharge process of the cell.55 Unlike a LE, SEs are not able to compensate volume changes by low-viscous flow. During discharge (lithium metal dissolution), lithium atoms from the lithium metal electrode are oxidized and transported away from the interface as ions through the SE. Thus, vacancies form at the LMA|SE interface, which can accumulate if the rate of replenishment is not sufficient to account for the rate of discharge.56,57 In addition, pore formation and contact loss lead to current focusing and high internal resistance, which limits the available discharge capacity (lithium depletion) despite the fact that there is still lithium left, which corresponds to a kinetic limitation at the anode surface.58 As significantly improving the vacancy transport within the lithium metal proves to be quite difficult, another strategy to suppress pore formation during discharge is the application of stack pressure in the MPa range. This strategy works by replenishing lithium vacancies at the interface via creep and plastic deformation.59 However, applying high stack pressure is undesirable in large-scale applications, and it remains unclear whether the suppression of pore formation during fast discharge can be achieved under low stack pressures. Another strategy is to modify the interface by inserting a thin functional interlayer that mitigates the pore formation and lithium depletion issue.60,61 An additional issue arises when the current direction is reversed and lithium metal is deposited at the porous interface, as the local current density gets too high to safely and homogeneously deposit lithium on a conformal interface. In this case, lithium dendrite formation leads to a cell failure coupled with a potential safety hazard is possible, in which the stored energy can be released within seconds (Issue #3).62,63 However, it is important to add that dendrite formation can also occur without prior contact loss if the current density exceeds a certain value for a variety of underlying reasons. Conditions that favor dendrite growth are low electrolyte density (e.g., due to interparticle voids, pores), inhomogeneous interfaces or an increased electronic conductivity within the SE, e.g., at grain boundaries.64 Therefore, dendrite formation is both an electrochemical and a mechanical issue, and is heavily influenced by the applied stack pressure. While it is important to keep an intact interface without pores, applying very high stack pressure can also lead to damage of the separator and lithium extrusion into SE, thus creating mechanically formed dendrites.65 This creates a dilemma when choosing the right stack pressure needed for a specific cell, potentially motivating the need for variable stack pressure systems. During prolonged cycling, there is also the issue of the formation of “dead lithium” (Issue #4). Dead lithium may form when lithium is locally isolated (electronically) within the separator, for example due to uneven lithium dissolution during the discharge, or due to partial degradation of lithium as a result of solid electrolyte reduction reactions. This lithium can no longer be electrochemically assessed, and the available capacity of the cell is reduced. However, it in turn also influences the mechanical properties and propensity for dendrite formation of the separator, as it may crack more easily as dead lithium can act similar to pre-existing separator defects. Additionally, manufacturing challenges arise when lithium metal foils need to be handled.66,67 It is difficult to prepare thin (10 MS cm−1 for All-Solid-State Batteries. Angew. Chem. Int. Ed. 2023, 62 (13), 1–5. https://doi.org/10.1002/anie.202217581. 33. Mindemark, J.; Lacey, M. J.; Bowden, T.; Brandell, D. Beyond PEO—Alternative Host Materials for Li + -Conducting Solid Polymer Electrolytes. Prog. Polym. Sci. 2018, 81, 114–143. https://doi.org/10.1016/j.progpolymsci.2017.12.004. 34. Hood, Z. D.; Zhu, Y.; Miara, L. J.; Chang, W. S.; Simons, P.; Rupp, J. L. M. A Sinter-Free Future for Solid-State Battery Designs. Energ. Environ. Sci. 2022, 15 (7), 2927–2936. https://doi.org/10.1039/D2EE00279E. 35. Miara, L.; Windmüller, A.; Tsai, C.-L.; Richards, W. D.; Ma, Q.; Uhlenbruck, S.; Guillon, O.; Ceder, G. About the Compatibility between High Voltage Spinel Cathode Materials and Solid Oxide Electrolytes as a Function of Temperature. ACS Appl. Mater. Interfaces 2016, 8 (40), 26842–26850. https://doi.org/10.1021/acsami.6b09059. 36. Beaupain, J. P.; Waetzig, K.; Otto, S.-K.; Henss, A.; Janek, J.; Malaki, M.; Pokle, A.; Müller, J.; Butz, B.; Volz, K.; Kusnezoff, M.; Michaelis, A. Reaction of Li1.3Al0.3Ti1.7 (PO4)3 and LiNi0.6Co0.2Mn0.2O2 in Co-Sintered Composite Cathodes for Solid-State Batteries. ACS Appl. Mater. Interfaces 2021, 13 (40), 47488–47498. https://doi.org/10.1021/ acsami.1c11750.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Overview

511

37. Ren, Y.; Danner, T.; Moy, A.; Finsterbusch, M.; Hamann, T.; Dippell, J.; Fuchs, T.; Müller, M.; Hoft, R.; Weber, A.; Curtiss, L. A.; Zapol, P.; Klenk, M.; Ngo, A. T.; Barai, P.; Wood, B. C.; Shi, R.; Wan, L. F.; Heo, T. W.; Engels, M.; Nanda, J.; Richter, F. H.; Latz, A.; Srinivasan, V.; Janek, J.; Sakamoto, J.; Wachsman, E. D.; Fattakhova-Rohlfing, D. Oxide-Based Solid-State Batteries: A Perspective on Composite Cathode Architecture. Adv. Energy Mater. 2022, 2201939. https://doi.org/10.1002/aenm.202201939. 38. Xiao, Y.; Wang, Y.; Bo, S.-H.; Kim, J. C.; Miara, L. J.; Ceder, G. Understanding Interface Stability in Solid-State Batteries. Nat. Rev. Mater. 2019, 5 (2), 105–126. https://doi.org/ 10.1038/s41578-019-0157-5. 39. Hatzell, K. B.; Chen, X. C.; Cobb, C. L.; Dasgupta, N. P.; Dixit, M. B.; Marbella, L. E.; McDowell, M. T.; Mukherjee, P. P.; Verma, A.; Viswanathan, V.; Westover, A. S.; Zeier, W. G. Challenges in Lithium Metal Anodes for Solid-State Batteries. ACS Energy Lett. 2020, 5 (3), 922–934. https://doi.org/10.1021/acsenergylett.9b02668. 40. Wenzel, S.; Randau, S.; Leichtweiß, T.; Weber, D. A.; Sann, J.; Zeier, W. G.; Janek, J. Direct Observation of the Interfacial Instability of the Fast Ionic Conductor Li10GeP2S12 at the Lithium Metal Anode. Chem. Mater. 2016, 28 (7), 2400–2407. https://doi.org/10.1021/acs.chemmater.6b00610. 41. Aktekin, B.; Riegger, L. M.; Otto, S.-K.; Fuchs, T.; Henss, A.; Janek, J. SEI Growth on Lithium Metal Anodes in Solid-State Batteries Quantified with Coulometric Titration Time Analysis. Nat. Commun. 2023, 14 (1), 6946. https://doi.org/10.1038/s41467-023-42512-y. 42. Wenzel, S.; Sedlmaier, S. J.; Dietrich, C.; Zeier, W. G.; Janek, J. Interfacial Reactivity and Interphase Growth of Argyrodite Solid Electrolytes at Lithium Metal Electrodes. Solid State Ion. 2018, 318, 102–112. https://doi.org/10.1016/j.ssi.2017.07.005. 43. Tan, D. H. S.; Wu, E. A.; Nguyen, H.; Chen, Z.; Marple, M. A. T.; Doux, J.-M.; Wang, X.; Yang, H.; Banerjee, A.; Meng, Y. S. Elucidating Reversible Electrochemical Redox of Li6PS5Cl Solid Electrolyte. ACS Energy Lett. 2019, 4 (10), 2418–2427. https://doi.org/10.1021/acsenergylett.9b01693. 44. Cheng, D.; Wynn, T. A.; Wang, X.; Wang, S.; Zhang, M.; Shimizu, R.; Bai, S.; Nguyen, H.; Fang, C.; Kim, M.; Li, W.; Lu, B.; Kim, S. J.; Meng, Y. S. Unveiling the Stable Nature of the Solid Electrolyte Interphase between Lithium Metal and LiPON Via Cryogenic Electron Microscopy. Joule 2020, 4 (11), 2484–2500. https://doi.org/10.1016/ j.joule.2020.08.013. 45. Walther, F.; Randau, S.; Schneider, Y.; Sann, J.; Rohnke, M.; Richter, F. H.; Zeier, W. G.; Janek, J. Influence of Carbon Additives on the Decomposition Pathways in Cathodes of Lithium Thiophosphate-Based All-Solid-State Batteries. Chem. Mater. 2020, 32 (14), 6123–6136. https://doi.org/10.1021/acs.chemmater.0c01825. 46. Liu, G.; Lu, Y.; Wan, H.; Weng, W.; Cai, L.; Li, Z.; Que, X.; Liu, H.; Yao, X. Passivation of the Cathode–Electrolyte Interface for 5 V-Class All-Solid-State Batteries. ACS Appl. Mater. Interfaces 2020, 12 (25), 28083–28090. https://doi.org/10.1021/acsami.0c03610. 47. Auvergniot, J.; Cassel, A.; Foix, D.; Viallet, V.; Seznec, V.; Dedryvère, R. Redox Activity of Argyrodite Li6PS5Cl Electrolyte in All-Solid-State Li-Ion Battery: An XPS Study. Solid State Ion. 2017, 300, 78–85. https://doi.org/10.1016/j.ssi.2016.11.029. 48. Aktekin, B.; Sedykh, A. E.; Müller-Buschbaum, K.; Henss, A.; Janek, J. The Formation of Residual Lithium Compounds on Ni-Rich NCM Oxides: Their Impact on the Electrochemical Performance of Sulfide-Based ASSBs. Adv. Funct. Mater. 2024, 2313252. https://doi.org/10.1002/adfm.202313252. 49. Schmuch, R.; Wagner, R.; Hörpel, G.; Placke, T.; Winter, M. Performance and Cost of Materials for Lithium-Based Rechargeable Automotive Batteries. Nat. Energy 2018, 3 (4), 267–278. https://doi.org/10.1038/s41560-018-0107-2. 50. Xu, H.; Zhang, H.; Ma, J.; Xu, G.; Dong, T.; Chen, J.; Cui, G. Overcoming the Challenges of 5 V Spinel LiNi0.5Mn1.5O4 Cathodes with Solid Polymer Electrolytes. ACS Energy Lett. 2019, 4 (12), 2871–2886. https://doi.org/10.1021/acsenergylett.9b01871. 51. Lee, Y.; Fujiki, S.; Jung, C.; Suzuki, N.; Yashiro, N.; Omoda, R.; Ko, D.; Shiratsuchi, T.; Sugimoto, T.; Ryu, S.; Ku, J. H.; Watanabe, T.; Park, Y.; Aihara, Y.; Im, D.; Han, I. T. High-Energy Long-Cycling all-Solid-State Lithium Metal Batteries Enabled by Silver-Carbon Composite Anodes. Nat. Energy 2020, 5, 299–308. https://doi.org/10.1038/s41560020-0575-z. 52. Holme, T. A Discussion of QuantumScape’s Battery Technology Performance Results; 2021. 14th of January 2021. 53. Wenzel, S.; Leichtweiss, T.; Krüger, D.; Sann, J.; Janek, J. Interphase Formation on Lithium Solid Electrolytes — An In Situ Approach to Study Interfacial Reactions by Photoelectron Spectroscopy. Solid State Ion. 2015, 278, 98–105. https://doi.org/10.1016/j.ssi.2015.06.001. 54. Otto, S.; Riegger, L. M.; Fuchs, T.; Kayser, S.; Schweitzer, P.; Burkhardt, S.; Henss, A.; Janek, J. In Situ Investigation of Lithium Metal–Solid Electrolyte Anode Interfaces with ToF-SIMS. Adv. Mater. Interfaces 2022, 9 (13), 2102387. https://doi.org/10.1002/admi.202102387. 55. Krauskopf, T.; Richter, F. H.; Zeier, W. G.; Janek, J. Physicochemical Concepts of the Lithium Metal Anode in Solid-State Batteries. Chem. Rev. 2020, 120 (15), 7745–7794. https://doi.org/10.1021/acs.chemrev.0c00431. 56. Krauskopf, T.; Hartmann, H.; Zeier, W. G.; Janek, J. Toward a Fundamental Understanding of the Lithium Metal Anode in Solid-State Batteries - An Electrochemo-Mechanical Study on the Garnet-Type Solid Electrolyte Li6.25Al0.25La3Zr2O12. ACS Appl. Mater. Interfaces 2019, 11 (15), 14463–14477. https://doi.org/10.1021/acsami.9b02537. 57. Schmalzried, H.; Janek, J. Chemical Kinetics of Phase Boundaries in Solids. Ber. Bunsen. Phys. Chem 1998, 143 (2), 127–143. 58. Eckhardt, J. K.; Fuchs, T.; Burkhardt, S.; Klar, P. J.; Janek, J.; Heiliger, C. 3D Impedance Modeling of Metal Anodes in Solid-State Batteries − Incompatibility of Pore Formation and Constriction Effect in Physical-Based 1D Circuit Models. ACS Appl. Mater. Interfaces 2022, 14, 42757–42769. https://doi.org/10.1021/acsami.2c12991. 59. Wang, M. J.; Choudhury, R.; Sakamoto, J. Characterizing the Li-Solid-Electrolyte Interface Dynamics as a Function of Stack Pressure and Current Density. Joule 2019, 3 (9), 2165–2178. https://doi.org/10.1016/j.joule.2019.06.017. 60. Fuchs, T.; Mogwitz, B.; Otto, S.; Passerini, S.; Richter, F. H.; Janek, J. Working Principle of an Ionic Liquid Interlayer During Pressureless Lithium Stripping on Li6.25Al0.25La3Zr2O12 (LLZO) Garnet-Type Solid Electrolyte. Batter. Supercaps 2021, 4 (7), 1145–1155. https://doi.org/10.1002/batt.202100015. 61. Wang, Z.; Xia, J.; Ji, X.; Liu, Y.; Zhang, J.; He, X.; Zhang, W.; Wan, H.; Wang, C. Lithium Anode Interlayer Design for All-Solid-State Lithium-Metal Batteries. Nat. Energy 2024. https://doi.org/10.1038/s41560-023-01426-1. 62. Kasemchainan, J.; Zekoll, S.; Spencer Jolly, D.; Ning, Z.; Hartley, G. O.; Marrow, J.; Bruce, P. G. Critical Stripping Current Leads to Dendrite Formation on Plating in Lithium Anode Solid Electrolyte Cells. Nat. Mater. 2019, 18 (10), 1105–1111. https://doi.org/10.1038/s41563-019-0438-9. 63. Flatscher, F.; Philipp, M.; Ganschow, S.; Wilkening, H. M. R.; Rettenwander, D. The Natural Critical Current Density Limit for Li7La3Zr2O12 Garnets. J. Mater. Chem. A 2020, 8 (31), 15782–15788. https://doi.org/10.1039/c9ta14177d. 64. Liu, X.; Garcia-Mendez, R.; Lupini, A. R.; Cheng, Y.; Hood, Z. D.; Han, F.; Sharafi, A.; Idrobo, J. C.; Dudney, N. J.; Wang, C.; Ma, C.; Sakamoto, J.; Chi, M. Local Electronic Structure Variation Resulting in Li ‘Filament’ Formation within Solid Electrolytes. Nat. Mater. 2021, 20, 1485–1490. https://doi.org/10.1038/s41563-021-01019-x. 65. Ham, S.-Y.; Yang, H.; Nunez-cuacuas, O.; Tan, D. H. S.; Chen, Y.-T.; Deysher, G.; Cronk, A.; Ridley, P.; Doux, J.-M.; Wu, E. A.; Jang, J.; Meng, Y. S. Assessing the Critical Current Density of all-Solid-State Li Metal Symmetric and Full Cells. Energy Storage Mater. 2023, 55, 455–462. https://doi.org/10.1016/j.ensm.2022.12.013. 66. Otto, S. K.; Fuchs, T.; Moryson, Y.; Lerch, C.; Mogwitz, B.; Sann, J.; Janek, J.; Henss, A. Storage of Lithium Metal: The Role of the Native Passivation Layer for the Anode Interface Resistance in Solid State Batteries. ACS Appl. Energy Mater. 2021, 4 (11), 12798–12807. https://doi.org/10.1021/acsaem.1c02481. 67. Otto, S.-K.; Moryson, Y.; Krauskopf, T.; Peppler, K.; Sann, J.; Janek, J.; Henss, A. In-Depth Characterization of Lithium-Metal Surfaces With XPS and ToF-SIMS: Toward Better Understanding of the Passivation Layer. Chem. Mater 2021, 33 (3), 859–867. https://doi.org/10.1021/acs.chemmater.0c03518. 68. Albertus, P.; Anandan, V.; Ban, C.; Balsara, N.; Belharouak, I.; Buettner-Garrett, J.; Chen, Z.; Daniel, C.; Doeff, M.; Dudney, N. J.; Dunn, B.; Harris, S. J.; Herle, S.; Herbert, E.; Kalnaus, S.; Libera, J. A.; Lu, D.; Martin, S.; McCloskey, B. D.; McDowell, M. T.; Meng, Y. S.; Nanda, J.; Sakamoto, J.; Self, E. C.; Tepavcevic, S.; Wachsman, E.; Wang, C.; Westover, A. S.; Xiao, J.; Yersak, T. Challenges for and Pathways toward Li-Metal-Based All-Solid-State Batteries. ACS Energy Lett. 2021, 6 (4), 1399–1404. https://doi.org/ 10.1021/acsenergylett.1c00445. 69. Riegger, L. M.; Mittelsdorf, S.; Fuchs, T.; Rueß, R.; Richter, F. H.; Janek, J. Evolution of the Interphase between Argyrodite-Based Solid Electrolytes and the Lithium Metal AnodedThe Kinetics of Solid Electrolyte Interphase Growth. Chem. Mater. 2023, 35 (13), 5091–5099. https://doi.org/10.1021/acs.chemmater.3c00676. 70. Wang, M. J.; Carmona, E.; Gupta, A.; Albertus, P.; Sakamoto, J. Enabling “Lithium-Free” Manufacturing of Pure Lithium Metal Solid-State Batteries through in Situ Plating. Nat. Commun. 2020, 11 (1), 5201. https://doi.org/10.1038/s41467-020-19004-4. 71. Fuchs, T.; Becker, J.; Haslam, C. G.; Lerch, C.; Sakamoto, J.; Richter, F. H.; Janek, J. Current-Dependent Lithium Metal Growth Modes in “Anode-Free” Solid-State Batteries at the Cu|LLZO Interface. Adv. Energy Mater. 2023, 13 (1), 2203174. https://doi.org/10.1002/aenm.202203174.

512

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Overview

72. Kazyak, E.; Wang, M. J.; Lee, K.; Yadavalli, S.; Sanchez, A. J.; Thouless, M. D.; Sakamoto, J.; Dasgupta, N. P. Understanding the Electro-Chemo-Mechanics of Li Plating in Anode-Free Solid-State Batteries with Operando 3D Microscopy Understanding the Electro-Chemo-Mechanics of Li Plating in Anode-Free Solid-State Batteries with Operando 3D Microscopy. Matter 2022, 5, 3912–3934. https://doi.org/10.1016/j.matt.2022.07.020. 73. Motoyama, M.; Hirota, M.; Yamamoto, T.; Iriyama, Y. Temperature Effects on Li Nucleation at Cu/LiPON Interfaces. ACS Appl. Mater. Interfaces 2020, 12 (34), 38045–38053. https://doi.org/10.1021/acsami.0c02354. 74. Alexander, G. V.; Shi, C.; O’Neill, J.; Wachsman, E. D. Extreme Lithium-Metal Cycling Enabled by a Mixed Ion- and Electron-Conducting Garnet Three-Dimensional Architecture. Nat. Mater. 2023,. https://doi.org/10.1038/s41563-023-01627-9. 75. Fu, K.; Gong, Y.; Hitz, G. T.; McOwen, D. W.; Li, Y.; Xu, S.; Wen, Y.; Zhang, L.; Wang, C.; Pastel, G.; Dai, J.; Liu, B.; Xie, H.; Yao, Y.; Wachsman, E. D.; Hu, L. Three-Dimensional Bilayer Garnet Solid Electrolyte Based High Energy Density Lithium Metal-Sulfur Batteries. Energ. Environ. Sci. 2017, 10 (7), 1568–1575. https://doi.org/10.1039/c7ee01004d. 76. Lee, J.; Lee, T.; Char, K.; Kim, K. J.; Choi, J. W. Issues and Advances in Scaling up Sulfide-Based All-Solid-State Batteries. Acc. Chem. Res. 2021, 54 (17), 3390–3402. https:// doi.org/10.1021/acs.accounts.1c00333. 77. Wandt, J.; Freiberg, A. T. S.; Ogrodnik, A.; Gasteiger, H. A. Singlet Oxygen Evolution from Layered Transition Metal Oxide Cathode Materials and Its Implications for Lithium-Ion Batteries. Mater. Today 2018, 21 (8), 825–833. https://doi.org/10.1016/j.mattod.2018.03.037. 78. Strauss, F.; Teo, J. H.; Schiele, A.; Bartsch, T.; Hatsukade, T.; Hartmann, P.; Janek, J.; Brezesinski, T. Gas Evolution in Lithium-Ion Batteries: Solid Versus Liquid Electrolyte. ACS Appl. Mater. Interfaces 2020, 12 (18), 20462–20468. https://doi.org/10.1021/acsami.0c02872. 79. Culver, S. P.; Koerver, R.; Zeier, W. G.; Janek, J. On the Functionality of Coatings for Cathode Active Materials in Thiophosphate-Based All-Solid-State Batteries. Adv. Energy Mater. 2019, 9 (24), 1–14. https://doi.org/10.1002/aenm.201900626. 80. Wang, L.; Mukherjee, A.; Kuo, C.-Y.; Chakrabarty, S.; Yemini, R.; Dameron, A. A.; DuMont, J. W.; Akella, S. H.; Saha, A.; Taragin, S.; Aviv, H.; Naveh, D.; Sharon, D.; Chan, T.-S.; Lin, H.-J.; Lee, J.-F.; Chen, C.-T.; Liu, B.; Gao, X.; Basu, S.; Hu, Z.; Aurbach, D.; Bruce, P. G.; Noked, M. High-Energy All-Solid-State Lithium Batteries Enabled by Co-Free LiNiO2 Cathodes with Robust Outside-in Structures. Nat. Nanotechnol. 2024, 19 (2), 208–218. https://doi.org/10.1038/s41565-023-01519-8. 81. Zuo, T.-T.; Walther, F.; Ahmed, S.; Rueß, R.; Hertle, J.; Mogwitz, B.; Volz, K.; Janek, J. Formation of an Artificial Cathode–Electrolyte Interphase to Suppress Interfacial Degradation of Ni-Rich Cathode Active Material with Sulfide Electrolytes for Solid-State Batteries. ACS Energy Lett. 2023, 8 (3), 1322–1329. https://doi.org/10.1021/ acsenergylett.2c02835. 82. Huang, Y.; Zhou, L.; Li, C.; Yu, Z.; Nazar, L. F. Waxing Bare High-Voltage Cathode Surfaces to Enable Sulfide Solid-State Batteries. ACS Energy Lett. 2023, 8 (11), 4949–4956. https://doi.org/10.1021/acsenergylett.3c01717. 83. Xiao, Y.; Miara, L. J.; Wang, Y.; Ceder, G. Computational Screening of Cathode Coatings for Solid-State Batteries. Joule 2019, 3 (5), 1252–1275. https://doi.org/10.1016/j. joule.2019.02.006. 84. Bielefeld, A.; Weber, D. A.; Rueß, R.; Glavas, V.; Janek, J. Influence of Lithium Ion Kinetics, Particle Morphology and Voids on the Electrochemical Performance of Composite Cathodes for All-Solid-State Batteries. J. Electrochem. Soc. 2022, 169 (2), 020539. https://doi.org/10.1149/1945-7111/ac50df. 85. Hendriks, T. A.; Lange, M. A.; Kiens, E. M.; Baeumer, C.; Zeier, W. G. Balancing Partial Ionic and Electronic Transport for Optimized Cathode Utilization of High-Voltage LiMn2O4/ Li3InCl6 Solid-State Batteries. Batter. Supercaps 2023, 6 (4). https://doi.org/10.1002/batt.202200544. 86. Clausnitzer, M.; Mücke, R.; Al-Jaljouli, F.; Hein, S.; Finsterbusch, M.; Danner, T.; Fattakhova-Rohlfing, D.; Guillon, O.; Latz, A. Optimizing the Composite Cathode Microstructure in All-Solid-State Batteries by Structure-Resolved Simulations. Batter. Supercaps 2023, 6 (11). https://doi.org/10.1002/batt.202300167. 87. Bielefeld, A.; Weber, D. A.; Janek, J. Modeling Effective Ionic Conductivity and Binder Influence in Composite Cathodes for All-Solid-State Batteries. ACS Appl. Mater. Interfaces 2020, 12 (11), 12821–12833. https://doi.org/10.1021/acsami.9b22788. 88. Bielefeld, A.; Weber, D. A.; Janek, J. Microstructural Modeling of Composite Cathodes for All-Solid-State Batteries. J. Phys. Chem. C 2019, 123 (3), 1626–1634. https://doi.org/ 10.1021/acs.jpcc.8b11043. 89. Minnmann, P.; Quillmann, L.; Burkhardt, S.; Richter, F. H.; Janek, J. Quantifying the Impact of Charge Transport Bottlenecks in Composite Cathodes of All-Solid-State Batteries. J. Electrochem. Soc. 2021, 168, 040537. https://doi.org/10.1149/1945-7111/abf8d7. 90. Koerver, R.; Zhang, W.; De Biasi, L.; Schweidler, S.; Kondrakov, A. O.; Kolling, S.; Brezesinski, T.; Hartmann, P.; Zeier, W. G.; Janek, J. Chemo-Mechanical Expansion of Lithium Electrode Materials-on the Route to Mechanically Optimized All-Solid-State Batteries. Energ. Environ. Sci. 2018, 11 (8), 2142–2158. https://doi.org/10.1039/c8ee00907d. 91. Koerver, R.; Aygün, I.; Leichtweiß, T.; Dietrich, C.; Zhang, W.; Binder, J. O.; Hartmann, P.; Zeier, W. G.; Janek, J. Capacity Fade in Solid-State Batteries: Interphase Formation and Chemomechanical Processes in Nickel-Rich Layered Oxide Cathodes and Lithium Thiophosphate Solid Electrolytes. Chem. Mater. 2017, 29 (13), 5574–5582. https://doi.org/ 10.1021/acs.chemmater.7b00931. 92. Ruess, R.; Schweidler, S.; Hemmelmann, H.; Conforto, G.; Bielefeld, A.; Weber, D. A.; Sann, J.; Elm, M. T.; Janek, J. Influence of NCM Particle Cracking on Kinetics of Lithium-Ion Batteries with Liquid or Solid Electrolyte. J. Electrochem. Soc. 2020, 167 (10), 100532. https://doi.org/10.1149/1945-7111/ab9a2c. 93. Li, H.; Liu, A.; Zhang, N.; Wang, Y.; Yin, S.; Wu, H.; Dahn, J. R. An Unavoidable Challenge for Ni-Rich Positive Electrode Materials for Lithium-Ion Batteries. Chem. Mater. 2019, 31 (18), 7574–7583. https://doi.org/10.1021/acs.chemmater.9b02372. 94. Mesnier, A.; Manthiram, A. Influence of Single-Crystalline Morphology on the Electrochemical Behavior of High-Nickel Layered Oxide Cathodes. J. Electrochem. Soc. 2023, 170 (8), 080509. https://doi.org/10.1149/1945-7111/aceca6.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes Peng Zhanga, Zhen Liua, Kang Xiaa, and He Jiab, aSchool of Materials Science and Engineering, PCFM Lab, Sun Yat-sen University, Guangzhou, PR China; bInstitute of condensed Matter and Nanoscience (IMCN), Bio- and Soft Matter (BSMA), Université catholique de Louvain, Louvain-la-Neuve, Belgium © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is a update of D. Golodnitsky, SECONDARY BATTERIES – LITHIUM RECHARGEABLE SYSTEMS | Electrolytes: Single Lithium Ion Conducting Polymers, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 112–128, ISBN 9780444527455, https://doi.org/10.1016/ B978-044452745-5.00890-X.

1 2 2.1 2.2 2.3 3 3.1 3.2 3.3 4 4.1 4.2 4.3 4.4 5 5.1 5.2 5.3 6 6.1 6.2 7 References

Introduction Basic concepts of polymer electrolytes Ion transport in electrolytes Types of polymer electrolytes Advantage and disadvantages of polymer electrolytes Synthesis and characterization of polymer electrolytes Polymer selection and synthesis Electrolyte preparation and processing techniques Characterization techniques Properties and performance of polymer electrolytes in ASSB Ionic conductivity and mobility Electrochemical stability Mechanical properties and durability Interface and interfacial phenomena Applications of polymer electrolytes in all solid-state batteries Lithium-ion batteries Sodium-ion batteries Other emerging battery technologies Challenges and future prospects of polymer electrolytes in all solid-state batteries Remaining challenges and limitations Future research directions and opportunities Conclusion

514 515 515 515 516 516 516 518 519 519 519 519 521 521 521 521 522 523 523 523 524 524 524

Abstract This chapter provides an overview of the use of polymer electrolytes as a key component in all solid-state secondary ion batteries. The chapter begins with an introduction to the fundamental concepts of solid-state batteries and the benefits of using polymer electrolytes. The chapter then delves into the different types of polymer electrolytes that can be used in these batteries, including their characteristics and advantages. The use of polymer electrolytes in solid-state secondary ion battery is also explored, with a focus on their role in enhancing battery performance and safety. Finally, the chapter discusses future directions for research in this field. It’s aimed to provide a comprehensive guide to the use of polymer electrolytes in all solid-state secondary ion batteries and highlights their potential to revolutionize the field of energy storage.

Objectives

• • • •

A general overview of the applications of polymer electrolytes in the all-solid-state secondary batteries, with a focus on lithium polymer battery. Synthesis and characterization of polymer electrolytes are surveyed. Addressing the structure and property correlations in polymer electrolytes. Challenges and future prospects of the polymer electrolytes in all-solid-state secondary batteries are discussed.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00128-6

513

514

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes

Glossary of symbols ASSB ESW FTIR GISAXS GIWAXS GPE HOMO KIBs Li-air LIBs LiDFOB Li-S LiTFSI LPEs LUMO MIBs NASICON NMR NVPF PAN PECA PEG PEGDA PEGDMA PEO PMMA PPC PPO PTMC PVC PVDF PVDF-HFP SAXS SEI SEM SIBs SPEs SSE TEGDME Tg WAXS XPCS XRD ZIBs

1

all-solid-state batteries electrochemical stability window Fourier transform infrared spectroscopy grazing incidence small-angle X-ray scattering grazing incidence wide-angle X-ray scattering gel polymer electrolytes highest occupied molecular orbital potassium-ion batteries lithium-air lithium-ion batteries lithium difluoro (oxalato) borate lithium-sulfur lithium bis (trifluoromethanesulfonyl) imide liquid polymer electrolytes lowest unoccupied molecular orbital magnesium-ion batteries sodium superionic conductor nuclear magnetic resonance spectroscopy sodium vanadium phosphate fluoride polyacrylonitrile poly(ethyl cyanoacrylate) poly(ethylene glycol) poly(ethylene glycol) diacrylate poly(ethylene glycol) dimethacrylate poly(ethylene oxide) poly(methyl methacrylate) poly(propylene carbonate) poly(phenylene oxide) poly(trimethylene carbonate) polyvinyl carbonate polyvinylidene fluoride poly(vinylidene fluoride-co-hexafluoropropylene) small-angle X-ray scattering solid-electrolyte interphase scanning electron microscopy sodium-ion batteries solid-polymer electrolytes solid-state electrolytes tetraethylene glycol dimethyl ether glass transition temperature wide-angle X-ray scattering X-ray photon correlation spectroscopy X-ray diffraction Zinc-ion batteries

Introduction

All-solid-state batteries (ASSB) are a type of battery that use solid-state electrolytes (SSE) instead of liquid or gel electrolytes found in conventional batteries. The SSE are typically made of ceramic or glass materials, and they provide a number of advantages over liquid or gel electrolytes, including higher energy density, improved safety, and longer battery life. ASSB are still in the development stage, but they have the potential to revolutionize the battery industry by providing more powerful and longer-lasting batteries for a wide range of applications, including electric vehicles,1,2 portable electronics, and renewable energy storage.3,4 Polymer electrolytes are a crucial component in the development of ASSB. Compared to ceramic or glass electrolytes, polymer electrolytes offer several advantages, such as flexibility, ease of manufacturing, and high ionic conductivity at room temperature. One of the main challenges in developing ASSB is achieving high ionic conductivity in the SSE. Polymer electrolytes, due to their

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes

515

unique structure and properties, can offer high ionic conductivity at room temperature, which is essential for practical applications. Additionally, polymer electrolytes can be easily processed into thin films, which can be used as an SSE in batteries. This flexibility in processing makes polymer electrolytes ideal for creating thin, uniform layers of solid-state electrolytes that can be incorporated into various battery designs, including thin-film and flexible batteries. This chapter overview several aspects of polymer electrolytes: firstly, the basic concepts, followed by the synthesis and characterization, then the properties/performance and applications of polymer electrolytes in ASSB, and the last addressing the challenges and future prospects for the applications of polymer electrolytes in ASSB.

2 2.1

Basic concepts of polymer electrolytes Ion transport in electrolytes

Ion transport in electrolytes refers to the movement of ions, through a solution or material that contains dissolved ions, known as an electrolyte. The transport of ions is critical for the generation of electrical energy in batteries. In an electrolyte, positively charged ions, known as cations, are attracted to negatively charged electrodes, while negatively charged ions, known as anions, are attracted to positively charged electrodes. When an electric potential is applied across an electrolyte, these ions will move toward the oppositely charged electrode, resulting in an electric current. The rate of ion transport in electrolytes is dependent on several factors, including the concentration and size of the ions, the viscosity of the electrolyte, and the magnitude of the applied electric potential. In some cases, ion transport can be hindered by factors such as the presence of other ions or the formation of a layer of ions at the surface of the electrode, known as an ion double layer. In addition, ion transport is critical for a wide range of applications, including battery and fuel cell technology, electrochemical sensing, and membrane separations. Research in this field is focused on developing new materials and techniques to optimize ion transport and improve the efficiency of these technologies.

2.2

Types of polymer electrolytes

Polymer electrolytes are materials that contain an ion-conducting polymer matrix and a dissolved salt that provides the ionic charge carriers. These materials have attracted significant attention for use in various electrochemical devices, including rechargeable batteries, fuel cells, and sensors, due to their low volatility, enhanced safety, and good mechanical properties by comparing to conventional liquid electrolytes. In general, there are three main types of polymer electrolytes, i.e., solid polymer electrolytes (SPEs), gel polymer electrolytes (GPEs) and liquid polymer electrolytes (LPEs) (Fig. 1). The polymer matrix is typically a crosslinked network that has good mechanical stability, and the dissolved salt provides high ionic conductivity. SPEs are attractive for use in lithium-ion batteries, as they are less flammable and more stable than liquid electrolytes.5 By comparison, GPEs are similar to SPEs, but they include a solvent, which swells the polymer network and creates a gel-like consistency. This leads to an increase in the ionic conductivity compared to SPEs. In contrast, LPEs are made of a liquid polymer matrix that contains a dissolved salt. LPEs have high ionic conductivity due to the mobility of the polymer chains (i.e., the solvents weaken the forces that hold the polymer chains together and enable them to move more freely), but they are often less stable and more flammable than SPEs or GPEs. LPEs are commonly used in electrochemical capacitors, where high power density is required. Ionic conductivity Interfacial contact

Flexibility

Stable SEI

Machanical strength

Processability

Dendrite suppression Low leakage risk Fig. 1 Performance profiles of different electrolytes.

LPEs GPEs SPEs

516

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes

Fig. 2 Arrhenius plot of the conductivities of various polymer-salt SPEs.7 Reproduced with permission. Copyright 1986, Annual Reviews, Inc.

The first SPE was reported by Wright in 1973.6 The ion conducting membrane was made of poly(ethylene oxide) (PEO) incorporating alkali metal salts. This discovery has sustained a great interest so far and inspired further research in several areas of material science behind the area of battery. To facilitate lithium ions dissociation, the polymer matrix should have a relatively high dielectric constant. Thus, the polymers with polar functionalities such as ethers, esters, aziridines, sulfides and phosphazenes have been developed as host for the preparation of SPE (Fig. 2).7 In practice, lower dielectric constant induces a weak salt dissociation and thus the formation of clusters or ion pairs, which impede the ion mobility and give rise to much lower ionic conductivities.8

2.3

Advantage and disadvantages of polymer electrolytes

Polymer electrolytes endow the ASSB several new features, which are not available in the conventional liquid electrolytes. First of all, unlike liquid electrolytes, which are highly flammable and can lead to thermal runaway and explosions, polymer electrolytes are less flammable that reduce the risk of safety issues. Secondly, despite to the steel cell that are popular for the liquid cell, aluminum-plastic flexible packaging is generally used for the assembly of polymer electrolyte based ASSB, bringing the ASSB cells good safety performance, light weight and customized design of the shape. Thirdly, polymer electrolytes can have good mechanical stability and flexibility, which decreases the electrolyte thickness and enables the production of thinner battery cells. Finally, polymer electrolytes show low internal resistance and large capacity, making them be in line with the current demand direction of various electrical equipment. In contrast, polymer electrolytes also have some limitations for the secondary ion battery applications. For example, although polymer electrolytes can have high ionic conductivity at room temperature, they typically exhibit lower ionic conductivity than liquid electrolytes at higher temperatures, limiting their performance in high-power applications; they usually show narrow electrochemical stability window values, which restricts their use with high-voltage electrode materials; they are sensitive to moisture and can degrade if exposed to high levels of humidity or water, which can reduce their overall performance and stability. In addition, although research on polymer electrolytes in ASSBs is rapidly growing, the commercial availability of these materials and batteries is still limited. Therefore, polymer electrolytes offer several advantages for use in ASSBs, but there are also limitations that need to be addressed to improve their performance and enable their widespread use in commercial applications.

3 3.1

Synthesis and characterization of polymer electrolytes Polymer selection and synthesis

Polymer electrolytes play a critical role in secondary batteries, and the design and synthesis of the appropriate polymer is crucial to achieve high performance and stability. In general, the polymer used for polymer electrolytes in ASSBs should have high ionic

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes

H

(

O

H

C

C

O

O

)n

H

H

(

H

C

C

H

H

O

(

O

)

O

O

O n

O

O

PEO

H

517

PEGDA

TEGDME

O O

C

O

O

PPC

)n

(O

(

)n

O

H

CH3

C

C

H

COOCH3

PTMC

)n

PMMA

(

H

CN

C

C

H

H

F

C

C

H

F

)n

PVDF

PAN

O

O

(

)n

H

H

F

C

C

H

F

F

F

C

C

F

CF3

O O

O

(

O )n

O

O

(

O

)n O

O

(

)x (

)y

O

O

O O

PEGDMA

O

PVC

PVDF-HFP

Fig. 3 Chemical structure of some polymers used for ASSBs.9 Reproduced with permission. Copyright 2023, Springer Nature.

conductivity, good mechanical properties, high thermal stability, good electrochemical stability and balanced compatibility with electrode materials. The polymer should have a high degree of ion mobility to facilitate the transport of ions through the electrolyte (Fig. 3).9 Polymers with high dielectric constants and low glass transition temperatures tend to exhibit high ionic conductivity. In addition, the polymer should be mechanically stable and flexible to withstand the stresses that occur during battery operation. Polymers with good mechanical properties, such as high elongation at break and high modulus, are desirable. Moreover, the polymer should have good thermal stability to prevent degradation or decomposition during battery operation. Meanwhile, the polymer should also have good electrochemical stability to prevent oxidation or reduction reactions that can degrade the electrolyte and reduce the battery performance. Polymers with high oxidation potentials and low reduction potentials are preferred. Furthermore, the polymer electrolyte should be compatible with the electrode materials used in the battery. For example, if the anode or cathode material undergoes a large volume change during cycling, the polymer should be able to accommodate this without cracking or losing conductivity. Polymers used in polymer electrolytes can be synthesized using a variety of methods depending on monomer chemistry, desired properties of the polymer, and the specific requirements as mentioned above for their applications in electrolytes. In general, radical polymerization, anionic polymerization and ring-opening polymerization are the most popular synthesis method for polymers used in electrolytes. These synthesis methods are briefly introduced in the following. Radical polymerization is a widely used method for synthesizing polymers from vinyl monomers. Typically, a radical initiator is used to initiate the polymerization reaction, and the monomer is polymerized to form a polymer chain. Radical polymerization is commonly used to synthesize polymers such as poly(methyl methacrylate) (PMMA) and poly(vinylidene fluoride-co-hexafluoropropylene) (PVDF-HFP), which have been widely used in polymer electrolytes for battery applications.10 Anionic polymerization is a technique for synthesizing polymers from anionic monomers such as ethylene oxide or propylene oxide. In this method, an initiator such as an alkali metal or alkaline earth metal is used to initiate the polymerization reaction, and the monomer is polymerized to form a polymer chain. Anionic polymerization is commonly used to synthesize polyethylene oxide (PEO) and its block copolymers, which are popular in polymer electrolytes for batteries and fuel cells.11 Ring-opening polymerization is a method for synthesizing polymers from cyclic monomers such as lactones or cyclic carbonates.12 In this method, a metal catalyst such as tin octoate or titanium isopropoxide is used to initiate the polymerization reaction, and the cyclic monomer is polymerized to form a polymer chain. Ring-opening polymerization is commonly used to synthesize nylon-6, which is also used in polymer electrolytes for lithium-ion batteries.

518

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes

3.2

Electrolyte preparation and processing techniques

Electrolyte preparation and processing techniques play crucial rules for the development of high-performance electrolytes in various electrochemical devices, including batteries, fuel cells, and supercapacitors. The choice of electrolyte preparation and processing technique depends on the specific requirements of the electrochemical device and the properties of the electrolyte materials. A thorough understanding of the advantages and limitations of each technique is necessary to choose the optimal method for a specific application. For example, solution casting (Fig. 4a)13 is a common method for preparing polymer electrolytes, where the polymer and salt are dissolved in a suitable solvent, cast onto a substrate, and then (vacuum) dried to form a solid film. Solution casting allows for precise control of the electrolyte composition and thickness, and it is suitable for the preparation of a wide range of polymer electrolyte systems. In-situ polymerization (Fig. 4b)14 is a technique for synthesizing polymer electrolytes directly in the presence of the electrolyte salt, which attracts the interests of many scientists recently.15,16 In this method, the monomer is mixed with the initiator and the electrolyte salt, and then polymerization is initiated using heat or light. In-situ polymerization can produce highly homogeneous polymer electrolytes with excellent ion transport properties due to the improved electrolyte-electrode interface contacts. Electrospinning (Fig. 5)13 is a technique for preparing polymer electrolyte nanofibers with high surface area and porosity. In this method, a polymer solution containing the electrolyte salt is fed through a syringe, and a high voltage is applied to the needle to generate a charged jet. The charged jet is then collected on a grounded substrate, forming a fibrous membrane. Electrospinning can produce highly porous electrolyte membranes with excellent ion transport properties. Hot pressing is a technique for preparing polymer electrolyte membranes by applying heat and pressure to a dry or pre-swollen polymer electrolyte film. Hot pressing can improve the mechanical stability and reduce the porosity of the membrane, which can increase the ionic conductivity and reduce the risk of short circuits. Wet coating is a versatile technique that can be applied to various components of a battery to improve performance, reliability, and safety. For the polymer electrolyte preparation, it involves coating a polymer electrolyte solution onto the surface of the electrode material or substrate, followed by a drying process to evaporate the solvent and leave behind a solid coating of the polymer electrolyte. The wet coating process can be carried out using various methods, such as doctor blade coating, spray coating, spin coating, and dip coating. Doctor blade coating involves spreading the polymer electrolyte solution onto the surface of the electrode material using a blade, while spray coating involves spraying the solution onto the electrode surface using a nozzle. Spin coating involves depositing the solution onto a spinning substrate, while dip coating involves immersing the substrate into the solution and withdrawing it at a

(a)

Salt/Polymer

Stirring

(b)

Solvent

Anode Electrolyte Cathode

Host polymer

Uniform mixing

Final Sample

Dried in Vacuum

Casting onto Substrate

Fig. 4 Schematic illustration showing the polymer electrolyte preparation via (a) solution casting method.13 Via (b) in-situ polymerization method.14 (a) Reproduced with permission. Copyright 2023, Elsevier. (b) Reproduced with permission. Copyright 2022, Elsevier.

Fig. 5 Schematic illustration showing the polymer electrolyte preparation via electro-spinning.13 Reproduced with permission. Copyright 2023, Elsevier.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes

519

controlled rate. The choice of wet coating method will depend on various factors such as the type of electrode material, the desired coating thickness, and the production volume. It is important to optimize the wet coating process to ensure uniform coating thickness and good adhesion between the electrolyte and electrode material, which can significantly affect battery performance and lifespan.

3.3

Characterization techniques

Polymer electrolytes are commonly characterized using various techniques to evaluate their properties and performance. In practice, a combination of these and other characterization techniques can provide a comprehensive understanding of the properties and performance of polymer electrolytes, which is critical for their development and optimization. Among which, electrochemical impedance spectroscopy (EIS) is a popular technique used to measure the ionic conductivity and charge transfer resistance of the polymer electrolyte. EIS can provide information on the ion transport properties of the polymer electrolyte and its ability to function as an effective electrolyte in a battery or other electrochemical device. Additionally, Fourier transform infrared spectroscopy (FTIR), nuclear magnetic resonance spectroscopy (NMR), X-ray diffraction (XRD) and scanning electron microscopy (SEM) are popular techniques to characterize the structure of polymer electrolytes. FTIR is a spectroscopic technique used to identify the functional groups present in the polymer electrolyte. It can provide information on the polymer structure and chemical bonding, which can be important for understanding its stability and ion transport properties. NMR is a technique used to study the molecular structure and dynamics of the polymer electrolyte. NMR can provide information on the mobility of the polymer segments, the degree of crosslinking, and the distribution of polymer chain lengths, which can be important for understanding its mechanical and ion transport properties. XRD is a technique used to study the crystal structure of the polymer electrolyte. SEM is a microscopy technique used to study the morphology and microstructure of the polymer electrolyte. It can provide information on the pore structure, surface morphology, particle size distribution, degree of crystallinity and crystal orientation of the polymer electrolyte, which can be important for understanding its mechanical properties and ion transport properties. Moreover, X-ray scattering, including small-angle X-ray scattering (SAXS), wide-angle X-ray scattering (WAXS) and X-ray photon correlation spectroscopy (XPCS), can be used to investigate the structure and morphology of polymer electrolytes. SAXS can provide information on the size and shape of polymer microdomains, while WAXS can reveal the crystalline structure.17,18 If they are working in grazing-incidence mode, i.e., GISAXS and GIWAXS, would be powerful in characterizing the surface and interface structure of polymer electrolytes.19 Meanwhile, XPCS provides dynamic information of ion motion and diffusion in the electrolyte at unprecedent time-resolution. Understanding the polymer morphology is important for optimizing the electrolyte properties and performance. Their combination provides splendid structure information about the polymer electrolytes covering from molecular structure through nanoscale structure to microscopic structure.20,21

4 4.1

Properties and performance of polymer electrolytes in ASSB Ionic conductivity and mobility

The ionic conductivity and mobility of polymer electrolytes are critical factors that determine their performance in electrochemical devices. To improve the ionic conductivity and mobility of polymer electrolytes, researchers have explored different approaches such as incorporating ionic liquids or nanoparticles into the polymer matrix, modifying the polymer structure, or adjusting the temperature and concentration of the electrolyte (Fig. 6).22–25 A better understanding of the relationship between polymer structure, ionic concentration, and mobility is essential for developing polymer electrolytes with improved performance for various electrochemical devices. The ionic conductivity of a polymer electrolyte is determined by the mobility of the charged species (ions) through the polymer matrix. The mobility of the ions in a polymer electrolyte is influenced by several factors, including the polymer structure, ionic concentration, temperature, and morphology. For example, a polymer with a highly cross-linked structure may have lower ion mobility due to restricted ion movement within the polymer matrix. On the other hand, a polymer with a more flexible, amorphous structure may have higher ion mobility due to the increased free volume for ion transport.26Note: cross-linking restricts the movement of polymer chains by physically tethering them to the network structure, which reduces the available free volume within the polymer. Additionally, increasing the concentration of the ionic species can enhance the ionic conductivity of the polymer electrolyte.17,18,24 The ionic mobility of a polymer electrolyte dependents on several factors such as the molecular weight and polydispersity of the polymer, the size and charge of the ionic species, the morphology of the polymer electrolyte, and the presence of additives. Generally, higher molecular weight polymers tend to have lower ionic mobility, while smaller ionic species tend to have higher mobility.24 The morphology of the polymer electrolyte, such as the presence of pores and channels, can also affect the ionic mobility.17

4.2

Electrochemical stability

Electrochemical stability refers to the ability of the electrolyte to resist chemical reactions and degradation under the operating conditions of the device, particularly during charging and discharging cycles. If the electrolyte undergoes undesirable chemical reactions, such as oxidation or reduction, it can result in the formation of unwanted byproducts, reduced ionic conductivity, and ultimately, failure of the device. In practice, the electrolytes’ electrochemical stability can be reflected in the electrochemical

520

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes

Fig. 6 General approaches have been used to prepare high performance SSE.22 Reproduced with permission. Copyright 2023, John Wiley and Sons.

Fig. 7 Schematic illustration of energy levels of polymer electrolytes in battery. (a) Schematic energy diagram of a polymer electrolyte. Eg is the energy difference between highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of polymer electrolyte. (b) HOMO and LUMO levels of common polymers and lithium salts in SPEs.27 Reproduced with permission. Copyright 2023, John Wiley and Sons.

stability window (ESW) value, which is the range of voltages within which the electrolyte remains stable and does not undergo any chemical reactions (Fig. 7a).27 It is typically defined as the voltage range between the redox potentials of the electrolyte and electrode materials in the device. For example, in lithium-ion batteries, the ESW of the electrolyte is typically between 0 and 4.5 V, depending on the specific electrolyte used (Fig. 7b).27 This means that the battery can operate at voltages up to 4.5 V without causing any unwanted electrochemical reactions in the electrolyte. Operating the battery above this range value can lead to electrolyte decomposition and the formation of unwanted solid-electrolyte interphase (SEI) layers on the electrode surfaces, which would reduce battery performance and lifespan.28 In general, the electrochemical stability and window of a polymer electrolyte are affected by several factors, including the polymer structure, functional groups, and morphology, as well as the choice of salts and additives in the electrolyte. For example, a polymer electrolyte with a more stable backbone structure, such as a cross-linked polymer, may have a higher electrochemical stability than a linear polymer. Similarly, the presence of functional groups that can undergo redox reactions, such as nitro or carbonyl groups, can lower the electrochemical stability of the electrolyte. Therefore, selecting a polymer electrolyte with good electrochemical stability and a wide ESW is important for ensuring the safety and long-term performance of secondary ion battery.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes 4.3

521

Mechanical properties and durability

Mechanical properties of a polymer electrolyte refer to its ability to withstand mechanical stress, such as bending, stretching, and compression. The mechanical properties of a polymer electrolyte depend on several factors, such as the polymer structure, cross-linking density, and molecular weight. A polymer electrolyte with a more flexible, amorphous structure may have better mechanical properties than a more rigid, crystalline structure. Cross-linking the polymer can increase its mechanical strength and rigidity but may also decrease its ionic conductivity. Durability of a polymer electrolyte refers to its ability to maintain its chemical and physical properties over a long period of time, particularly under harsh operating conditions. The durability of a polymer electrolyte is influenced by several factors, such as the polymer stability, resistance to chemical degradation, and thermal stability. Exposure to high temperatures, humidity, and oxidative environments can cause the polymer electrolyte to degrade, resulting in a decrease in its ionic conductivity and mechanical properties. To improve the mechanical properties and durability of polymer electrolytes, researchers have developed different approaches such as incorporating fillers or reinforcing agents, modifying the polymer structure, or cross-linking the polymer. For example, incorporating inorganic fillers, such as silica or alumina, into the polymer matrix can increase the mechanical strength and rigidity of the polymer electrolyte.29 Cross-linking the polymer can also improve its mechanical properties and durability but may adversely affect its ionic conductivity.

4.4

Interface and interfacial phenomena

Interfacial phenomena that occur at this interface between the electrodes and the polymer electrolyte, such as charge transfer, ion transport, and chemical reactions, play an important role in the overall behavior of the battery.30,31 One important interfacial phenomenon is the interfacial charge transfer between the electrodes and the polymer electrolyte. For electrochemical reactions to occur at the electrodes, ions must be able to transfer through the electrolyte to reach the electrode surface. This transfer of charge is affected by several factors, including the ionic conductivity of the electrolyte, the electronic conductivity of the electrodes, and the interfacial resistance between the two materials. The interfacial resistance can be reduced by modifying the surface chemistry of the electrodes or by introducing additives to the electrolyte that can enhance ion transfer. Another important interfacial phenomenon is the transport of ions and solvent molecules across the interface. The behavior of these ions and molecules can be affected by the surface chemistry of the electrodes and the polymer electrolyte, as well as by the electrochemical potential gradient across the interface. In some cases, ion and solvent transfer across the interface can lead to the formation of a thin interfacial layer that can affect the overall performance of the device. In addition to charge transfer and ion transport, interfacial chemical reactions can also occur at the interface between the electrodes and the polymer electrolyte. These reactions can affect the stability and durability of the electrolyte, and can lead to the formation of unwanted byproducts that can interfere with device performance. Minimizing interfacial chemical reactions requires careful selection of the electrode materials and the electrolyte components, as well as optimization of the device operating conditions. Therefore, understanding and controlling interfacial phenomena is important for optimizing the performance and stability of electrochemical devices that use polymer electrolytes. A deeper understanding of the interfacial behavior can lead to the development of new materials and design strategies that can enhance battery performance and lifetime.

5 5.1

Applications of polymer electrolytes in all solid-state batteries Lithium-ion batteries

SPEs have attracted considerable attention as a potential replacement for liquid electrolytes in lithium-ion batteries (LIBs). SPEs offer several advantages over liquid electrolytes, including improved safety, high ionic conductivity, and compatibility with high-voltage cathode materials. One of the most widely studied polymers for SPEs in LIBs is polyethylene oxide (PEO), which has been shown to exhibit high lithium-ion conductivity (>10−4 S/cm) at room temperature. Other polymers, such as polyethylene glycol (PEG), polyvinylidene fluoride (PVDF), and their copolymers, have also been explored for use in SPEs. In addition to polymer selection, researchers have focused on developing novel processing techniques to optimize the mechanical properties and electrochemical performance of SPEs. For example, the use of block copolymers has been shown to improve the mechanical stability and flexibility of SPEs, while the incorporation of inorganic fillers and ceramic particles can enhance the thermal stability and mechanical strength of SPEs (Fig. 8).23–25,32 The development of high-capacity cathode materials, such as lithium-rich layered oxides and sulfur-based compounds, has also spurred interest in the use of SPEs in LIBs. However, the relatively low electrochemical stability of SPEs compared to ceramic electrolytes remains a challenge for their practical application in high-voltage LIBs. Recent studies have also explored the use of hybrid electrolyte systems that combine SPEs with small amounts of ceramic electrolytes. These hybrid systems can offer improved electrochemical performance while maintaining some of the advantages of SPEs. Therefore, the development of SPEs for LIBs is an active area of research, and significant progress has been made in recent years. However, further studies are needed to optimize the performance and durability of SPE-based LIBs for practical applications. To relieve these obstacles, recent studies have aimed to overcome the challenges of low room-temperature ionic conductivity, poor interface contacts, and limited electrochemical stability

522

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes

All solid-state Li battery

Addition of nanofiller to polymer matrix

Polymer Blend Polymer Plasticizing

Polymer structure mdification

Cathode active material

Li salt

Carbon

Li anode

Polymer chain

Fig. 8 Design of polymer-based solid electrolytes.32 Reproduced with permission. Copyright 2022, Elsevier.

of polymer electrolytes, as well as to explore new battery architectures and applications. The development of high-performance and durable polymer electrolytes is crucial for the advancement of all solid-state lithium-ion batteries for future energy storage and transportation.

Fig. 9 Temperature-dependent ionic conductivities of NASICON materials.33 Reproduced with permission. Copyright 2018, Elsevier.

5.2

Sodium-ion batteries

Similar to that of LIBs, PEO is one of the most commonly used polymer matrices for SPEs in sodium-ion batteries (SIBs) due to its high ionic conductivity and good mechanical stability. PEO-based SPEs have shown high sodium-ion conductivity (>10−4 S/cm) at room temperature and can support the reversible intercalation and deintercalation of sodium ions in various cathode and anode materials. Other polymers, such as PVDF, polyacrylonitrile (PAN), and their copolymers, have also been studied for their potential use in SIBs. These polymers can be modified with various functional groups, cross-linking agents, and plasticizers to improve their mechanical properties, thermal stability, and electrochemical performance. Recent studies have focused on developing SPEs with high ionic conductivity and good electrochemical stability, as well as exploring new cathode and anode materials for SIBs. For example, researchers have demonstrated the use of NVPF (sodium vanadium phosphate fluoride) and NASICON (sodium superionic conductor) (Fig. 9)33 as promising cathode materials for SIBs with SPEs.34 In addition, the use of inorganic fillers and ceramic particles as additives in SPEs has been shown to improve their mechanical strength and thermal stability. Overall, compared to that of LIBs, more work is needed to improve the performance and durability of SIBs with SPEs, as well as to address challenges such as the limited availability of high-capacity sodium cathode materials.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes 5.3

523

Other emerging battery technologies

Zinc-ion batteries (ZIBs) are a type of rechargeable battery that utilize zinc ions as the charge carrier instead of lithium and sodium ions. Some recent studies have focused on developing new SPEs with improved properties for use in ZIBs.35 One study reported the synthesis and characterization of a novel SPE based on a copolymer of vinylidene fluoride and hexafluoropropylene.36 The SPE exhibited high ionic conductivity, excellent mechanical properties, and good electrochemical stability in contact with zinc electrodes. Other studies have investigated the use of composite electrolytes, where the SPE is combined with other materials such as ceramics or ionic liquids to improve the ionic conductivity and overall performance of the ZIB.37 One study reported the development of a composite SPE based on polyethylene oxide and lithium bis(trifluoromethanesulfonyl)imide, which exhibited high ionic conductivity and good stability in contact with zinc electrodes.38 However, the development of SPEs for use in ZIBs is still in the early stages of research, and significant challenges remain in terms of optimizing the electrolyte properties and improving the performance of the ZIBs. Potassium-ion batteries (KIBs) are a promising alternative to LIBs due to the abundance and low cost of potassium. Li et al. reported that the development of a composite SPE based on polyethylene oxide and potassium bis(fluorosulfonyl)imide, which exhibited high ionic conductivity and good electrochemical stability in contact with potassium electrodes.39 Meanwhile, the other studies have investigated the use of different polymer chemistries and processing techniques to improve the properties of the SPEs. Despite these promising results, the development of KIBs with SPEs still faces several challenges, including the optimization of electrode materials and the improvement of cycling stability. SPEs have also been investigated as a potential electrolyte for Magnesium-ion batteries (MIBs) due to their high ionic conductivity and good mechanical stability. Sánchez et al. reported the development of a composite SPE based on PEO and magnesium perchlorate, which exhibited high ionic conductivity and good electrochemical stability in contact with magnesium electrodes.40 Acosta et al. investigated the use of a new SPE based on a copolymer of PEO and PPO for use in MIBs.41 The SPE exhibited high ionic conductivity and good stability in contact with magnesium electrodes, and the battery showed promising performance in terms of capacity and cycling stability. However, the development of MIBs with SPEs is still in the early stages, and there are several challenges that need to be addressed. One of the main challenges is the development of suitable cathode materials that can intercalate magnesium ions. Another challenge is the low diffusion coefficient of magnesium ions in SPEs, which can limit the overall performance of the battery. Lithium-sulfur (Li-S) batteries are attractive candidates for next-generation energy storage systems due to their high theoretical energy density and low cost. However, the practical implementation of Li-S batteries is hindered by several challenges, such as low sulfur utilization, poor cycling stability, and safety issues related to the use of liquid electrolytes. SPEs have been proposed as a potential solution to these challenges by providing a stable and safe electrolyte interface and enabling the use of lithium metal anodes. Recently, there has been a growing interest in the use of SPEs for Li-S batteries. Several studies have reported the use of SPEs based on various polymers, such as PEO, PEG and PVdF-HFP, for Li-S batteries. These SPEs have shown promising properties, such as high ionic conductivity, good mechanical stability, and excellent compatibility with lithium metal anodes. In addition, some studies have investigated the use of composite SPEs that incorporate functional materials, such as ceramic nanoparticles and carbon nanotubes, to enhance the performance of Li-S batteries. These composite SPEs have shown improved properties, such as increased ionic conductivity, better mechanical strength, and enhanced sulfur utilization. Despite these promising results, there are still several challenges that need to be addressed for the practical implementation of Li-S batteries with SPEs. These include the optimization of the properties of the SPEs, the development of suitable cathode materials that can effectively utilize sulfur, and the mitigation of the formation of lithium dendrites that can cause safety issues. Lithium-air (Li-air) batteries are another type of next-generation energy storage system that has attracted considerable attention due to their high theoretical energy density. However, the practical implementation of Li-air batteries is hindered by several challenges, such as the limited availability of suitable cathode materials and the high reactivity of lithium metal anodes with air. SPEs have been proposed as a potential solution to some of these challenges by providing a stable and safe electrolyte interface and enabling the use of lithium metal anodes. Recently, Zhang et al. and Liu et al. have summarized the progress and challenges of polymer electrolytes for Li-air applications in their review papers.42,43 In general, there are still several challenges that need to be addressed for the practical implementation of Li-air batteries with SPEs. These include the development of suitable cathode materials that can effectively utilize oxygen, the mitigation of the formation of lithium dendrites that can cause safety issues, and the optimization of the properties of the SPEs to improve their performance and stability.

6 6.1

Challenges and future prospects of polymer electrolytes in all solid-state batteries Remaining challenges and limitations

While polymer electrolytes show great promise for use in ASSB, there are still some challenges and limitations that need to be addressed. Some of the remaining challenges include: a. Low ionic conductivity: one of the main limitations of polymer electrolytes is their relatively low ionic conductivity compared to liquid electrolytes. While there has been significant progress in developing high-conductivity polymer electrolytes, further improvements are needed to promote commercial viability.

524

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes

b. Poor interfacial stability: the interface between the polymer electrolyte and the electrode can be prone to degradation, which can limit the performance and cycle life of the battery. Strategies to improve the interfacial stability are an active area of research. c. Limited electrochemical stability: some polymer electrolytes can be unstable at high voltages, which can lead to decomposition and reduced battery performance. Developing polymer electrolytes with improved electrochemical stability is an ongoing challenge. d. Limited commercially available material options: the range of materials that can be used as solid-state electrolytes is limited, which can restrict the design and performance of all-solid-state batteries. Developing new materials with the desired properties and improving the synthesis and processing methods for existing materials is an area of active research. e. Scale-up and commercialization: while there have been significant advances in the development of polymer electrolytes for ASSB, scaling up production and achieving commercial viability remains a challenge.

6.2

Future research directions and opportunities

By addressing the bottlenecks mentioned above, there are several future research directions and opportunities for polymer electrolytes in ASSB, including high-conductivity polymer electrolytes, interfacial engineering, electrochemical stability, new materials and advanced characterization techniques. Developing polymer electrolytes with even higher ionic conductivity is a key area of research. This can be achieved through the development of new polymer chemistries, advanced processing techniques, and the incorporation of new additives or fillers. Improving the stability and performance of the interface between the polymer electrolyte and the electrode is also an important research direction. Strategies such as surface modification, interfacial layers, and advanced electrode materials can be used to improve the interfacial stability. Developing polymer electrolytes with improved electrochemical stability at high voltages is another important research direction. This can be achieved through the development of new polymer chemistries or the incorporation of new additives. In addition, there is also a need for the development of new solid-state polymer electrolyte materials. This can include the design, synthesis and scaling-up production of new polymer, hybridization with other organic or inorganic materials that offer higher ionic conductivity, improved stability, or other desirable properties. Furthermore, developing new techniques for characterizing the structure and properties of polymer electrolytes is also an area of opportunity. This can include advanced imaging techniques, such as scanning probe microscopy, or in situ spectroscopic techniques to study the behavior of polymer electrolytes under operating conditions.

7

Conclusion

Polymer electrolytes are promising candidates for use in all-solid-state batteries, which have the potential to offer improved safety, energy density, and cycle life compared to conventional liquid electrolyte batteries. Some of the advantages of polymer electrolytes include light weight, high capacity and safety, and good mechanical properties. However, there are also some challenges and limitations that need to be addressed, such as low electrochemical stability and poor interface stability with electrodes. Recent research has focused on developing new polymer chemistries, processing techniques, and additives to improve the performance of polymer electrolytes in ASSB. Additionally, the development of advanced characterization techniques is helping researchers better understand the behavior of polymer electrolytes under operating conditions. There are also opportunities to explore the use of polymer electrolytes in other types of solid-state batteries, such as zinc-ion, potassium-ion, and magnesium-ion batteries. Overall, continued research and development in this field has the potential to enable the widespread adoption of high-performance all-solid-state batteries.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

Sun, Y.-K. Promising All-Solid-State Batteries for Future Electric Vehicles. ACS Energy Lett. 2020, 5 (10), 3221–3223. Motavalli, J. Technology: A Solid Future. Nature 2015, 526 (7575), S96–S97. Wang, C.; Sun, X. The Promise of Solid-State Batteries for Safe and Reliable Energy Storage. Engineering 2023, 21, 32–35. Schlem, R.; Burmeister, C. F.; Michalowski, P.; Ohno, S.; Dewald, G. F.; Kwade, A.; Zeier, W. G. Energy Storage Materials for Solid-State Batteries: Design by Mechanochemistry. Adv. Energy Mater. 2021, 11 (30), 2101022. Choudhury, S.; Stalin, S.; Vu, D.; Warren, A.; Deng, Y.; Biswal, P.; Archer, L. A. Solid-State Polymer Electrolytes for High-Performance Lithium Metal Batteries. Nat. Commun. 2019, 10 (1), 4398. Fenton, D. E.; Parker, J. M.; Wright, P. V. Complexes of Alkali Metal Ions with Poly(Ethylene Oxide). Polymer 1973, 14 (11), 589. Armand, M. B. Polymer Electrolytes. Annu. Rev. Mater. Sci. 1986, 16, 245–261. Agrawal, R. C.; Pandey, G. P. Solid Polymer Electrolytes: Materials Designing and All-Solid-State Battery Applications: An Overview. J. Phys. D. Appl. Phys. 2008, 41 (22), 223001. Yang, X.; Liu, J.; Pei, N.; Chen, Z.; Li, R.; Fu, L.; Zhang, P.; Zhao, J. The Critical Role of Fillers in Composite Polymer Electrolytes for Lithium Battery. Nano Micro Lett. 2023, 15 (1), 74. Yang, C. L.; Li, Z. H.; Li, W. J.; Liu, H. Y.; Xiao, Q. Z.; Lei, G. T.; Ding, Y. H. Batwing-Like Polymer Membrane Consisting of PMMA-Grafted Electrospun PVdF–SiO2 Nanocomposite Fibers for lithium-Ion Batteries. J. Membr. Sci. 2015, 495, 341–350. Butzelaar, A. J.; Röring, P.; Mach, T. P.; Hoffmann, M.; Jeschull, F.; Wilhelm, M.; Winter, M.; Brunklaus, G.; Théato, P. Styrene-Based Poly(Ethylene Oxide) Side-Chain Block Copolymers as Solid Polymer Electrolytes for High-Voltage Lithium-Metal Batteries. ACS Appl. Mater. Interfaces 2021, 13 (33), 39257–39270. Plummer, C. M.; Li, L.; Chen, Y. Ring-Opening Polymerization for the Goal of Chemically Recyclable Polymers. Macromolecules 2023, 56 (3), 731–750.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolytes

525

13. Laxmiprasanna, N.; Sandeep Reddy, P.; Shiva Kumar, G.; Balakrishna Reddy, M.; Ganta, K. K.; Jeedi, V. R.; Praveen, B. V. S. A Review on Nano Composite Polymer Electrolytes for High-Performance Batteries. Mater. Today Proc. 2023, 72, 286–292. 14. Wang, Q.; Xu, X.; Hong, B.; Bai, M.; Li, J.; Zhang, Z.; Lai, Y. Molecular Engineering of a Gel Polymer Electrolyte Via In-Situ Polymerization for High Performance lithium Metal Batteries. Chem. Eng. J. 2022, 428, 131331. 15. Yu, J.; Lin, X.; Liu, J.; Yu, J. T. T.; Robson, M. J.; Zhou, G.; Law, H. M.; Wang, H.; Tang, B. Z.; Ciucci, F. In Situ Fabricated Quasi-Solid Polymer Electrolyte for High-Energy-Density Lithium Metal Battery Capable of Subzero Operation. Adv. Energy Mater. 2022, 12 (2), 2102932. 16. Sun, M.; Zeng, Z.; Peng, L.; Han, Z.; Yu, C.; Cheng, S.; Xie, J. Ultrathin Polymer Electrolyte Film Prepared by In Situ Polymerization for lithium Metal Batteries. Mater. Today Energy 2021, 21, 100785. 17. Imholt, L.; Dörr, T. S.; Zhang, P.; Ibing, L.; Cekic-Laskovic, I.; Winter, M.; Brunklaus, G. Grafted Polyrotaxanes As Highly Conductive Electrolytes for Lithium Metal Batteries. J. Power Sources 2019, 409, 148–158. 18. Zhou, Z.; Tao, Z.; Zhang, L.; Zheng, X.; Xiao, X.; Liu, Z.; Li, X.; Liu, G.; Zhao, P.; Zhang, P. Scalable Manufacturing of Solid Polymer Electrolytes with Superior Room-Temperature Ionic Conductivity. ACS Appl. Mater. Interfaces 2022, 14 (29), 32994–33003. 19. Zhou, Z.; Tao, Z.; Chen, R.; Liu, Z.; He, Z.; Zhong, L.; Li, X.; Chen, G.; Zhang, P. Elastomeric Electrolyte for High Capacity and Long-Cycle-Life Solid-State Lithium Metal Battery. Small Methods 2023, 7 (4), 2201328. 20. Li, X.; He, Z.; Liu, Z.; Chen, Y.; Zhou, Z.; Chen, G.; Qi, W.; Rauber, D.; Kay, C. W. M.; Zhang, P. High-Performance Post-Treatment-Free PEDOT Based Thermoelectric with the Establishment of Long-Range Ordered Conductive Paths. Chem. Eng. J. 2023, 454, 140047. 21. Liang, P.; Li, Q.; Chen, L. M.; Tang, Z. J.; Li, Z. T.; Wang, Y.; Tang, Y. C.; Han, C. P.; Lan, Z. W.; Zhi, C. Y.; Li, H. F. The Magnetohydrodynamic Effect Enables a Dendrite-Free Zn Anode in Alkaline Electrolytes. J. Mater. Chem. A 2022, 10 (22), 11971–11979. 22. Li, Z.; Fu, J. L.; Zhou, X. Y.; Gui, S. W.; Wei, L.; Yang, H.; Li, H.; Guo, X. Ionic Conduction in Polymer-Based Solid Electrolytes. Adv. Sci. 2023, 10 (10), 2201718. 23. Dörr, T. S.; Pelz, A.; Zhang, P.; Kraus, T.; Winter, M.; Wiemhöfer, H.-D. An Ambient Temperature Electrolyte with Superior Lithium Ion Conductivity Based on a Self-Assembled Block Copolymer. Chem. A Eur. J. 2018, 24 (32), 8061–8065. 24. Pelz, A.; Dörr, T. S.; Zhang, P.; de Oliveira, P. W.; Winter, M.; Wiemhöfer, H.-D.; Kraus, T. Self-Assembled Block Copolymer Electrolytes: Enabling Superior Ambient Cationic Conductivity and Electrochemical Stability. Chem. Mater. 2019, 31 (1), 277–285. 25. Zhou, Z.; Zou, R.; Liu, Z.; Zhang, P. Deciphering the Role of Tetrahydrofuran Residue in the Poly(Ethylene Oxide)/LiTFSI Hybrid Used for Secondary Battery Electrolyte. Giant 2021, 6, 100056. 26. Miyamoto, T.; Shibayama, K. Free-Volume Model for Ionic Conductivity in Polymers. J. Appl. Phys. 1973, 44 (12), 5372–5376. 27. Huo, S. D.; Sheng, L.; Xue, W. D.; Wang, L.; Xu, H.; Zhang, H.; He, X. M. Challenges of Polymer Electrolyte with Wide Electrochemical Window for High Energy Solid-State lithium Batteries. InfoMat 2023, 5 (3), e12394. 28. Xiao, Y.; Wang, Y.; Bo, S.-H.; Kim, J. C.; Miara, L. J.; Ceder, G. Understanding Interface Stability in Solid-State Batteries. Nat. Rev. Mater. 2020, 5 (2), 105–126. 29. Zhang, T.; Li, J.; Li, X.; Wang, R.; Wang, C.; Zhang, Z.; Yin, L. A Silica-Reinforced Composite Electrolyte with Greatly Enhanced Interfacial Lithium-Ion Transfer Kinetics for High-Performance Lithium Metal Batteries. Adv. Mater. 2022, 34 (41), 2205575. 30. Banerjee, A.; Wang, X.; Fang, C.; Wu, E. A.; Meng, Y. S. Interfaces and Interphases in All-Solid-State Batteries with Inorganic Solid Electrolytes. Chem. Rev. 2020, 120 (14), 6878–6933. 31. Lou, S. F.; Zhang, F.; Fu, C. K.; Chen, M.; Ma, Y. L.; Yin, G. P.; Wang, J. J. Interface Issues and Challenges in All-Solid-State Batteries: Lithium, Sodium, and beyond. Adv. Mater. 2021, 33 (6), 2000721. 32. Reinoso, D. M.; Frechero, M. A. Strategies for Rational Design of Polymer-Based Solid Electrolytes for Advanced lithium Energy Storage Applications. Energy Storage Mater. 2022, 52, 430–464. 33. Hou, W. R.; Guo, X. W.; Shen, X. Y.; Amine, K.; Yu, H. J.; Lu, J. Solid Electrolytes and Interfaces in all-Solid-State Sodium Batteries: Progress and Perspective. Nano Energy 2018, 52, 279–291. 34. Gu, Z.-Y.; Guo, J.-Z.; Cao, J.-M.; Wang, X.-T.; Zhao, X.-X.; Zheng, X.-Y.; Li, W.-H.; Sun, Z.-H.; Liang, H.-J.; Wu, X.-L. An Advanced High-Entropy Fluorophosphate Cathode for Sodium-Ion Batteries with Increased Working Voltage and Energy Density. Adv. Mater. 2022, 34 (14), 2110108. 35. Wu, K.; Huang, J.; Yi, J.; Liu, X.; Liu, Y.; Wang, Y.; Zhang, J.; Xia, Y. Recent Advances in Polymer Electrolytes for Zinc Ion Batteries: Mechanisms, Properties, and Perspectives. Adv. Mater. 2020, 10 (12), 1903977. 36. Gao, C.; Wang, J.; Huang, Y.; Li, Z.; Zhang, J.; Kuang, H.; Chen, S.; Nie, Z.; Huang, S.; Li, W.; Li, Y.; Jin, S.; Pan, Y.; Long, T.; Luo, J.; Zhou, H.; Wang, X. A High-Performance Free-Standing Zn Anode for Flexible Zinc-Ion Batteries. Nanoscale 2021, 13 (22), 10100–10107. 37. Candhadai Murali, S. P.; Samuel, A. S. Zinc Ion Conducting Blended Polymer Electrolytes Based on Room Temperature Ionic Liquid and Ceramic Filler. J. Appl. Polym. Sci. 2019, 136 (24), 47654. 38. Li, J.; Li, F.; Li, D.; Cheng, D.; Wang, Z.; Liu, X.; Wang, H.; Zeng, X.; Huang, Y.; Xu, H. Negatively Charged Laponite Sheets Enhanced Solid Polymer Electrolytes for Long-Cycling Lithium-Metal Batteries. ACS Appl. Mater. Interfaces 2023, 15 (3), 4044–4052. 39. Li, L.; Zhao, S.; Hu, Z.; Chou, S. L.; Chen, J. Developing Better Ester- and Ether-Based Electrolytes for Potassium-Ion Batteries. Chem. Sci. 2021, 12 (7), 2345–2356. 40. Guzmán-Torres, J.; González-Juárez, E.; de la Luz Hernández-Nieto, M.; Espinosa-Roa, A.; Sánchez, E. M. Solid Polymer Electrolyte Based on PEO/PVDF/Mg(ClO4)2-[EMIM] [ESO4] System for Rechargeable Magnesium Ion Batteries. Ionics 2023, 29 (6), 2341–2349. 41. Acosta, J. L.; Morales, E. Synthesis and Characterization of Polymeric Electrolytes for Solid State Magnesium Batteries. Electrochim. Acta 1998, 43 (7), 791–797. 42. Zhou, D.; Shanmukaraj, D.; Tkacheva, A.; Armand, M.; Wang, G. Polymer Electrolytes for Lithium-Based Batteries: Advances and Prospects. Chem 2019, 5 (9), 2326–2352. 43. Wang, J.; Li, S.; Zhao, Q.; Song, C.; Xue, Z. Structure Code for Advanced Polymer Electrolyte in Lithium-Ion Batteries. Adv. Funct. Mater. 2021, 31 (12), 2008208.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes Kazunori Takada, National Institute for Materials Science, Tsukuba, Ibaraki, Japan © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is an update of K. Takada, SECONDARY BATTERIES – LITHIUM RECHARGEABLE SYSTEMS – LITHIUM-ION | Electrolytes: Solid Oxide, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 328–336, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00211-2

1 2 3 4 5 6 7 8 References Further readings

Introduction g-Li3PO4-type oxysalts NASICON-type phosphates Perovskite-type oxides Garnet-type oxides Other oxides Application of lithium-ion-conductive oxide electrolytes to electrochemical power sources Concluding remarks

527 527 529 531 533 534 534 536 536 537

Abstract Ceramic electrolytes are expected to be a fundamental solution for the safety issue of lithium-ion batteries arising from the combustible organic electrolytes. Many researches aiming at the development of highly ion-conductive solid electrolytes have given us several types of oxide solid electrolytes: especially g-Li3PO4-type, NASICON-type, perovskite-type, and garnet-type structures. Some of these have been applied to thin-film batteries. In some cases, these solid electrolytes have high Li+-ion conductivity, comparable to liquid electrolytes. Moreover, they are chemically stable so that they will fit production of batteries. However, they have not become alternatives to the organic electrolytes. This chapter describes the characteristics of some oxide solid electrolytes and the future prospects.

Key points

• • •

This chapter summarizes the development history of oxide-based lithium-ion conductive solid electrolytes, their characteristics, and their application to solid-state batteries. The highest ionic conductivities achieved among oxide-based solid electrolytes are on the order of 10−3 S cm−1. Assembling battery materials with highly conductive interfaces and good interfacial contact is the key to high performance in oxide-based solid-state batteries.

Nomenclature

Symbols and units

ap Ea V xc m s s0 s300K

Perovskite parameter Activation energy for conduction (kJmol−1, eV) Unit cell volume Percolation threshold for a simple cubic lattice Exponent Specific ionic conductivity (Scm−1) Preexponential factor in Arrhenius equation for conduction Specific ionic conductivity at 300 K

Abbreviations and acronyms LISICON NASICON PEO

526

Lithium superionic conductor Sodium super ionic conductor Poly(ethylene)oxide

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00021-9

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

1

527

Introduction

Lithium-ion batteries (LIBs) are a key technology in today’s advanced information society, which are used to power portable information equipment including mobile phones and notebook PCs. In addition, they are required to contribute to the realization of an environment-friendly society as the power source for electric vehicles and load-leveling apparatus. However, safety issue is intrinsic for the lithium-ion batteries, because they contain combustible substances, which are organic solvents used in the electrolytes. Moreover, the safety issue is becoming more serious. We need batteries to power electric vehicles and store renewable energy from photovoltaics and wind power generators for realizing a low carbon society. Batteries for such applications are much larger than those used in consumer electronics, and increasing the battery size increases the amount of combustible electrolytes and lowers the heat radiation that easily lead the batteries to thermal runaway. Solid electrolytes are expected to be a fundamental solution for this safety issue, and all-ceramic solid-state batteries are desired to satisfy the global energy storage needs. Another advantage of solid-state batteries is that they show long life. Every species other than Li+ ions, for example, counter-anions, solvent molecules, and impurities, migrate in liquid electrolyte. When such a species migrates to the surface of cathode, it may be oxidatively decomposed; or one reaching the surface of the anode may be reduced. That is, the migration can lead to side reactions that degrade the battery performance. Such electrochemical decomposition is particularly prevalent in lithium-ion batteries, because their high energy densities originate from the high cell potential generated by the combination of a highly oxidizing cathode and a highly reducing anode. In solid electrolytes, only lithium ions diffuse. This means that side reactions associated with the movement and subsequent decomposition of other species is prevented. Therefore, solid-state batteries show long cycle life, long shelf life, and extremely small self-discharge. Such high durability is beneficial for batteries in a future low carbon society, because batteries for vehicle application and stationary use need much longer life than those for consumer electronics. Solid electrolytes with high ionic conductivity are a key to the development of the solid-state battery. There are several types of solid electrolytes: nitrides, halides, sulfides, and oxides, among which oxides have the following advantages. The first advantage is their chemical stability. Other solid electrolytes are, for example, hygroscopic, and difficult to handle in ambient atmosphere, which is a major obstacle for assembling batteries or battery components. One can employ a variety of processing methods in the assembly because of the stability. Electrolyte layers should be as thin as possible in order to reduce the internal resistances and increase the energy stored in the batteries by increasing fraction of active materials. Such thin electrolyte layers are, in general, fabricated by thin-film production process, including evaporation process. However, such processes may not be suitable for the production of batteries. On the contrary, oxide solid electrolytes can be formed into thin films by wet process, including sol-gel method, which will be much better for the battery production. Finally, oxide ions are highly electronegative, meaning that their electrons are difficult to remove. The superior oxidative stability of these electrolytes make them compatible with high-voltage cathode. In general, oxides are relatively easy to be synthesized, and thus the exploration of oxide-type solid electrolytes has been going on for a long time. The first oxide electrolyte to demonstrate lithium-ion conduction was Li-b-alumina, and this finding was followed by many studies aiming at the development of highly conductive oxide electrolytes. Around 1980, most of the studies were on lithium-ion conduction in lithium-rich oxysalts, and some oxysalts, many of which have g-Li3PO4 structure, were developed. Because their ionic conductivities are of the orders of 10−7 to 10−5 Scm−1 at room temperature and lower than that of liquids by several orders of magnitude, these solid electrolytes were applied only as thin-film batteries in order to make the internal resistance as small as possible. Drastic increase of the conductivity in oxides was achieved in the 1990s. Two additional types of oxide electrolytes were developed: one has a sodium super ionic conductor (NASICON)-type structure, and the other has a perovskite-type structure. Both show ionic conductivities of 10−3 Scm−1. While this conductivity is lower than that of nonaqueous liquid electrolytes used in current lithium batteries by one order of magnitude; however, transport number for lithium ion in the liquid electrolytes is below 0.5. That is, conductivity for lithium ion in liquid electrolytes is of the order of 10−3 Scm−1. On the other hand, lithium transport number in solid electrolytes is unity. Therefore, ionic conductivity of 10−3 Scm−1 can be considered a threshold value for solid electrolytes to be viable for use in lithium-ion batteries. Such conductivity was achieved also in garnet-type structure and LiTa2PO8 in the 20th century. This chapter describes details of these oxide solid electrolytes and their application to solid-state batteries.

2

g-Li3PO4-type oxysalts

Some oxysalts including Li2SO4 show fast ionic conduction in their high-temperature phases, and many attempts were made to stabilize such highly conducting phases down to ambient temperature. In 1977, R. D. Shannon and coworkers synthesized Li2+xC1−xBxO3, Li3−xB1−xCxO3, Li4−xSi1−xPxO4, Li4−2xSi1−xSxO4, Li4+xSi1−xAlxO4, and Li5−xAl1−xSixO4 and reported that ionic conduction is enhanced when two types of oxysalts are formed into solid solutions.1 This study was followed by the discovery of fast ionic conduction in Li14Zn(GeO4)4 in 1978.2 This material was named LISICON (lithium superionic conductor), motivated by such studies, and many lithium-ion-conducting oxysalts were found around 1980: one group comprises solid solutions based on lithium nesosilicate (Li4SiO4), and the other based on g-Li3PO4.

528

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

s0 (S cm−1)

Si P

Ge Si P As

Si V

Ge Ge Ti Ti As V As V

6

10

5

10

4

10

Ea (kJ mol−1)

s300 K (S cm−1)

-4

10

-5

10

-6

10

-7

10

52 48 44 40 330

340

350

V (A) Fig. 1 Specific ionic conductivity at 300 K, s300K; activation energy for conduction, Ea; and preexponential factor, s0, as a function of unit cell volume for xLi4MIVO−4 (1−x)Li3MVO4 system (MIV ¼ Ge, Ti and MV ¼ As, V) systems with x ¼ 0.5.

Besides R. D. Shannon’s work, systematic study was also done by A. R. West and coworkers on xLi4MIVO–4(1−x)Li3MVO4 systems (MIV ¼ Ge, Ti and MV ¼ As, V). The conductivities of the solid solutions always showed maxima when x ¼ 0.4–0.6, which was consistent with R. D. Shannon’s results. On the contrary, comparison between the ionic conductivities of different types of oxysalts subsequently revealed another strong correlation between unit cell volume and the conductivity.3 Fig. 1 shows the relationship between the cell volume and the ion-conducting properties of the eight systems. The ionic conductivity increases and the activation energy for conduction decreases with increasing cell volume, while the preexponential factor remained unchanged. Since all the solid solutions plotted in the figure had a composition of x ¼ 0.5, carrier concentrations will almost be the same, which was also supported by the constant preexponential factor. On the contrary, the activation energy and the ionic conductivity are strongly correlated with the cell volume. Large cell volume provides wider conduction channels for the Li+ ions and make the Li+ ions more mobile, which lowers the activation energy and increases the conductivity. Most of the oxysalts do not contain any transition metal elements unlike NASICON-type or perovskite-type oxides described below. When the solid electrolyte contains transition metal elements, they are reduced in contact with lithium metal anodes. In other words, solid electrolytes containing transition metals are not compatible with lithium anodes. Therefore, g-Li3PO4-type oxysalts have been used in solid-state lithium batteries for a long time, despite their low ionic conductivities (below 10−5 S cm−1) at room temperature. The first lithium battery based on oxysalt electrolytes was fabricated in thin-film form back in 1983.4 T. Kudo and coworkers assembled this thin-film battery with a lithium anode and a titanium sulfide (TiS2) cathode. The electrolyte used in this battery was Li3.6Si0.6P0.4O4, which was used in the battery in its amorphous state. Although the specific conductivity at 25  C was only 5  10−6 S cm−1, thin-film formation shortened the diffusion length in the electrolyte layer, enabling the battery to operate at a current density of 16 m A cm−2. Although g-Li3PO4-type oxysalts including LISICON were available mainly in thin-film batteries, they have made a big contribution to recent research on solid-state lithium batteries; it has evolved into two important related materials: Lipon and thio-LISICON. J. B. Bates and coworkers found that when lithium silicates, lithium phosphates, or lithium phosphosilicates are formed into thin films in nitrogen-containing atmosphere by radio frequency (r.f.) magnetron sputtering, nitrogen is incorporated into the films and enhances the ionic conductivity. When lithium phosphate is sputtered in pure nitrogen atmosphere, its specific ionic conductivity increases to 2  10−6Scm−1 (25  C) in its amorphous phase, which is called Lipon.5 Although the Lipon itself is amorphous, the crystalline counterpart has g-Li3PO4 structure, and thus it can be categorized into g-Li3PO4-type oxysalts. Lipon gave birth to a variety of thin-film batteries because of its relatively high stability and wide electrochemical window. Moreover, the gLi3PO4-type structure presented solid electrolytes exhibiting much higher ionic conductivities among sulfide analogs,

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

529

Since sulfide ions more weakly attract lithium ions than oxide ions do, sulfides should generally have higher conductivities than oxides. It may suggest that a sulfide isostructural to an oxide electrolyte will show very high ionic conductivity. However, only sulfide glasses had been reported to show high ionic conductivities among sulfides in the 20th century. Of course, their specific ionic conductivities were of the order of 10−3 Scm−1, much higher than oxide glasses; there were no crystalline sulfides with fast Li+-ion conduction. In 2000, R. Kanno and coworkers reported that Li4GeS4 exhibiting the g-Li3PO4 structure. Although its conductivity is of the order of 10−7 Scm−1 and not very high, the finding led to many lithium-ionic conductors derived from the materials by aliovalent substitution. Such sulfides are categorized into thio-LISICON family. The highest conductivity of 2.2  10−3 Scm−1 (25  C) with an activation energy for conduction of 20 kJmol−1 was achieved at a composition of Li3.25Ge0.25P0.75S4, which is one of the highest conductivities in solid electrolytes. After 5 years, M. Tatsumisago and coworkers found that a precipitated phase from Li2S–P2S5 glasses shows high ionic conductivities. Recent structural analysis has revealed that the glass ceramics also have thioLISICON-related structure. When the glass ceramics were precipitated from a sulfide glass with a composition of 70Li2S–30P2S5, they show a very high conductivity of 3.2  10−3Scm−1 at ambient temperature and a remarkably low activation energy of 12 kJmol−1.

3

NASICON-type phosphates

The highest ionic conductivities of g-Li3PO4-type oxysalts are of the order of 10−5 Scm−1. Although they are relatively good ionic conductors, their conductivities are much lower than those of sulfides and nitrides. One reason for this the is high LidO bonding energy; oxide ions strongly attract lithium ions, leading to lower Li-ion mobility. Oxide ions form a close-packed array in gLi3PO4 structure, and lithium ions should migrate through the narrow conducting channels among the oxide ions. Relation between the cell volume and the conductivity shown in Fig. 1 also clearly demonstrates it; wider channel is necessary for higher ionic conductivity. In 1976, Na1+xZr2P3−xSixO12 was found to have fast Na+-ion conduction, which is called NASICON.6 In NASICON, ZrO6 octahedra are linked by PO4 tetrahedra to form three-dimensional skeleton structure as shown in Fig. 2. Since the open structure was considered to cause fast ionic conduction, many attempts have been made to obtain lithium-ion-conducting counterparts. However, simple substitution of sodium with lithium did not give a good ionic conductor; ionic conductivity of LiZr2(PO4)3 was lower than 10−9 Scm−1. On the other hand, fast ionic conduction in NASICON structure was achieved in a Ti2(PO4)3 network. In systematic studies by J.-M. Winand and coworkers as well as H. Aono and coworkers, it was revealed that ionic conductivity in the NASICON structure is strongly correlated with the size of the diffusion channels in the skeleton network. NaM2(PO4)3 compounds (M ¼ Ge, Sn, Ti, Zr, and Hf ) show the highest conductivity when M ¼ Zr.7 That is, the conductivity for Na+ ion is maximized, when the framework is composed of the largest Zr4+. On the other hand, the lithium counterparts show the highest conductivity in LiTi2(PO4)3. It is concluded that since Li+ ion has smaller ionic radius than Na+ ion, the skeleton framework consisting of ZrO6 octahedra is too large for Li+ ion to show high ionic conductivity. Fig. 3 shows the activation energies for conduction in the bulk obtained for various types of NASICON-type phosphates plotted as a function of the cell volume. Although it includes the data for different types of phosphates, which were derived from LiTi2(PO4)3, LiGe2(PO4)3, and LiHf2(PO4)3, activation energies are on a single curve regardless of the octahedral cations with a minimum at the cell volume of 1310 A˚ 3. It is

ZrO6 PO4

Na

Fig. 2 Crystal structure of sodium super ionic conductor (NASICON).

530

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

LiM2(PO4)3 M = Ge

M = Ti

M = Hf

0.45

Ea (eV)

0.40

0.35

0.30

0.25 1200

1300

1400

1500

3

V (Å ) Fig. 3 Activation energy for conduction, Ea, of LiTi2(PO4)3, LiGe2(PO4)3, LiSn2(PO4)3,LiHf2(PO4)3, and their derivatives plotted against the unit cell volume, V. Circles, inverted triangles, squares, and triangles indicate Ea for Ti, Ge, Sn, and Hf systems.

very clear from the figure that the ionic conductivity of LiM2(PO4)3 is the highest when M ¼ Ti, which is a smaller tetravalent cation than Zr4+.8 In addition to lattice size, the carrier concentration is also an important factor contributing to high ionic conductivity. Carrier concentration is, in general, controlled by doping aliovalent cations. Indeed, when part of the tetravalent titanium was substituted with trivalent cations including aluminum, scandium, yttrium, and lanthanum resulting in the formation of Li1+xAxTi2−x(PO4)3, they induced interstitial ions and enhanced the ionic conduction. The conductivities showed maxima at x ¼ 0.3, and the highest conductivity of almost 10−3 Scm−1 was achieved when A ¼ Al. Further studies revealed that the enhancement of the conductivity does not come from the optimization of the carrier density but from the improved sinterability of the resulting solid electrolytes. The cation substitution increased the ionic conductivity of LiTi2(PO4)3 from 10−6 to 10−4 Scm−1 regardless of types of the trivalent cations. That is, the conductivity increased, regardless of whether the trivalent cation was larger or smaller than Ti4+ ion, or whether the lattice was expanded or shrunk. In addition, the conductivities enhanced not only by doping the aliovalent cations but also by adding lithium phosphate (Li3PO4), lithium borate (Li3BO3), and lithium oxide (Li2O). Many samples prepared through the studies revealed strong correlation between the porosity and the observed conductivity, which suggests that the conductivity enhancement originated from the densification of the sintered pellets. Today, many studies on oxide solid electrolytes are focused on the joining solid electrolyte particles, because grain boundary resistance is a big issue for realizing oxide-type solid-state batteries, as described later. This finding has pioneered studies on grain boundaries. The contribution of bulk and grain boundary resistance can be separately evaluated by complex impedance analysis. This is because the characteristic frequencies for ionic conduction in the bulk and at the grain boundary are different. When conductivity is measured by the complex impedance method, responses from the bulk and grain boundary conductions appear as two semicircles in the Nyquist plot. Deconvoluted semicircle appearing in the higher-frequency region is always attributable to a response from bulk ion conduction, while that in the lower frequency region originates from the conduction at the grain boundary. The contribution of the bulk and the grain boundary evaluated by the above method is shown in Fig. 4. Activation energies for bulk conduction were 0.38, 0.30, and 0.42 eV for germanium, titanium, and hafnium systems, respectively. They are not changed by the aliovalent substitutions or by doping lithium oxide. On the contrary, activation energies for conduction at the grain boundary decreased with the substitution or the doping and showed minima at x ¼ 0.1–0.3. This result strongly suggests that neither the substitution nor the doping affects the ionic conduction in the bulk. They simply enhance the sinterability of the material, yielding decreases in grain boundary resistance.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

531

0.55

Ea (eV)

0.50

Hf grain boundary

0.45 Hf bulk

0.40

Ge grain boundary

Ge bulk

0.35

Ti grain boundary Ti bulk

0.30

0.0

0.2 x in

0.4

III IV Li1+xM xM 2-x(PO4)3

Fig. 4 Activation energies for conduction of bulk and grain boundary components for sodium super ionic conductor (NASICON)-type phosphates. Circles, triangles, and inverted triangles in (a) are for Li1+xAlxGe2–x(PO4)3, Li1+xAlxTi2–x(PO4)3, and Li1+xFexHf2–x(PO4)3, respectively, and those in (b) are for LiGe2(PO4)3, LiTi2(PO4)3, and LiHf2(PO4)3 doped with Li2O, respectively. Open and closed symbols correspond to bulk and grain boundary contribution, respectively.

4

Perovskite-type oxides

Perovskite-type alkaline-earth titanates (ATiO3, A ¼ Ca, Sr, Ba) had been attracting much interest for their dielectricity and ferroelectricity. J. Brous and coworkers succeeded in substituting the alkaline-earth ions with a trivalent rare earth (lanthanum) and monovalent alkali ions (lithium, sodium, and potassium) and found dielectric hysteresis and a dielectric anomaly in Li1/2La1/2TiO3.9 Y. Inaguma and coworkers observed increasing capacitance upon heating, large dielectric loss, and dielectric relaxation that originated from ionic conduction, and reported that the ionic conductivity of the system at room temperature is as high as 10−3 Scm−1.10 Perovskite oxides can be represented by a chemical formula of Li3xLa2/3−xTiO3. Fig. 5 shows the basic crystal structure of the perovskite-type titanates. Titanium atoms octahedrally coordinated with oxygen atoms occupy the corner of the cube (B-site), and the center of the cube (A site) is occupied by La3+ ion, Li+ ion, or vacancy. When the La3+ ions, the Li+ ions, and the vacancies are randomly distributed over the A sites, the lattice belongs to cubic symmetry (space group: Pm-3m). It should be noted, however, that this representation is too simplified. Recent studies based on neutron diffraction revealed that Li+ ions do not reside at the center of the cubes but occupy the center of the bottlenecks, and maximum entropy method used in one of these studies also revealed the Li+-ion conduction paths as density distribution of lithium nuclei. Although the basic structure is simple, the perovskite-type solid electrolytes show many polymorphs that originated from cation ordering or tilting of the TiO6 octahedra. When Li0.5La0.5TiO3 is synthesized by heating at 1350  C and then quenching, the lattice is

TiO6

ap

Bottle neck Fig. 5 Basic crystal structure of perovskite-type solid electrolyte.

La, Li, vacancy

532

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

cubic. On the other hand, when it is slowly cooled, broad reflections appear, suggesting a superstructure. In the slowly cooled phase, Li+ ions and La3+ ions are somewhat ordered to form alternate lithium-rich and lanthanum-rich planes, which lowers the symmetry from cubic to tetragonal to form ap  ap  2ap superlattice and gives the superstructure reflections, where ap is a perovskite parameter defined as cube root of the perovskite unit. Tilting of the TiO6 octahedra also distorts the cubic lattice and lowers the symmetry. Hexagonal symmetry was reported for Li0.5La0.5TiO3, in this case, diagonal distortion originating from cation ordering along the c-direction and tilting of TiO6 octahedra results in √2ap  √2ap  2ap superstructure. When the lithium content is very small (x < 0.08), the unit cell sometimes becomes orthorhombic. It is believed that the conductivity in perovskite-type oxides is mainly governed by two factors: bottleneck size and site percolation. When a Li+ ion migrates from one A site to a neighboring one, it should go through a bottleneck surrounded by four oxygen atoms. When a Li+ ion approaches the bottleneck (3d site in the space group of Pm-3m), the oxygen atoms attract the Li+ ions, which acts as a potential barrier for the conduction. Therefore, perovskite-type oxides also show correlation between ionic conductivity and lattice size, similar to the NASICON-type phosphates. When the conductivities of Li0.34LnTiO3 are compared for Ln ¼ La, Pr, Nd, and Sm, a clear relation between the conductivity and the lattice parameter can be found as indicated in Fig. 6.11 The larger the lattice parameter, the higher the conductivity and the lower the activation energy for conduction. That is, the larger bottleneck makes Li+ ions more mobile. Although the tendency indicated in Fig. 6 suggests that the conductivity is increased, when the A site is occupied by larger trivalent cations, La3+ is the largest trivalent cation. In order to further enlarge the bottleneck size, Y. Inaguma and coworkers introduced divalent Sr2+ ions, which are larger than La3+, to the A site in place of the lanthanides. The resultant [(Li1/2La1/2)1−xSrx] TiO3 showed a conductivity of 1.5  10−3Scm−1 at 300 K, which was slightly higher than that in the (Li,La)TiO3 system.12 Li+ ions in the perovskite structure migrate through the A sites, which are partially occupied by other Li+ ions and La3+ ions, with the residuals A site being left vacant. Because the La3+ ions are immobile and block the migration of the Li+ ions, La3+ should distribute to leave a conduction path through the system, as schematically illustrated in Fig. 7a. In other words, a group of neighboring Li+ ions and vacancies should percolate through the system to provide ionic conduction. According to the percolation theory, conductivity, s, is generally represented as s∝ðx −x c Þm

(1)

where the percolation threshold for a simple cubic lattice, xc, is 0.3117, and the exponent, m, is 2.0 for all three-dimensional lattices. Fig. 7b shows the conductivity data for LiTaO3–SrTiO3 system, where part of A sites accommodate Li+ ions and Sr2+ ions, and B sites are occupied by tantalum and titanium atoms. The best fit for the conductivity data was obtained by m ¼ 2.2  0.2 on the assumption of xc ¼ 0.3117. This value agreed with the theoretical value within the statistical error, supporting three-dimensional -1

10

-2

0.7

Sm (470 K)

10

0.6

-3

-4

10

Nd (300K) 0.5

Ea (eV)

s (S cm−1)

10

Pr (300 K) -5

10

0.4

-6

10

La (300 K) -7

10

0.380 0.382 0.384 0.386 0.388 ap (nm) Fig. 6 Specific ionic conductivity at 400 K (open circles) and activation energy for conduction (closed circles) plotted against perovskite parameter for Li0.34La0.51TiO2.94, Li0.34Pr0.56TiO3.01, Li0.34Nd0.55TiO3.00, and Li0.38Sm0.52TiO2.97.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

(a)

(b)

533

Threshold

10

-3

O La

10

-4

s (S·cm−1)

Li

10

10

-5

-6

Vacancy

10

-7

0.30 0.35 0.40 0.45 0.50 x in xLiTaO3 – (1−x)SrTiO3 Fig. 7 Schematic drawing of percolation model in perovskite-type solid electrolyte (a) and specific ionic conductivity measured for xLiTaO−3 (1−x) SrTiO3 and that calculated by percolation theory. Open and closed circles indicate the conductivity measured with Au and Li electrodes, respectively.

ionic conduction in perovskite-type oxides. Figs. 6 and 7 clearly indicate a dilemma in achieving high ionic conductivity in perovskite structure. Introduction of lower-valent cations are effective to enlarge the lattice size and widen the bottleneck for high ionic conductivity, whereas it increases the immobile cations at the A-sites that block the ionic conduction.

5

Garnet-type oxides

One of the oxide electrolytes that are developed in the 21st century has garnet-type structure. The first report was published in 2003, where Li5La3M2O12 (M ¼ Nb, Ta) were disclosed.13 Garnets are orthosilicates with a general formula of AII3BIII 2 (SiO4)3, where A and B cations are coordinated with eight and six oxygen atoms, respectively. In Li5La3M2O12, La and M occupy eight- and six-coordinated sites in the garnet-like structure as shown in Fig. 8. Although the conductivities of the garnet-type oxides are lower by an order of magnitude than those of NASICON-type phosphates and perovskite-type oxides, they show remarkable advantages: namely small grain boundary resistance and stability against lithium metal. In NASICON-type phosphates and perovskite-type oxides, contribution of grain boundaries to the resistance is much larger than that of the bulks, even when the powdered samples are sintered into pellets at temperatures higher than 1200  C. On the contrary, the contribution of the grain boundaries is in the same order of magnitude or less than that of the bulk in garnet-type Li6ALa2Ta2O12 (A ¼ Sr, Ba), even when the samples were sintered at 900  C.

NbO6 or TaO6

La Li

Fig. 8 Crystal structure of garnet-type oxide.

534

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

Another advantage of garnet-type oxides is their compatibility with lithium metal anode. Although NASICON-type phosphates and perovskite-type oxides have high specific ionic conductivity of the order of 10−3 Scm−1, they are unstable to electrochemical reduction. Lithium ions are intercalated into the NASICON structure with the reduction of Ti4+ to Ti3+ at 2.5 V versus Li+/Li.14 Perovskite oxides also contain Ti4+, which is similarly reduced at 1.5 V versus Li+/Li accompanied by insertion of Li+ ions.15 On the contrary, Li6BaLa2Ta2O12 is reported to be stable against lithium metal. No reaction is observed, even when the materials are immersed in molten lithium.16 These advantages initiated intensive studies to achieve high ionic conductivities in the garnet-type framework. A significant improvement was achieved in Li7La3Zr2O12, which exhibited a room temperature conductivity of 3  10−4 Scm−1.17 The Li7La3Zr2O12 in this study was reported to have a cubic garnet-type structure, whereas the following studies revealed that Li7La3Zr2O12 originally has a tetragonal lattice and unintentional Al-doping from the alumina crucible stabilizes the cubic phase, and the ionic conductivity of the tetragonal phase is as low as 10−6 Scm−1.18 In spite of the low ionic conductivity of the tetragonal phase, Li7La3Zr2O12 gave a variety of related materials having cubic structures and higher conductivities. Partial substitution of Zr with Nb increased the ionic conductivity to 8  10−4 Scm−1.19 and that with Ta further to 1  10−3 Scm−1.20 On the other hand, elemental substitution in the Li sites also enhanced the ionic conductivity; for example, introduction of Ga into the Li sites increased the ionic conductivity to 1.3  10−3 Scm−1.21

6

Other oxides

Since oxides are relatively stable and can be handled in ambient air, exploration of oxide solid electrolytes has a long history. Thus, there seems to be less room for discovering novel solid electrolytes among oxides than sulfides. In fact, ionic conductivities of the order of 10−3 Scm−1 have been achieved only in three kinds of frameworks, i.e., in NASICON, perovskite, and garnet. However, a new framework that provides high ionic conductivity appeared unexpectedly. LiTa2PO8 was the recently emerged solid electrolyte.22 It has an unprecedented anionic framework made of TaO6 octahedra connected by PO4 tetrahedra, as illustrated in Fig. 9 and shows three-dimensional ionic diffusion with bulk conductivity of 1.6  10−3 Scm−1.

7

Application of lithium-ion-conductive oxide electrolytes to electrochemical power sources

Solid electrolytes with g-Li3PO4-type structure have been applied to thin-film batteries as mentioned above, and development of Lipon gave birth to a variety of thin-film batteries. Thin-film batteries are now expected to store energy from solar cells and supply power to smart cards, RFID tags, and implantable medical devices, wireless sensors, etc. In addition to such applications in the coming smart society, thin-film batteries have proved potentials of solid-state batteries. One of the most important features of solid electrolytes is single-ion conduction: only lithium ions are mobile in the solid electrolytes. Since there are no species that diffuse to the electrode surface and cause side reactions, solid electrolytes are anticipated to provide high durability to batteries. For example, a thin-film battery employing Lipon with a Li anode and a LiCoO2 cathode exhibits excellent cycling performance, with only 3% capacity fade after 30,000 cycles.23 The absence of side reactions in solid-state batteries is believed to pave the way for high-voltage cathodes. Although several kinds of 5-V cathodes, e.g., LiNixMn2−xO4, LiCoxMn2−xO4, and LiCoPO4, were developed in 1990s to increase theoretical energy densities, they have never been applied to lithium-ion batteries, because their high cathode potentials decompose most nonaqueous electrolytes. On the other hand, mobility of oxide ions, which are the conceivable reactant to be oxidized in oxide-based solid electrolytes, is so low at room temperature that the solid electrolytes hardly undergo oxidative decomposition. Therefore, solid-state batteries could utilize high-voltage cathodes. LiNi0.5Mn1.5O4 was reported to show stable cycling in a thin-film battery.24 the capacity retention for this cell was 90% after 10,000 cycles; moreover, its accumulated irreversible capacity for 1000 cycles, which is a charge loss accumulated during the cycle, is only 7.9 mAh g−1 based on the weight of the cathode material suggesting that oxidative decomposition of the electrolyte hardly proceed in the solid-state system. Li

TaO6

PO4

Fig. 9 Crystal structure of LiTa2PO8.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

535

As described above, thin-film batteries have clearly demonstrated the advantages of solid-state systems; however, solid-state is strongly required for large-sized lithium batteries. Although lithium-ion batteries, which are typical high-performance rechargeable batteries, are expected to play an important role in the coming low-carbon society, the safety issue is much more serious, and the durability must be enhanced. When such bulky lithium batteries are fabricated with oxide solid electrolytes, the biggest difficulty is a large resistance at grain boundaries. High ionic conductivities of the electrolytes are necessary for the bulky batteries, because lithium ions must diffuse long distance. The highest conductivities among oxide solid electrolytes have achieved the order of 10−3 S cm−1 in NASICON-type, perovskite-type, garnet-type, and LiTa2PO8 solid electrolytes, however, they show the large grain boundary resistances that increase the internal resistance, which lowers the power density of solid-state batteries. Of course, sintering process at high temperatures will reduce the resistance to some extent; however, such high temperature heat treatment induces interdiffusion between the electrode and electrolyte materials. The interdiffusion results in the formation of impurity phases at the interface, which is a critical problem to overcome for attaining practical battery performance. One way to overcome this problem is lowering sintering temperature during the battery construction. Candidate solid electrolytes that are sinterable at low temperatures are g-Li3PO4-type oxysalts. Although their ionic conductivities of oxysalts had been of the order of 10−5 S cm−1, employment of spark-plasma sintering has increased that for Li3.5Ge0.5V0.5O4 to 9.6  10−5 S cm−1 at a low sintering temperature of 700  C. In addition, Li3.5Ge0.5V0.5O4 is compatible with typical cathode materials with layered structures (LiCoO2 and LiNi1/3Mn1/3Co1/3O2): co-sintering of these materials at 700  C does not form impurity phases at their interfaces. The high sinterability and thermodynamic compatibility with cathode materials enable us to fabricate bulk-type solid-state batteries; a solid-state half-cell using the Li3.5Ge0.5V0.5O4 was successfully constructed by spark-plasma sintering at 450  C.25 The g-Li3PO4-type solid electrolytes seem good candidates for bulk-type solid-state batteries owing to the high sinterability and cathode compatibility with the cathode materials. However, the conductivities are below 10−4 S cm−1, and thus the operation temperature of the solid-state battery is limited to an elevated temperature of 60  C. Although increasing ionic conductivity is necessary to realize solid-state batteries that are operatable at room temperature, the relationship between the lattice volume and ionic conductivity shown in Fig. 1 suggests that the conductivity of g-Li3PO4-type solid electrolyte is already approaching its highest value. On the other hand, compositional complexity is proposed as a new strategy for enhancing ionic conductivity. g-Li3PO4-type solid electrolytes in quasi-ternary oxysalt systems tend to show higher ionic conductivities and lower activation energy for conduction than those in quasi-binary systems with the same lattice volume. On the basis of this concept, ionic conductivity of Li3.5Ge0.5V0.5O4 has been increased to 1.5  10−4 S cm−1 with a reduced activation energy of 0.26 eV in a quasi-ternary oxysalt, Li3.68(Ge0.6V0.36Ga0.04)O4.26 The above approaches for realizing bulk-type solid-state batteries are employing solid electrolytes with high sinterability and increasing their ionic conductivity. Another approach is improving sinterability of solid electrolytes with high ionic conductivity. Li7La3Zr2O12 with garnet structure is regarded as a promising solid electrolyte for this approach because it is stable against lithium metal and has a conductivity as high as 10−3 S cm−1 in the Ta or Ga-substituted systems. However, it needs higher temperature than 1000  C for sintering. The sintering temperature was lowered by combination of Bi substitution for Ta and introducing compositional deviation from the stoichiometry. The introduction of Bi and La deficiency caused by the compositional deviation forms a molten phase composed of Li2O–Bi2O3, which lowers the sintering temperature of the garnet-type solid electrolyte to 775  C. The low sintering temperature enables the construction of solid-state cells by co-sintering.27 However, the Bi substitution decreases the conductivity of the garnet phase to 1.8  10−4 S cm−1, and thus the cell is operatable only at 60  C. Formation of molten phases seems to be effective in lowering sintering temperature. A drastic reduction in the sintering temperature of Li7La3Zr2O12 was attained by a unique sintering process under the presence of a molten phase.28 In this sintering technique, a Nb-substituted garnet-type solid electrolyte with a composition of Li6.4La3Zr1.4Nb0.6O12 was first converted to (Li5.4H1.0)La3Zr1.4Nb0.6O12 by H+/Li+ ion-exchange first, and then the protonated (Li5.4H1.0)La3Zr1.4Nb0.6O12 was heated with a mixture of LiOH and LiNO3 with a molar ratio of 1:1 as a Li+ provider in the sintering step. The low melting point of 191  C of the Li+ provider forms a molten phase that promotes the sintering, and high relative density of 90% was achieved at a remarkably low sintering temperature of 400  C. A bulk-type solid-state battery was fabricated at the low sintering temperature of 400  C by employing this sintering technique. The battery constructed with LiNi1/3Mn1/3Co1/3O2 and lithium metal as the cathode and the anode, respectively, could be cycled at 25  C; however, the operating current density was limited to 40 mA cm−2 (0.02  C), because the ionic conductivity of the Li6.4La3Zr1.4Nb0.6O12 prepared by this method was only 2  10−4 S cm−1. Recently, some private companies are prototyping solid-state batteries. These are chip-type batteries fabricated by applying the processes used to produce multilayer ceramic capacitor (MLCC). According to their press releases, the materials with the highest performance are not always employed in the solid-state batteries. For example, most of the batteries use Li1+xAlxGe2−x(PO4)3 as the solid electrolyte, which is categorized as a NASICON-type solid electrolytes. However, the highest ionic conductivity among NASICON-type solid electrolytes is achieved at Li1+xAlxTi2−x(PO4)3 as mentioned in this chapter. Its ionic conductivity is of the order of 10−3 S cm−1, while it remains of the order of 10−4 S cm−1 in Li1+xAlxGe2−x(PO4)3. The use of Li1+xAlxGe2−x(PO4)3in spite of its lower conductivity, is probably due to its higher sinterability than that of Li1+xAlxTi2−x(PO4)3. In addition, the cathode materials employed in most of the solid-state batteries are LiCoPO4, Li3V2(PO4)3, or Li2CoP2O7, although their specific capacities are lower than that of layered oxides used in the current liquid-electrolyte batteries, which will be due to the compatibility with the solid electrolyte. Because battery materials showing the highest performance are not combined in the solid-state batteries, their performance has not become comparable to that of lithium-ion batteries.

536

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

Although practical performance for general purpose has not been attained in bulk-type solid-state batteries with oxide solid electrolytes yet, oxide solid electrolytes have successfully improved performance of other types of solid-state batteries as interfacial modification layers. Since modification layers can be thin, electrochemical stability is much more important than ionic conductivity, and oxide solid electrolytes meet this requirement. The following are some examples of such interfacial modification. Poly(ethylene)oxide (PEO)-based polymer electrolyte is a promising candidates for use in lithium polymer batteries because of its low glass transition temperature and flexibility, which are intended to apply to the large-sized batteries. However, as it is not stable to electrochemical oxidation, it cannot be combined with 4 V cathodes, for example, lithium cobalt oxide (LiCoO2) and lithium manganese oxide (LiMn2O4). Therefore, 3 V cathodes, e.g., vanadium pentaoxide (V2O5) and lithium iron phosphate (LiFePO4), have been used in such batteries. On the contrary, a research group of Central Research Institute of Electric Power Industry proposed a new type of battery with a PEO-based polymer electrolyte using a 4 V cathode, LiCoO2. They coated the surface of LiCoO2 particles with a thin film of an oxide solid electrolyte, Li3PO4, in order to use LiCoO2 in the PEO electrolyte. The oxide electrolyte layer prevents the direct contact of the polymer electrolyte with the LiCoO2 and suppresses oxidative decomposition of the PEO-based polymer electrolyte, which was observed as the reduction in the electrode impedance. Moreover, they demonstrated that the technique enables us to use a 5 V cathode, LiNi0.5Mn1.5O4, in the PEO electrolyte.29 Another good example of the availability of oxide solid electrolytes is an interfacial modification between oxide cathodes and sulfide solid electrolytes. Sulfide solid electrolytes have higher performance in ionic conduction than oxide ones; the ionic conductivities have reached 10−2 Scm−1, and activation energies for conduction are less than 0.2 eV. In addition, grain boundary resistances are extremely low even without the sintering process; one can make a solid electrolyte layer with low resistance only by cold pressing owing to their deformability. In contrast to these advantages, sulfide solid electrolytes have a drawback: when they are connected to high-voltage cathodes, forming a highly resistive layer at the interface. Weak attraction between sulfide ions and Li+ ions is the reason for their high ionic conductivity in sulfides. On the other hand, when it is contacted to an oxide, or a high-voltage cathode, the high-voltage decreases lithium-ion concentration of the sulfide solid electrolyte at the interface to the cathode. Furthermore, the weak attraction between sulfide ions and Li+ ions makes the decrease in the lithium ion concentration intense to form a lithium-depleted layer on the electrolyte side of the interface. Because charge carriers are depleted at the interface, the conductivity is drastically decreased there, bringing about a high interfacial resistance. When an oxide solid electrolyte layer is interposed at the interface, it prevents the formation of the lithium-depleted layer and reduces the interfacial resistance. The interfacial modification was shown to improve the high-rate capability of solid-state lithium battery to be comparable to that of commercialized lithium-ion cells.30

8

Concluding remarks

A variety of oxide electrolytes have been developed as shown here, and high ionic conductivities have been achieved in these electrolytes; however, in most cases they are only applied to thin-film batteries. Of course, the ionic conductivities of oxide solid electrolytes are somewhat lower than those of sulfides and liquid electrolytes. The highest specific conductivity of liquid systems used in commercialized lithium-ion cells reaches 10−2 Scm−1; however, this value contains the contribution of the anions. When taking into account that transport number for Li+ ions is unity in the oxide solid electrolytes, it can be concluded that the specific ionic conductivity of 10−3 Scm−1 will be high enough for battery application. Sulfide solid electrolytes have higher ionic conductivities than oxides, and practical performance has been attained in solid-state batteries with sulfide solid electrolytes. Although they are expected to solve the safety issue of lithium-ion batteries, and development of the solid-state batteries with oxide solid electrolytes is under progress. Because the sulfide solid electrolytes are not so chemically stable and difficult to handle, oxide solid electrolytes are preferable for mass production of solid-state batteries to sulfides. The biggest challenge in realizing oxide-type solid-state batteries are the large grain boundary resistance and poor compatibility between solid electrolytes and electrode materials at the sintering temperature. Collaborative research between materials and process science on combining the battery materials via ionically conductive interfaces is necessary to overcome these problems.

References 1. Shannon, R. D.; Taylor, B. E.; English, A. D.; Berzins, T. New Li Solid Electrolytes. Electrochim. Acta 1977, 22, 783–796. 2. Hong, H. Y.-P. Crystal Structure and Ionic Conductivity of Li14Zn(GeO4)4 and Other New Li+ Superionic Conductors. Mater. Res. Bull. 1978, 13, 117–127. 3. Rodger, A. R.; Kuwano, J.; West, A. R. Li+ ion Conducting Solid Solutions in the Systems Li4XO4–Li3YO4: X¼Si, Ge, Ti; Y¼P, As, V; Li4XO4–LiZO2: Z¼Al, Ga, Cr and Li4GeO4–Li2CaGeO4. Solid State Ion. 1985, 15, 186–198. 4. Kanehori, K.; Matsumoto, K.; Miyauchi, K.; Kudo, T. Thin Film Solid Electrolyte and Its Application to Secondary Lithium Cell. Solid State Ion. 1983, 9-10, 1445–1448. 5. Bates, J. B.; Dudney, N. J.; Gruzalski, G. R. Fabrication and Characterization of Amorphous Lithium Electrolyte Thin Films and Rechargeable Thin-Film Batteries. J. Power Sources 1993, 43, 103–110. 6. Goodenough, J. B.; Hong, H. Y.-P.; Kafalas, J. A. Fast Na+-Ion Transport in Skeleton Structure. Mater. Res. Bull. 1976, 11, 203–220. 7. Winand, J.-M.; Rulmont, A.; Tarte, P. Nouvelles Solutions Solides LI(MIV)2−x(NIV)x(PO4)3 (L¼Li, Na; M, N,¼Ge, Sn, Ti, Zr, Hf ) Synthésis et étude par Diffraction x et Conductivité Ionique. J. Solid State Chem. 1991, 93, 341–349. 8. Aono, H.; Sugimoto, E.; Sadaoka, Y.; Imanaka, N.; Adachi, G. The Electrical Properties of Ceramic Electrolytes for LiMxTi2−x(PO4)3; yLi2O, M¼Ge, Sn, Hf, Zr Systems. J. Electrochem. Soc. 1993, 140, 1827–1833.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Oxide Electroltyes

537

9. Belous, A. G.; Novitskaya, G. N.; Polyanetskaya, S. V.; Gornikov, Y. I. Study of Complex Oxides With the Composition Li2/3−xLi3xTiO3. Inorg. Mater. 1987, 23, 412–415. 10. Inaguma, Y.; Liquan, C.; Itoh, M.; Nakamura, T.; Uchida, T.; Ikuta, H.; Wakihara, M. High Ionic Conductivity in Lithium Lanthanum Titanate. Solid State Commun. 1993, 86, 689–693. 11. Itoh, M.; Inaguma, Y.; Jung, W.-H.; Chen, L.; Nakamura, T. High Lithium Ion Conductivity in the Perovskite-Type Compounds Ln1/2Li1/2TiO3 (Ln¼La, Pr, Nd, Sm). Solid State Ion. 1994, 70-71, 203–207. 12. Inaguma, Y.; Matsui, Y.; Shan, Y.-J.; Itoh, M.; Nakamura, T. Lithium Ion Conductivity in the Perovskite-Type LiTaO3–SrTiO3 Solid Solution. Solid State Ion. 1995, 79, 91–97. 13. Thangadurai, V.; Kaack, H.; Weppner, W. Novel Fast Lithium Ion Conduction in Garnet-Type Li5La3M2O12. J. Am. Ceram. Soc. 2003, 86, 437–440. 14. Aatiq, A.; Ménétrier, M.; Croguennec, L.; Suard, E.; Delmas, C. On the Structure of Li3Ti2(PO4)3. J. Mater. Chem. 2002, 12, 2971–2978. 15. Bohnke, O.; Bohnke, C.; Fourquet, J. L. Mechanism of Ionic Conduction and Electrochemical Intercalation of Lithium Into the Perovskite Lanthanum Lithium Titanate. Solid State Ion. 1996, 91, 21–31. 16. Thangadurai, V.; Weppner, W. Li6ALa2Ta2O12 (A¼Sr, Ba): Novel Garnet-Like Oxides for Fast Lithium Ion Conduction. Adv. Funct. Mater. 2005, 15, 107–112. 17. Murugan, R.; Thangadurai, V.; Weppner, W. Fast Lithium Ion Conduction in Garnet-Type Li7La3Zr2O12. Angew. Chem. Int. Ed. 2007, 46, 7778–7781. 18. Awaka, J.; Kijima, N.; Hayakawa, H.; Akimoto, J. Synthesis and Structure Analysis of Tetragonal Li7La3Zr2O12 With the Garnet-Related Type Structure. J. Solid State Chem. 2009, 182, 2046–2052. 19. Ohta, S.; Kobayashi, T.; Asaoka, T. High Lithium Ionic Conductivity in the Garnet-Type Oxide Li7−XLa3(Zr2−X, NbX) O12 (X¼ 0–2). J. Power Sources 2011, 196, 3342–3345. 20. Li, Y.; Han, J.-T.; Wang, C.-A.; Xie, H.; Goodenough, J. B. Optimizing Li+ Conductivity in a Garnet Framework. J. Mater. Chem. 2012, 22, 15357–15361. 21. Bernuy-Lopez, C.; Manalastas, W., Jr.; Amo, J. M. L.; Aguadero, A.; Aguesse, F.; Kliner, J. A. Atmosphere Controlled Processing of Ga-Substituted Garnets for High Li-Ion Conductivity Ceramics. Chem. Mater. 2014, 26, 3610–3617. 22. Kim, J.; Kim, J.; Avdeev, M.; Yun, H.; Kim, S. J. LiTa2PO8: A Fast Lithium-Ion Conductor With New Framework Structure. J. Mater. Chem. A 2018, 6, 22478–22482. 23. Wang, B.; Bates, J. B.; Hart, F. X.; Sales, B. C.; Zuhr, R. A.; Robertson, J. D. Characterization of Thin-Film Rechargeable Lithium Batteries With Lithium Cobalt Oxide Cathodes. J. Electrochem. Soc. 1996, 143, 3203–3213. 24. Li, J.; Ma, C.; Chi, M.; Liang, C.; Dudney, N. J. Solid Electrolyte; The Key for High-Voltage Lithium Batteries. Adv. Energy Mater. 2015, 5, 1401408. 25. Okumura, T.; Takeuchi, T.; Kobayashi, H. All-Solid-State Batteries With LiCoO2-Type Electrodes: Realization of An Impurity-Free Interface by Utilizing a Cosinterable Li3.5Ge0.5V0.5O4 Electrolyte. ACS Appl. Energy Mater. 2021, 4, 30–34. 26. Zhao, G.; Suzuki, K.; Okumura, T.; Takeuchi, T.; Hirayama, M.; Kanno, R. Extending the Frontiers of Lithium-Ion Conducting Oxides: Development of Multicomponent Materials With g-Li3PO4-Type Structures. Chem. Mater. 2022, 34, 3948–3959. 27. Watanabe, K.; Tashiro, A.; Ichinose, Y.; Takeno, S.; Suematsu, K.; Mitsuishi, K.; Shimanoe, K. Lowering the Sintering Temperature of Li7La3Zr2O12 Electrolyte for Co-Fired AllSolid-State Batteries Via Partial Bi Substitution and Precise Control of Compositional Deviation. J. Cerma. Soc. Jpn. 2022, 130, 416–423. 28. Ohta, S.; Kawakami, M.; Nozaki, H.; Yada, C.; Saito, T.; Iba, H. Li+ Conducting Garnet-Type Oxide Sintering Triggered by an H+/Li+ Ion-Exchange Reaction. J. Mater. Chem. A 2020, 8, 8989–8996. 29. Seki, S.; Kobayashi, Y.; Miyamoto, H.; Mita, Y.; Iwahori, T. Fabrication of High-Voltage, High-Capacity All-Solid-State Lithium Polymer Secondary Batteries by Application of the Polymer Electrolyte/Inorganic Electrolyte Composite Concept. Chem. Mater. 2005, 17, 2041–2045. 30. Ohta, N.; Takada, K.; Zhang, L.; Ma, R.; Osada, M.; Sasaki, T. Enhancement of the High-Rate Capability of Solid-State Lithium Batteries by Nanoscale Interfacial Modification. Adv. Mater. 2006, 18, 2226–2229.

Further readings 1. Aono, H.; Imanaka, N.; Adachi, G. High Li+ Conducting Ceramics. Acc. Chem. Res. 1994, 27, 265–270. 2. Bachman, J. C.; Muy, S.; Grimaud, A.; Chang, H.-H.; Pour, N.; Lux, S. F.; Paschos, O.; Maglia, F.; Lupart, S.; Lamp, P.; Giordano, L.; Shao-Horn, Y. Inorganic Solid-State Electrolytes for Lithium Batteries: Mechanisms and Properties Governing Ion Conduction. Chem. Rev. 2016, 116, 140–162. 3. Knauth, P. Inorganic Solid Li Ion Conductors: An Overview. Solid State Ion. 2009, 180, 911–916. 4. Ren, Y.; Chen, K.; Chen, R.; Liu, T.; Zhang, Y.; Nan, C.-W. Oxide Electrolytes for Lithium Batteries. J. Am. Ceram. Soc. 2015, 98, 3603–3623. 5. Stramare, S.; Thangadurai, V.; Weppner, W. Lithium Lanthanum Titanates: A Review. Chem. Mater. 2003, 15, 3974–3990. 6. Tangadurai, V.; Narayanan, S.; Pinzaru, D. Garnet-Type Solid-State Li Ion Conductors for Li Batteries: Critical Review. Chem. Soc. Rev. 2014, 43, 4714–4727.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes Ryoji Kanno, Institute of Innovative Research, Tokyo Institute of Technology, Yokohama City, Kanagawa, Japan © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 3 4 5 5.1 5.2 5.3 5.4 5.5 5.6 6 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 7 7.1 7.2 8 8.1 8.2 8.3 8.4 8.5 8.5.1 8.5.2 8.6 8.6.1 8.6.2 8.7 9 9.1 9.2 9.3 9.4 9.5 9.6 9.6.1 9.6.2 9.6.3 9.6.4 9.6.5 9.7 9.8 9.9 9.10 9.11 10 References

538

Introduction Overview of lithium sulfide electrolytes Structural characters of glass and crystalline compounds in the Li2S—P2S5 binary system Comparison of ionic conductivity with major lithium-ion conductors Sulfide glass system History Materials of sulfide glass Ionic conductivity Glass and crystalline phases Structure Synthesis Thio-LISICON General comments Materials The Li2S–GeS2–P2S5 system The Li2S–SiS2–Al2S3 and Li2S–SiS2–P2S5 systems The Li2S–P2S5 system Material design diagram Synthesis Structure Li7P3S11 Synthesis and conductivity Crystal structure Argyrodite sulfides General comments Structure Lithium ion diffusion pathway Factors that affect ionic conduction Synthesis Solid-state synthesis Liquid-phase synthesis Several topics related to argyrodite Effect of grain boundaries on conductivity measurements Stability with electrodes Solid-state battery characteristics LGPS-type sulfides General comments Material diversity Synthesis and phase diagram Single crystals Solution methods Other chemistries Si and Sn systems LidPdS systems Halogen doping Oxygen substitution Other chemistries Structure Conduction pathways Characteristics as ionic conductor Lithium conduction mechanism Solid-state batteries Concluding remarks

Encyclopedia of Electrochemical Power Sources, 2nd Edition

539 539 541 543 543 543 544 544 544 545 546 546 546 546 547 547 547 547 548 548 548 548 549 550 550 550 550 551 552 552 552 552 552 553 553 554 554 555 555 557 557 557 557 558 558 558 558 558 558 558 560 561 562 564

https://doi.org/10.1016/B978-0-323-96022-9.00273-5

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

539

Abstract This section describes an overview of lithium-based sulfide solid electrolytes. Sulfide solid electrolytes have relatively high lithium ionic conductivity. Taking advantage of their conductivity, their use as solid electrolytes has been proposed for solid-state batteries with properties comparable to those of lithium-ion batteries. Sulfide glasses were first found to have high ionic conductivity. In 2000, thio-LISICON was found to have ionic conductivity higher than that of sulfide glasses. Subsequently, materials such as Argyrodite, Li7P3S11, and LGPS were discovered. The material variety, synthesis, and properties of the sulfide glass, Argyrodite, thio-LISICON, and LGPS systems are described.

Glossary All solid-state batteries Batteries composed of all solid-state cells. All solid-state cell Cells composed of solid-state materials. Solid-state materials are used for positive electrode, negative electrode, and electrolyte. Argyrodite Lithium ion conductors based on the Argyrodite-type structure. Chemical stabilities Chemically stable materials. For example, non-sensitive to moisture, air, O2, etc. No chemical reaction between the components used for batteries. Electrochemical stabilities Electrochemical potential window. LGPS Lithium ion conductors based on the LGPS-type structure. The original composition is the LGPS system is Li10GeP2S12. Lithium battery General definition of a battery that uses lithium as a charge carrier. Lithium-ion battery Name of a 4 V-class rechargeable battery with intercalation electrodes and organic solvent-based electrolyte. Lithium-ion conductor Materials which show lithium ionic diffusion in solid materials. Thio-LISICON LIthium Super Ionic CONductor. “Thio” indicates sulfide system.

Key points • • • • • •

1

Sulfide solid electrolytes have made rapid progress as electrolytes for practical batteries. The development of sulfide-based solid electrolytes has a long history. From the initial pioneering research, material development has progressed until they were applied as electrolytes in practical solid-state batteries. The electrolytes that have been attempted to be used as practical materials can be classified into sulfide glass, thio-LISICON, Argyrodite, LGPS, and Li7P3S11 systems, and their structures are related to each other. Each material system has its own ionic conduction mechanism and guidelines for improving its properties. All-solid-state batteries using LGPS and Argyrodite as electrolytes exhibit excellent characteristics superior to those of liquid-based batteries. Materials with excellent properties can be used as electrolytes for lithium batteries, regardless of whether they are liquid or solid forms.

Introduction

The development of new materials leads to the invention of new devices. Introduced in 1991, lithium-ion batteries (LIBs) have groundbreaking performance and are still evolving. Research aimed at creating new electrochemical materials and contributing to the development of novel energy storage devices faces numerous hurdles. The performance limitations of rechargeable batteries are primarily determined by their material properties, necessitating the development of novel materials with superior characteristics. In the pursuit of diligently understanding the phenomenon of ion diffusion within solids, exploring new substances via research and attempting their application in batteries have enabled the successful exploration of lithium-based systems, and the goal of producing battery solids is finally beginning to yield results. Particularly, the exploitation of high ionic conductivity materials with the sulfide type has facilitated the emergence of a new category of energy storage devices, including the all-solid-state battery. This chapter reviews the history of the development of lithium solid electrolytes with the sulfide type and their application in all-solidstate batteries. Particular focus is given to the development process of these electrolytes, which surpasses the conductivity characteristics of liquid-electrolyte systems targeted by lithium-ion conductors, and its application to solid-state batteries is described.

540

2

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

Overview of lithium sulfide electrolytes

The starting point for the development of lithium-based sulfide solid electrolytes was the sulfide glass electrolyte.1,2 Glass electrolytes were investigated as silver and copper ionic conductors, and their development was subsequently extended to the lithium-based electrolytes. Fig. 1 shows typical material systems of sulfide-based lithium ionic conductors, their ionic conductivities, the year of the material was found, and how the ionic conductivity improved as the research progressed. Sulfide glasses were first found to have high ionic conductivity, and for a long time it was thought that glass systems had higher ionic conductivity than crystalline systems. In 2000, thio-LISICON was discovered3,4 and shown to be a crystalline material that exceeded the ionic conductivity of sulfide glasses.5 Subsequently, materials such as Argyrodite,6,7 Li7P3S11,8,9 and LGPS10 were discovered, and it became clear that materials with ionic conductivity higher than that of the liquid electrolytes used in lithium-ion batteries commonly exist. The crystal structures of typical materials are shown in Fig. 2. In addition, the compositions of the representative compounds that exist in the currently reported material systems and their ionic conductivity values are summarized in Table 1. All of them have unique crystal structures suitable for lithium ion diffusion in the solid materials.

LPGS Li7P3S11

Sulfide-glass Argyrodite Thio-LISICON

Li2MS3

Fig. 1 Typical material systems of sulfide-based lithium ionic conductors, their ionic conductivities, the year the material was found, and how the ionic conductivity improved as the research progressed.

LiS

4

PS Li2S-P2S5 glass

4

E-Li3PS4 (Thio LISICON)

C PS PS

2 7

LiS

4

4

Li10GeP2S12 Argyrodite

Li7P3S11

Fig. 2 The crystal structures of typical lithium solid electrolytes in the sulfide system. Li2S-P2S5 glass,11 b-Li3PS4 (Thio LISICON), Argyrodite, Li7P3S11, and LGPS.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes Table 1

Typical sulfide-based lithium solid electrolytes.

Structure Type Sulfide Crystal LGPS

Chemical composition system

Composition

Conductivity at rt. s/mS cm−1

References

Li–M–P–S–X (M ¼ Ge, Si, Sn; X ¼ Cl, O)

Li10GeP2S12 Li9.54Si1.74P1.44S11.7Cl0.3 Li9.42Si1.02P2.1S9.96O2.04 Li9.54[Si0.6Ge0.4]1.74P1.44S11.1Br0.3O0.6 Li6PS5Cl Li6.6Ge0.6P0.4S5I Li7Ge3PS12 Li6-xPS5-xCl1+x Li4.275Ge0.61Ga0.25S4 Li4SnS4 b-Li3PS4 Li3.25Ge0.25P0.75S4 Li7P3S11 Li2S-P2S5 Li2S-P2S5-LiI M2S-GeS2-MI (M ¼ Li, Ag) 0.03Li3PO4–0.59Li2S-0.38SiS2

12 25 0.32 32 1.9 18.4 0.11 17 6.5  10−2 7.0  10−5 0.16 2.2 17 0.1 1

10 12 13 14 6,7 15 16 17 3,4 18 19 20 8,9 21 2 1 22

Argyrodite

Li–M–P–S–X (M ¼ Si, Ge; X ¼ Cl, Br, I)

Thio-LISICON

Li–M’–M”–S (M’, M” ¼ P, Si, Ge, Sn, Al, Ga, etc.)

Li7P3S11 Sulfide Glass

Li–P–S Li2S-P2S5

3

541

0.69

Structural characters of glass and crystalline compounds in the Li2S—P2S5 binary system

There are common structural features in the glass and crystalline phases with high lithium ion conductivity. The structure of these sulfides changes systematically with the composition both in the glassy and crystalline phases. Within the Li2S-P2S5 system, for example, various compounds including Li2P2S6, Li4P2S6, Li7P3S11, a-Li3PS4, b-Li3PS4, g-Li3PS4, LT-Li7PS6, Li7P3S11, and HT-Li7PS6 are available, which can be categorized based on the arrangement pattern of the PS4 tetrahedra.23 Fig. 3 shows the

Argyrodite

Thio-LISICON

LGPS

Fig. 3 Crystal structures observed in the in the Li2S-P2S5 system. Arrangements of MX4 tetrahedra in the a, b, and g-phases in Li3PS4. From Homma, K.; Yonemura, M.; Kobayashi, T.; Nagao, M.; Hirayama, M.; Kanno, R. Crystal Structure and Phase Transitions of the Lithium Ionic Conductor Li3PS4. Solid State Ion 2011, 182 (1), 53–58. doi: 10.1016/j.ssi.2010.10.001.; Kudu, Ö. U.; Famprikis, T.; Fleutot, B.; Braida, M.-D.; Le Mercier, T.; Islam, M. S.; Masquelier, C. A Review of Structural Properties and Synthesis Methods of Solid Electrolyte Materials in the Li2S−P2S5 Binary System. J. Power Sources 2018, 407, 31–43. https://doi.org/10.1016/j.jpowsour.2018.10.037.

542

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

Fig. 4 The anionic species in the Li2S-P2S5 system.24 From Kudu, Ö. U.; Famprikis, T.; Fleutot, B.; Braida, M.-D.; Le Mercier, T.; Islam, M. S.; Masquelier, C. A Review of Structural Properties and Synthesis Methods of Solid Electrolyte Materials in the Li2S−P2S5 Binary System. J. Power Sources 2018, 407, 31–43. https://doi.org/10.1016/j.jpowsour.2018.10.037.

crystalline phases observed in the Li2S - P2S5 binary system. Fig. 4 shows the anion units found in both the crystalline and glassy phases. Five anion species are observed. - The isolated PS43− tetrahedra form a structure with a composition of Li2S (> 75 mol%). - The (P2S74−), consisting of PS43− tetrahedra sharing two corners, is observed in a structure with a low amount of Li2S (< 75 mol %) composition. - The P2S64− unit is formed by two PS3 units and a PdP bond between them. The oxidation state of the phosphorus in the P2S64− anion is +4. - In the 60 mol% Li2S composition, there is meta-thiodiphosphate (P2S62−) with PS4 units sharing two edges, and meta-thiophosphate with PS−3 forming chains linked by corner sharing. Table 2 summarizes the structures that appear in the glassy and crystalline phases and the constituent units. Various structural units (P2S64−, P2S74−, PS43−) are present in glassy materials with Li2S compositions above 60 mol%. The relative proportions of these units depend on the initial composition of the material. For each composition, the structural units with the highest content are shown in bold in Table 2. The type and relative ratio of the structural unit in each crystalline phase in the Li2S-P2S5 system can be characterized by the crystal structure.

Table 2 Summary of the crystalline phases and the local structures observed for the crystalline and glassy phases in the Li2S-P2S5 system. Mol% Li2S

Crystalline phases observed After heat-treatment

Local P-S units in glass

Local P-S units in the crystal

50 60 67 70

Li2P2S6 Li2P2S6, Li4P2S6 Li2P2S6 Li7P3S11 g-Li3PS4 b-Li3PS4 a-Li3PS4 b-Li3PS4 d-Li3PS4 Thio-Lisicon analogue III (b- Li3PS4) Li7PS6, Li3PS4, Li2S Thio-Lisicon analogue II (Li9.6P3S12) Thio-Lisicon analogue III (b-Li3PS4)

PS−3 P2S62−, P2S64−, P2S74− P2S64−, P2S74−, PS43− P2S62−, P2S74−, PS43− P2S62−, P2S74−, PS43−

P2S62− P2S62−, P2S64− P2S64− P2S74−, PS43−

75

80

PS43−

P2S62−, P2S74−, PS43−

The species with the highest abundance was marked bold in the local structures.24

PS43−

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

4

543

Comparison of ionic conductivity with major lithium-ion conductors

Fig. 5 shows Arrhenius plots of major lithium ionic conductors together with oxides, sulfides, hydrides and halides, all of which show the highest-class conductivity values among each material group.25 This figure shows the position of the sulfide systems among the general ionic conductors. The sulfides show high ionic conductivity among these lithium ionic conductors and these conductivity values are similar to those of liquid electrolytes used for lithium ion battery. However, silver and copper ion conductors have much higher ionic conductivities. (a)

(b)

LPGS Li7P3S11

Argyrodite

Liq-Elctrolyte

(c)

Fig. 5 Arrhenius plots of major lithium ionic conductors. (a), (b), and (c), respectively, display data for oxides, sulfides, and others, such as hydrides and halides, all of which show the highest- class conductivity values among each material group. For reference, data for several polymer and liquid electrolytes are included. From reference Kato, Y.; Hori, S.; Kanno, R. Li10GeP2S12-type Superionic Conductors: Synthesis, Structure, and Ionic Transportation. Adv. Energy Mater. 2020, 10(42), 2002153. https://doi.org/10.1002/aenm.202002153.

5 5.1

Sulfide glass system History

In the 1970s, glass-based sulfides were explored initially with silver and copper systems, expanded to lithium-based systems, leading to the discovery of a series of Li2S-P2S5-based glasses.2 Li2S-P2S5-LiI glasses (with Li2S/P2S5 ¼ 2) exhibited high conductivity of approximately 10−3 Scm−1 at 25  C. Research on solid-state batteries using M2S-GeS2-MI (M ¼ Li, Ag) glass electrolytes was conducted primarily in France.1 However, solid electrolytes composed of Li2S, SiS2, GeS2, LiI, and other similar constituents had limited electrochemical stability, which hindered the development of solid-state batteries. Kondo et al.22 reported a significant improvement in electrochemical stability with a glass electrolyte composition of 0.03Li3PO4–0.59Li2S-0.38SiS2, exhibiting a conductivity of 6.9  10−4 Scm−1. Subsequently, extensive research on batteries using this glass electrolyte was conducted. The energy density of the solid-state batteries was enhanced with the use of 4 V positive electrodes,26,27 and challenges such as self-discharge and cycle characteristics were improved. Furthermore, batteries utilizing graphite as negative electrodes were reported,28 leading to the realization of lithium-ion type solid-state batteries.29

544

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

Fig. 6 Conductivity Arrhenius plots of the LPS glasses. Figure from reference Dietrich, C.; Weber, D. A.; Sedlmaier, S. J.; Indris, S.; Culver, S. P.; Walter, D.; Janek, J.; Zeier, W. G. Lithium Ion Conductivity in Li2S–P2S5 Glasses—Building Units and Local Structure Evolution During the Crystallization of Superionic Conductors Li3PS4, Li7P3S11 and Li4P2S7. J. Mater. Chem. A 2017, 5(34), 18111–18119. https://doi.org/10.1039/C7TA06067J.

5.2

Materials of sulfide glass

Lithium ion-conductive inorganic glasses were considered until the 2000s to have higher ionic conductivity than crystalline materials due to the open structure and large free volume characteristic of glasses. Inorganic glasses with high lithium-ion conductivity are based on sulfide systems. Based on Li2S, glasses exist in the compositions with P2S5, SiS2, GeS2, B2S3. Among them, glasses with particularly high ionic conductivity are xLi2S-(100-x)P2S5-based glasses with various Li2S contents. Fig. 6 shows the Arrhenius conductivity plots of the xLi2S-(100-x)P2S5-based glasses. 75Li2S-25P2S5 glass with only PS4(3−) units showed the highest ionic conductivity of 2.8  10−4 S cm−1 at room temperature.30,31 Binary glasses other than P2S5-based glasses, such as xLi2S-(100-x)B2S3 and xLi2S-(100-x)SiS2, exhibit ionic conductivity of 10−4 S cm−1 at room temperature.32,33 On the other hand, xLi2S-(100-x)GeS2 exhibits low ionic conductivity of 10−5–10−7 S cm−1.34

5.3

Ionic conductivity

In the glass systems, ionic conductivity can be improved by increasing the concentration of charge carrier ions.35 Attempts to increase the Li+ concentration by doping lithium salts have been examined in the Li2S-P2S5 system. Oxide systems Li3PO4, Li4SiO4, Li4GeO4, LiBO3, LiAlO3, halide systems LiX (X: halogen), and hydride systems LiBH4 are typical additive compounds. Oxide systems: xLi2S-(100−x)SiS2 doped with Li3PO4, Li4SiO4, and Li4GeO4 exhibit a wide electrochemical window and conductivity significantly higher than 10−3 S cm−1.22 The addition of LiBO3 and LiAlO3 improves thermal stability against crystallization while keeping the conductivity of 10−3 S cm−1.36 Hydride systems: 77(75Li2S-25P2S5)-33LiBH4 exhibits high lithium ionic conductivity of 1.6  10−3 S cm−1 and electrochemical stability up to 5 V for Li+/Li.37 Halide systems: effective in improving ionic conductivity; the larger the halide ion, the higher the ionic conductivity. Both xLi2S(100−x)B2S3 and xLi2S-(100−x)SiS2 glasses have ionic conductivities of 10−4 S cm−1 at room temperature.32,33 LiI-doped 30Li2S-26B2S3-33LiI glass exhibits an ionic conductivity of 1.7  10−3 S cm−1 at room temperature,38 and 40Li2S-28SiS2-30LiI glass has an ionic conductivity of 1.8  10−3 S cm−1 at room temperature.39

5.4

Glass and crystalline phases

Sulfide crystals are crystallized from the LPS glass system by heat treatment. These include Li2P2S6 (50Li2S-50P2S5),30 Li2P2S6 (50Li2S-50P2S5),30 Li7P3S11 (70Li2S-30P2S5),40 Li3PS4 (75Li2S-25P2S5),41 Li7PS6 (88Li2S-12P2S5),42 and Li4P2S6 (67Li2S-33P2S5).30 Li7P3S11 is a so-called glass-ceramics. Before the discovery of crystalline sulfide compounds with higher ionic conductivity than glass systems, glass was usually considered to decrease ionic conductivity by crystallization. However, when a crystalline phase is produced from the amorphous state by heat treatment at a relatively low temperature (around 200  C), a phase different from that obtained at the usual sintering temperature of around 500  C can be obtained. Although it is believed that a metastable state material can be synthesized, it should be noted that a crystalline phase that exists thermodynamically stable in the vicinity of 200  C and decomposes by raising the temperature to higher than that is produced. If this stable phase at low temperatures has a crystalline structure suitable for ionic conductivity, it will exhibit much higher ionic conductivity than the corresponding glass phase of the same composition. Li7P3S11 is a good example of such a phase. This phase exists as a pure crystalline phase, as can be seen from the observation that it is produced by sintering at low temperatures after obtaining the precursor from solution.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes 5.5

545

Structure

The xLi2S-(100-x)P2S5-based glasses are composed of different building blocks such as PS4(3−), P2S7(4−), P2S6(4−), and PS3(1−) as shown in Fig. 4. Fig. 7 shows Raman and 31P MAS NMR spectra of LPS glasses of different stoichiometries. Glasses with low Li2S content (x  60) are dominated by P2S7(4−) units, with one bridging S atom and three terminal S atoms in each unit. On the contrary, glasses with high Li2S (x  70) have more tetrahedral PS4(3−) units with all S atoms terminating. The highest ionic conductivity was found for 75Li2S-25P2S5 glass with only PS4(3−) units.30,31 A typical glass structure of high ionic conducting glasses is shown in Fig. 8. The atomic and electronic structures of Li2S-P2S5 binary glasses have been determined by a combination of density functional theory (DFT) and reverse Monte Carlo (RMC) simulations using synchrotron X-ray diffraction, neutron diffraction, and Raman spectroscopic data.11 In 67Li2S-33P2S5 (67Li2S), 70Li2S-30P2S5 (70Li2S), and 75Li2S-25P2S5 (75Li2S) glasses, there exist P2S64−, PS34−, and P2S74− ions as indicated above. The ratio of PSx polyhedral anions is reflected in these glass structures and affects the lithium ion distribution around them. At high Li2S content, the free volume around the PSx polyhedral anions allows for the distribution of lithium ions, and the lithium ions around the PSx polyhedron are influenced by the polarization of the anions. The presence of P2S7 anions may also suppress lithium ion conduction. High ionic conduction in these glasses can be controlled by edge sharing between the PSx and LiSy polyhedra without electron transfer between the P and bridging sulfur ions.

Fig. 7 Raman (a) and 31P MAS NMR (b) spectra of LPS glasses of different stoichiometries in the Li2S-P2S5 system. (c) The building units in the glass electrolyte; PS43− monomers (green) (for high Li2S content region), P2S74− dimers (orange) (lower Li2S content region), P2S64− dimers (purple) (over the compositional range), PS3 chains (rose) (low fraction of Li2S of 60 mol%).30 From Dietrich, C.; Weber, D. A.; Sedlmaier, S. J.; Indris, S.; Culver, S. P.; Walter, D.; Janek, J.; Zeier, W. G. Lithium Ion Conductivity in Li2S–P2S5 Glasses—Building Units and Local Structure Evolution During the Crystallization of Superionic Conductors Li3PS4, Li7P3S11 and Li4P2S7. J. Mater. Chem. A 2017, 5(34), 18111–18119, https://doi.org/10.1039/C7TA06067J.

Fig. 8 DFT/RMC model of 70Li2S glass. Green, Li; Purple, P and PS polyhedral anions; Yellow, S. Figure from Ohara, K., Mitsui, A., Mori, M., Onodera, Y., Shiotani, S., Koyama, Y., Orikasa, Y., Murakami, M., Shimoda, K., Mori, K., et al. Structural and electronic features of binary Li2S-P2S5 glasses. Sci. Rep. 2016, 6 (1), 21302. https://doi.org/10.1038/srep21302.

546

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

Fig. 9 The amorphous composition regions obtained by high-energy ball-milling process in the Li2S-GeS-P2S5 system; hollow circles indicate amorphous samples and solid circles indicate partially crystalline samples. The hatched area is the glass-forming region by conventional melt-quenching methods. Figure from reference Yamamoto, H.; Machida, N.; Shigematsu, T. A Mixed-former Effect on Lithium-ion Conductivities of the Li2S–GeS2–P2S5 Amorphous Materials Prepared by a High-energy Ball-milling Process. Solid State Ion. 2004, 175(1), 707–711. https://doi.org/10.1016/j.ssi.2004.08.028.

5.6

Synthesis

Because sulfides decompose in humid atmosphere, they should be handled under inert atmospheres. There are melt-quenching methods involving high-temperature treatment and prolonged mechanical mixing, ball-milling methods, and wet chemical processes. Melt-quenching method: Starting materials such as Li2S and P2S5 are sealed in quartz tubes, melted at 900–1100  C, and quenched in ice water or a twin-roller quenching machine.32,43 Ball milling method: A high-energy ball mill is used to grind, amorphize, and mix the material. The process is usually performed at room temperature. In the Li2S-P2S5-GeS2 ternary system, the amorphous region formed by ball milling is wider than by conventional melt quenching (Fig. 9).44 Most sulfide glasses can be synthesized by ball milling. Annealing after ball milling provides a crystalline phase, which also contributes to the processes for synthesizing high-purity crystalline phases. Wet chemical reactions: Melt-quenching and ball-milling methods are difficult to scale up. Wet method synthesis is advantageous in reducing reaction time, synthesizing homogeneous materials, and forming a close electrode-electrolyte interface43.45 The high reactivity of sulfide precursors limits solvents to nonpolar solvents or non-protic solvents with low polarity.45,46 Glasses, Li7P3S11, Li3PS4, and Li4SnS4 have been synthesized in solvents such as N-methyl-formamide (NMF), tetrahydrofuran (THF), and ethyl acetate. Li7P3S11 was synthesized in THF, acetonitrile (ACN), and mixed solutions of THF and ACN.45,47 In MeOH solution, 0.6LiI ∙ 0.4 Li4SnS4 glass, which is stable in dry air, was synthesized.48

6 6.1

Thio-LISICON General comments

In the thio-LSICON group, Li3.25Ge0.25P0.75S4 was reported as the first crystalline sulfide whose lithium conductivity exceeded 1  10−3 S cm−1 at room temperature. The LISICONs are a group of materials termed Lithium Superionic Conductors (LISICONs),49 which were one of the most prominent discoveries in the 1970s. A systematic synthesis of the Li2S–GeS2–Ga2S3 pseudo-ternary system revealed that the phases within a specific composition range showed high conductivity, and these conductive phases had a crystal structure similar to the LISICON-type oxides. This study was the first example in which the oxygen sublattice was replaced with sulfur in LISICONs. This replacement resulted in a significant increase in the lithium conductivity due to the larger ionic radius and polarizability of sulfur. After the discovery of thio-LISICON, further studies in the Li3PS4-Li4GeS4 system have led to the discovery of the LGPS system with even higher ionic conductivity. This section mainly describes the thio-LISICON before the discovery of LGPS. A thio-LISICON (Li3.25Ge0.25P0.75S4) was applied for all-solid-state cells.50,51

6.2

Materials

Examples of materials that belong to the thio-LISICON family exist in the Li2S–GeS2–P2S5 system, the Li2S–SiS2–Al2S3 and Li2S–SiS2–P2S5 systems, and the Li2S–P2S5 system. The materials design is based on the modification of Li4SiS4 and Li4GeS4.52,53

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

547

Li4−xSi1−xPxS4,53 Li4−xGe1−xPxS4,5 and Li4−2xZnxGexS43 were used as vacancy doped-systems (Li+ + M 4+ ! □ + M 5+ or 2Li+ ! □ + Zn2+), whereas Li4+xSi1−xAlxS453 and Li4+xGe1−xGaxS43 were used as interstitial ion-doped systems (M4+ ! Li+ + M3+).

6.3

The Li2S–GeS2–P2S5 system

The crystalline phase Li4−xGe1−xPxS4 was found based on Li4GeS4 with the substitution Ge4++Li+$P5+.5 The solid solution member x ¼ 0.75 in Li4−xGe1−xPxS4 shows high conductivity of 2.2  10−3 Scm−1 at 25  C, together with negligible electronic conductivity. The system has a much higher ionic conductivity of 2.4  10−3 Scm−1 at 25  C when synthesized under Ar gas flow.54 This system is later studied precisely by drawing phase diagram.

6.4

The Li2S–SiS2–Al2S3 and Li2S–SiS2–P2S5 systems

The Li4+xSi1−xAlxS4 and Li4−xSi1−xPxS4 systems were found to have high ionic conductivity.53 The host material is Li4SiS4, and when the Li+ interstitials or Li+ vacancies were created by the partial substitutions of Al3+ or P5+ for Si4+, large increases in conductivity occur. The solid solution member x ¼ 0.6 in Li4−xSi1−xPxS4 showed a high conductivity of 6.4  10−4 Scm−1 at 27  C with negligible electronic conductivity.

6.5

The Li2S–P2S5 system

Thio-LISICON related materials were discovered in Li2S–P2S5 binary systems.55 The solid solution with the composition range 0.0 < x < 0.27 in Li3+5xP1−xS4 was synthesized at 700  C.55 The solid solution member at x ¼ 0.065 in Li3+5xP1−xS4 showed the highest conductivity value of 1.5  10−4 Scm−1 at 27  C with negligible electronic conductivity and an activation energy of 22 kJmol−1. The extra lithium ions in Li3PS4 created by partial substitution of P5+ for Li+ led to the large increase in ionic conductivity. Although Li3+5xP1−xS4 has slightly lower ionic conductivity than that of the thio-LISICON (Li4−xGe1−xPxS4 and Li4 −xSi1−xPxS4), absence of metal elements in the structure might lead to higher electrochemical and chemical stabilities, which is an important condition as a solid electrolyte of all solid-state lithium batteries.

6.6

Material design diagram

The thio-LISICON family is expressed by the general formula Li4−xM1−yM0 yS4 with M ¼ Si, Ge, P and M0 ¼ P, Al, Zn, Ga, etc.5,56 Plots of composition versus conductivity are shown in Fig. 10. All the end members of the solid solutions, Li4MS4, Li4MS4, and Li3MS4, showed rather low ionic conductivity and higher activation energy for conduction. The conductivity increased with the solid solutions in all the systems, which indicates that a disordered arrangement caused by cation mixing is an important factor for high ionic conduction. Fig. 11 summarizes the materials diagram of the thio-LISICON together with the diagram for the oxide LISICON system. The ionic conduction properties are strongly dominated by the size and polarizability of constituent ions, or the interstitial-vacancy character caused by the substitutions. Note that the conductivities of the thio-LISICON are higher than those of the LISICON. This Li2S-SiS2-P2S5 Li2S-SiS2-Al2S3 Li2S-GeS2-Sb2S5 Li2S-GeS2-P2S5 Li2S-GeS2-Ga2S3 Li2S-GeS2-ZnS Li2S-P2S5

-3

logV Scm-1

-4 -5 Li4P0.8S4

-6 -7

Li3PS4

Li4GeS4

-8

Li4SiS4

3.0

3.5

4.0

Li5AlS4

4.5

5.0

x in LixM1-yM'yS4 Fig. 10 Relationship between the conductivities and the compositions of the thio-LISICON family. The maximum ionic conductivity was obtained in the non-stoichiometric compositions between 3.0 and 4.0, and 4.0 and 5.0.55 From 1st version Murayama, M.; Sonoyama, N.; Yamada, A.; Kanno, R. Material Design of New Lithium Ionic Conductor, Thio-LISICON, in the Li2S-P2S5 System. Solid State Ion. 2004, 170(3–4), 173–180. https://doi.org/10.1016/j.ssi.2004.02.025.

548

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

Oxides Zn2GeO 4

Sulfides Li 3PO 4

Oxides to sulfides

Li3PS 4

Li2ZnGeS 4

interstitial Li

6.0 x

10 -7

so-called "LISICON" system

Scm -1

3.0 x

Li vacancies

Li4GeO 4

10 -6

Scm -1

3.0 x 10 -7 Scm -1

6.4 x 10 -4 Scm -1 Li vacancies

Li vacancies

2.2 x 10 -3 Scm -1

Li 4SiS 4 interstitial Li

1.0 x 10 -8 Scm -1

2.3 x 10 -7 Scm -1

Li vacancies

4.5 x 10 -6 Scm -1

Li4GeS 4

Oxides to sulfides interstitial Li

4.4 x 10 -6 Scm -1

Li vacancies

Li vacancies

Li4SiO 4

Li3SbS 4

interstitial Li

6.5 x 10 -5 Scm -1

Li vacancies

Li3VS 4

Li vacancies

1.4 x 10 -7 Scm -1

Li vacancies

Li 5AlS 4

Li 5AlO 4

Li5GaS 4

LiAlGeS 4

Oxides to sulfides

Fig. 11 Materials diagram of the thio-LISICON family. The diagram is described in comparison with the oxide LISICON system. The highest conductivity is obtained in the solid solution between Li4GeS4 and Li3PS4.5 From 1st version Kanno, R.; Murayama, M. Lithium Ionic Conductor Thio-LISICON: The Li[sub 2]S-GeS [sub 2]-P[sub 2]S[sub 5] System. J. Electrochem. Soc. 2001, 148(7), A742. https://doi.org/10.1149/1.1379028.

indicates clearly that the replacement of larger and more polarizable S for O improved conductivity, or larger and more polarizable framework is preferable for ionic conduction. The important characteristic of this family of materials is its wide range of solid solution obtained by aliovalent substitutions, which introduced interstitial lithium ions or vacancies and improved ionic conductivities. Based on the parent compounds, Li4MS4 (M ¼ Si, Ge, P), lithium vacancy was introduced either by the substitution of pentavalent cations for Ge or Si in the systems Li4−xGe1−xPxS4, Li4−xSi1−xPxS4, and Li4−xGe1−xSbxS4 or by the substitution of divalent cations for lithium in Li4−2xZnxGeS4; extra lithium ion was inserted by the substitution of trivalent cations for Ge or Si in the systems Li4+xGe1−xGaxS4 and Li4+xSi1−xAlxS4. When doped to create either Li+ vacancies or Li+ interstitials, dramatic increases in conductivity occur.

6.7

Synthesis

Synthesis of these sulfides is based on a rather simple procedure. Two synthesis methods are reported. Synthesis in a sealed tube: Starting materials of binary sulfides such as Li2S, SiS2, GeS2, Al2S3, and P2S5 were weighed and mixed in appropriate molar ratios in an argon-filled glove box. The mixture was sealed in a carbon-coated quartz tube and heated to a reaction temperature of 700  C for 8 h. After the reaction, the tube was slowly cooled to room temperature. Synthesis under flowing argon: Li3.25Ge0.25P0.75S4 was prepared by heating appropriate amounts of Li2S, GeS2, and P2S5 at under flowing argon gas at 800  C. A 10% excess of sulfides P2S5 and S was necessary in order to obtain the required products.

6.8

Structure

The host materials Li4GeS4, Li4SiS4, and Li3PS4 have the same structure as the g-Li3PO4 type and are composed of hexagonal close-packed sulfide-ion arrays. Ge, Si, and P are distributed over the tetrahedral sites, and the MS4 (M ¼ Ge, Si, and P) tetrahedra are isolated from each other. The framework structure was formed by the LiS4, MS4 tetrahedra, and LiS6 octahedra, which are connected with each other in three dimensions (see Fig. 1255). The lithium ions are distributed over these tetrahedral sites and octahedral site. The structural consideration of the host materials Li4GeS4, Li4SiS4, and Li3PS4 indicates that the lithium ions in the octahedra, which are connected in one dimension by sharing edges, and the interstitial tetrahedral sites, which share faces to the Li octahedra, participate in the ionic conduction.

7 7.1

Li7P3S11 Synthesis and conductivity

Li7P3S11 was synthesized during the materials developments on the Li2S-P2S5 glass system. After ball milling, Li7P3S11 underwent crystallization through heat treatment, exhibiting an ionic conductivity of 3.2  10−3 Scm−1 at room temperature.57 Subsequently, a value of 1.7  10−2 Scm−1 at room temperature was determined.9

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

549

LiS4

(a) c a

LiS6

MS4

b

GeS4 g=1

Li4GeS4

Li(3)S4 g=1

Li(5)S6 g=1 Li(1)S4 g=1

Li(2)S4 g=0.602

(b)

Li4SiS4

SiS4 g=1 Li(3)S4 g=0.481

Li(5)S6 g=0.893

Li(1)S4 g=0.988 Li(2)S4 g=1

Li3PS4

PS4 g=1 Li(4)S4 g=0.28

Li(5)S6 g=0.68

Fig. 12 (a) Structure of the thio-LISICON. (b) Schematic drawing of the lithium environment in the thio-LISICON structure.55 From 1st version Murayama, M.; Sonoyama, N.; Yamada, A.; Kanno, R. Material Design of New Lithium Ionic Conductor, Thio-LISICON, in the Li2S-P2S5 System. Solid State Ion. 2004, 170(3–4): 173–180. https://doi.org/10.1016/j.ssi.2004.02.025.

7.2

Crystal structure

The material had a structure similar to Ag7P3S11, where PS4 tetrahedra was connected through corner sharing, forming a framework structure consisting of P2S7 units.40 Fig. 13 shows the arrangement of P2S7 di-tetrahedra and PS4 tetrahedra of Li7P3S11 is similar to that of Ag7P3S11, but the two-fold, screw and c glide symmetries are lost in the structure of Li7P3S11, due to a different relative orientation of neighboring PS43− anions. This was distinctive compared to other LiPS-based materials where PS4 tetrahedra existed in isolated form. Challenges regarding stability compared to PS4 units exist.

Fig. 13 Structures of Li7P3S11 with the pseudo-monoclinic supercell (left) and Ag7P3S11 (right) viewed along the [010] direction. Figure after Yamane, H.; Shibata, M.; Shimane, Y.; Junke, T.; Seino, Y.; Adams, S.; Minami, K.; Hayashi, A.; Tatsumisago, M. Crystal Structure of a Superionic Conductor, Li7P3S11. Solid State Ion. 2007, 178(15–18), 1163–1167. https://doi.org/10.1016/j.ssi.2007.05.020.

550

8 8.1

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

Argyrodite sulfides General comments

In 2008, Li6PS5X (X ¼ Cl, Br, I) compounds were reported to have high lithium-ion conductivity. The Argyrodite sulfides are based on the basic structure of mineral argyrodite (Ag8GeS6).6 The sulfides Li6PS5X (X ¼ Cl, Br, I) adopted the same crystal structure as the cubic Cu- and Ag-argyrodite compounds (see Fig. 2). Li-ions were randomly distributed in the tetrahedral positions (48 h and 24 g sites in space group F-43 m No 216), while P and S occupied the tetrahedral position (4b site) and 16e site, respectively, forming isolated PS4 tetrahedra that extended in three dimensions. Additionally, X-ions formed a face-centered cubic (fcc) lattice (4a and 4c sites). Li-ions diffused through the partially occupied positions. The Li6PS5I compound exhibited lower ionic conductivity compared to Li6PS5Cl and Li6PS5Br. Introduction of defects increased the ionic conductivity, and the defective argyrodite Li6-xPS5-xCl1+x exhibited high ionic conductivity of 17 mS cm−1.58 Solid-state batteries utilizing argyrodite-type sulfides (Li6PS5X, X ¼ Cl, Br) were reported in 2010,59 and significant progress has been made in subsequent technological developments. Among sulfide solid electrolytes, lithium Argyrodite has various advantages. These include high ionic conductivity, the formation of stable interfacial phases during battery operation, the availability of cost-effective precursors, and the existence of synthesis methods suitable for mass synthesis. In 2020, a battery using an Argyrodite solid electrolyte was reported with an AgdC composite anode and a high Ni-content layer oxide cathode. The solid-state battery exhibited a high energy density exceeding 900 Wh L−1, a stable efficiency exceeding 99.8%, and a long cycle life (1000 cycles).60

8.2

Structure

Argyrodite structures are first discovered in the mineral Ag8GeS6 which shows high Cu+ and Ag+ ionic conductivities.6 Lithium Argyrodite has two polymorphisms and its phase transition temperature depends on the chemical composition.61 The cubic structure is stabilized when sulfide ions are replaced by halide ions with the composition, Li6PS5X (X ¼ Cl, Br, I). The X− anion forms a face-centered cubic (fcc) framework, with non-bonded “free” S2− ions and PS43− units occupying the tetrahedral and octahedral voids, respectively (Fig. 14). A disorder in the site occupancy ratio of the S2−/X− positions has been observed. In Li6PS5I, the I- and S2− anions occupy two separate sites (4a and 4d). Site disorder exists at X ¼ Br− and Cl−.62 This disordered arrangement plays an important role in the Li+ ion transport properties. The cubic phase forms a close packing of the tetrahedra, providing many inequivalent tetrahedral voids suitable for lithium to occupy.61 The voids can be classified into five different types based on the number of corners, edges, and faces of each tetrahedron shared with adjacent PS43−units. Type 5 (T5, 48 h) and T5a sites (24 g) form a cage-like structure around the sulfide ion position (4d).

8.3

Lithium ion diffusion pathway

There are three jump mechanisms in the Argyrodite structure.63 (1) A doublet jump, involving T5a(24 g). This corresponds to localized motion between T5(48 h) tetrahedra (Fig. 15a). (2) A jump in the cage. This is the lithium motion between pairs of T5 tetrahedra in the same lithium cage. In this jump, the T2(48 h) site connects with the T5 site in the same cage (Fig. 15b). There is a T5-T2-T5 pathway within the cage, with the T2 site in the middle position.

Fig. 14 Crystal structure a of Li6PS5X (X ¼ Cl, Br, I). The local coordination of the five types of tetrahedral interstitial sites, alongside the trigonal coordinated type 5a site. The X− anions occupy Wyckoff 4a positions, while S2− ions Wyckoff 4d. P and S part of the PS43− units occupy Wyckoff 4b and 16e, respectively. Figure from Ref. Minafra, N.; Kraft, M. A.; Bernges, T.; Li, C.; Schlem, R.; Morgan, B. J.; Zeier, W. G. Local Charge Inhomogeneity and Lithium Distribution in the Superionic Argyrodites Li6PS5X (X ¼ Cl, Br, I). Inorg. Chem. 2020, 59 (15), 11009–11019. https://doi.org/10.1021/acs.inorgchem.0c01504.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

551

Fig. 15 Three jump mechanisms in the Argyrodite structure. Figure from Ref. Minafra, N.; Kraft, M. A.; Bernges, T.; Li, C.; Schlem, R.; Morgan, B. J.; Zeier, W. G. Local Charge Inhomogeneity and Lithium Distribution in the Superionic Argyrodites Li6PS5X (X ¼ Cl, Br, I). Inorg. Chem. 2020, 59 (15), 11009–11019. https://doi. org/10.1021/acs.inorgchem.0c01504; Zhou, L. D.; Minafra, N.; Zeier, W. G.; Nazar, L. F. Innovative Approaches to Li-Argyrodite Solid Electrolytes for All-Solid-State Lithium Batteries. Acc. Chem. Res. 2021, 54 (12), 2717–2728. https://doi.org/10.1021/acs.accounts.0c00874.

(3) An inter-cage jump. This is a lithium transfer between T5 sites in different cages; T2 is the middle of two T5 sites belonging to different cages. The transfer between different cages is possible by a continuous pathway (T5-T2-T2-T5) through face-sharing polyhedra (Fig. 15b). Long-distance lithium diffusion is possible only if all three classes of jumps can take place.64–66 Beside the above pathway, lithium diffusion through the T4 site also exists: there is a continuous face-sharing tetrahedral pathway connecting different lithium cages through T5-T4-T5 (Fig. 15c).

8.4

Factors that affect ionic conduction

(1) Disordered structure and ionic conductivity Fig. 16a shows the disorder of anion sites as a result of composition change and the effect on ionic conductivity. The change in anion distribution causes a corresponding change in anion charge distribution, which affects the lithium distribution. The higher the degree of site disorder, the more delocalized the Li+ density, resulting in better cage-to-cage connectivity and faster Li diffusion. (Fig. 16b and c).64,66,67 (2) Lithium disorder and effects on ionic conductivity A disordered Li ion sub-lattice is necessary to obtain high conductivity. In addition to disordered anion sites, aliovalent cation substitution is efficient in generating interstitial Li ions or Li vacancies.20,68,69 The Li excess Li6+xMxSb1-xS5I (M ¼ Si, Ge, Sn) has high

Fig. 16 (a) Ionic conductivity as a function of the degree of X−/S2− site disorder in Li6PS5X (X ¼ Cl, Br, I). (b and c) Lithium distribution determined by MEM analysis of Li6PS5Br showing the relationship between the degrees of anion site disorder and lithium density. Figure from Ref. Zhou, L. D.; Minafra, N.; Zeier, W. G.; Nazar, L. F. Innovative Approaches to Li-argyrodite Solid Electrolytes for All-solid-state Lithium Batteries. Acc. Chem. Res. 2021, 54 (12), 2717–2728. https://doi. org/10.1021/acs.accounts.0c00874.

552

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

Fig. 17 (a) Li+ content of each site as a function of Si4+ content in Li6+xSixSb1-xS5I. (b) BVSE model of the migration barrier by Li6+xSixSb1-xS5I (x ¼ 0, 0.7). The activation energy of Li diffusion decreases as the Li ion carrier concentration increases. Figure from Ref. Zhou, L. D.; Minafra, N.; Zeier, W. G.; Nazar, L. F. Innovative Approaches to Li-argyrodite Solid Electrolytes for All-solid-state Lithium Batteries. Acc. Chem. Res. 2021, 54 (12), 2717–2728. https://doi.org/10.1021/ acs.accounts.0c00874.

conductivity of up to 24 mS-cm−1 with only slight S2−/I− site disorder. However, there is a disordered distribution of Li on the four Li sites. Excess Li ions are distributed to two new Li sites (Fig. 17a). The new Li sites (T2) correspond to the interstitial sites for Li ion diffusion. With increasing (Li++Si4+) content, the T5-T5a-T5 site occupancy decreases and the new T2 site occupancy increases. This rearrangement of Li increases the site energy and decreases the activation energy of lithium diffusion (Fig. 17b). Other examples: in Li6+x(Si/Ge)x(P/Sb)1-xS5I,15,68,70 the increase in lithium carrier density reduces the activation energy and increases ionic conductivity; in Li6PS5Br, Si4+/P5+ substitution increases lithium carrier density and improves ionic conductivity.71 Argyrodite Li7+xMxP1-xS6 (M ¼ Si, Ge), with excess Li and no halides, shows high ionic conductivity of 3 mS-cm−1 without any anion site disorder.72,73 The interstitial sites that contribute to diffusion flatten the migration barrier for Li ion diffusion, resulting in cooperative ion migration.

8.5

Synthesis

8.5.1 Solid-state synthesis The Argyrodite sulfides are synthesized by a solid-state reaction. Precursors (Li2S, P2S5, LiX) are grinded and mixed, pelletized, and calcined at a certain temperature. Precursors milled in a high-energy ball mill are uniformly mixed and partially reacted by a milling reaction to yield the product. High temperature treatment is used for crystallization. This is a normal solid-state synthesis reaction.

8.5.2 Liquid-phase synthesis After dissolving Li3PS4-3THF, Li2S and LiX in dry ethanol, Li6PS5X (X ¼ Cl, Br, I), Li6PS5ClxBr1-x (0  x  1) and Li6-yPS5-yCl1+y (y ¼ 0–0.5) are synthesized by heat treatment at 550  C.74 The products (Li6PS5Cl and Li6PS5Br) exhibit high ionic conductivity of about 2.4 mS cm−1 at room temperature. The Cl/Br mixed and Cl-rich phases exhibit about 4 mS-cm−1. To crystallize and obtain high ionic conductivity, high-temperature calcination at 550  C is required.75 Li7PS6 is obtained by dissolving b- Li3PS4 in acetonitrile, reacting with Li2S in ethanol, and annealing at 200  C.76 The solution synthesis of Li6+xMx(P/Sb)1-xS5I (M ¼ Si, Ge, Sn) has not been established.

8.6

Several topics related to argyrodite

8.6.1 Effect of grain boundaries on conductivity measurements The ionic conductivity of solid electrolytes is measured by electrochemical impedance spectroscopy (EIS) of cold-pressed pellets of samples. This technique is used for sulfide solid electrolytes because of their softness. However, a high pellet density is required to make close contact between the current collector and the solid electrolyte and to lower the grain boundary resistance. For example, cold-pressed Li6.6Si0.6Sb0.5S5I shows an ionic conductivity of 14.8 mS cm−1 at room temperature, while the sintered pellets show a conductivity of 24 mS cm−1.68 The contribution of the grain boundaries to the ionic conductivity should be determined by low temperature measurements. The ionic conductivity and activation energy values of solid electrolytes are affected by the synthesis process, including purity, crystallinity, and homogeneity. The pressure applied to the sample for the measurement, pellet density, and contact between the pellet and current collector also affect them.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

553

8.6.2 Stability with electrodes The advantage of solid-state batteries over liquid electrolyte batteries is that they enable the selection of the suitable electrolyte for each of the cathode and anode. 8.6.2.1 Stability for oxidation (cathode) The stability of solid electrolytes is considered in the decomposition pathway of the electrolyte. Electrolytes are utilized at potentials outside the thermodynamically stable range, resulting in electrochemical decomposition. Interfacial reactions with the cathode active material lead to chemical decomposition (Fig. 18); Argyrodite exhibits a relatively narrow potential window77–79 and decomposes during charge and discharge cycling. For example, Li6PS5Cl is oxidized above about 2.5 V vs. Li/Li+. The potential window determined by electrochemical measurements shows a value that is kinetically stable. Therefore, they are usually higher than the thermodynamically calculated oxidation potentials.79,80 The decomposition of the solid electrolyte causes the formation of a new phase between the cathode and the electrolyte, which increases the cell impedance and causes a capacity decrease81.82 Not only electrochemical stability to the cathode active material, but also decomposition is an issue for practical use by oxidation with water and carbonate adsorbed on the surface of the oxide particles. 8.6.2.2 Stability to reduction (anode) At low potentials, Li6PS5Cl is reduced by P at1.08 V vs. Li/Li+.77,79 The thermodynamically calculated reduction potential is 1.71 V (vs. Li/Li+).79 Therefore, Argyrodite electrolytes are not, in principle, suitable for Li metal anodes. However, forming a stable passivated solid electrolyte interphase (SEI) at the interface,83 they exhibit relatively stable charge-discharge behavior at high current densities up to 1 mA cm−2. For example, Li6.7Si0.7Sb0.3S5I can be charged and discharged for 1000 h at 0.6 mA cm−2 (Fig. 19).

8.7

Solid-state battery characteristics

The Argyrodite is a promising solid electrolyte for solid state batteries. The advantage of this material is that its composition can be varied to improve the properties of the solid electrolyte, such as higher ionic conductivity, to enhance the performance of solid-state

Fig. 18 Decomposition process of Argyrodite (Li6PS5Cl). Figure from Ref. Zhou, L. D.; Minafra, N.; Zeier, W. G.; Nazar, L. F. Innovative Approaches to Li-argyrodite Solid Electrolytes for All-solid-state Lithium Batteries. Acc. Chem. Res. 2021, 54 (12), 2717–2728. https://doi.org/10.1021/acs.accounts.0c00874.

Fig. 19 Voltage profiles of Li plating/stripping ni a Li/Li6.7Si0.7Sb0.3S5I/Li symmetric cell cycled at different current densities.68 Figure from Ref. Zhou, L.; Assoud, A.; Zhang, Q.; Wu, X.; Nazar, L. F. New Family of Argyrodite Thioantimonate Lithium Superionic Conductors. J. Am. Chem. Soc. 2019, 141 (48), 19002–19013. https://doi.org/10.1021/jacs.9b08357.

554

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes (a) 101 100

80

>99.8% average

99

60 98 40

97

20

96

0 0

100

200

300

400

500 Cycle number

600

700

800

900

Coulombic efficiency (%)

Capacity retention (%)

100

95 1,000

(b)

Fig. 20 (a) Electrochemical performance of all-soli-state battery using the Argyrodite electrolyte. Cycling performance and Coulombic efficiency of the Ag–C-anode|Argyrodite-solid-electrolyte |NMC-cathode prototype pouch cell (0.6 Ah). (b) Specific power and energy of the solid-state battery using multilayer design. The Ragone plot of the batteries using Li metal or equivalent (for example, anode-free) as the anode. (a) Figure from Ref. Lee, Y.-G.; Fujiki, S.; Jung, C.; Suzuki, N.; Yashiro, N.; Omoda, R.; Ko, D.-S.; Shiratsuchi, T.; Sugimoto, T.; Ryu, S.; et al. High-energy Long-cycling All-solid-state Lithium Metal Batteries Enabled by Silver–carbon Composite Anodes. Nat. Energy 2020, 5 (4), 299–308. https://doi.org/10.1038/s41560-020-0575-z.; (b) Figure from Ref. Ye, L.; Li, X. A Dynamic Stability Design Strategy for Lithium Metal Solid State Batteries. Nature 2021, 593 (7858), 218–222. https://doi.org/10.1038/s41586-021-03486-3.

batteries. The interfacial decomposition process at the electrode-electrolyte interface does not interfere significantly with the operation of the solid electrolyte in solid-state batteries. The followings are examples of a solid electrolyte battery using the Argyrodite electrolytes (Fig. 20). A high-performance all-solid-state lithium metal battery with this sulfide electrolyte was reported using a AgdC composite anode. The thin AgdC layer can effectively regulate Li deposition, which leads to a genuinely long electrochemical cyclability. Using a high-Ni layered oxide cathode with a high specific capacity (>210 mAh g−1) and high areal capacity (>6.8 mAh cm−2) and an argyrodite-type sulfide electrolyte. A prototype pouch cell (0.6 Ah) exhibited a high energy density (>900 Wh L−1), stable Coulombic efficiency over 99.8% and long cycle life (1000 times).60 To solve the problem of lithium dendrite growth, a multilayer structure of electrolyte (Li5.5PS4.5Cl1.5/Li10Ge1P2S12/ Li5.5PS4.5Cl1.5) has also been reported.84 The cycle stability is extremely high; for example, the capacity retention after 10,000 cycles at a rate of 20  C is 82% and after 2000 cycles at a rate of 1.5  C is 81.3%. Thus, the Argyrodite and various other solid electrolytes are becoming the electrolytes of choice for batteries, where it is no longer necessary to distinguish between solid and liquid state.

9 9.1

LGPS-type sulfides General comments

Ever since the first report on Li10GeP2S12 (LGPS) in 2011,10 its unique structure and unexceptionally high lithium conductivity >1  10−2 S cm−1 have attracted extensive interest, especially for applications in solid-state battery. The modifications of LGPS, Li9.54Si1.74P1.44S11.7Cl0.3 and Li9.54[Si0.6Ge0.4]1.74P1.44S11.1Br0.3O0.6, were reported to show even higher conductivities. This unexceptionally high conductivity is ascribed to the unique crystal structure. LGPS and its modifications are described with a focus on the synthesis, structure, and ionic transportation. Battery applications are briefly summarized to indicate the potential merits of using LGPS-type phases as solid electrolytes with high lithium ion conductivity.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes 9.2

555

Material diversity

Fig. 21 shows the materials diagram of the LGPS-type crystals.25 Table 3 summarizes the materials variety belonging to the LGPS family. In the LGPS-type structure, the tetrahedral units form the framework of the entire crystal. There are two crystallographic positions in space group P42/nmc (No. 137)—M 1(2b) and M 2(4d)—in the tetrahedral center, each of which is partially occupied by cations. Sulfur atoms are typically positioned at the apexes of the tetrahedral unit and sulfur atoms can be replaced by oxygen, selenium, and halogen atoms. Na-based derivatives also exist. It is possible to modify the framework tetrahedron by various cations and anions, thus enabling us to design a variety of materials with the LGPS-type structure, as schematically shown in Fig. 21.

Fig. 21 Various materials categorized in the LGPS family. (a) Constituent elements. (b) Materials map of the electrolyte with the LGPS-type structure. Derivatives are categorized into two large groups: cation-substituted materials and materials with an anion-substituted composition. (c) Concept of material design for the LGPS phase from thio-LISICONs. Figures from Ref. Kato, Y.; Hori, S.; Kanno, R. Li10GeP2S12-type Superionic Conductors: Synthesis, Structure, and Ionic Transportation. Adv. Energy Mater. 2020, 10 (42), 2002153. https://doi.org/10.1002/aenm.202002153.

9.3

Synthesis and phase diagram

The discovery of LGPS dates back to the development of thio-LISICON in 2000.3 After the thio-LISICON material was found, these phases were used as an electrolyte after being subjected to vibration milling for reducing the particle size for applying all-solid-state

556

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes Table 3

LGPS-type solid electrolytes. Composition, ionic conductivity, and activation energy.

Composition

sRT/S cm−1 −2

Li9.54Si1.74P1.44S11.7Cl0.3 Li10(Si0.5Ge0.5)P2S12 Li10(Ge0.5Sn0.5)P2S12 Li10(Si0.5Sn0.5)P2S12 Li10GeP2S11.7O0.3 Li9.6P3S12

2.5  10 4.2  10−3 6.0  10−3 4.3  10−3 1.2  10−2 1.2  10−3

Li9P3S9O3 Li10GeP2S12

4.3  10−5 1.2  10−2

Li10.35Si1.35P1.65S12

6.7  10−3

Li9.81Sn0.81P2.19S12

5.5  10−3

Li9.42Si1.02P2.1S9.96O2.04 Li10GeP2S12 Single crystal [110]

1.0  10−4 7.0  10−3

Li10GeP2S12 Single crystal [001]

2.8  10−2

Li10GeP2S12 Li10Ge2/3Sn1/3P2S12 Li10Ge1/3Sn2/3P2S12 Li10SnP2S12 Li10GeP2S12 grain Li10GeP2S12 powder Li10(Si0.3Sn0.7)P2S12 Li10SnP2S12 Li10.35[Sn0.27Si1.08]P1.65S12 Li9.54[Si0.6Ge0.4]1.74P1.44S11.1Br0.3O0.6

7.6  10−3 6.4  10−3 4.8  10−3 3.8  10−3 1.0  10−2 9.5  10−3 7.6  10−3 5.6  10−3 1.2  10−2 3.2  10−2

Ea/eV

Temperature range T/K

References

0.26 0.29 0.28 0.31 0.18 0.36 0.24 0.35 0.31 0.17 0.30 0.14 0.29 0.19 0.27 0.34 0.43 0.33 0.40 0.27 0.28 0.28 0.30 0.32 0.32 0.30 0.32 0.19 0.24

243–298 243–298 243–298 243–298 298–398 173–253 253–373 298–473 193–298 322–673 173–298 322–673 173–298 322–573 298–523 243–293 200–243 243–293 200–243 233–333 233–333 233–333 233–333 143–213 143–333 143–333 143–333 298–393 230–300

12 85 85 85 86 12 87 10 88 88 13 89 89 90 90 90 90 91 91 91 92 93 14

cells.50,51 After the milling process, its conductivity remained unchanged. This was an anomalous phenomenon, which led to the phase-diagram study and the LGPS phase was clarified. Fig. 22 shows the phase diagram for the Li4GeS4–Li3PS4 system.94 Thio-LISICON Region II was found to be a mixture of the orthorhombic thio-LISICON phase. Li4GeS4 (b’ phase) and Li3PS4 (b phase) exist as the main crystal phases with g-Li3PO4-type crystal structures and Li10GeP2S12 (LGPS phase). The LGPS phase was determined to be stable up to 600  C, with a slight increase in the formation compositional range as temperature increases. In a certain composition range, there was a two-phase mixture of

Fig. 22 Phase diagram for the Li3PS4–Li4GeS4 pseudo-binary system. The molar ratio in Li3PS4 corresponds to x in Li4xGe4−xPxS4 or d in Li10+dGe1+dP1−dS12 where d ¼ 2 − 3x, which represent the chemical formula for thio-LISICONs or LGPS-type solid solutions, respectively. The L, a, b, g, G, and b, represents the liquid phase, a-Li3PS4, b-Li3PS4, g-Li3PS4, Li10GeP2S12, and Li4GeS4 phase, respectively. Figure from Ref. Kato, Y.; Hori, S.; Kanno, R. Li10GeP2S12-type Superionic Conductors: Synthesis, Structure, and Ionic Transportation. Adv. Energy Mater. 2020, 10 (42), 2002153. https://doi.org/10.1002/aenm.202002153.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

557

LGPS and the liquid phase between 600 and 650  C. Above 650  C, a two-phase region of b’ and the liquid phase existed. Further increasing the temperature resulted in complete melting. Depend on the synthesis conditions, compounds in the Li2S–P2S5–GeS2 pseudo-ternary system appear as impurity phases. Examples of these impurity phases include Li3PS4 and Li4GeS4, both of which are the end members of the phase diagram, as well as Li4P2S7 (Li4P2S6), Li7PS6, and Li2GeS3.3,23,42,52,95,96 The amorphous phase contained influences the ionic conductivity when synthesis temperatures below 550  C are used.97 These secondary phase components in the sample are partially responsible for differences in the reported conductivity and thermal stability.94,98–100 The Li10GeP2S12 was then usually synthesized by mechanical mixing, followed by heat treatment for a mixture of lithium sulfide, phosphorous sulfide, and germanium sulfide.10 Numerous synthesis methods have been developed. A subsequent study reported that synthesis temperatures more than 400  C are required for the crystallization of the LGPS phase from precursors obtained by mechanochemical treatment.94,97

9.4

Single crystals

The phase-diagram shows that the LGPS phase at a compositional range of 45–67 mol% Li3PS4 partially melts above 550  C and decomposes incongruently above 650  C. It is difficult to grow a single crystal in the melt at the original composition (Li10GeP2S12). However, small single crystals of LGPS were picked up from the multi-phase mixture.101 Li10SnP2S12 single crystals with small size were also used for the structure analysis.92 Large single crystals of LGPS greater than 1 mm3 in size (Fig. 23)89 were grown from an Li3PS4-rich (85 mol%) composition using the liquidus line on the phase diagram. LGPS crystals form at 630  C upon melt cooling, followed by continuous growth at the expense of the melt until the crystals finally quench to room temperature to avoid b-phase formation.

9.5

Solution methods

In addition to single crystal growth, there are also approaches for synthesis that include mixing in solvent without mechanical mixing.102,103 Two different solvents were used to obtain LGPS due to differences in the reactivity of P2S5 and GeS2 in organic solvents.102 The film form of LGPS was synthesized using hydrazine as solvents.103 The high ionic conductive Li9.54Si1.74P1.44S11.7Cl0.3 was also synthesized by the liquid-phase method. The phase showed an ionic conductivity of 6.6 mS cm−1 at 298 K and the all-solid-state cell prepared using this phase as the solid electrolyte exhibits stable cycling, with a discharge capacity retention of >97% after 100 cycles.104

9.6

Other chemistries

Many derivatives have been reported as members of the LGPS-type materials group (as shown in Fig. 21).

9.6.1 Si and Sn systems Sn analogue (Li10SnP2S12) of LGPS88,92 and the Si analogue (Li11Si2PS12)88,105 were synthesized either by high pressure or the conventional synthesis route88 with the composition of Li10+dM1+dP2−dS12 (M ¼ Si, Sn, Ge and 0.20 < d < 0.43 (Si), 0 < d < 0.35 (Ge), and − 0.25 < d < 0 (Sn)). The oxygen substitution was possible for the LidSidPdS (O) systems.106 The existence of a small

1mm Fig. 23 Single crystals of Li10GeP2S12.89 From Iwasaki, R.; Hori, S.; Kanno, R.; Yajima, T.; Hirai, D.; Kato, Y.; Hiroi, Z. Weak Anisotropic Lithium-ion Conductivity in Single Crystals of Li10GeP2S12. Chem. Mater. 2019, 31 (10), 3694–3699. https://doi.org/10.1021/acs.chemmater.9b00420.

558

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

amount of oxides that potentially contaminated the starting materials and might influence the final products after synthesis. The framework ([M]S4) can comprise two or more M elements: M ¼ Ge/Si,85 Ge/Sn,85,90 and Si/Sn.91,93

9.6.2 LidPdS systems Li9.6PS1212 with the LGPS structure was synthesized based on the notion that the valency of P changes from pentavalent to tetravalent.107 The Li9.6PS12 phase can be considered identical to the thio-LISICON analogue II phase (80Li2S • 20P2S5).108 This phase was synthesized by over 10 h of ball milling, followed by the heat treatment at temperatures as low as 240–260  C. When the valences of P and S are +5 and −2 in the crystal, respectively, the structural composition can be explained approximately by Li3+5dP1−dS4.

9.6.3 Halogen doping Halogen help to form the LGPS phase when used as additives, thus providing a highly ionic conductive phase with reduced grain boundary resistance. In the Si system, Li9.54Si1.74P1.44S11.7Cl0.3 exhibited the high ion conductivity at room temperature (2.5  10−2 S cm−1).12 This material was synthesized at 475  C with a highly crystallized phase and a small amount of argyrodite as the secondary phase.12 The halogen doped LidPdS system similar to those for LGPS-type Li9.6PS12 phase were synthesized.109

9.6.4 Oxygen substitution Oxygen atom substitutes for sulfide atom in the structure.13,86,110 Oxygen-substituted LGPS phases show the improved electrochemical stability against reduction at low potential, with slightly low conductivity13,86,110,111 The substitution range was, for example, (0  x  0.6) in Li10GeP2S12−xOx. The solid-state cell with LidIn anode and LiCoO2 showed improved cycle performance. Oxygen substitution was observed in Li10SiP2S12−xOx,106 and Li9P3S9O3.87,110,112

9.6.5 Other chemistries The 4d site of LGPS can be occupied by Al3+ and Sb5+.91,113 Sb (x ¼ 0.075 in Li10Ge(P1 −xSbx)2S12) substitution reduce the generation of hydrogen disulfide under ambient atmosphere, with no deterioration of the ionic conductivities.

9.7

Structure

The crystal structure of LGPS can be described using two components: (1) the skeleton composed of tetrahedral units and (2) the Li ions distributed among the channels. Fig. 24 depicts the crystal structure of LGPS.89 In the tetragonal lattice (P42/nmc No. 137) with a ¼ 8.66402 and c ¼ 12.5830 A˚ , Ge and P share the Ge1/P1 position (4d), while the remaining are located at the P2(2b) position. These elements form [Ge1/P1]S4 and [P2]S4 tetrahedral units with four surrounding sulfur atoms, thus providing a rigid scaffold where the Li atoms distribute. The four Li sites were identified by diffraction studies: Li1(16 h), Li2(4d), Li3(16 h), and Li4(4c). In the first report,10 ab initio solution was used for determining the arrangement of [Ge]S4 and [P]S4 (see in Fig. 24) by analyzing the high-flux synchrotron XRD powder data of Li10GeP2S12 using Fox program,114 followed by the Rietveld refinement of powder neutron diffraction data to detect Li positions. The Li4 site was missing in the initial stage,10 which was first highlighted by theoretical calculations.115 Its position was confirmed by subsequent measurements using XRD on a single-crystal that was picked up101 and Rietveld refinement of powder neutron diffraction combined with the maximum entropy method (MEM) to visualize the Li positions.99 A minor change in the site splitting of Li2 (4d to 8f ) or Li3 (8f to 16i) was proposed by subsequent studies: a powder neutron diffraction study using bond valence analysis100 or an XRD study on a high-quality single-crystal, respectively.89

9.8

Conduction pathways

Among all the four Li sites, Li1 and Li3 are coordinated by four neighboring sulfur atoms, thereby forming an edge-shared tetrahedral chain of -[Li1]S4-[Li1]S4-[Li3]S4- along the c-axis (Fig. 24, a2). On the contrary, Li2 and Li4 sites are considered sixcoordinated,99 consisting of [Li2]S6 and [Li4]S6 octahedral units, respectively, each of which connect by edge-sharing [Li3]S4 or [Li1] S4, thus forming chains over the ab-plane (Fig. 24, a3 and a4). The initial report confirmed that the pathway along the c-axis because the Li1 and Li3 sites demonstrate larger atomic displace parameters (ADPs) than the Li2 sites.10 The neutron diffraction at a temperature as low as 100 K, where the thermal vibration of Li is quite small, identified the exact Li positions99 and the Li4 position was suggested to be the migration pathway.99

9.9

Characteristics as ionic conductor

LGPS is unique crystal structure type in contrast to known mother structures such as Ag7P3S11, silver Argyrodite (Ag8GeS6), and LISICON (g-Li3PO4) in the sulfide-based lithium ionic conductors Li7P3S11, lithium Argyrodite, and thio-LISICON, respectively. In a tetragonal lattice (P42/nmc No. 137, a ¼ 8.66402, c ¼ 12.5830 A˚ ), germanium and phosphorus share Ge1/P1 (4d position), with the remaining phosphorus atoms in the P2(2b) position; together with four sulfur [Ge1/P1]S4 and [P2]S4 tetrahedral units are formed. Four Li positions, Li1(16 h), Li2(4d), Li3(16 h), and Li4(4c), exist. The tetrahedral arrangement structure of LGPS differs from that of the a, b, and g phases of Li3PS4 or any of the other thiophosphates.24,95 The tetrahedral arrangement of LGPS is different from that of Li3PS4. In the Li4GeS4-Li3PS4 system, as Ge content increases, the PS4 tetrahedra change from the g-Li3PS4 type,

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

559

Fig. 24 The LGPS-type structure. (a1) Perspective views of the tetrahedral skeleton and Li. (a2) Connections between polyhedral units with Li sites ([Li1]S4 and [Li3]S4). (a3) [Li3]S4 and [Li2]S6. (a4) [Li1]S4 and [Li4]S6. (b1 and b2) LGPS-type structural unit found in multiple unit cells in Na11Sn2PS12. (b1) Perspective view and (b2) top view. (c1–c3) Difference in the arrangements of [P]S4 and [M]S4 units observed in Li10GeP2S12 with M ¼ Ge (c1), Li11Si2PS12 with M ¼ Si (c2), and Na11Sn2PS12 with M ¼ Sn (c3). In the Na11Sn2PS12, [P]S4 indicated by a blue arrow points in a direction different from the corresponding ones in the Li10GeP2S12 and Li11Si2PS12. Figures from Ref. Kato, Y.; Hori, S.; Kanno, R. Li10GeP2S12-type Superionic Conductors: Synthesis, Structure, and Ionic Transportation. Adv. Energy Mater. 2020, 10 (42), 2002153. https://doi.org/10.1002/aenm.202002153.

560

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

in which they are arranged in the same direction, to the b-Li3PS4 type, in which the tetrahedra are disordered, while in LGPS the GeS4 and PS4 tetrahedra are arranged more disordered (see Fig. 3). The introduction and mixing of tetrahedra of different sizes is the factor that produces LGPS-type structures at intermediate compositions. Na11Sn2PS12 has a different orientation of [P] S4 tetrahedra.116 Similar to the mixed polyanion effect in the LISICON system, where the presence of many types of tetrahedral units in the crystal improves conductivity, the LGPS-type phase containing [Si/Sn/P]S4 improves conductivity.93 The explanation of the phenomena of increased ionic conductivity due to polyanion mixing has since become a more general concept as the high entropy effect117 and even higher ionic conductivity is achieved in LGPS.14 Thio-LISICON and LGPS have different anion arrangements: the former is hexagonally close-packed (hcp),23 while LGPS has a body-centered cubic (bcc) lattice.118 AgX and CuX (X ¼ Cl, Br, I) in the bcc arrangement show high ionic conductivity, while AgX and CuX in the hcp arrangement show high ionic conductivity. In hcp, three types of Li transfer are possible: tetrahedral-tetrahedral, tetrahedral-octahedral, and octahedral-octahedral. In bcc, Li transfer occurs between face-sharing tetrahedra with small barriers. The bcc arrangement at the Li1 and Li3 positions in LGPS is advantageous for Li migration, and also suggests that the main conduction pathway in LGPS is a one-dimensional chain along the c-axis through Li1 and Li3.

9.10 Lithium conduction mechanism The ionic conductivity measurements and precise structural analysis of single crystals have revealed the ionic conductivity phenomenon. An Arrhenius plot of the ionic conductivity of the single crystal is shown in Fig. 25. The ionic conductivity of the bulk in the [001] and [110] directions, determined from impedance measurements, is s ¼ 27 and 7 mS cm−1 at room temperature, respectively, higher in the [001] direction than the 14 mS cm−1 and 5 mS cm−1.99,100 obtained for the polycrystalline sample, indicating the presence of a one-dimensional pathway. The conductivity in the [110] direction is not small, suggesting a two-dimensional pathway. At low temperatures below about 250 K, the slope increases and the diffusion process of Li changes. The activation energies Ea are nearly equal in the two directions above and below 250 K at 0.29 and 0.39 eV in [001] and 0.30 and 0.40 eV in [110], respectively. The values at room temperature are close to those of the polycrystalline samples. The ions diffuse by hopping between the positions where they are present. The positions are linked by a small energy barrier to form a diffusion pathway, and long-range migration is achieved. For long-range ion diffusion, which corresponds to ionic conductivity, the ions move along the pathway with the lowest energy, and the highest value of the energy barrier along that pathway is the activation energy Ea of that diffusion. The concentration n of mobile ion carriers such as vacancies and interstitial atoms is also important. A low Ea and large n are necessary to obtain high ionic conductivity. Usually, at temperature T, s ¼ nem0 exp:ð − Ea=kBTÞ:

Fig. 25 Conductivity Arrhenius plots for single crystals of Li10GeP2S12. Conductivity along c-direction shows higher conductivity with similar activation energies as the ab-plane.89 From Iwasaki, R.; Hori, S.; Kanno, R.; Yajima, T.; Hirai, D.; Kato, Y.; Hiroi, Z. Weak Anisotropic Lithium-ion Conductivity in Single Crystals of Li10GeP2S12. Chem. Mater. 2019, 31 (10): 3694–3699. https://doi.org/10.1021/acs.chemmater.9b00420.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

561

Fig. 26 (a) Crystal structure of Li10GeP2S12 at 10 K. (b) MEM map Li site positions determined by single crystal neutron diffraction measurements. (c) Potential barrier along the conduction pathway deduced by one particle potential (OPP) method.119 From Yajima, T.; Hinuma, Y.; Hori, S.; Iwasaki, R.; Kanno, R.; Ohhara, T.; Nakao, A.; Munakata, K.; Hiroi, Z. Correlated Li-ion Migration in the Superionic Conductor Li10GeP2S12. J. Mater. Chem. A 2021, 9 (18): 11278–11284. https://doi. org/10.1039/d1ta00552a.

There are two diffusion pathways in the LGPS: a one-dimensional pathway in [001] involving the Li1 and Li3 positions and a two-dimensional pathway in [110] involving the Li1 and Li4 positions. The thermal vibration parameter of Li2 is small and the [110] 2D pathway via the Li2 position does not contribute to diffusion. The Ea values for these two pathways involving Li1- Li3 ([001]) and Li1-Li4 ([110]) were determined from molecular dynamics calculations and NMR experiments to be 0.16–0.19 eV and 0.26–0.30 eV for [001] and [110], respectively. Impedance measurements of single crystals show Ea values of about 0.3 eV in both of the two directions, close to the local Ea value in [110]. This suggests that the long-range diffusion that dominates the bulk conductivity is determined by the bottleneck in the [110] direction. Fig. 26a and b shows the crystal structure determined by single-crystal neutron diffraction. The scattering length density distributions at the Li2 and Li4 positions are elliptical, while those at the Li1 and Li3 positions extend in the [001] direction. The Li3 position also splits into three at 10 K. At 10 K, Li stops at three slightly adjacent positions, but at room temperature, the single elongated density means that Li ions can easily move between the three positions by thermal energy, and the potential barriers between each other are very small. According to the effective single-particle potential (OPP) calculations (Fig. 26c), the migration barrier for single ions in the 1D and 2D pathways is 0.09 eV, which is much smaller than the experimental value for the 1D pathway ([001]). On the other hand, the Li1dLi4dLi1 potential barrier in the [110] direction is 0.35 eV, which is comparable to the experimental value of 0.30 eV, indicating that the single ion jump mechanism contributes to migration in the 2-D pathway. Diffusion in the 1D channel is thought to occur in a correlated and coordinated manner, rather than by classical single ion jumps between positions. The migration barrier calculated by NEB assuming that the three Li atoms move cooperatively in one direction along the pathway is 0.33 eV, which is larger than the value calculated by OPP and consistent with the experimental value of 0.29 eV based on AC impedance measurements.89 On the other hand, the two-dimensional pathways Li4-Li1(Li3)dLi4 and Li2dLi3dLi2 are 0.41 eV and 0.59 eV, respectively, which are close to those obtained by the OPP calculation, suggesting a single ion hopping mechanism at work in the two-dimensional pathway. The collective motion of Li ions is discussed to explain why the activation energy of 0.16 eV observed in the LGPS by NMR is much smaller than the Ea ¼ 0.47 eV calculated by classical single ion hopping.120 The reason why the migration along the 1D pathway is unlikely to occur despite the flat energy barrier of the 1D pathway may be due to the correlation between the Li ions in the 1D channel. This mechanism is different from the usual known cooperative migration mechanism. Quasi-elastic neutron scattering (QENS) experiments of LGPS have revealed ion dynamics in the angstrom-scale spatial region and in the picosecond to nanosecond timescale.121 LGPS shows a high lithium diffusion coefficient at 298 K on the order of 10−6 cm2 s−1, higher than the diffusion coefficient determined by pulsed field gradient nuclear magnetic resonance (PFG-NMR) (see Fig. 27). This difference indicates that there is an impediment to ion motion on longer time scales. The guidelines for material design concept can be obtained from the results of the ion conductivity mechanism elucidation. The one-dimensional channel of LGPS is almost filled with Li, and the potential between the positions is rather small, indicating that the structure is suitable for ionic conductivity. However, the measured activation energy is relatively large, suggesting that the collective motion is inhibited by the interaction between Li in the channels. In other words, higher ionic conductivity can be expected by decreasing the amount of Li in the channels or by disturbing the environment of the conduction pathway using high entropy effect.14

9.11 Solid-state batteries This section describes the characteristics of solid-state batteries using LGPS as solid electrolyte. The advantages of using solid-state batteries are as follows.

562

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

QENS

PFG-NMR

Fig. 27 DPFG-NMR/10−7 cm−2 s −1 ¼ 1.5 measured by PFG-NMR (time sale: 20 ms) at 353 K, and DQENS / 10−7 cm−2 s −1 ¼ 8 measured by QENS (time scale: 100 ps) at 338 K, indicating DQENS > DPFG-NMR. Higher diffusivity in short time scale indicates Li-ion motion is more significant in shorter length scale (site to site jump).121 From Hori, S.; Kanno, R.; Kwon, O.; Kato, Y.; Yamada, T.; Matsuura, M.; Yonemura, M.; Kamiyama, T.; Shibata, K.; Kawakita, Y. Revealing the Ion Dynamics in Li10GeP2S12 by Quasi-elastic Neutron Scattering Measurements. J. Phys. Chem. C 2022, 126 (22), 9518–9527. https://doi.org/10.1021/acs.jpcc.2c01748.

(1) Wide temperature range and simplified cooling system, (2) Bipolar stacking improves the volume efficiency of the battery system, (3) Solid electrolytes have an ionic transport number of 1, which simplifies battery reactions and enables high currents. In a liquid electrolyte, cations and anions move toward opposite sides, but in a solid electrolyte, only lithium carries the charge, so the concentration gradient that causes the diffusion limit seen in a liquid electrolyte is not generated. High power output, rapid charging, and high energy density battery single cells can therefore be expected, and both energy density and input/output density are possible. These characteristics were demonstrated in LGPS-based solid-state batteries. Research on the solid-state battery has progressed rapidly since then. Solid-state batteries were prepared using LGPS solid electrolytes (12 mS cm−1 at room temperature) or Li9.54Si1.74P1.44S11.7Cl0.3 (25 mS cm−1 at room temperature), LiCoO2 positive electrode, In/Li negative electrode or Li4Ti5O12 negative electrode. Their output characteristics were shown to be extremely excellent.12 The output characteristics surpassed even those of capacitors, and it became clear that the advantage of making the battery solid was charge and discharge at high current densities. Fig. 28 shows a Ragon plot.12 It takes output density on the vertical axis and energy density on the horizontal axis; the higher the output and energy density, the better the device. Solid-state batteries have advantages in both output characteristics and energy density. The advantages of using solid batteries are obvious, and this has since prompted the research of solid-state batteries. In 2023, Li9.54[Si0.6Ge0.4]1.74P1.44S11.1Br0.3O0.6 with an ionic conductivity of 32 mS cm−1 was also found.14 Using this electrolyte, it was demonstrated that lithium metal anodes can be charged and discharged at high current densities with a cathode thicknesses of 1 mm in this battery structures. Fig. 29 shows the performance of all-solid-state cells. Discharge curves for the thick electrode presented herein (800 mm) under 0.587 mA cm−2 (0.025  C) at different temperatures ranging from 25  C to −10  C.14 Next to solid-state batteries with the battery configurations similar to lithium-ion batteries, the possibility of second- and third-generation solid-state batteries with even better properties was demonstrated.

10

Concluding remarks

Sulfide solid electrolytes with relatively high lithium ion conductivity have been found in the sulfide glass, Argyrodite, thio-LISICON, and LGPS systems. These materials are not only good examples for material exploration of ionic conductors, but also for use as solid electrolytes in all-solid-state batteries. These sulfides exhibit high ionic conductivity among lithium-ion

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

563

Fig. 28 The Ragone plot. The Ragone plots of the all-solid-state-cells and previously reported batteries and capacitors. The red dashed line indicates the specific energy E ¼ 102 Wh kg−1 and specific power P ¼ 10 kW kg−1. The all-solid-state cells simultaneously achieved high energy and power (E > 102 Wh kg−1 and P > 10 kW kg−1), which is difficult to achieve for conventional devices.12 From Kato, Y.; Hori, S.; Saito, T.; Suzuki, K.; Hirayama, M.; Mitsui, A.; Yonemura, M.; Iba, H.; Kanno, R. High-Power All-Solid-State Batteries Using Sulfide Superionic Conductors. Nat. Energy 2016, 1, 16030. https://doi.org/10.1038/nenergy.2016.30.

Fig. 29 Performance of all-solid-state cells. Discharge curves for the thick electrode presented herein (800 mm) under 0.587 mA cm−2 (0.025  C) at different temperatures ranging from 25  C to −10  C. Inset: cross sectional scanning electron microscopy image of the cathode composite pellet with a capacity loading of 23.5 mAh cm−2.14 From Li, Y.; Song, S.; Kim, H.; Nomoto, K.; Kim, H.; Sun, X.; Hori, S.; Suzuki, K.; Matsui, N.; Hirayama, M.; et al. A Lithium Superionic Conductor for Millimeter-thick Battery Electrode. Science 2023, 381 (6653), 50–53. https://doi.org/10.1126/science.add7138.

conductors, and their conductivity values are similar to those of liquid electrolytes used in lithium-ion batteries. These solid electrolytes are becoming the electrolyte of choice for batteries that no longer need to distinguish between solid and liquid electrolytes. These sulfide solid electrolytes even indicate a future direction for batteries: the possibility of second- and third-generation solid-state batteries with even better properties next to solid-state batteries with battery configurations similar to those of lithium-ion batteries.

564

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

References 1. Carette, B.; Maurin, M.; Ribes, M.; Duclot, M. Ionic Conductive Sulfide-based M2SGeS2MI(M¼Li,Ag) Glass Systems. Their Use in Solid State Batteries. Solid State Ion. 1983, 9–10, 655–658. https://doi.org/10.1016/0167-2738(83)90310-7. 2. Mercier, R.; Malugani, J.-P.; Fahys, B.; Robert, G. Superionic Conduction in Li2S-P2S5-LiI-Glasses. Solid State Ion. 1981, 5, 663–666. https://doi.org/10.1016/0167-2738(81) 90341-6. 3. Kanno, R.; Hata, T.; Kawamoto, Y.; Irie, M. Synthesis of a New lithium Ionic Conductor, Thio-LISICON-lithium Germanium Sulfide System. Solid State Ion. 2000, 130 (1–2), 97–104. https://doi.org/10.1016/s0167-2738(00)00277-0. 4. Leube, B. T.; Inglis, K. K.; Carrington, E. J.; Sharp, P. M.; Shin, J. F.; Neale, A. R.; Manning, T. D.; Pitcher, M. J.; Hardwick, L. J.; Dyer, M. S.; et al. Lithium Transport in Li4.4M0.4M’S’0.6(4) (M ¼ Al3+, Ga3+, and M’ ¼ Ge4+, Sn4+): Combined Crystallographic, Conductivity, Solid State NMR, and Computational Studies. Chem. Mater. 2018, 30 (20), 7183–7200. https://doi.org/10.1021/acs.chemmater.8b03175. 5. Kanno, R.; Marayama, M. Lithium Ionic Conductor Thio-LISICON-the Li2S-GeS2-P2S5 System. J. Electrochem. Soc. 2001, 148 (7), A742–A746. https://doi.org/ 10.1149/1.1379028. 6. Deiseroth, H. J.; Kong, S. T.; Eckert, H.; Vannahme, J.; Reiner, C.; Zaiss, T.; Schlosser, M. Li6PS5X: A Class of Crystalline Li-rich Solids with an Unusually High Li+ Mobility. Angew. Chem. Int. Ed. 2008, 47 (4), 755–758. https://doi.org/10.1002/anie.200703900. 7. Rao, R. P.; Adams, S. Studies of Lithium Argyrodite Solid Electrolytes for all-Solid-State Batteries. Phys. Status Solidi A 2011, 208 (8), 1804–1807. https://doi.org/10.1002/ pssa.201001117. 8. Mizuno, F.; Hayashi, A.; Tadanaga, K.; Tatsumisago, M. New, Highly Ion-conductive Crystals Precipitated from Li2S-P2S5 Glasses. Adv. Mater. 2005, 17 (7), 918. https://doi. org/10.1002/adma.200401286. 9. Seino, Y.; Ota, T.; Takada, K.; Hayashi, A.; Tatsumisago, M. A Sulphide Lithium Super Ion Conductor is Superior to Liquid Ion Conductors for Use in Rechargeable Batteries. Energ. Environ. Sci. 2014, 7 (2), 627–631. https://doi.org/10.1039/c3ee41655k. 10. Kamaya, N.; Homma, K.; Yamakawa, Y.; Hirayama, M.; Kanno, R.; Yonemura, M.; Kamiyama, T.; Kato, Y.; Hama, S.; Kawamoto, K.; Mitsui, A. A Lithium Superionic Conductor. Nat. Mater. 2011, 10 (9), 682–686. https://doi.org/10.1038/nmat3066. 11. Ohara, K.; Mitsui, A.; Mori, M.; Onodera, Y.; Shiotani, S.; Koyama, Y.; Orikasa, Y.; Murakami, M.; Shimoda, K.; Mori, K.; et al. Structural and electronic features of binary Li2S-P2S5 glasses. Scientific Reports 2016, 6 (1), 21302. https://doi.org/10.1038/srep21302. 12. Kato, Y.; Hori, S.; Saito, T.; Suzuki, K.; Hirayama, M.; Mitsui, A.; Yonemura, M.; Iba, H.; Kanno, R. High-Power All-Solid-State Batteries Using Sulfide Superionic Conductors. Nat. Energy 2016, 1, 16030. https://doi.org/10.1038/nenergy.2016.30. 13. Hori, S.; Suzuki, K.; Hirayama, M.; Kato, Y.; Kanno, R. Lithium Superionic Conductor Li9.42Si1.02P2.1S9.96O2.04 with Li10GeP2S12-type Structure in the Li2S-P2S5-SiO2 Pseudoternary System: Synthesis, Electrochemical Properties, and Structure Composition Relationships. Front. Energy Res. 2016, 4, 38. https://doi.org/ 10.3389/fenrg.2016.00038. 14. Li, Y.; Song, S.; Kim, H.; Nomoto, K.; Kim, H.; Sun, X.; Hori, S.; Suzuki, K.; Matsui, N.; Hirayama, M.; et al. A Lithium Superionic Conductor for Millimeter-thick Battery Electrode. Science 2023, 381 (6653), 50–53. https://doi.org/10.1126/science.add7138. 15. Kraft, M. A.; Ohno, S.; Zinkevich, T.; Koerver, R.; Culver, S. P.; Fuchs, T.; Senyshyn, A.; Indris, S.; Morgan, B. J.; Zeier, W. G. Inducing High Ionic Conductivity in the Lithium Superionic Argyrodites Li6+x P(1-x)Ge(x)S(5)l for All-solid-state Batteries. J. Am. Chem. Soc. 2018, 140 (47), 16330–16339. https://doi.org/10.1021/jacs.8b10282. 16. Inoue, Y.; Suzuki, K.; Matsui, N.; Hirayama, M.; Kanno, R. Synthesis and Structure of Novel Lithium-ion Conductor Li7Ge3PS12. J. Solid State Chem. 2017, 246, 334–340. https://doi.org/10.1016/j.jssc.2016.12.001. 17. Feng, X.; Chien, P.-H.; Wang, Y.; Patel, S.; Wang, P.; Liu, H.; Immediato-Scuotto, M.; Hu, Y.-Y. Enhanced Ion Conduction by Enforcing Structural Disorder in Li-deficient Argyrodites Li6−xPS5−xCl1+x. Energy Stor. Mater. 2020, 30, 67–73. https://doi.org/10.1016/j.ensm.2020.04.042. 18. Kaib, T.; Haddadpour, S.; Kapitein, M.; Bron, P.; Schroder, C.; Eckert, H.; Roling, B.; Dehnen, S. New Lithium Chalcogenidotetrelates, LiChT: Synthesis and Characterization of the Li+-Conducting Tetralithium Ortho-sulfidostannate Li4SnS4. Chem. Mater. 2012, 24 (11), 2211–2219. https://doi.org/10.1021/cm3011315. 19. Liu, Z. C.; Fu, W. J.; Payzant, E. A.; Yu, X.; Wu, Z. L.; Dudney, N. J.; Kiggans, J.; Hong, K. L.; Rondinone, A. J.; Liang, C. D. Anomalous High Ionic Conductivity of Nanoporous Beta-Li3PS4. J. Am. Chem. Soc. 2013, 135 (3), 975–978. https://doi.org/10.1021/ja3110895. 20. Huang, W. Z.; Yoshino, K.; Hori, S.; Suzuki, K.; Yonemura, M.; Hirayama, M.; Kanno, R. Superionic Lithium Conductor With a Cubic Argyrodite-type Structure in the Li-Al-Si-S System. J. Solid State Chem. 2019, 270, 487–492. https://doi.org/10.1016/j.jssc.2018.12.015. 21. Robert, G.; Malugani, J. P.; Saida, A. Fast Ionic Silver and Lithium Conduction in Glasses. Solid State Ion. 1981, 3–4, 311–315. https://doi.org/10.1016/0167-2738(81) 90104-1. 22. Kondo, S.; Takada, K.; Yamamura, Y. New Lithium Ion Conductors Based on Li2S-SiS2 System. Solid State Ion. 1992, 53, 1183–1186. https://doi.org/10.1016/0167-2738(92) 90310-l. 23. Homma, K.; Yonemura, M.; Kobayashi, T.; Nagao, M.; Hirayama, M.; Kanno, R. Crystal Structure and Phase Transitions of the Lithium Ionic Conductor Li3PS4. Solid State Ion. 2011, 182 (1), 53–58. https://doi.org/10.1016/j.ssi.2010.10.001. 24. Kudu, Ö. U.; Famprikis, T.; Fleutot, B.; Braida, M.-D.; Le Mercier, T.; Islam, M. S.; Masquelier, C. A Review of Structural Properties and Synthesis Methods of Solid Electrolyte Materials in the Li2S−P2S5 Binary System. J. Power Sources 2018, 407, 31–43. https://doi.org/10.1016/j.jpowsour.2018.10.037. 25. Kato, Y.; Hori, S.; Kanno, R. Li10GeP2S12-type Superionic Conductors: Synthesis, Structure, and Ionic Transportation. Adv. Energy Mater. 2020, 10 (42), 2002153. https://doi. org/10.1002/aenm.202002153. 26. Iwamoto, K.; Aotani, N.; Takada, K.; Kondo, S. Application of Li3PO4-Li2S-SiS2 Glass to the Solid State Secondary Batteries. Solid State Ion. 1995, 79, 288–291. https://doi.org/ 10.1016/0167-2738(95)00076-I. 27. Takada, K.; Aotani, N.; Iwamoto, K.; Kondo, S. Solid State Lithium Battery with Oxysulfide Glass. Solid State Ion. 1996, 86-88, 877–882. https://doi.org/10.1016/0167-2738 (96)00199-3. 28. Takada, K.; Inada, T.; Kajiyama, A.; Sasaki, H.; Kondo, S.; Watanabe, M.; Murayama, M.; Kanno, R. Solid-state Lithium Battery with Graphite Anode. Solid State Ion. 2003, 158 (3–4), 269–274. https://doi.org/10.1016/s0167-2738(02)00823-8. 29. Seino, Y.; Takada, K.; Kim, B. C.; Zhang, L. Q.; Ohta, N.; Wada, H.; Osada, M.; Sasaki, T. Synthesis of Phosphorous Sulfide Solid Electrolyte and All-solid-state Lithium Batteries With Graphite Electrode. Solid State Ion. 2005, 176 (31–34), 2389–2393. https://doi.org/10.1016/j.ssi.2005.03.026. 30. Dietrich, C.; Weber, D. A.; Sedlmaier, S. J.; Indris, S.; Culver, S. P.; Walter, D.; Janek, J.; Zeier, W. G. Lithium Ion Conductivity in Li2S–P2S5 Glasses—Building Units and Local Structure Evolution During the Crystallization of Superionic Conductors Li3PS4, Li7P3S11 and Li4P2S7. J. Mater. Chem. A 2017, 5 (34), 18111–18119. https://doi.org/10.1039/ C7TA06067J. 31. Ohtomo, T.; Hayashi, A.; Tatsumisago, M.; Tsuchida, Y.; Hama, S.; Kawamoto, K. All-solid-state Lithium Secondary Batteries Using the 75Li2S25P2S5 Glass and the 70Li2S30P2S5 Glass–Ceramic as Solid Electrolytes. J. Power Sources 2013, 233, 231–235. https://doi.org/10.1016/j.jpowsour.2013.01.090. 32. Zhang, Z.; Kennedy, J. H. Synthesis and Characterization of the B2S3 Li2S, the P2S5 Li2S and the B2S3 P2S5 Li2S Glass Systems. Solid State Ion. 1990, 38 (3), 217–224. https:// doi.org/10.1016/0167-2738(90)90424-P. 33. Pradel, A.; Ribes, M. Ionically Conductive Chalcogenide Glasses. J. Solid State Chem. 1992, 96 (1), 247–257. https://doi.org/10.1016/S0022-4596(05)80318-4. 34. Souquet, J. L.; Robinel, E.; Barrau, B.; Ribes, M. Glass Formation and Ionic Conduction in the M2S GeS2 (M ¼ Li, Na, Ag) Systems. Solid State Ion. 1981, 3-4, 317–321. https:// doi.org/10.1016/0167-2738(81)90105-3.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

565

35. Tatsumisago, M.; Hayashi, A. Superionic Glasses and Glass–ceramics in the Li2S–P2S5 System for All-solid-state Lithium Secondary Batteries. Solid State Ion. 2012, 225, 342–345. https://doi.org/10.1016/j.ssi.2012.03.013. 36. Hayashi, A.; Komiya, R.; Tatsumisago, M.; Minami, T. Characterization of Li2S–SiS2–Li3MO3 (M¼B, Al, Ga and In) Oxysulfide Glasses and their Application to Solid State Lithium Secondary Batteries. Solid State Ion. 2002, 152–153, 285–290. https://doi.org/10.1016/S0167-2738(02)00313-2. 37. Yamauchi, A.; Sakuda, A.; Hayashi, A.; Tatsumisago, M. Preparation and Ionic Conductivities of (100 − x)(0.75Li2S0.25P2S5)xLiBH4 Glass Electrolytes. J. Power Sources 2013, 244, 707–710. https://doi.org/10.1016/j.jpowsour.2012.12.001. 38. Wada, H.; Menetrier, M.; Levasseur, A.; Hagenmuller, P. Preparation and Ionic Conductivity of New B2S3-Li2S-LiI Glasses. Mater. Res. Bull. 1983, 18 (2), 189–193. https://doi. org/10.1016/0025-5408(83)90080-6. 39. Kennedy, J. H.; Zhang, Z.; Eckert, H. Ionically Conductive Sulfide-Based Lithium Glasses. J. Non Cryst. Solids 1990, 123 (1), 328–338. https://doi.org/10.1016/0022-3093 (90)90804-U. 40. Yamane, H.; Shibata, M.; Shimane, Y.; Junke, T.; Seino, Y.; Adams, S.; Minami, K.; Hayashi, A.; Tatsumisago, M. Crystal Structure of a Superionic Conductor, Li7P3S11. Solid State Ion. 2007, 178 (15–18), 1163–1167. https://doi.org/10.1016/j.ssi.2007.05.020. 41. Mercier, R.; Malugani, J. P.; Fahys, B.; Robert, G.; Douglade, J. Structure Du Tetrathiophosphate De Lithium. Acta Crystallogr. Sect. B 1982, 38 (7), 1887–1890. https://doi. org/10.1107/S0567740882007535. 42. Kong, S. T.; Gün, Ö.; Koch, B.; Deiseroth, H. J.; Eckert, H.; Reiner, C. Structural Characterisation of the Li Argyrodites Li7PS6 and Li7PSe6 and Their Solid Solutions: Quantification of Site Preferences by MAS-NMR Spectroscopy. Chem. – A Eur. J. Dermatol. 2010, 16 (17), 5138–5147. https://doi.org/10.1002/chem.200903023. 43. Tatsumisago, M.; Yoneda, K.; Machida, N.; Hinami, T. Ionic Conductivity of Rapidly Quenched Glasses with High Concentration of Lithium Ions. J. Non Cryst. Solids 1987, 95-96, 857–864. https://doi.org/10.1016/S0022-3093(87)80691-9. 44. Yamamoto, H.; Machida, N.; Shigematsu, T. A Mixed-former Effect on Lithium-ion Conductivities of the Li2S–GeS2–P2S5 Amorphous Materials Prepared by a High-energy Ball-milling Process. Solid State Ion. 2004, 175 (1), 707–711. https://doi.org/10.1016/j.ssi.2004.08.028. 45. Yao, X.; Liu, D.; Wang, C.; Long, P.; Peng, G.; Hu, Y.-S.; Li, H.; Chen, L.; Xu, X. High-energy All-solid-state Lithium Batteries with Ultralong Cycle Life. Nano Lett. 2016, 16 (11), 7148–7154. https://doi.org/10.1021/acs.nanolett.6b03448. 46. Inada, T.; Takada, K.; Kajiyama, A.; Kouguchi, M.; Sasaki, H.; Kondo, S.; Watanabe, M.; Murayama, M.; Kanno, R. Fabrications and Properties of Composite Solid-state Electrolytes. Solid State Ion. 2003, 158 (3–4), 275–280. https://doi.org/10.1016/s0167-2738(02)00889-5. 47. Xu, R. C.; Xia, X. H.; Yao, Z. J.; Wang, X. L.; Gu, C. D.; Tu, J. P. Preparation of Li7P3S11 Glass-ceramic Electrolyte by Dissolution-evaporation Method for All-solid-state Lithium ion Batteries. Electrochim. Acta 2016, 219, 235–240. https://doi.org/10.1016/j.electacta.2016.09.155. 48. Choi, Y. E.; Park, K. H.; Kim, D. H.; Oh, D. Y.; Kwak, H. R.; Lee, Y.-G.; Jung, Y. S. Coatable Li4SnS4 Solid Electrolytes Prepared from Aqueous Solutions for All-solid-state Lithium-Ion Batteries. ChemSusChem 2017, 10 (12), 2605–2611. https://doi.org/10.1002/cssc.201700409. 49. Hong, H. Y.-P. Crystal Structure and Ionic Conductivity of Li14Zn(GeO4)4 and Other New Li+ Superionic Conductors. Mater. Res. Bull. 1978, 13 (2), 117–124. https://doi.org/ 10.1016/0025-5408(78)90075-2. 50. Inada, T.; Kobayashi, T.; Sonoyama, N.; Yamada, A.; Kondo, S.; Nagao, M.; Kanno, R. All Solid-state Sheet Battery Using Lithium Inorganic Solid Electrolyte, Thio-LISICON. J. Power Sources 2009, 194 (2), 1085–1088. https://doi.org/10.1016/j.jpowsour.2009.06.100. 51. Kobayashi, T.; Yamada, A.; Kanno, R. Interfacial Reactions at Electrode/electrolyte Boundary in all Solid-state Lithium Battery Using Inorganic Solid Electrolyte, Thio-LISICON. Electrochim. Acta 2008, 53 (15), 5045–5050. https://doi.org/10.1016/j.electacta.2008.01.071. 52. Murayama, M.; Kanno, R.; Kawamoto, Y.; Kamiyama, T. Structure of the Thio-LISICON, Li4GeS4. Solid State Ion. 2002, 154–155, 789–794. https://doi.org/10.1016/S01672738(02)00492-7. 53. Murayama, M.; Kanno, R.; Irie, M.; Ito, S.; Hata, T.; Sonoyama, N.; Kawamoto, Y. Synthesis of New Lithium Ionic Conductor Thio-LISICON—Lithium Silicon Sulfides System. J. Solid State Chem. 2002, 168 (1), 140–148. https://doi.org/10.1006/jssc.2002.9701. 54. Matsumura, T.; Nakano, K.; Kanno, R.; Hirano, A.; Imanishi, N.; Takeda, Y. Nickel Sulfides as a Cathode for All-solid-state Ceramic Lithium Batteries. J. Power Sources 2007, 174 (2), 632–636. https://doi.org/10.1016/j.jpowsour.2007.06.168. 55. Murayama, M.; Sonoyama, N.; Yamada, A.; Kanno, R. Material Design of New Lithium Ionic Conductor, Thio-LISICON, in the Li2S-P2S5 System. Solid State Ion. 2004, 170 (3–4), 173–180. https://doi.org/10.1016/j.ssi.2004.02.025. 56. Kanno, R.; Murayama, M.; Sakamoto, K. New lithium solid electrolytes, thio-lisicon: Materials design concept and application to solid state battery. Solid State Ionics, Trends in the New Millennium Proceedings of the 8th Asian Conference, 15–19 December 2002: Langkawi, Malaysia, 2002. https://doi.org/10.1142/5175. 57. Minami, K.; Mizuno, F.; Hayashi, A.; Tatsumisago, M. Lithium Ion Conductivity of the Li2S-P2S5 Glass-based Electrolytes Prepared by the Melt Quenching Method. Solid State Ion. 2007, 178 (11−12), 837–841. https://doi.org/10.1016/j.ssi.2007.03.001. 58. Peng, X. Y.; Chien, P. H.; Wang, Y.; Patel, S.; Wang, P. B.; Liu, H. Y.; Immediato-Scuotto, M.; Hu, Y. Y. Enhanced Ion Conduction by Enforcing Structural Disorder in Li-Deficient Argyrodites Li6-xPS5-xCl1-x. Energy Stor. Mater. 2020, 30, 67–73. https://doi.org/10.1016/j.ensm.2020.04.042. 59. Stadler, F.; Fietzek, C. Crystalline Halide Substituted Li-Argyrodites as Solid Electrolytes for Lithium Secondary Batteries. ECS Trans. 2010, 25 (36), 177–183. Proceedings Paper https://doi.org/10.1149/1.3393854. 60. Lee, Y.-G.; Fujiki, S.; Jung, C.; Suzuki, N.; Yashiro, N.; Omoda, R.; Ko, D.-S.; Shiratsuchi, T.; Sugimoto, T.; Ryu, S.; et al. High-energy Long-cycling All-solid-state Lithium Metal Batteries Enabled by Silver–carbon Composite Anodes. Nat. Energy 2020, 5 (4), 299–308. https://doi.org/10.1038/s41560-020-0575-z. 61. Kong, S.-T.; Deiseroth, H.-J.; Reiner, C.; Gün, Ö.; Neumann, E.; Ritter, C.; Zahn, D. Lithium Argyrodites With Phosphorus and Arsenic: Order and Disorder of Lithium Atoms, Crystal Chemistry, and Phase Transitions. Chemistry 2010, 16 (7), 2198–2206. https://doi.org/10.1002/chem.200902470 (Acccessed on January 15, 2024). 62. Kraft, M. A.; Culver, S. P.; Calderon, M.; Böcher, F.; Krauskopf, T.; Senyshyn, A.; Dietrich, C.; Zevalkink, A.; Janek, J.; Zeier, W. G. Influence of Lattice Polarizability on the Ionic Conductivity in the Lithium Superionic Argyrodites Li6PS5X (X ¼ Cl, Br, I). J. Am. Chem. Soc. 2017, 139 (31), 10909–10918. https://doi.org/10.1021/jacs.7b06327. 63. Zhou, L. D.; Minafra, N.; Zeier, W. G.; Nazar, L. F. Innovative Approaches to Li-argyrodite Solid Electrolytes for All-solid-state Lithium Batteries. Acc. Chem. Res. 2021, 54 (12), 2717–2728. https://doi.org/10.1021/acs.accounts.0c00874. 64. Minafra, N.; Kraft, M. A.; Bernges, T.; Li, C.; Schlem, R.; Morgan, B. J.; Zeier, W. G. Local Charge Inhomogeneity and Lithium Distribution in the Superionic Argyrodites Li6PS5X (X ¼ Cl, Br, I). Inorg. Chem. 2020, 59 (15), 11009–11019. https://doi.org/10.1021/acs.inorgchem.0c01504. 65. de Klerk, N. J. J.; Rosłon, I.; Wagemaker, M. Diffusion Mechanism of Li Argyrodite Solid Electrolytes for Li-Ion Batteries and Prediction of Optimized Halogen Doping: The Effect of Li Vacancies, Halogens, and Halogen Disorder. Chem. Mater. 2016, 28 (21), 7955–7963. https://doi.org/10.1021/acs.chemmater.6b03630. 66. Morgan, B. J. Mechanistic Origin of Superionic Lithium Diffusion in Anion-disordered Li6PS5X Argyrodites. Chem. Mater. 2021, 33 (6), 2004–2018. https://doi.org/10.1021/ acs.chemmater.0c03738. 67. Gautam, A.; Sadowski, M.; Ghidiu, M.; Minafra, N.; Senyshyn, A.; Albe, K.; Zeier, W. G. Engineering the Site-disorder and Lithium Distribution in the Lithium Superionic Argyrodite Li6PS5Br. Adv. Energy Mater. 2021, 11 (5), 2003369. https://doi.org/10.1002/aenm.202003369 (Acccessed on January 15, 2024). 68. Zhou, L.; Assoud, A.; Zhang, Q.; Wu, X.; Nazar, L. F. New Family of Argyrodite Thioantimonate Lithium Superionic Conductors. J. Am. Chem. Soc. 2019, 141 (48), 19002–19013. https://doi.org/10.1021/jacs.9b08357. 69. Huang, W. Z.; Cheng, L. D.; Hori, S.; Suzuki, K.; Yonemura, M.; Hirayama, M.; Kanno, R. Ionic Conduction Mechanism of a lithium Superionic Argyrodite in the Li-Al-Si-S-O System. Mater. Adv. 2020, 1 (3), 334–340. https://doi.org/10.1039/d0ma00115e. 70. Ohno, S.; Helm, B.; Fuchs, T.; Dewald, G.; Kraft, M. A.; Culver, S. P.; Senyshyn, A.; Zeier, W. G. Further Evidence for Energy Landscape Flattening in the Superionic Argyrodites Li6+xP1–xMxS5I (M ¼ Si, Ge, Sn). Chem. Mater. 2019, 31 (13), 4936–4944. https://doi.org/10.1021/acs.chemmater.9b01857.

566

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

71. Minafra, N.; Culver, S. P.; Krauskopf, T.; Senyshyn, A.; Zeier, W. G. Effect of Si Substitution on the Structural and Transport Properties of Superionic Li-argyrodites. J. Mater. Chem. A 2018, 6 (2), 645–651. https://doi.org/10.1039/C7TA08581H. 72. Schneider, H.; Du, H.; Kelley, T.; Leitner, K.; ter Maat, J.; Scordilis-Kelley, C.; Sanchez-Carrera, R.; Kovalev, I.; Mudalige, A.; Kulisch, J.; et al. A Novel Class of Halogen-free, Super-Conductive Lithium Argyrodites: Synthesis and Characterization. J. Power Sources 2017, 366, 151–160. https://doi.org/10.1016/j.jpowsour.2017.09.012. 73. Zhang, Z.; Sun, Y.; Duan, X.; Peng, L.; Jia, H.; Zhang, Y.; Shan, B.; Xie, J. Design and Synthesis of Room Temperature Stable Li-argyrodite Superionic Conductors Via Cation Doping. J. Mater. Chem. A 2019, 7 (6), 2717–2722. https://doi.org/10.1039/C8TA10790D. 74. Zhou, L.; Park, K.-H.; Sun, X.; Lalère, F.; Adermann, T.; Hartmann, P.; Nazar, L. F. Solvent-engineered Design of Argyrodite Li6PS5X (X ¼ Cl, Br, I) Solid Electrolytes With High Ionic Conductivity. ACS Energy Lett. 2019, 4 (1), 265–270. https://doi.org/10.1021/acsenergylett.8b01997. 75. Yubuchi, S.; Uematsu, M.; Hotehama, C.; Sakuda, A.; Hayashi, A.; Tatsumisago, M. An Argyrodite Sulfide-based Superionic Conductor Synthesized by a Liquid-phase Technique With Tetrahydrofuran and Ethanol. J. Mater. Chem. A 2019, 7 (2), 558–566. https://doi.org/10.1039/C8TA09477B. 76. Ziolkowska, D. A.; Arnold, W.; Druffel, T.; Sunkara, M.; Wang, H. Rapid and Economic Synthesis of a Li7PS6 Solid Electrolyte from a Liquid Approach. ACS Appl. Mater. Interfaces 2019, 11 (6), 6015–6021. https://doi.org/10.1021/acsami.8b19181. 77. Tan, D. H. S.; Wu, E. A.; Nguyen, H.; Chen, Z.; Marple, M. A. T.; Doux, J.-M.; Wang, X.; Yang, H.; Banerjee, A.; Meng, Y. S. Elucidating Reversible Electrochemical Redox of Li6PS5Cl Solid Electrolyte. ACS Energy Lett. 2019, 4 (10), 2418–2427. https://doi.org/10.1021/acsenergylett.9b01693. 78. Dewald, G. F.; Ohno, S.; Kraft, M. A.; Koerver, R.; Till, P.; Vargas-Barbosa, N. M.; Janek, J.; Zeier, W. G. Experimental Assessment of the Practical Oxidative Stability of Lithium Thiophosphate Solid Electrolytes. Chem. Mater. 2019, 31 (20), 8328–8337. https://doi.org/10.1021/acs.chemmater.9b01550. 79. Zhu, Y. Z.; He, X. F.; Mo, Y. F. Origin of Outstanding Stability in the Lithium Solid Electrolyte Materials: Insights from Thermodynamic Analyses Based on First-principles Calculations. ACS Appl. Mater. Interfaces 2015, 7 (42), 23685–23693. https://doi.org/10.1021/acsami.5b07517. 80. Richards, W. D.; Miara, L. J.; Wang, Y.; Kim, J. C.; Ceder, G. Interface Stability in Solid-state Batteries. Chem. Mater. 2016, 28 (1), 266–273. https://doi.org/10.1021/acs. chemmater.5b04082. 81. Xu, H. J.; Yu, Y. R.; Wang, Z.; Shao, G. S. A Theoretical Approach to Address Interfacial Problems in All-Solid-State Lithium Ion Batteries: Tuning Materials Chemistry for Electrolyte and Buffer Coatings Based on Li6PA5Cl Halichalcogenides. J. Mater. Chem. A 2019, 7 (10), 5239–5247. https://doi.org/10.1039/c8ta11151k. 82. Walther, F.; Koerver, R.; Fuchs, T.; Ohno, S.; Sann, J.; Rohnke, M.; Zeier, W. G.; Janek, J. Visualization of the Interfacial Decomposition of Composite Cathodes in Argyrodite-based All-solid-state Batteries Using Time-of-flight Secondary-ion Mass Spectrometry. Chem. Mater. 2019, 31 (10), 3745–3755. https://doi.org/10.1021/acs. chemmater.9b00770. 83. Schwietert, T. K.; Arszelewska, V. A.; Wang, C.; Yu, C.; Vasileiadis, A.; de Klerk, N. J. J.; Hageman, J.; Hupfer, T.; Kerkamm, I.; Xu, Y.; et al. Clarifying the Relationship between Redox Activity and Electrochemical Stability in Solid Electrolytes. Nat. Mater. 2020, 19 (4), 428–435. https://doi.org/10.1038/s41563-019-0576-0. 84. Ye, L.; Li, X. A Dynamic Stability Design Strategy for Lithium Metal Solid State Batteries. Nature 2021, 593 (7858), 218–222. https://doi.org/10.1038/s41586-021-03486-3. 85. Kato, Y.; Saito, R.; Sakano, M.; Mitsui, A.; Hirayama, M.; Kanno, R. Synthesis, Structure and Lithium Ionic Conductivity of Solid Solutions of Li10(Ge1-xMx)P2S12 (M ¼ Si, Sn). J. Power Sources 2014, 271, 60–64. https://doi.org/10.1016/j.jpowsour.2014.07.159. 86. Sun, Y.; Suzuki, K.; Hara, K.; Hori, S.; Yano, T.-a.; Hara, M.; Hirayama, M.; Kanno, R. Oxygen Substitution Effects in Li10GeP2S12 Solid Electrolyte. J. Power Sources 2016, 324, 798–803. https://doi.org/10.1016/j.jpowsour.2016.05.100. 87. Takada, K.; Osada, M.; Ohta, N.; Inada, T.; Kajiyama, A.; Sasaki, H.; Kondo, S.; Watanabe, M.; Sasaki, T. Lithium Ion Conductive Oxysulfide, Li3PO4—Li3PS4. Solid State Ion. 2005, 176 (31), 2355–2359. https://doi.org/10.1016/j.ssi.2005.03.023. 88. Hori, S.; Suzuki, K.; Hirayama, M.; Kato, Y.; Saito, T.; Yonemura, M.; Kanno, R. Synthesis, Structure, and Ionic Conductivity of Solid Solution, Li10+dM1+dP2-dS12 (M ¼ Si, Sn). Faraday Discuss. 2014, 176, 83–94. https://doi.org/10.1039/C4FD00143E. 89. Iwasaki, R.; Hori, S.; Kanno, R.; Yajima, T.; Hirai, D.; Kato, Y.; Hiroi, Z. Weak Anisotropic Lithium-ion Conductivity in Single Crystals of Li10GeP2S12. Chem. Mater. 2019, 31 (10), 3694–3699. https://doi.org/10.1021/acs.chemmater.9b00420. 90. Krauskopf, T.; Culver, S. P.; Zeier, W. G. Bottleneck of Diffusion and Inductive Effects in Li10Ge1–xSnxP2S12. Chem. Mater. 2018, 30 (5), 1791–1798. https://doi.org/10.1021/ acs.chemmater.8b00266. 91. Bron, P.; Dehnen, S.; Roling, B. Li10Si0.3Sn0.7P2S12—A Low-cost and Low-grain-boundary-resistance Lithium Superionic Conductor. J. Power Sources 2016, 329 (Supplement C), 530–535. https://doi.org/10.1016/j.jpowsour.2016.08.115. 92. Bron, P.; Johansson, S.; Zick, K.; Schmedt auf der Günne, J.; Dehnen, S.; Roling, B. Li10SnP2S12: An Affordable Lithium Superionic Conductor. J. Am. Chem. Soc. 2013, 135 (42), 15694–15697. https://doi.org/10.1021/ja407393y (Acccessed on January 08, 2014). 93. Sun, Y.; Suzuki, K.; Hori, S.; Hirayama, M.; Kanno, R. Superionic Conductors: Li10+d [SnySi1–Y]1+dP2−dS12 With a Li10GeP2S12-type Structure in the Li3PS4–Li4SnS4–Li4SiS4 Quasi-ternary System. Chem. Mater. 2017, 29 (14), 5858–5864. https://doi.org/10.1021/acs.chemmater.7b00886. 94. Hori, S.; Kato, M.; Suzuki, K.; Hirayama, M.; Kato, Y.; Kanno, R. Phase Diagram of the Li4GeS4–Li3PS4 Quasi-binary System Containing the Superionic Conductor Li10GeP2S12. J. Am. Ceram. Soc. 2015, 98 (10), 3352–3360. https://doi.org/10.1111/jace.13694. 95. Homma, K.; Yonemura, M.; Nagao, M.; Hirayama, M.; Kanno, R. Crystal Structure of High-temperature Phase of Lithium Ionic Conductor, Li3PS4. J. Physical Soc. Japan 2010, 79. https://doi.org/10.1143/jpsjs.79sa.90. 96. Mercier, R.; Malugani, J. P.; Fahys, B.; Douglande, J.; Robert, G. Synthese, Structure Cristalline et Analyse Vibrationnelle de l’hexathiohypodiphosphate de Lithium Li4P2S6. J. Solid State Chem. 1982, 43 (2), 151–162. https://doi.org/10.1016/0022-4596(82)90224-9. 97. Tsukasaki, H.; Mori, S.; Shiotani, S.; Yamamura, H.; Iba, H. Direct Observation of a Non-isothermal Crystallization Process in Precursor Li10GeP2S12 Glass Electrolyte. J. Power Sources 2017, 369, 57–64. https://doi.org/10.1016/j.jpowsour.2017.09.070. 98. Kato, Y.; Shiotani, S.; Morita, K.; Suzuki, K.; Hirayama, M.; Kanno, R. All-solid-state Batteries With Thick Electrode Configurations. J. Phys. Chem. Lett. 2018, 9 (3), 607–613. https://doi.org/10.1021/acs.jpclett.7b02880. 99. Kwon, O.; Hirayama, M.; Suzuki, K.; Kato, Y.; Saito, T.; Yonemura, M.; Kamiyama, T.; Kanno, R. Synthesis, Structure, and Conduction Mechanism of the Lithium Superionic Conductor Li10+dGe1+dP2-dS12. J. Mater. Chem. A 2015, 3 (1), 438–446. https://doi.org/10.1039/C4TA05231E. 100. Weber, D. A.; Senyshyn, A.; Weldert, K. S.; Wenzel, S.; Zhang, W.; Kaiser, R.; Berendts, S.; Janek, J.; Zeier, W. G. Structural Insights and 3D Diffusion Pathways within the Lithium Superionic Conductor Li10GeP2S12. Chem. Mater. 2016, 28 (16), 5905–5915. https://doi.org/10.1021/acs.chemmater.6b02424. 101. Kuhn, A.; Köhler, J.; Lotsch, B. V. Single-Crystal X-ray Structure Analysis of the Superionic Conductor Li10GeP2S12. Phys. Chem. Chem. Phys. 2013, 15 (28), 11620–11622. https://doi.org/10.1039/C3CP51985F. 102. Machida, N.; Hashimoto, S.; Kinoshita, C. Abstract of Spring Meeting of Japan Society of Powder and Powder Metallurgy, 2019. 103. Wang, Y.; Liu, Z.; Zhu, X.; Tang, Y.; Huang, F. Highly lithium-ion Conductive Thio-LISICON Thin Film Processed by Low-temperature Solution Method. J. Power Sources 2013, 224, 225–229. https://doi.org/10.1016/j.jpowsour.2012.09.097. 104. Ito, T.; Hori, S.; Hirayama, M.; Kanno, R. Liquid-phase Synthesis of the Li10GeP2S12-type Phase in the Li–Si–P–S–Cl System. J. Mater. Chem. A 2022, 10 (27), 14392–14398. https://doi.org/10.1039/D2TA02834D. 105. Kuhn, A.; Gerbig, O.; Zhu, C.; Falkenberg, F.; Maier, J.; Lotsch, B. V. A New Ultrafast Superionic Li-conductor: Ion Dynamics in Li11Si2PS12 and Comparison with Other Tetragonal LGPS-type Electrolytes. Phys. Chem. Chem. Phys. 2014, 16 (28), 14669–14674. https://doi.org/10.1039/C4CP02046D. 106. Kim, K.-H.; Martin, S. W. Structures and Properties of Oxygen-substituted Li10SiP2S12–xOx Solid-state Electrolytes. Chem. Mater. 2019, 31 (11), 3984–3991. https://doi.org/ 10.1021/acs.chemmater.9b00505. 107. Feng, X.; Chien, P.-H.; Zhu, Z.; Chu, I.-H.; Wang, P.; Immediato-Scuotto, M.; Arabzadeh, H.; Ong, S. P.; Hu, Y.-Y. Studies of Functional Defects for Fast Na-ion Conduction in Na3−yPS4−xClx with a Combined Experimental and Computational Approach. Adv. Funct. Mater. 2019, 29 (9), 1807951. https://doi.org/10.1002/adfm.201807951.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Sulfide Electroltyes

567

108. Hayashi, A.; Hama, S.; Minami, T.; Tatsumisago, M. Formation of Superionic Crystals from Mechanically Milled Li2S–P2S5 Glasses. Electrochem. Commun. 2003, 5 (2), 111–114. https://doi.org/10.1016/S1388-2481(02)00555-6. 109. Ujiie, S.; Hayashi, A.; Tatsumisago, M. Preparation and Ionic Conductivity of (100−x)(0.8Li2S0.2P2S5)xLiI Glass–ceramic Electrolytes. J. Solid State Electrochem. 2012, 17 (3), 675–680. https://doi.org/10.1007/s10008-012-1900-7. 110. Suzuki, K.; Sakuma, M.; Hori, S.; Nakazawa, T.; Nagao, M.; Yonemura, M.; Hirayama, M.; Kanno, R. Synthesis, Structure, and Electrochemical Properties of Crystalline Li–P–S–O Solid Electrolytes: Novel Lithium-conducting Oxysulfides of Li10GeP2S12 Family. Solid State Ion. 2016, 288, 229–234. https://doi.org/10.1016/j.ssi.2016.02.002. 111. Neveu, A.; Pelé, V.; Jordy, C.; Pralong, V. Exploration of Li–P–S–O Composition for Solid-state Electrolyte Materials Discovery. J. Power Sources 2020, 467. https://doi.org/ 10.1016/j.jpowsour.2020.228250. 112. Daikuhara, S.; Satoshi, H.; Suzuki, K.; Hirayama, M.; Kanno, R. Abstract of the 43rd Solid State Ionics Japan, 2017. 113. Liang, J.; Chen, N.; Li, X.; Li, X.; Adair, K. R.; Li, J.; Wang, C.; Yu, C.; Norouzi Banis, M.; Zhang, L.; et al. Li10Ge(P1–xSbx)2S12 Lithium-ion Conductors with Enhanced Atmospheric Stability. Chem. Mater. 2020, 32 (6), 2664–2672. https://doi.org/10.1021/acs.chemmater.9b04764. 114. Favre-Nicolin, V.; Cerny, R. FOX, ‘Free Objects for crystallography’: A Modular Approach to Ab Initio Structure Determination from Powder Diffraction. J. Appl. Cryst. 2002, 35 (6), 734–743. https://doi.org/10.1107/S0021889802015236. 115. Adams, S.; Prasada Rao, R. Structural Requirements for Fast lithium Ion Migration in Li10GeP2S12. J. Mater. Chem. 2012, 22 (16), 7687–7691. https://doi.org/10.1039/ C2JM16688G. 116. Zhang, Z.; Ramos, E.; Lalere, F.; Assoud, A.; Kaup, K.; Hartman, P.; Nazar, L. F. Na11Sn2PS12: A New Solid State Sodium Superionic Conductor. Energ. Environ. Sci. 2018, 11 (1), 87–93. https://doi.org/10.1039/c7ee03083e. 117. Zeng, Y.; Ouyang, B.; Liu, J.; Byeon, Y.-W.; Cai, Z.; Miara, L. J.; Wang, Y.; Ceder, G. High-entropy Mechanism to Boost Ionic Conductivity. Science 2022, 378 (6626), 1320–1324. https://doi.org/10.1126/science.abq1346. 118. Wang, Y.; Richards, W. D.; Ong, S. P.; Miara, L. J.; Kim, J. C.; Mo, Y. F.; Ceder, G. Design Principles for Solid-state Lithium Superionic Conductors. Nat. Mater. 2015, 14 (10), 1026. https://doi.org/10.1038/nmat4369. 119. Yajima, T.; Hinuma, Y.; Hori, S.; Iwasaki, R.; Kanno, R.; Ohhara, T.; Nakao, A.; Munakata, K.; Hiroi, Z. Correlated Li-ion Migration in the Superionic Conductor Li10GeP2S12. J. Mater. Chem. A 2021, 9 (18), 11278–11284. https://doi.org/10.1039/d1ta00552a. 120. He, X. F.; Zhu, Y. Z.; Mo, Y. F. Origin of Fast Ion Diffusion in Super-ionic Conductors. Nat. Commun. 2017, 8, 15893. https://doi.org/10.1038/ncomms15893. 121. Hori, S.; Kanno, R.; Kwon, O.; Kato, Y.; Yamada, T.; Matsuura, M.; Yonemura, M.; Kamiyama, T.; Shibata, K.; Kawakita, Y. Revealing the Ion Dynamics in Li10GeP2S12 by Quasi-elastic Neutron Scattering Measurements. J. Phys. Chem. C 2022, 126 (22), 9518–9527. https://doi.org/10.1021/acs.jpcc.2c01748.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Halides and Oxy-Halide Electrolytes Artur Tron, Palanivel Molaiyan, Marcus Jahn, and Andrea Paolella, Austrian Institute of Technology (AIT), Battery Technologies, Wien, Austria © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 Introduction 2 Lithium halides 3 Lithium oxyhalides 4 Lithium metals halides 5 Interface evolution between solid electrolytes and electrodes 6 Conclusions Acknowledgments References

568 569 570 571 573 574 574 574

Abstract Halide-based ceramics have been considered promising new solid electrolyte (SE) materials for all-solid-state batteries (ASSBs) due to their high chemical and electrochemical stabilities, high Li+ conductivity, and good mechanical properties. In this chapter, recent progress in halide and oxyhalide preparation are summarized. Furthermore, the interface evolution of halide solid electrolytes contacting cathode electrodes and lithium metal anodes is presented.

Key points

• • •

1

A summary of the chemical and physical properties of lithium-based halides and oxyhalides. An overview on battery preparation based on halide and oxyhalide electrolytes. The chapter underlines the importance of interfaces influenced by the possible side reactions occurring at the anode and cathode side.

Introduction

Lithium batteries represent a promising technology for enabling a worldwide green transition to electric vehicles (EVs),1 which can help reduce CO2 emissions from the combustion of gasoline ICEs. A widespread deployment of battery technologies is required without compromising on battery safety, reliability, and electrochemical performance. Lithium metal is considered to be of the best alternative anodes compared to the traditional graphite for the next-generation of rechargeable batteries owing to its high energy density (
4000 mAh g−1 vs
370 mAh g−1). Currently, the use of lithium metal as an anode with a flammable liquid electrolyte2,3 is limited by the formation of lithium dendrites, which may cause battery short circuits and safety issues. The use of a solid electrolyte (SE) represents an alternative since it both acts as a physical barrier to prevent short circuits and increases the lithium metal battery’s energy density.4 In recent years, a wide range of new solid electrolytes based on polymer (solid polymer electrolyte SPE),5 ceramic,6,7 or hybrid polymer–ceramic8,9 materials have been proposed in the literature. More in detail, polymer electrolytes generally exhibit an ionic conductivity of 10–3 S/cm only at temperatures above 60  C.10 Unfortunately, polymer-based electrolytes are not able to inhibit dendrites growth,11 probably due to low transport Li+ numbers and salt inhomogeneities. Among ceramics, garnet lithium lanthanum zirconate Li7La3Zr2O12 (LLZO) shows high ionic conductivity12 but high temperatures are required for synthesis13 and sintering14 in order to densify the primary LLZO particles well. The high resistance between lithium metal and LLZO necessitates the use of an interlayer to improve the lithium metal wettability.15 When LLZO is used as ceramic filler in a polymer electrolyte, the final membranes generally show higher ionic conductivity at room temperature (10−4 S/cm) compared to bare SPE due to amorphization reactions of the polymeric matrix.16 Among oxide-electrolytes, perovskite lithium lanthanum titanate Li0.34La0.56TiO3 (LLTO)17 was explored as SE due to its high ionic conductivity, although it may decompose when in contact with lithium metal.18 NASICON phosphate-based electrolytes (lithium aluminum germanium phosphate Li1.5Al0.5Ge1.5(PO4)3 (LAGP)19 and lithium aluminum titanium phosphate Li1.5Al0.5Ti1.5(PO4)3 (LATP))8 are also unstable toward lithium metal20 while still exhibiting high ionic conductivity at RT.21 Sulfide solid electrolytes like argyrodite lithium chloro-thiophosphate Li6PS5Cl22 and lithium thiophosphates based on Li2SP2S523 are synthesized at low temperatures and show high ionic conductivities. Unfortunately, sulfide materials need a high pressure24 to cycle, and their processability is problematic due to moisture decomposition forming toxic H2S.25 The scientific community is continuously searching for new materials to satisfy the following requirements: (a) low-cost synthesis, (b) low pressure during cycling, (c) low toxicity under

568

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00069-4

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Halides and Oxy-Halide Electrolytes

569

Fig. 1 schematic representing the lithium halides presented in this chapter: lithium halides (LiF, LiCl, LiBr), lithium oxyhalides (Li3OCl, Li2OHCl), and lithium metal halides (Li3InCl6, Li3YCl6).

moisture exposure, (d) a wide electrochemical stability window (in a range of 0 up to 5 V). In this chapter, we present the halide and oxyhalide based materials as solid electrolytes: they represent a new attractive class of electrolyte due to the nature of halogen anions that interact more weakly with lithium ions than with divalent sulfur or oxygen anions. The high polarizability of the ionic bond facilitates lithium-ion transport. Interestingly, some inorganic halide salts having a high ionicity and are stable in dry air, making them a suitable class of materials for ASSB manufacturing. It should be noted that in the last decades, oxide- and sulfide-based electrolytes have been the most attractive for ASSB research; however, due to the critical issues mentioned above, halide electrolytes are more promising candidates for solid-state systems.26,27 Currently, halide solid electrolytes for ASSBs have several advantages compared to the sulfide electrolytes, including: (a) higher stability at high voltage and (b) higher stability under air exposure. In this chapter we will resume the recent progress about (a) lithium halides, (b) lithium oxyhalides and (c) mixed metal halides solid electrolytes as shown in Fig. 1.

2

Lithium halides

In lithium-ion batteries, lithium fluoride (LiF) is a major component of the inorganic layers of solid electrolyte interface (SEI; forms naturally in the wake of natural degradation of organic electrolytes exposed to lithium metal, e.g., LiPF6 in EC-DEC).28 Unfortunately, LiF shows a low ionic conductivity (10−9 S/cm)29 compared to other solid electrolyte materials, but it has attracted much interest due to its high stability toward lithium metal, which facilitates its plating in liquid electrolyte30: Wang et al. were able to prepare an ultrathin LiF/Li3Sb hybrid interface layer after reaction of SbF3 with metallic lithium, showing excellent electrochemical stability.31 Chen et al.32 could deposit an in-situ thin LiF-rich layer by using a thin cellulose acetate membrane with a Lithium bis(trifluormethylsulfonyl) amide (LiTFSI) salt, cycling a Li//LiFePO4 cell at 1 C for over 1000 cycles. LiAlF4 is a solid electrolyte stable at high voltages, widely used to protect lithium-rich cathode materials as shown by Zhao et al.,33 who synthesized a lithiumion-conductive LiAlF4 nanolayer coated onto Li1.2Ni0.2Mn0.6O2 (LNMO) showing a capacity of 246 mAh/g after 100 cycles at 0.1 C. Feinauer et al.34 were able to use Li3AlF6 as solid electrolyte at 100  C with 2% of Al2O3 as inert filler: the Li//Li3AlF6// LiMn2O4 cell showed rapid capacity fading after only 10 cycles. Lithium chloride was explored in a molten salt configuration35,36: a Lithium metal//LLZO electrolyte//AlCl3-LiCl cathode cell (where LiCl becomes a conductive electrolyte at 110  C) showed an energy density of 150 Wh/kg after 100 cycles.37 A mixture of Li3Bi and LiCl in a molar ratio of 2:3 showed a ionic conductivity of
10−4 S/cm as previously reported by Yang et al.38 Unemoto et al.39 proposed a binary system LiBH4 and LiCl with an ionic conductivity of
10−3 S/cm at RT. Galvez-Aranda et al.40 have conducted a modeling study of the interface of Li9N2Cl3 solid electrolyte without detecting any side reactions when the electrolyte is in contact with lithium metal. Li5NCl2 solid electrolyte was synthesized by Landgraf et al.41: the material exhibited an ionic conductivity higher than 10−4 S/cm and an activation energy Ea
0.25 eV after 7Li NMR characterization. Li3P-LiCl is another promising class of solid electrolyte: at 558  C Li3P-LiCl may form an eutectic when they are mixed in a molar ratio of 9:1.42 Lithium chloride can be in combination with lithium sulfide Li2S and phosphorus sulfide P2S5 argyrodite Li6PS5Cl43 system. When LiCl is present in excess, the argyrodite can deliver a better chemical and electrochemical stability toward metallic lithium: the final Li||Li symmetric cells were stable over 1000 h at a current density of 0.2 mA/cm2 as shown by Li et al.44 Xu et al.45 have demonstrated that lithium bromide LiBr shows the lowest melting point (
800 K) and the highest diffusion coefficient (1.5A˚ 2/ps) at the molten state, giving better electrochemical performance than LiCl and LiI. LiBr can be used in combination with poly(butadiene-acrylonitrile): Yaroslavtseva et al.46 observed the existence of metastable behavior leading to LiBr segregation and reaching a maximum ionic conductivity value of
10–4 S/cm (>50  C) at a salt concentration of 1.12 mol/kg for the final heterogeneous PBAN–LiBr electrolyte. Lithium bromide can be also used as an additive in combination with borohydrides and sulfides, forming new compounds such as in 0.6LiBH4–0.4LiBr,47 Li2S-P2S5-LiBr48 and Li6PS5Br.49 Recently, mixed halides doped argyrodites were proposed, e.g., a Li5.3PS4.3ClBr0.7 glass ceramic, which showed the highest ionic conductivity value for this class of solid electrolyte(5.2  10−3 S/cm).50

570

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Halides and Oxy-Halide Electrolytes

Lithium iodide LiI’s ionic conductivity is generally quite low (10−8–10−10 S/cm).51 Its hydrate form has a higher ionic conductivity. Dunkin et al. explored Lithium iodide (3-hydroxypropionitrile)2 (LiI(HPN)2) as solid electrolyte achieving an ionic conductivity of 2.36  10−6 S/cm. Takahashi et al.52 used LiI as an additive in Li3PS4 solid electrolyte synthesis in a molar ratio 30:70. When the final material was annealed at 200  C it showed a better stability toward lithium metal. Raganata et al.53 have proposed a solid electrolyte mixed at a ratio of Li3PO4:LiI:LiCl 0.5:0.25:0.25, achieving an ionic conductivity value of 8  10−4 S/cm at RT, around three orders of magnitudes higher than Li3PO4, which showed an ionic conductivity value of 3  10−7 S/cm.

3

Lithium oxyhalides

Lithium oxychloride Li3OCl is a ceramic with an anti-perovskite structure (LRAP) and interesting properties,54,55 including: (a) low electronic conductivity, (b) high ionic conductivity (>10−3 S/cm at RT), and (c) low activation enthalpy. Unfortunately, the low intrinsic oxidative stability can be the origin of decomposition reactions at high voltages (3.17 V vs Li/Li+).56 Li3OCl shows metastability with respect to the precursors Li2O and LiCl as per the study by Emly et al.57 The mechanical properties of Li3OCl as calculated by Wu et al.58 suggest that Li3OCl is brittle: this aspect makes it less than an ideal SSE for ASSBs. It’s softer than LLTO and LLZO, but stiffer than a sulfide-based electrolyte such as LGPS. Due to its low melting point it has been used as sintering agent to lower the densification temperature of oxide ceramics such as LLZO as reported by Tian et al.59 Li3OCl can show a high stability when in contact with lithium metal as demonstrated by Lu et al.60 in symmetrical cell cycling. A thin Li3OCl layer can reduce the polarization of a 4.5 V Lithium metal//LiNi0.85Co0.1Al0.05O2 (NCA) cell, stabilizing the interface by controlling liquid electrolyte decomposition reactions.61 By using simulations, Wu et al.62 suggested that the lithium diffusion is fast and predominately at interfacial boundaries in agreement with the Lu et al. study63: it’s a low-energy process, yielding a high self-diffusivity of 0.88  10−5 cm2/s and a conductivity of 1.60 S/cm at 300 K. The diffusion can be increased through a regular distribution of LiCl Schottky defects as proposed by Baktash et al.64 Dawson et al.65 were able to calculate the Li-ion migration activation energy: they reported a value of 0.40–0.56 eV and 0.29 eV for grain boundaries and the bulk respectively. Van Duong et al.66 proposed the existence of S(111) grain boundaries to explain the low experimental vs theoretical ionic conductivity value. Zhang et al.67 have shown that the possibility for anti-perovskite to incorporate calcium ions inside the structure forming Li3−xCax/2OCl after mix Li2O:CaO:LiCl. Stegmaier et al.68 have explored what happens at the interface cathode/Li3OCl, observing vacancy accumulation through modeling. Although Li3OBr is more polarizable than Li3OCl,69 it shows a lower ionic conductivity: in simulations,70 the addition of bromide results in higher conductivity Li3OCl1−xBrx structures in a range of 0.235  x  0.395. When LRAP is annealed over 220  C the ionic conductivity value may increase by two orders of magnitude as reported by Zhao et al.71 Clarke et al.72 investigated the role of fluoride ions doping: the authors observed the fluoride ions were able to limit the long-range Li-ions migration in agreement with an observed decrease in Li-ion conductivity with respect to bare Li3OCl sample. Lu et al.73 have deposited a 200 nm thick Li3OCl layer by pulsed laser deposition (PLD) showing an ionic conductivity of 2.0  10−4 S cm−1 at room temperature. The final full cell LiCoO2//graphite with LRAP as electrolyte separator has shown a capacity of 60 mAh/g after 20 cycles at 10 mA/g between 2.2 and 4.2 V. In 2014, Braga et al.74 have reported creating a first modified Li3OCl anti-perovskite by addition of calcium: the final Li3–2⁎0.005Ca0.005ClO showed superionic conduction up to 25 mS/cm at room temperature. Braga et al.75 reported for glass Li2.99Ba0.005OCl1−x(OH)x an activation energy comparable to liquid electrolytes (e.g., 1.6  10−20 J for 1 M LiPF6 in EC-DEC liquid electrolyte). In 2016, Braga et al.76 reported an unusual theory for Li2.99Ba0.005O1+xCl1−2x glass, proposing a lithium metal plating at the cathode side during battery charging in order to explain an infinite discharge superior to battery’s theoretical capacity: as a further example, Braga et al.77 have reported an experimental capacity of 600 mAh/g for a Li//Li[LixNi0.5−yMn1.5−z]O4−x−dFx (F:LNMO) cell by using a mixture of Li2.99Ba0.005Cl1−2xO1+x and a commercial liquid electrolyte. Concerning this aspect, Heenen et al.78 explained the reported low activation energy as a consequence of the possible formation of hydrated LiClxH2O impurities. Thermodynamically it is easier to synthesize the anti-perovskite Li2OHCl than Li3OCl in agreement with computational study of Liu et al.79 Dowson et al.80 underlined the difficulty to prepare a moisture- free Li3OCl antiperovskite due to the high hydration enthalpy for Li3OCl (0.74 eV). The role of protons in antiperovskites is considered a research subject of great interest to the scientific community81: Song et al.82 have suggested that protons H+ enhance the Li+ conductivity of Li2+xOH1−xCl due to an OH− group rotation, creating Frenkel defects near the OH− groups. By using NMR experiments, Song et al.83 determined that only Li+ contributes to long-range ion transport while H+ protons are fixed in an incomplete isotropic rotation of the OH− groups. Ling et al.84 have observed that the H+ content in Li2OHCl increases the energy barrier for Li ion transport due to the existence of repulsive electrostatic forces between H+ and Li+ along the Li–O plane. Hood et al.85 have investigated the reaction occurring between Li2OHCl and lithium metal, showing the formation of H2 gas and Li2O/LiCl mix as protector layer avoiding further decomposition of Li2OHCl. Ye et al.86 have synthesized Li2OHCl at high temperatures: at first, precursors were heated at 320  C, then Li2OHCl powder was pressed at 600 MPa, and then heated to 280  C for 2 h. The final electrolyte showed a high cycling rate capability at 0.8 mA/cm2 in symmetrical Li//Li cells due to the suppression of voids as source of lithium diffusion inhomogeneities. Fig. 2 summarizes the ionic conductivity values for different antiperovskite composition.87 As for new structures, Gao et al.88 observed that partial fluorination of Cl sites induces a conversion of orthorhombic Li2OHCl into cubic Li2OHCl0.9F0.1, leading to a rise in ionic conductivity from 3.7  10−7 S/cm to 1.4  10−6 S/cm. Li et al.89 have synthesized a fluoride doped Li2(OH)0.9F0.1Cl antiperovskite with a measured Li+ ionic conductivity of 3.5  10−5 S/cm at 25  C and a good stability in a lithium metal/LiFePO4 cell with a capacity of 100 mAh/g after 40 cycles at 0.2 C (loading of 5 mg/cm2) at 65  C. Xu et al.90 have synthesized a new anti-perovskite Li6.5OS1.5I1.5 showing an impressive ionic conductivity of 2.28  10−2 mS/cm at 75  C.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Halides and Oxy-Halide Electrolytes

571

Fig. 2 Plot of reported Arrhenius plots for Li antiperovskites. From Zheng, J., Perry, B. & Wu, Y. Antiperovskite Superionic Conductors: A Critical Review. ACS Mater. Au 2021, 1, 92–106. Copyright American Chemical Society.

4

Lithium metals halides

Lithium indium chloride Li3InCl6 solid electrolyte is generally synthesized by a solid-state91 or ball-milling92 process; however Lu et al. synthesized Li3InCl6xH2O by mixing LiCl and InCl3, obtaining an ionic conductivity of 2.04  10−3 S/cm at 25  C after water removal. Subsequently, Luo et al.93 have proposed an interesting ethanol-based solution synthesis followed by vacuum drying: the authors were able to cycle a Li//LiNi0.8Co0.1Mn0.1O2 cell for over 200 cycles at 0.1 C and RT. Li3InCl6 can easily decompose under moisture exposure forming InCl3 and LiCl, while InCl3 further hydrolyzes, forming In2O394 (see Fig. 4a). Fortunately, the hydrated Li3InCl6 intermediate sample can be re-converted into anhydrous high conductive crystal by annealing at 200  C under vacuum as shown by Zheng et al.95 (see Fig. 4b). The dehydration process was explored by Sacci et al.96: neutron powder diffraction (NPD) analysis showed the existence of a three phased transformation process through progressive loss of water. When Li3InCl6 is blended with Li3N and Li6PS5Cl,97 a Li//NMC811 solid-state battery delivered high discharge capacity of
170 mAh/g after 80 cycles at 0.1 C rate. An aqueous solution of Li3InCl698 was dropped on a LAGP-carbon nanotubes cathode and then annealed at 150  C. Li3InCl6 could reduce the interfacial resistance of solid-state lithium air batteries and promote the decomposition of discharge products like Li2O2.98 The addition of fluoride can improve the stability toward moisture as

572

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Halides and Oxy-Halide Electrolytes

demonstrated by Chen et al.99 Kim et al.100 demonstrated that the energy barrier for Li-ion migration decreases near the position of fluoride ions in metastable F-doped Li3InCl6 by using first-principle simulation methods. The authors have also shown the thermodynamic possibility for LiF to segregate by considering the metastable nature of the electrolyte. A dual halide Li3InCl4.8F1.2 reported by Zhang et al.101 exhibited an extended stability window up to 6 V, improving cycling toward LiCoO2 after formation of a LiF rich interface. Li3InCl6 can form a stable solid electrolyte interface (SEI) when it’s in contact with lithium metal: Zhang et al.102 observed the formation of a protective layer composed by LixIny and LiCl allowing a stable cycling in Li//LiFePO4 cell at 1 C for over 400 cycles with a commercial liquid electrolyte. Riegger et al.103 discovered that the interface between Lithium metal and Li3InCl6 (and Li3YCl6 too) is thermodynamically unstable, leading to a continuous growth in resistance, with SEI resistance dominated by LiCl. In addition, Li3InCl6 solid electrolyte suffers of degradation process against NMC622 (
3.7 V vs Li+/LiIn), needing an interlayer protection.104 A temperature-dependent decomposition reaction between the aluminum current collector and Li3InCl6 may occur as well, forming InCl3105 (see Fig. 4c). Interestingly, it’s possible to observe good cycling performance at low temperatures (−10  C) when Li3InCl6 is used as solid electrolyte in an In//LiCoO2 cell.106 Recently, a freeze-dried synthesized Li3InCl6 showed impressive electrochemical performance107: the NCM90 -Li3InCl6/Li3InCl6/Li6PS5Cl/Li cell showed a capacity of 60 mAh/g after 30.000 cycles at a 20 C charge rate. Li3InCl6 has a monoclinic structure (space group: C2/m): indium occupies the Wyckoff 2a (M1) position, while lithium occupies the Wyckoff 4 h (Li1) and Wyckoff 2d (Li2) positions. Interestingly, when Zr4+ ions replace In3+ ions in Li3InCl6, forming Li3−xIn1−xZrxCl6 (0  x(Zr)  0.5), a cation-site disorder on M2/Li4 is promoted, inducing a decreasing volume of the unit cell can be observed with the incorporation of zirconium, which may be due to the difference in ionic radii (r(Zr4+) ¼ 0.72 A˚ < r(In3+) ¼ 0.8 A˚ ).108 Luo et al.109 have reported the synthesis of Li2.9In0.9Zr0.1Cl6 with an ionic conductivity (1.54  10−3 S/cm at 20  C) higher than bare Li3InCl6 (0.88  10−3 S/cm at 20  C), while Van der Maas et al.110 have measured an ionic conductivity of 2  10−3 S/cm for Li2.7In0.7Zr0.3Cl6 solid electrolyte. A partial scandium Sc3+ replacement can form a crystal with a formula Li2Sc1/3In1/3Cl4 having a better chemical stability toward Ni-rich LiNi0.85Co0.1Mn0.05O2 (NCM85) with respect to Li3InCl6 and Li2.5Y0.5Zr0.5Cl6.111 Zhang et al.112 analyzed the heterostructure Li3ScCl6/LiF via first-principles-based density functional theory (DFT): the presence of LiF can limit the dendrites growth problem between Li3ScCl6 as a solid electrolyte and the Li metal anode interface. A zirconium and hafnium substitution of Li3ScCl6 can improve ionic conductivity: Li2.6Sc0.6Zr0.4Cl6 and Li2.6Sc0.6Hf0.4Cl6 exhibited an ionic conductivity of 1.61  10−3 S/cm and 1.33  10−3 S/cm with a low activation energies: 0.326 and 0.323 eV.113 About the processability of a thin film fabrication, Zhao et al.114 have proposed a Li3InCl6//glass-fiber solid composite electrolyte in a Li//LiCoO2 cell, showing a capacity of 80 mAh/g after 20 cycles at 0.1 C. Alternatively, Wang et al.115 have used polytetrafluoroethylene (PTFE) as binder to prepare a 20 mm thick membrane with Li3InCl6 and Li6PS5Cl showing a capacity of
100 mAh/after 50 cycles at 0.1 C and RT, generally better than PMMA or SBR binders.116 The monoclinic lithium zirconium chloride Li2ZrCl6 solid electrolyte can be a low-cost suitable candidate for lithium halide electrolytes,117 although its lithium-ion conductivity is low at room temperature, where it presents an hexagonal close-packed (hcp) anionic structure (5.7  10−6 S/cm at 30  C). After aliovalent indium substitution, the ionic conductivity increases up to 1.08  10−3 S/cm at 30  C due to the formation of a monoclinic structure (space group C2/m) having a cubic close packed (ccp) anionic structure.118 As shown in Fig. 3, Li2ZrCl6 can deliver a capacity of 150 mAh/g after

Fig. 3 (a) The initial charge/discharge curves at 0.1 C, with the Coulombic efficiency ZCoulomb denoted, (b–c) Rate capability at 0.2, 0.33, 0.5, 1 and 2 C. (d) Long-term cycling performance at 0.5 C of Li-In | LPSCl-LZC | NMC811 cell. From Wang, K. et al. A Cost-Effective and Humidity-Tolerant Chloride Solid Electrolyte for lithium Batteries. Nat. Commun. 2021, 12. Copyright Springer Nature.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Halides and Oxy-Halide Electrolytes

573

200 cycles with high voltage NMC811 cathode material at 1 C.119 An aliovalent Fe3+ substitution can effect an improvement as well: the composition Li2.25Zr0.75Fe0.25Cl6 showed an ionic conductivity value of 0.98 S/cm at RT.120 Liang et al. have explored lithium hafnium chloride Li2.73Hf1.09Cl6, observing good cycling stability in an In//NMC811 cell with a capacity of 100 mAh/g after 180 cycles at 0.1 C and RT. Li3HoCl6 is a stable inorganic solid electrolyte in Li//Se cell: Li et al.121 observed a capacity of
400 mAh/g after 700 cycles at 0.1 C with a Coulombic efficiency near 100%. Lithium metal bromides are also of interest122: Shi et al.123 have synthesized holmium-based Li3HoBr6 solid electrolytes with high ionic conductivity (1.25  10−3 S/cm at RT) via the vacuum evaporation assisted method. Li3ErBr6 exhibited an ionic conductivity of 1  10−3 S/cm at RT and a reversible capacity of 720 mAh/g after 300 cycles at 0.2 C in LiIn//SeS2 cells (80% capacity retention).124 Li3TbBr6 displayed an ionic conductivity of 1.7  10−3 S/cm at RT and a capacity of 250 mAh/g after 600 cycles at 0.5 C in a Li-In/Te battery (loading 4 mg/cm2).125

5

Interface evolution between solid electrolytes and electrodes

The formation of a good interface between solid electrolytes and electrodes in SSBs is a difficult challenge that scientists are trying to solve by increasing the total ionic conductivity and limiting the side reactions occurring at lithium metal anode and positive electrodes.126,127 High voltage cathode chemistries are pivotal in SSBs to reaching the high energy densities necessary for large-scale applications (450 Wh/kg128). Compared to halide-based electrolytes, sulfide electrolytes have poor chemical compatibility with high voltage cathodes; therefore, much effort has been directed at improving the interface via LiNbO3 coatings to enable cathode stability for cycle life.129,130 Asano et al.131 have shown pioneering work where the pristine LiCoO2 cathode in contact with halide SE Li3YCl6 exhibited a high coulombic efficiency of 94.8% with low interfacial resistance of 16.8 Ohm cm2—compare to the sulfide-based SSB of 84.0% and a resistance of 128.4 Ohm/cm2. Moreover, Han et al.132 and Park et al.133 demonstrated that halide electrolytes could be stable in contact with high voltage cathodes and have high electrochemical oxidation stability up to 4 V. Liu et al.134 coated NMC523 by Li3InCl6 via a calcination process: NMC523 was added to an ethanol solution containing LiCl and InCl3. The final powder was calcinated at 400  C under nitrogen atmosphere to obtain the final LIC@NCM523. The cathode LIC@NCM523 delivered a capacity of 140 mAh/g after 200 cycles at 1 C charge (loading cathode mass: 2 mg/cm2). Recently Tanaka et al.135 have synthesized the oxychlorides LiNbOCl4 (LNOC) and LiTaOCl4 (LTOC), stable up to 5 V: these materials have shown a high ionic conductivity (>1.0  10−3 S/cm) and a capacity retention of 80% in LiCoO2/LiIn cells in combination with an argyrodite solid electrolyte. However, the other potential high-voltage cathode candidates of LiNi0.5Mn1.5O2 and LiMePO4 (Me–Co, Mn, Ni) with halide solid electrolytes have not been reported on yet. Therefore, development of future halide electrolytes with high ionic conductivities and wide electrochemical oxidation windows is of particular interest in stabilizing their interface, reaching high specific capacities and energy densities, and is considered a potential foray into the new technology of halide SSBs. In fast charge applications, a full- cell136 composed of VCl3-Li3InCl6-C could be charged at 6 C for over 200 cycles, showing a discharge capacity of
100 mAh/g with loading up to 25 mg/cm2. Another noteworthy aspect of the halide SSB is related to the anode interface137: halide electrolytes have an unstable interface in contact with lithium metal anodes due to reduction reactions of metals (e.g., Y, In) that lead to side reaction products causing cell degradation.103–138 Wang et al.139 could improve interface stability by using a buffer layer based on argyrodite Li6PS5Cl between the halide electrolyte and lithium anode (see Fig. 4). The compatibility between sulfide and halide electrolytes is controversial: Riegger et al.103 proposed the use of Li6PSCl5 as suitable solution to stabilize Li-metal interface, while and Koç et al.140 reported the

Fig. 4 (a) Time-resolved EIS spectra of Li/Li3YCl6/Li symmetric cells, (b) Time-resolved EIS spectra of Li/Li6PS5Cl/Li3YCl6/Li6PS5Cl/Li symmetric cells, (c) The symmetric cell performance comparison, (d) The mixed electronic and ionic interface between lithium metal and Li3YCl6, (e) The Li+ -conductive interface between Li3YCl6 and Li enabled by a thin layer of Li6PS5Cl. From Wang, C. et al. A Universal Wet-Chemistry Synthesis of Solid-State Halide Electrolytes for all-Solid-State lithium-Metal Batteries. Sci. Adv. 2021, 7: 1–10 (2021). Copyright Science Magazine).

574

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Halides and Oxy-Halide Electrolytes

existence of a resistant passivation layer. Recently, Tanibata et al.141 have shown that LiFePO4 cathodes exhibited a low discharge capacity of 0.5 mAh/g with an LiAlCl4 solid electrolyte,142 whereas an Li2FeCl4 cathode could show a capacity of 100 mAh/g after 50 cycles at C/20. More research and studies of cathode and anode interfaces is still required to understand the mechanisms of the degradation of the interfaces between electrolyte, cathode, and anode.

6

Conclusions

LIBs currently on the market use highly volatile and flammable organic solvent-based liquid electrolytes; safety is a major concern. The decomposition of the liquid electrolytes in LIBs that causes internal short circuits by the formation of dendrites and the performance were underestimated. Therefore, ASSBs are attractive to researchers and industrial sectors as a next-generation battery that can store and deliver a high amount of energy with high safety and sustainability. The ASSB technology of solid electrolytes is currently one of the methods to prevent the growth of dendrites in LIBs. The crucial components of ASSBs are electrolytes like oxides, polymers, sulfides, and halides. There are several key advantages and challenges: (a) HSEs are gaining more attention due to their high ionic conductivity at ambient temperature, high voltage stability, processibility, and good electrochemical performance. (b) HSEs are easy to synthesize and scalable in an industrial application. More attention should be paid to industrially relevant upscaling of processing technologies. (c) Interfaces in HSEs still a significant challenge to overcome. Introducing new coatings on the cathode and anode would be key to overcoming these challenges in the future. (d) Advanced in situ characterization tools would be more beneficial to understanding fundamental electrochemical reactions of HSEs-based ASSBs, which will help to formulate better chemistries. Undoubtedly, further innovation in HSEs fabrication, processing, and applicability is essential for commercial SSB applications.

Acknowledgments The authors want to thank Austrian Institute of Technology for the support. A.P. thanks Prof. Ashok Vijh of Institut de Recherche d’Hydro-Quebec IREQ (Varennes, Canada) for his useful suggestions.

References 1. Armand, M.; Tarascon, J. M. Building better batteries. Nature 2008, 451, 652–657. 2. Wang, J.; et al. Fire-Extinguishing Organic Electrolytes for Safe Batteries. Nat. Energy 2018, 3, 22–29. 3. Zhao, Q.; Liu, X.; Stalin, S.; Khan, K.; Archer, L. A. Solid-State Polymer Electrolytes with in-Built Fast Itnerfacial Transport for Secondary lithium Batteries. Nat. Energy 2019, 4, 365–373. 4. Cavers, H.; Molaiyan, P.; Abdollahifar, M.; Lassi, U.; Kwade, A. Perspectives on Improving the Safety and Sustainability of High Voltage Lithium-Ion Batteries Through the Electrolyte and Separator Region. Adv. Energy Mater. 2022, 12, 2200147. 5. Zhou, D.; et al. Stable Conversion Chemistry-Based Lithium Metal Batteries Enabled by Hierarchical Multifunctional Polymer Electrolytes with Near-Single Ion Conduction. Angew. Chem. Int. Ed. 2019, 58, 6001–6006. 6. Murugan, R.; Thangadurai, V.; Weppner, W. Fast Lithium Ion Conduction in Garnet-Type Li7La3Zr2O12. Angew. Chem. Int. Ed. 2007, 46, 7778–7781. 7. Campanella, D.; Belanger, D.; Paolella, A. Beyond Garnets, Phosphates and Phosphosulfides Solid Electrolytes: New Ceramic Perspectives for all Solid lithium Metal Batteries. J. Power Sources 2021, 482, 228949. 8. Lee, W.; et al. Ceramic–Salt Composite Electrolytes from Cold Sintering. Adv. Funct. Mater. 2019, 29. 9. Zhang, Z.; et al. Review on Composite Solid Electrolytes for Solid-State lithium-Ion Batteries. Mater. Today Sustain. 2023, 21. 10. Commarieu, B.; Paolella, A.; Daigle, J.; Zaghib, K. Toward High lithium Conduction in Solid Polymer and Polymer—Ceramic Batteries. Curr. Opin. Electrochem. 2018, 9, 56–63. 11. Golozar, M.; et al. In Situ Scanning Electron Microscopy Detection of Carbide Nature of Dendrites in Li-Polymer Batteries. Nano Lett. 2018, 18, 7583–7589. 12. Murugan, R.; Thangadurai, V.; Weppner, W. Fast Lithium Ion Conduction in Garnet-Type Li7La3Zr2O12. Angew. Chem. Int. Ed. Engl. 2007, 4, 7778–7781. 13. Paolella, A.; et al. Discovering the Influence of Lithium Loss on Garnet Li7La3Zr2O12Electrolyte Phase Stability. ACS Appl. Energy Mater. 2020, 3, 3415–3424. 14. Kubicek, M.; et al. Oxygen Vacancies in Fast Lithium-Ion Conducting Garnets. Chem. Mater. 2017, 29, 7189–7196. 15. Duan, J.; et al. Lithium–Graphite Paste: An Interface Compatible Anode for Solid-State Batteries. Adv. Mater. 2019, 31, 1–7. 16. Zheng, J.; Dang, H.; Feng, X.; Chien, P. H.; Hu, Y. Y. Li-Ion Transport in a Representative Ceramic-Polymer-Plasticizer Composite Electrolyte: Li7La3Zr2O12-Polyethylene Oxide-Tetraethylene Glycol Dimethyl Ether. J. Mater. Chem. A 2017, 5, 18457–18463. 17. Jiang, Z.; et al. Tape-Casting Li0.34La0.56TiO3 Ceramic Electrolyte Films Permit High Energy Density of Lithium-Metal Batteries. Adv. Mater. 2020, 32, 1–7. 18. Han, Y.; et al. Interface Issues of lithium Metal Anode for High-Energy Batteries: Challenges, Strategies, and Perspectives. InfoMat 2021, 3, 155–174. 19. Paolella, A.; et al. Unveiling the Cation Exchange Reaction between the NASICON Li1.5Al0.5Ge1.5(PO4)3Solid Electrolyte and the pyr13TFSI Ionic Liquid. J. Am. Chem. Soc. 2022, 144, 3442–3448. 20. Paolella, A.; et al. Understanding the Reactivity of a Thin Li1.5Al0.5Ge1.5(PO4)3 Solid-State Electrolyte toward Metallic Lithium Anode. Adv. Energy Mater. 2020, 10. 21. Hartmann, P.; et al. Degradation of NASICON-Type Materials in Contact with Lithium Metal: Formation of Mixed Conducting Interphases (MCI) on Solid Electrolytes. J. Phys. Chem. C 2013, 117, 21064–21074. 22. Yu, C.; Zhao, F.; Luo, J.; Zhang, L.; Sun, X. Recent Development of lithium Argyrodite Solid-State Electrolytes for Solid-State Batteries: Synthesis, Structure, Stability and Dynamics. Nano Energy 2021, 83, 105858.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Halides and Oxy-Halide Electrolytes

575

23. Dietrich, C.; et al. Lithium ion Conductivity in Li2S-P2S5 Glasses-Building Units and Local Structure Evolution during the Crystallization of Superionic Conductors Li3PS4, Li7P3S11 and Li4P2S7. J. Mater. Chem. A 2017, 5, 18111–18119. 24. Spencer Jolly, D.; et al. Temperature Dependence of Lithium Anode Voiding in Argyrodite Solid-State Batteries. ACS Appl. Mater. Interfaces 2021, 13, 22708–22716. 25. Lee, D.; et al. Critical Role of Zeolites as H2S Scavengers in Argyrodite Li6PS5Cl Solid Electrolytes for all-Solid-State Batteries. J. Mater. Chem. A 2021, 9, 17311–17316. 26. Kwak, H.; et al. Emerging Halide Superionic Conductors for All-Solid-State Batteries: Design, Synthesis, and Practical Applications. ACS Energy Lett. 2022, 1776–1805. https:// doi.org/10.1021/acsenergylett.2c00438. 27. Jena, A.; et al. Controlling Cell Components to Design High-Voltage All-Solid-State Lithium-Ion Batteries. ChemSusChem 2023,. https://doi.org/10.1002/cssc.202202151. 28. Ko, J.; Yoon, Y. S. Recent Progress in LiF Materials for Safe lithium Metal Anode of Rechargeable Batteries: Is LiF the Key to Commercializing Li Metal Batteries? Ceram. Int. 2019, 45, 30–49. 29. Li, X.; et al. Investigation the Electrochemical Properties of LiCl-LiBr-LiF-Doped Li7La3Zr2O12 Electrolyte for lithium Thermal Batteries. Ionics (Kiel) 2019, 26, 3875–3882. 30. Gao, T.; Wang, B.; Gao, J.; Wang, D. Lithium Fluoride Additive for Inorganic LiAlCl43SO2 Electrolyte toward Stable lithium Metal Anode. Electrochim. Acta 2020, 345. 31. Wang, A.; et al. Stable all-Solid-State lithium Metal Batteries Enabled by Ultrathin LiF/Li3Sb Hybrid Interface Layer. Energy Storage Mater. 2022, 49, 246–254. 32. Chen, M.; et al. Marrying Ester Group with Lithium Salt: Cellulose-Acetate-Enabled LiF-Enriched Interface for Stable Lithium Metal Anodes. Adv. Funct. Mater. 2021, 31, 1–10. 33. Zhao, S.; et al. Aegis of Lithium-Rich Cathode Materials via Heterostructured LiAlF4 Coating for High-Performance Lithium-Ion Batteries. ACS Appl. Mater. Interfaces 2018, 10, 33260–33268. 34. Feinauer, M.; Euchner, H.; Fichtner, M.; Anji Reddy, M. Unlocking the Potential of Fluoride-Based Solid Electrolytes for Solid-State Lithium Batteries. ACS Appl. Energy Mater. 2019, 2, 7196–7203. 35. Fujiwara, S.; Inaba, M.; Tasaka, A. New Molten Salt Systems for High Temperature Molten Salt Batteries: Ternary and Quaternary Molten Salt Systems Based on LiF-LiCl, LiF-LiBr, and LiCl-LiBr. J. Power Sources 2011, 196, 4012–4018. 36. Fujiwara, S.; Inaba, M.; Tasaka, A. New Molten Salt Systems for High-Temperature Molten Salt Batteries: LiF-LiCl-LiBr-Based Quaternary Systems. J. Power Sources 2010, 195, 7691–7700. 37. Lang, J.; et al. A Molten Battery Consisting of Li Metal Anode, AlCl3-LiCl Cathode and Solid Electrolyte. Energy Storage Mater. 2020, 24, 412–416. 38. Yang, J.; Jia, Y.; Yao, Z. Study on the lithium Solid Electrolytes of Li3N-Li3Bi-LiCl Ternary System - 2Li3Bi 3LiCl lithium Solid Electrolyte. Solid State Ion. 1997, 96, 215–218. 39. Unemoto, A.; et al. Pseudo-Binary Electrolyte, LiBH4-LiCl, for Bulk-Type all-Solid-State lithium-Sulfur Battery. Nanotechnology 2015, 26. 40. Galvez-Aranda, D. E.; Seminario, J. M. Ab Initio Study of the Interface of the Solid-State Electrolyte Li 9 N 2 Cl 3 with a Li-Metal Electrode. J. Electrochem. Soc. 2019, 166, A2048–A2057. 41. Landgraf, V.; et al. Li5NCl2: A Fully-Reduced, Highly-Disordered Nitride-Halide Electrolyte for Solid-State Batteries with Lithium-Metal Anodes. ACS Appl. Energy Mater. 2023, 6, 1661–1672. 42. Nazri, G. A.; Conell, R. A.; Julien, C. Preparation and Physical Properties of lithium Phosphide-lithium Chloride, A Solid Electrolyte for Solid State lithium Batteries. Solid State Ion. 1996, 86–88, 99–105. 43. Wang, S.; et al. High-Conductivity Argyrodite Li 6 PS 5 Cl Solid Electrolytes Prepared via Optimized Sintering Processes for All-Solid-State Lithium-Sulfur Batteries. ACS Appl. Mater. Interfaces 2018, 10, 42279–42285. 44. Li, Y.; et al. Interface Stability of LiCl-Rich Argyrodite Li6PS5Cl with Propylene Carbonate Boosts High-Performance lithium Batteries. Electrochim. Acta 2020, 363, 137128. 45. Xu, L.; et al. Electronic, Thermodynamic and Ion Transport Properties of the LiCl, LiBr and LiF Electrolytes of Liquid Metal Batteries. Chem. Phys. Lett. 2020, 744. 46. Yaroslavtseva, T. V.; et al. Solid Polymer Electrolytes in a Poly(Butadiene-Acrylonitrile)–LiBr System. Ionics (Kiel) 2017, 23, 3347–3363. 47. Gulino, V.; Dematteis, E. M.; Corno, M.; Palumbo, M.; Baricco, M. Theoretical and Experimental Studies of LiBH4-LiBr Phase Diagram. ACS Appl. Energy Mater. 2021, 4, 7327–7337. 48. Wang, Y.; et al. X-Ray Photoelectron Spectroscopy for Sulfide Glass Electrolytes in the Systems Li2S-P2S5 and Li2S-P2S5-LiBr. J. Ceram. Soc. Japan 2016, 124, 597–601. 49. Hanghofer, I.; et al. Substitutional Disorder: Structure and Ion Dynamics of the Argyrodites Li6PS5Cl, Li6PS5Br and Li6PS5I. Phys. Chem. Chem. Phys. 2019, 21, 8489–8507. 50. Cronau, M.; Szabo, M.; Roling, B. Single-Step Ball Milling Synthesis of Highly Li+Conductive Li5.3PS4.3ClBr0.7glass Ceramic Electrolyte Enables Low-Impedance All-Solid-State Batteries. Mater. Adv. 2021, 2, 7842–7845. 51. Dunkin, M. R.; et al. Improved Ionic Conductivity and Battery Function in a lithium Iodide Solid Electrolyte Via Particle Size Modification. Electrochim. Acta 2021, 388, 138569. 52. Takahashi, M.; et al. Investigation of the Suppression of Dendritic lithium Growth With a Lithium-Iodide-Containing Solid Electrolyte. Chem. Mater. 2021, 33, 4907–4914. 53. Paksi Raganata, G.; et al. Synthesize & Characterization of Li3PO4(0.5)LiI(0.25)LiCl(0,25)Solid Electrolyte for Lithium Ion Battery. IOP Conf. Ser. Mater. Sci. Eng. 2020, 924, 0–6. 54. Mouta, R.; Melo, M.Á. B.; Diniz, E. M.; Paschoal, C. W. A. Concentration of Charge Carriers, Migration, and Stability in Li3OCl Solid Electrolytes. Chem. Mater. 2014, 26, 7137–7144. 55. Fabian, M.; et al. Structural Characterization of Oxyhalide Materials for Solid-State Batteries. Phys. Status Solidi Appl. Mater. Sci. 2021, 218, 1–9. 56. Schwietert, T. K.; Vasileiadis, A.; Wagemaker, M. First-Principles Prediction of the Electrochemical Stability and Reaction Mechanisms of Solid-State Electrolytes. JACS Au 2021, 1, 1488–1496. 57. Emly, A.; Kioupakis, E.; Van Ven, A.; Der. Phase Stability and Transport Mechanisms in Antiperovskite Li3OCl and Li3OBr Superionic Conductors. Chem. Mater. 2013, 25, 4663–4670. 58. Wu, M.; Xu, B.; Lei, X.; Huang, K.; Ouyang, C. Bulk Properties and Transport Mechanisms of a Solid State Antiperovskite Li-Ion Conductor Li3OCl: Insights from First Principles Calculations. J. Mater. Chem. A 2018, 6, 1150–1160. 59. Tian, Y.; et al. Li6.75La3Zr1.75Ta0.25O12@Amorphous Li3OCl Composite Electrolyte for Solid State lithium-Metal Batteries. Energy Storage Mater. 2018, 14, 49–57. 60. Lü, X.; et al. Li-Rich Anti-Perovskite Li3OCl Films with Enhanced Ionic Conductivity. Chem. Commun. 2014, 50, 11520–11522. 61. Yan, C.; et al. 4.5 V High-Voltage Rechargeable Batteries Enabled by the Reduction of Polarization on the Lithium Metal Anode. Angew. Chem. 2019, 131, 15379–15382. 62. Wu, M. S.; Xu, B.; Luo, W. W.; Sun, B. Z.; Ouyang, C. Y. Interfacial Properties and Li-Ion Dynamics Between Li3OCl Solid Electrolyte and Li Metal Anode for All Solid State Li Metal Batteries from First Principles Study. Electrochim. Acta 2020, 334. 63. Lu, Z.; Chen, C.; Baiyee, Z. M.; Chen, X.; Ciucci, C. N.; F. Defect Chemistry and lithium Transport in Li3OCl Anti-Perovskite Superionic Conductor. Phys. Chem. Chem. Phys. 2015, 17, 32547–32555. 64. Baktash, A.; Demir, B.; Yuan, Q.; Searles, D. J. Effect of Defects and Defect Distribution on Li-Diffusion and Elastic Properties of Anti-Perovskite Li3OCl Solid Electrolyte. Energy Storage Mater. 2021, 41, 614–622. 65. Dawson, J. A.; Canepa, P.; Famprikis, T.; Masquelier, C.; Islam, M. S. Atomic-Scale Influence of Grain Boundaries on Li-Ion Conduction in Solid Electrolytes for All-Solid-State Batteries. J. Am. Chem. Soc. 2018, 140, 362–368. 66. Van Duong, L.; Nguyen, M. T.; Zulueta, Y. A. Unravelling the Alkali Transport Properties in Nanocrystalline A3OX (A ¼ Li, Na, X ¼ Cl, Br) Solid State Electrolytes. A theoretical prediction. RSC Adv. 2022, 12, 20029–20036. 67. Zhang, J.; et al. High Pressure-High Temperature Synthesis of lithium-Rich Li3O(Cl, Br) and Li3—XCax/2OCl Anti-Perovskite Halides. Inorg. Chem. Commun. 2014, 48, 140–143. 68. Stegmaier, S.; Voss, J.; Reuter, K.; Luntz, A. C. Li+ Defects in a Solid-State Li Ion Battery: Theoretical Insights with a Li3OCl Electrolyte. Chem. Mater. 2017, 29, 4330–4340. 69. Chen, R.; et al. Influence of Structural Distortion and Lattice Dynamics on Li-Ion Diffusion in Li3OCl1- xBr xSuperionic Conductors. ACS Appl. Energy Mater. 2021, 4, 2107–2114.

576

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Halides and Oxy-Halide Electrolytes

70. Deng, Z.; Radhakrishnan, B.; Ong, S. P. Rational Composition Optimization of the lithium-Rich Li3OCl1- xBrx Anti-Perovskite Superionic Conductors. Chem. Mater. 2015, 27, 3749–3755. 71. Zhao, Y.; Daemen, L. L. Superionic Conductivity in lithium-Rich Anti-Perovskites. J. Am. Chem. Soc. 2012, 134, 15042–15047. 72. Clarke, M. J.; Dawson, J. A.; Mays, T. J.; Islam, M. S. Atomistic Insights into the Effects of Doping and Vacancy Clustering on Li-Ion Conduction in the li3ocl Antiperovskite Solid Electrolyte. ACS Appl. Energy Mater. 2021, 4, 5094–5100. 73. Lü, X.; et al. Antiperovskite Li 3 OCl Superionic Conductor Films for Solid-State Li-Ion Batteries. Adv. Sci. 2016, 6, 1500359. 74. Braga, M. H.; Ferreira, J. A.; Stockhausen, V.; Oliveira, J. E.; El-Azab, A. Novel Li3ClO Based Glasses with Superionic Properties for lithium Batteries. J. Mater. Chem. A 2014, 2, 5470–5480. 75. Helena Braga, M.; Murchinson, A.; Ferreira, J.; Singh, P.; Goodenoguh, J. B. Glass-Amorphous Alkali-Ion Solid Electrolytes and their Performance in Symmetrical Cells. Energ. Environ. Sci. 2016, 9, 948–954. 76. Braga, M. H.; Grundish, N. S.; Murchison, A. J.; Goodenough, J. B. Alternative Strategy for a Safe Rechargeable Battery. Energ. Environ. Sci. 2017, 10, 331–336. 77. Braga, M. H.; M Subramaniyam, C.; Murchison, A. J.; Goodenough, J. B. Nontraditional, Safe, High Voltage Rechargeable Cells of Long Cycle Life. J. Am. Chem. Soc. 2018, 140, 6343–6352. 78. Heenen, H. H.; Voss, J.; Scheurer, C.; Reuter, K.; Luntz, A. C. Multi-Ion Conduction in Li 3 OCl Glass Electrolytes. J. Phys. Chem. Lett. 2019, 10, 2264–2269. 79. Liu, B.; et al. Computational Insights into the Ionic Transport Mechanism and Interfacial Stability of the Li2OHCl Solid-State Electrolyte. J. Mater. 2022, 8, 59–67. 80. Dawson, J. A.; et al. Elucidating lithium-Ion and Proton Dynamics in Anti-Perovskite Solid Electrolytes. Energ. Environ. Sci. 2018, 11, 2993–3002. 81. Guan, C.; et al. Enhanced Ionic Conductivity of Protonated Antiperovskites Via Tuning Lattice and Rotational Dynamics. J. Mater. Chem. A 2023,. https://doi.org/10.1039/ d2ta08307h. 82. Song, A. Y.; et al. Protons Enhance Conductivities in Lithium Halide Hydroxide/Lithium Oxyhalide Solid Electrolytes by Forming Rotating Hydroxy Groups. Adv. Energy Mater. 2018, 8. 83. Song, A. Y.; et al. Understanding Li-Ion Dynamics in Lithium Hydroxychloride (Li2OHCl) Solid State Electrolyte Via Addressing the Role of Protons. Adv. Energy Mater. 2020, 10. 84. Ling, S.; et al. Revisiting the Role of Hydrogen in Lithium-Rich Antiperovskite Solid Electrolytes: New Insight in Lithium Ion and Hydrogen Dynamics. Adv. Energy Mater. 2023, 13, 1–8. 85. Hood, Z. D.; Wang, H.; Samuthira Pandian, A.; Keum, J. K.; Liang, C. Li2OHCl Crystalline Electrolyte for Stable Metallic Lithium Anodes. J. Am. Chem. Soc. 2016, 138, 1768–1771. 86. Ye, Y.; et al. Elucidating the Suppression of lithium Dendrite Growth with a Void-Reduced Anti-Perovskite Solid-State Electrolyte Pellet for Stable lithium Metal Anodes. J. Energy Chem. 2023, 77, 62–69. 87. Zheng, J.; Perry, B.; Wu, Y. Antiperovskite Superionic Conductors: A Critical Review. ACS Mater. Au 2021, 1, 92–106. 88. Gao, L.; et al. Effects of Fluorination on Crystal Structure and Electrochemical Performance of Antiperovskite Solid Electrolytes. J. Energy Chem. 2023, 77, 521–528. 89. Li, Y.; et al. Fluorine-Doped Antiperovskite Electrolyte for All-Solid-State Lithium-Ion Batteries. Angew. Chem. Int. Ed. 2016, 55, 9965–9968. 90. Xu, H.; et al. Lithium Ion Conductivity in Double Antiperovskite Li6.5OS1.5I1.5: Alloying and Boundary Effects. ACS Appl. Energy Mater. 2019, 2, 6288–6294. 91. Li, X.; et al. Air-Stable Li3InCl6 Electrolyte with High Voltage Compatibility for all-Solid-State Batteries. Energ. Environ. Sci. 2019, 12, 2665–2671. 92. Molaiyan, P.; et al. Investigation of the Structure and Ionic Conductivity of a Li 3 InCl 6 Modified by Dry Room Annealing for Solid-State Li-Ion Battery Applications. Mater. Des. 2023, 227, 111690. 93. Luo, X.; et al. A Novel Ethanol-Mediated Synthesis of Superionic Halide Electrolytes for High-Voltage All-Solid-State Lithium-Metal Batteries. ACS Appl. Mater. Interfaces 2022, 14, 29844–29855. 94. Wang, S.; et al. Air Sensitivity and Degradation Evolution of Halide Solid State Electrolytes upon Exposure. Adv. Funct. Mater. 2022, 32, 1–8. 95. Zheng, X.; et al. Li3InCl6-Coated LiCoO2for High-Performance All Solid-State Batteries. Appl. Phys. Lett. 2022, 121. 96. Sacci, R. L.; Benneta, T. H.; Drews, A. R.; Venkataramani Anandan, M. J. K.; Daemen, L. L.; Nanda, J. Phase Evolution during lithium-Indium Halide Superionic Conductor Dehydration. J. Mater. Chem. A 2021, 9, 990–996. 97. Subramanian, Y.; Rajagopal, R.; Ryu, K. S. Blending a Li3N/Li3YCl6 Solid Electrolyte With Li6PS5Cl Argyrodite Structure to Improve Interface Stability and Electrochemical Performance in Lithium Solid-State Batteries. J. Alloys Compd. 2023, 940, 168867. 98. Zhao, C.; et al. Halide-Based Solid-State Electrolyte as an Interfacial Modifier for High Performance Solid-State Li–O2 Batteries. Nano Energy 2020, 75, 105036. 99. Chen, X.; et al. Improved Stability against Moisture and Lithium Metal by Doping F into Li3InCl6. J. Power Sources 2022, 545, 231939. 100. Kim, Y.; Choi, S. Investigation of the Effect of F-Doping on the Solid-Electrolyte Property of Li3InCl6. J. Power Sources 2023, 567, 232962. 101. Liang, J.; et al. A Series of Ternary Metal Chloride Superionic Conductors for High-Performance all-Solid-State Lithium Batteries. Adv. Energy Mater. 2022, 12. 102. Zhang, Y.; Sun, C. Composite Lithium Protective Layer Formed in Situ for Stable Lithium Metal Batteries. ACS Appl. Mater. Interfaces 2021,. https://doi.org/10.1021/ acsami.1c00745. 103. Riegger, L. M.; Schlem, R.; Sann, J.; Zeier, W. G.; Janek, J. Lithium-Metal Anode Instability of the Superionic Halide Solid Electrolytes and the Implications for Solid-State Batteries. Angew. Chem. Int. Ed. 2021, 60, 6718–6723. 104. Koç, T.; et al. Toward Optimization of the Chemical/Electrochemical Compatibility of Halide Solid Electrolytes in All-Solid-State Batteries. ACS Energy Lett. 2022, 2979–2987. https://doi.org/10.1021/acsenergylett.2c01668. 105. Deng, S.; et al. Fast-Charging Halide-Based All-Solid-State Batteries by Manipulation of Current Collector Interface. Adv. Funct. Mater. 2022, 32, 1–10. 106. Deng, S.; et al. Regulating Electronic Conductivity at Cathode Interface for Low-Temperature Halide-Based All-Solid-State Batteries. Adv. Funct. Mater. 2022, 2205594, 1–10. 107. Ma, T.; et al. High-Areal-Capacity and Long-Cycle-Life all-Solid-State Battery Enabled by Freeze Drying Technology. Energ. Environ. Sci. 2023,. https://doi.org/10.1039/ d3ee00420a. 108. Helm, B.; et al. Exploring Aliovalent Substitutions in the Lithium Halide Superionic Conductor Li3- xIn1- xZrxCl6(0  x  0.5). Chem. Mater. 2021, 33, 4773–4782. 109. Luo, X.; et al. Heterovalent Cation Substitution to Enhance the Ionic Conductivity of Halide Electrolytes. ACS Appl. Mater. Interfaces 2021, 13, 47610–47618. 110. van der Maas, E.; et al. Re-Investigating the Structure-Property Relationship of the Solid Electrolytes Li 3−xIn1−xZrxCl6 and the Impact of In-Zr(Iv) Substitution. J. Mater. Chem. A 2023, 11, 4559–4571. 111. Kochetkov, I.; et al. Different Interfacial Reactivity of Lithium Metal Chloride Electrolytes with High Voltage Cathodes Determines Solid-State Battery Performance. Energ. Environ. Sci. 2022,. https://doi.org/10.1039/d2ee00803c. 112. Zhang, C.; Zhuang, H.; Qi, Z.; Liu, X.; Ren, Y. The Effect of Defects for the Ion Transport of Li3ScCl6 and Li3InCl6 with the Interface of Lithium Metal Anode: A First-Principles Study. Mater. Today Commun. 2023, 35, 105764. 113. Zhang, H.; Du, Y.; Shi, X. Doping engineering of Scandium-Based Solid-State Electrolytes Toward Superior Ionic Conductivity. EcoMat 2023, 1–10. https://doi.org/10.1002/ eom2.12315. 114. Zhao, B.; Lu, Y.; Yuan, B.; Wang, Z.; Han, X. Preparation of Free-Standing Li3InCl6 Solid Electrolytes Film with Infiltration-Method Enable Roll-to-Roll Manufacture. Mater. Lett. 2022, 310, 131463. 115. Wang, C.; et al. Solvent-Free Approach for Interweaving Freestanding and Ultrathin Inorganic Solid Electrolyte Membranes. ACS Energy Lett. 2022, 7, 410–416. 116. Wang, K.; et al. Halide Electrolyte Li3InCl6-Based All-Solid-State Lithium Batteries With Slurry-Coated LiNi0.8Co0.1Mn0.1O2 Composite Cathode: Effect of Binders. Front. Mater. 2021, 8, 1–9. 117. Li, F.; et al. Stable All-Solid-State Lithium Metal Batteries Enabled by Machine Learning Simulation Designed Halide Electrolytes. Nano Lett. 2022, 22, 2461–2469. 118. Chen, S.; et al. Enabling Ultrafast lithium-Ion Conductivity of Li2ZrCl6 by Indium Doping. Chin. Chem. Lett. 2022, 33, 4635–4639.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Halides and Oxy-Halide Electrolytes 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142.

577

Wang, K.; et al. A Cost-Effective and Humidity-Tolerant Chloride Solid Electrolyte for lithium Batteries. Nat. Commun. 2021, 12. Kwak, H.; et al. New Cost-Effective Halide Solid Electrolytes for All-Solid-State Batteries: Mechanochemically Prepared Fe3+-Substituted Li2ZrCl6. Adv. Energy Mater. 2021, 11. Li, X.; et al. Highly Stable Halide-Electrolyte-Based All-Solid-State Li–Se Batteries. Adv. Mater. 2022, 34. Wang, S.; Liu, Y.; Mo, Y. Frustration in Super-Ionic Conductors Unraveled by the Density of Atomistic States. Angew. Chem. 2023, 202215544. Shi, X.; et al. Gram-Scale Synthesis of Nanosized Li3HoBr6 Solid Electrolyte for All-Solid-State Li-Se Battery. Small Methods 2021, 5, 1–9. Shi, X.; et al. Encapsulating and Operating a Stable Li3ErBr6-Based Solid Li–SeS2 Battery at Room Temperature. Adv. Funct. Mater. 2023,. https://doi.org/10.1002/adfm.202213638. Zeng, Z.; Shi, X.; Sun, M.; et al. Stable All-Solid-State Li-Te Battery with Li3TbBr6 Superionic Conductor. Nano Res 2023. https://doi.org/10.1007/s12274-023-5559-4. Nikodimos, Y.; Su, W. N.; Hwang, B. J. Halide Solid-State Electrolytes: Stability and Application for High Voltage All-Solid-State Li Batteries. Adv. Energy Mater. 2023, 13, 1–36. Yang, K.; et al. Determining the Role of Ion Transport Throughput in Solid-State Lithium Batteries. Angew. Chem. 2023, 202302586. Liu, L.; et al. Practical Evaluation of Energy Densities for Sulfide Solid-State Batteries. eTransportation 2019, 1, 100010. Ohta, N.; et al. LiNbO3-Coated LiCoO2 as Cathode Material for all Solid-State lithium Secondary Batteries. Electrochem. Commun. 2007, 9, 1486–1490. Hendriks, T. A.; Lange, M. A.; Kiens, E. M.; Baeumer, C.; Zeier, W. G. Balancing Partial Ionic and Electronic Transport for Optimized Cathode Utilization of High-Voltage LiMn2O4/ Li3InCl6 Solid-State Batteries. Batteries Supercaps 2023,. https://doi.org/10.1002/batt.202200544. Asano, T.; et al. Solid Halide Electrolytes with High Lithium-Ion Conductivity for Application in 4 V Class Bulk-Type All-Solid-State Batteries. Adv. Mater. 2018, 30, 1–7. Han, Y.; et al. Single- or Poly-Crystalline Ni-Rich Layered Cathode, Sulfide or Halide Solid Electrolyte: Which Will Be the Winners for All-Solid-State Batteries? Adv. Energy Mater. 2021, 11. Park, K. H.; et al. High-Voltage Superionic Halide Solid Electrolytes for All-Solid-State Li-Ion Batteries. ACS Energy Lett. 2020, 533–539. https://doi.org/10.1021/acsenergylett.9b02599. Liu, C.; et al. Optimized Layered Ternary LiNi0.5Co0.2Mn0.3O2 Cathode Materials Modified with Ultrathin Li3InCl6 Fast Ion Conductor Layer for lithium-Ion Batteries. J. Power Sources 2023, 566, 232961. Tanaka, Y.; et al. New Oxyhalide Solid Electrolytes with High Lithium Ionic Conductivity >10 mS cm −1 for All-Solid-State Batteries. Angew. Chem. 2023, 135, 1–5. Liang, J.; et al. Halide Layer Cathodes for Compatible and Fast-Charged Halides-Based All-Solid-State Li Metal Batteries. Angew. Chem. 2023, 135, 1–7. Seymour, I. D.; Quérel, E.; Brugge, R. H.; Pesci, F. M.; Aguadero, A. Understanding and Engineering Interfacial Adhesion in Solid-State Batteries with Metallic Anodes. ChemSusChem 2023,. https://doi.org/10.1002/cssc.202202215. Wang, S.; et al. Lithium Chlorides and Bromides as Promising Solid-State Chemistries for Fast Ion Conductors with Good Electrochemical Stability. Angew. Chem. Int. Ed. 2019, 58, 8039–8043. Wang, C.; et al. A Universal Wet-Chemistry Synthesis of Solid-State Halide Electrolytes for all-Solid-State lithium-Metal Batteries. Sci. Adv. 2021, 7, 1–10. Koç, T.; Marchini, F.; Rousse, G.; Dugas, R.; Tarascon, J. M. In Search of the Best Solid Electrolyte-Layered Oxide Pairing for Assembling Practical All-Solid-State Batteries. ACS Appl. Energy Mater. 2021, 4, 13575–13585. Tanibata, N.; Takimoto, S.; Aizu, S.; Takeda, H.; Nakayama, M. Applying the HSAB Design Principle to the 3.5 V-Class all-Solid-State Li-Ion Batteries with a Chloride Electrolyte. J. Mater. Chem. A 2022, 20756–20760. https://doi.org/10.1039/d2ta05152d. Flores-González, N.; et al. Understanding the Effect of Lattice Polarisability on the Electrochemical Properties of lithium Tetrahaloaluminates, LiAlX4 (X ¼ Cl, Br, I ). J. Mater. Chem. A 2022, 10, 13467–13475.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes Akiko Tsurumakia,b, Graziano Di Donatoa,c,d, and Maria Assunta Navarraa,b, aDepartment of Chemistry, Sapienza University of Rome, Rome, Italy; bHydro-Eco Research Center, Sapienza University of Rome, Rome, Italy; cHelmholtz Institute Ulm (HIU), Ulm, Germany; d Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 1.1 1.2 2 2.1 2.2 2.3 3 References

Li metal batteries Physical and chemical stability at Li metal and SE interface Improvement of interface property and stability Li-ion batteries Development of anode materials for ASS LIBs Insertion-type anode Alloying-type anode Conclusion

579 579 580 582 582 583 584 586 586

Abstract All-solid-state (ASS) batteries have been recognized as a next-generation energy storage system. The same anode materials adopted in liquid-electrolyte batteries are generally tested in ASS batteries, such as lithium metal, graphite, Sn, and Si. However, because of the solid of the electrolyte, the control of electrode-electrolyte interphases becomes crucial for high-performance and stable ASS batteries. In this chapter, issues remaining in the state-of-the-art ASS batteries are summarized and strategies for the anode development are discussed from viewpoint of both materials engineering and system optimization.

Key points

• • • •

Structural and volumetric changes of Li metal, formation of the SEI, and decomposition of SEs are crucial aspects for the effective implementation of Li metal anode in ASSBs. For the development of powder-based composite anodes, their composition as well as homogeneity in terms of component distributions, thickness, interfacial contacts are crucial factors. Graphite is a versatile material that acts as the active material but also as the electronic conductor, structural template, or buffer space in combination with other active materials. ASS environment is able to counteract volume changes of alloying-type anodes and lead to stable and long-term cycling of LIBs without complicated structuring of active materials.

Abbreviations ASS EEI LIB LMB OLE SE SEI

all-solid-state electrolyte electrode interphase Li-ion battery Li metal battery organic liquid electrolyte solid-state electrolyte solid electrolyte interphase

ASS batteries are recognized as a next-generation energy storage system exhibiting enhanced safety and energy density compared to traditional LIB systems. Although the active material itself is the same in ASS and conventional battery systems adopting liquid electrolytes, completely different interfacial properties are found due to the solid environment within the battery. The failure related to the EEI can be divided into chemical-electrochemical processes and physical-mechanical processes. The former related to the thermodynamical stability of SEs, and the latter arising from continuous volume change of anode materials, causing an inhomongeneous contact. Unlike OLEs, SEs do not flow into the pores in the anode, thus the contact cannot be improved spontaneously and is rather worsen during cycling by accumulation of stresses. Thus, intensive research has been carried out to design homogeneous and extended contact between SEs and anode materials.

578

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00222-X

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes

1 1.1

579

Li metal batteries Physical and chemical stability at Li metal and SE interface

Lithium has been considered as the holy grail of anode materials by virtue of its lightweight (6.941 g mol−1) and low density (0.534 g cm−3 at room temperature) resulting in a remarkable high theoretical gravimetric capacity (3860 mAh g−1) and volumetric capacity (2061 mAh cm−3), together with a favorable combination with a low reduction potential (E0 ¼ −3.045 V vs. standard hydrogen electrode, SHE). However, the practical use of Li metal anode in ASS batteries requires significant efforts to establish a stable EEI, defined by a combination of physical and chemical processes that govern the compatibility between the Li electrode and the SEs. More specifically, attention needs to be paid on the decomposition of SEs, the formation of a solid electrolyte interphase (SEI), and counteraction of the structural and volumetric changes of lithium, which are crucial for the effective implementation of lithium metal anode in achieving a high reversibility, low resistance, and enhanced stability in ASS batteries. Differing from OLEs wetting the electrodes, the solid nature of both SE and lithium metal does not allow a close and extended contact between these two interphases, leading a point-to-point contact that reduces the effective migration area of Li+ and increases interface impedances.1 In addition to these pre-existing cavities, the lithium stripping occurring upon cycling leads to their expansion and also the formation of new voids. The repetitive stripping/plating of lithium creates significant mechanical stress, resulting in complex chemo-mechanical failures within the SEs.1 Sulfide-type SEs exhibit a higher deformability, compared to oxide-type SEs, and are expected to exhibit a better contact between the SE and electrode. However, as observed through operando synchrotron X-ray tomography conducted by Sun et al., even ductile and malleable sulfide-type SEs suffer from physical disconnection between the SE and Li-based anodes, such as the formation of voids at the interface after discharge.2 Such undesirable morphological instability of the interface causes an increase in charge transfer resistance over the course of cycles, reducing the overall electrochemical performance of the cell. Above mentioned defects also create preferential sites for the formation of metallic lithium nuclei, which promote lithium infiltration into SEs and potentially cause short circuits. This growth mechanism due to the infiltration of metallic lithium into SEs cannot be accurately described by the term “dendrite”. As well summarized by Krauskopf and co-workers in a recent review, metallic lithium infiltrates at various defect points, including (a) pre-existing interconnected pores in SEs, (b) grain-boundary regions, (c) newly formed cracks in polycrystalline or single crystalline SEs, as well as (d) isolated pores in SEs and (e) macroscopic cracks in SE pellets.3 Hence, terms such as “Li filaments” or “Li whiskers” are more suitable for these complex mechanisms and structures of metallic lithium.3 According to the Monroe–Newman theory, the growth of Li whiskers is possible to be hindered by employing a SE having roughly twice the shear modulus of the lithium.4 Although this requirement is met by most inorganic SEs, several works have already demonstrated that even in stiffer SEs, such as Li7La3Zr2O12 (LLZO), the lithium penetration along the bulk of the electrolyte occurs. This discrepancy also highlights the Li filaments growth based on the infiltration mechanism. The current density threshold, above which metal filaments propagate across the SE leading to cell failure, is called critical current density (CCD). CCD is an essential parameter toward practical ASS batteries as it determines the power density of the cell. Because of a high interfacial impedance resulting in a larger iR loss, ASS batteries typically have a CCD lower than 1 mA cm−2, whereas a target value for competitive LMB is 3–10 mA cm−2.5 In the case of ASS batteries, the value of CCD is affected by interfacial roughness, bulk microscopic properties, surface and volume defect distribution, as well as elastic anisotropy, rather than intrinsic ionic conductivity of SEs. In addition to the physical contact failure at the EEI, the thermodynamic instability between SEs and electrodes generates parasitic (electro-)chemical interfacial reaction, accompanying a formation of SEI. The generated SEI layer serves to stabilize SEs by bridging the chemical potential gap between Li metal and SEs, but sometime causes sluggish kinetics of Li+ stripping and plating. Based on the intrinsic properties of different kinds of SEs, it is possible to distinguish three different EEI in ASS batteries (Fig. 1).6 (a) An EEI with a suitable thermodynamic stability and ability of Li+ ion transfer: this is the ideal scenario, and SEs are protected and do not react even in contact with Li metal. Such a “thermodynamically stable interface” is highly desirable for ASS batteries. (b) An EEI without an ability to protect SEs from decompositions: this is formed when there is a thermodynamic driving force for a chemical reaction between the two constituents of the interface. When the formed EEI possess both electronic and ionic conductivities, so-called “mixed conducting interphase” (MCI), degradation of SEs continues. The progression of this degradation front continuously alters the inherent properties of EEI, consumes Li+ undesirably, and eventually leads to cell

Fig. 1 Three types of interphases between lithium metal and a SE.6

580

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes

failure. This behavior is typical of the SEs with Ti4+ and Ge4+, which are less tolerant to reduction, such as NASICON-like Li1+xAlxGe2-x(PO4)3 (LAGP) and Li1+xAlxTi2-x(PO4)3 (LATP), as well as Li10GeP2S12 (LGPS) and perovskite-type Li3xLa2/3xTiO3 (LLTO). A formation of MCI and continuous decompositions are visible by combining in-situ X-ray photoelectron spectroscopy (XPS), AC impedance spectroscopy (for ionic conductivity), and DC polarization measurements (for electronic conductivity). Hartmann et al. suggested that Ti4+ and Ge4+ in the SEs are not suited as their constituents for their use with Li because after exposure to lithium, they undergo to a partial (Ti4+ to Ti3+ and Ge4+ to Ge2+) or even a complete (Ge4+ to Ge0) reduction.7 The formation of metallic species with electronic conductivity on the EEI unfavorably promotes the degradation process. Similarly, the decomposition products of LGPS are known to contain lithium-germanium alloy (Li-Ge), an electron conducting phase, in addition to the formation of lithium sulfide (Li2S) and lithium phosphide (Li3P).8 (c) An EEI with poor thermodynamic stability but with a low electronic conductivity sufficient to passivate the EEI: this layer is formed when the decomposition compounds form an electronically insulating (but still ionically conductive) layer at the interface between SE and Li metal, resulting in a “stable SEI” that suppresses electric transfer to the SE and protect them from continuous decompositions. This interphase layer is typically found in LLZO, LixPOyNz (LiPON) and Li7P3S11. LLZO is known as one of the SEs that can be combined with Li metal anodes. Upon contact with Li metal, LLZO undergoes reduction decomposition, which is accompanied by the simultaneous implantation of Li+, resulting in a tetragonal-like LLZO interphase that suppresses further interfacial reactions without compromising the ionic conductivity.9 Similarly, the EEI formed at Li|LiPON is known to be passivating, and in this case, the layer is based on nitrogen containing alkali phosphate glasses, specifically Li3PO4, Li3P, Li3N and Li2O.10 Wenzel et al. carried out a comparative study on the growth rate of the interphase generated between Li metal and argyrodite-type sulfide SEs (Li6PS5X, LPSX with X ¼ Cl, Br, and I) and compared these results with those of LGPS and Li7P3S11.11 From XPS analyses, a series of LPSX was found to form Li2S and Li3P form together with LiX, in contact with Li metal. From time-resolved impedance spectroscopy, chloride and bromide-containing LPSX was found to exhibit slow SEI formation rate comparable to Li7P3S11 and much lower than LGPS and LPSI.11

1.2

Improvement of interface property and stability

All the aforementioned complex processes of EEI formation affect and limit the overall performance of ASS batteries adopting lithium metal anodes, thus creating an obstacle to the practical use of ASS batteries. To date, the strategy toward a feasible ASSB consists of structural and/or chemical modifications of both the bulk materials and the interface, specifically (i) applying appropriate internal and external pressures to the battery, (ii) replacing lithium metal with lithium alloys, (iii) employing composite electrolytes, and (iv) forming protective interlayers. During an assembly of bulk-type ASS batteries, pressure is applied in two different stages: a high “fabrication” pressure (hundreds of MPa to several GPa) applied internally to the cell components and a low “stack” pressure (usually less than 10 MPa) applied externally to the whole cell. For the assembly of ASS batteries, SE powders are pressed under a high pressure to yield pellets. The pressure applied for a pellet directly affects the porosity and ionic conductivity of the SE. With respect to LiPS4, the porosity is about 80% when the pellet is prepared at 200 MPa, which is improved to 99.9% at a fabrication pressure of 700 MPa.12 As the LiPS4 density increases, its ionic conductivity increases linearly and the initial overpotential of the Li||Li symmetric cell decreases when cycling at a current density of 0.2 mA cm−2. The Li filament growth in pores within the SE was able to be suppressed when the pores become isolated and small in high density pellets above the critical relative density (>95%). The pressure to combine a SE pellet and Li metal foil also needs to be adjusted carefully from multiple aspects including pressure value, duration, and temperature, to reduce Li deformation.13 After preparation of cell components and their assembly, a lower “stack” pressure is externally applied and kept constant to ensure the contact at the EEI and also to counteract the volume change of electrode materials during charge and discharge, which are important parameters to enhance CCD. During battery cycling, Li metal anodes undergo the volume changes, and these becomes more pronounced in a high capacity cell, i.e., a higher cathode loading. When the stack pressure is applied by fixing the gap between cathode and anode, the volume expansion and shrinkage of the cell cannot be compensated, resulting in a lower CCD. In contrast to this, when a constant stack pressure is applied in response to the volume change of the cell, a higher CCD can be achieved.13 The value of applying pressure must be carefully chosen, taking into account the different physical characteristics of the selected SE and electrode. In particular, a certain pressure is necessary to reinforce Li|SE contact, but an excess pressure causes deformation of Li metal, owing to its high ductility, thus triggering short circuit of the cell. Lithium alloys provide an attractive advantage to construct a stable EEI enabling long-term cycling of ASS batteries. To date, several Li-based alloys have been demonstrated to be an ideal alternative for pure Li metal anode, such as Li-In, Li-Al, Li-Si, Li-Ge, and Li-Mg. Among various lithium alloys, Li-In alloys are one of the most utilized ones because of their mechanical ductility and constant redox potential (0.62 V vs. Li+/Li) over a wide stoichiometry range.14 The possibility of dendrite growth is not completely eliminated even in the case of Li-In anodes, especially when cells are cycled at high currents and loads, but unlike Li dendrite growth that penetrates the SE vertically, Li-In filaments grow laterally in the SE by forming a network, alleviating structural damage to the SE (Fig. 2).14 The Li-based alloy anodes also show an improved physical contact with the SE. LLZO is one of the SEs that can be combined with Li because of its cathodic stability. However, it is lithiophobic and exhibits poor wettability with Li metal. A thin layer formation of Al, Si, and Ge on Li metal enhances the contact with LLZO and Li metal anode, reducing the interface impedance, facilitating the charge transfer across the interface, and remarkably improving the performance of cells.15 An alloy based on Li-Mg is not only compatible with LLZO but also enables Li stripping and plating within the alloy framework to form Li-deficient and Li-rich alloys, minimizing the volume changes that occur during cycling.16

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes

581

Fig. 2 Schematic illustration of filament growth of Li and Li-In anodes at the interface with the Li6PS5Cl (LPSCl) SE.14

Regarding composite electrolytes or protective interlayers, these approaches combine inorganic SE with organic materials. Over the past few decades, different composite electrolytes with polymers have been proposed, and according to the content ratio of inorganic SE and organic polymer, they can be divided into two categories: (i) “ceramic-in-polymer (CIP)” and (ii) “polymerin-ceramic (PIC).” Polymer matrices are generally poly(vinylidene fluoride) (PVDF), poly(vinylidene fluoride-co-hexafluoropropylene) (PVDF–HFP), poly(ethylene oxide) (PEO), polyacrylonitrile (PAN). These composite electrolytes are able to combine polymer-derived properties, including good and homogeneous adhesion on Li surface and better flexibility to the mechanical stress, as well as SE-derived features, such as robust mechanical strength against dendrite growth and desirable ionic conductivity. In a systematic study ranging from CIP to PIC with different Li6.4La3Zr1.4Ta0.6O12 (LLZTO) particle sizes, it was found that CIPs with smaller LLZTO particles showed higher ionic conductivities and PICs with larger LLZTO particles exhibited higher mechanical strengths.17 The PICs were favorable for blocking Li filament growth but unfavorable for wettability with Li metal anodes due to less flexibility. The authors also prepared a PIC sandwiched by CIPs to simultaneously achieve dendrite suppression and excellent interfacial contact with Li metal. By employing the sandwich type electrolytes, LiFePO4||Li cells exhibited an excellent cycle performance (99.1 mAh g−1 at 0.1C at room temperature). In addition to the composite electrolytes with polymers, most recently SEs are combined with OLE or ionic liquids (ILs), and they are specifically referred to as quasi-solid-state electrolytes (QSSEs). The resulting QSSEs retain the solid-state nature, i.e., no fluidity, self-standing, and no liquid leakage, whereas the liquid content contributes to solve the problems of poor contact with anode materials and the sluggish interfacial kinetics.8 To completely retain the safety advantages of SEs, the ones with ILs, which are non-volatile and non-flammable, have attracted significant attention. Even when these electrolytes contain flammable OLE, thermal stability is desirable because the amount of OLE is small, and the OLE is trapped in the pores of the SE. For alleviation of the interfacial issues, the use of flexible and protective layer between Li anode and SEs is also promising. Hu et al. developed a composite based on lithium bis(trifluoromethylsulfonyl)imide (LiTFSI), poly(vinylene carbonate) (PVC), and SiO2 nanoparticles, used to stabilize the Li|LAGP interface. While the polymer matrix with LiTFSI ensures high ionic conductivity and low interfacial resistance, the silica particles act as flame retardants as well as a physical barrier to side reactions at the interface. These synergistic effect enables stable Li stripping/plating for more than 1500 h at room temperature.18 Very recently, Wu et al. proposed a novel ultrathin layer (20 mm), composed of bis(fluorosulfonyl)imide-based ILs and PEO, which shows a remarkable ionic conductivity of 1.25  10−3 S cm−1 at room temperature. The presence of this thin interlayer at Li|LAGP interface led to outstanding interface stability for more than 2000 h of continuous Li plating/stripping cycles in symmetric Li|LAGP|Li cells.19 In a similar manner to QSSEs based on SEs and liquid materials, it is also possible to form an interlayer with the liquid materials. In this case, since a liquid is fluid and can fill the cavity of solid materials, it also serves a function as infiltration material. Gao et al. obtained an optimized stabilized interface by in-situ electrochemical reduction of 1.0 M LiTFSI in dioxolane (DOL)/1,2-dimethoxyethane (DME) (1:1, v/v) on Li metal anode. The interface formed by the reduction decomposition products was able to protect LGPS, which enabled a remarkable longer lifespan and lower overpotentials of Li stripping/plating.20 Similar results were obtained also by employing a small amount (ml) of 1.0 M LiPF6 in ethylene carbonate (EC)/dimethyl carbonate (DMC)

582

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes

on the LATP surface.8 Also, ILs can be used as the interlayer. Xiong et al. designed a composite based on LAGP and 1-butyl-3-methylimidazolium bis(fluorosulfonyl)imide ([Bmim][FSI])/LiFSI mixture and utilized this composite as the interlayer.21 This layer suppressed the reduction of Ge4+ and Li filament growth, allowing enhanced long-term cycling stability and a low interfacial resistance. As a result, Li||LiFePO4 cells exhibited ultra-stable cycle performances with specific capacities higher than 110 mAh g−1 at 2C.

2 2.1

Li-ion batteries Development of anode materials for ASS LIBs

The use of an anode that works at a potential slightly higher than that of Li is an effective means to reduce Li dendrite formation and improve battery safety. However, in contrast to Li metal anodes, other anode materials are often porous or pulverized during cycling. Unlike OLEs, SEs cannot spontaneously fill the cavities of electrodes, leaving the contact area between the electrode and the electrolyte non-uniform. This inconsistently filled contact area forms one of the most critical challenges in the development of ASS LIBs, particularly in the use of powder-based electrodes in bulk-type cells at a high mass loading. One possible solution is the preparation of composite electrodes, based on an active material with SE as a binder and carbon as an electrical conductor to effectively enhance the contact area available for lithiation and de-lithiation. For the development and use of these composite electrodes, attention needs to be paid to the following points: (1) (2) (3) (4)

optimization of active material-to-SE ratio to reinforce the formation of lithium-ion conduction pathways; preparation of homogeneous composite materials and high density electrodes; ensuring that the thickness of solid materials is consistent to avoid uneven deposition of lithium and stress distribution; application of adequate pressure to the cell to ensure sufficient contact between materials, thereby inhibiting the formation of voids and cracks during cycling.

These are general practices for the development of ASS LIBs and applicable not only to anode but also cathode composite materials. To achieve the first criterion, a composition study must be carried out for each electrode composite, which generally includes active materials, SE, and carbon-based electrical conductors. Although the addition of carbon can be avoided when graphite is used as an active material, carbon is still often included to improve the electrical conductivity of graphite. The optimal weight composition will vary depending on the density and morphological properties such as the diameter or aspect ratio of the components. The concentration of carbon can be determined based on the percolation threshold concept which is mainly affected by its dimension. The electrical percolation threshold for carbon is at most 15 wt% and decreases as its aspect ratio is increased. This value also changes depending on the density of other components. In ASS LIBs, the amount of carbon conductive additives is generally set to 3–5 wt%.22 In contrast, the amount of SE needs to be above 30 wt% to form successive Li+ conduction pathways.22 While a larger SE fraction in the composite is favorable for Li+ conduction, it is unfavorable for the energy density of batteries because it reduces the relative amount of active material in the composite electrodes. Thus, optimization of active material-to-SE ratio is crucial. According to a compositional study for cathode composite electrodes, a large ratio of cathode-to-SE particle size enables a higher cathode loading.23 By decreasing the diameter of SE, the Li+ conduction pathway becomes narrower but more widely spread in the active material. Although this study was carried out for cathode materials, the same concept should be applicable for anode materials as well. Overall, the composition study for electrodes in ASS LIBs is very complicated because electrodes contain different components, e.g., active material, SE, and a carbon-based conductor, and the performance of the electrode is defined cooperatively by these components. The mixture design approach, in which the component ratio is depicted as a coordinate of a triangular simplex plot, is a useful means to optimize the ratio of 3 components and to obtain the best performance of the composites through regression analyses (Fig. 3).24 This approach also enables to discriminate the effect of each component and the combined effect of the components on the end-performance.

Fig. 3 Triangular simplex plots for composite anodes based on Si (active material), LPSI (SE), and MAG (carbon-based conductor), with capacity performance at the 20th cycle.24

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes

583

For the preparation of electrodes, it has been found that homogeneity is strongly influenced by the choice of mixing procedure.25 Mixing procedures can be divided into wet and dry processes depending on whether a solvent is used. For industrial applications, sheet-type electrodes using polymeric binders are preferred because of their mechanical flexibility. To dissolve or disperse polymeric binders homogeneously, a wet mixing process with an appropriate solvent is used. In the case of sulfide-based SEs which are reactive to common polar solvents, applicable solvents are restricted to apolar ones, such as xylene and toluene, or less-polar ones, such as THF. In addition, the porosity of the electrodes after solvent removal needs to be taken into consideration because the presence of voids and cracks suppresses ionic conductivity, especially in the direction of the depth. Dry mixing, on the other hand, is a more simple, practical, and cost-effective technology, which is also used for the premixing of solid materials prior to wet processes. The resulting material is a powder, which is pressed at room temperature (so-called “cold pressing”) and used as a pellet-shaped electrode. It is possible to mix materials manually using mortar and pestle or, preferably, mechanically using a tumbler ball mill or high-energy ball mill. The milling time and energy need to be determined so as not to affect the structure of the active materials. This technique is utilized more for sulfide-based SEs rather than oxide-based ones because the latter are not amorphous and their grain boundary resistance cannot be eliminated after cold pressing. The formation of homogeneous electrodes as well as a uniform distribution of electrodes on the SE, i.e., even thickness, allow the avoidance of inhomogeneous current distribution and electrode utilization, which lead to undesired side reactions, accelerated degradation, lithium plating, and an uneven mechanical expansion of the anode.22 The effects of electrode heterogeneity are more severely pronounced for thicker electrodes and cycling at higher C-rates. Homogeneity of EEI is also a concern and is one of the motivations for the development of quasi-solid-state batteries that use OLEs, including ionic liquids, as an interlayer between the SE and electrolyte. It is essential that interfacial uniformness be constant throughout the course of cycling, even after the formation of SEI. However, many SEI components such as LiOH, Li2O and Li2CO3 form non-uniform interfaces, resulting in non-uniform current distribution on the anode, causing dendrite formation in areas of highest current density. Finally, the application of adequate pressure to the cell to ensure contact between materials and inhibit the formations of voids and cracks during cycling, also helps to maintain the homogeneity of the materials and their interfaces, especially for those affected by large volume changes such as Si and Sn. In addition, application of high pressure is known to reduce decomposition of SEs and allows the combination of anode materials that work at lower potentials and various SE materials. As can be explained by the isotherm Gibbs–Duhem equation describing the relationship between changes in chemical potential and thermodynamics, a higher pressure induces an increase in chemical potential of materials. In the case of ASS batteries, the effect of pressure is not negligible and mechanical constriction introduces energy barriers that prevent bulk and interfacial decomposition of SSEs.

2.2

Insertion-type anode

Graphite is one of the most commonly used negative electrode for LIBs because of its advantages, such as low production cost, non-toxicity, high stability, and excellent electrical conductivity. It has sp2 hybridized graphene layers that are linked by substantially weak van der Waals forces and p–p interactions of the delocalized electron orbitals. The layered structure enables the intercalation of ionic and molecular species through expansion of the interlayer distance, accompanied by re-stacking of the graphene layers during de-intercalation.26 The maximum lithium content is as much as LiC6, resulting in a theoretical capacity of 372 mAh g−1. Graphite anodes can be considered fundamental materials for the development of ASS LIBs because of their versatility: they are used not only as the active material but also as the electronic conductor, structural template, or buffer space in combination with other active materials, such as lithium, Si, and Sn. Its low lithiation/de-lithiation potential (i.e., between 0.25 and 0.01 V vs. Li+/Li) also makes it a promising anode for ASS LIBs. This advantage, however, comes with a caveat: there is an increased risk of electrolyte decomposition as well as metal plating, especially at higher charging currents. Conventional OLEs exhibit their electrochemical stability limit at approximately 0.8 V vs. Li+/Li; below this potential, the electrolytes undergo reduction decomposition, resulting in the formation of SEI with thickness of 10–50 nm.27 This decomposition continues as long as new and bare graphite surfaces are formed from continuous exfoliation, consuming Li+ and electrolyte solvent, leading to high cell impedance. ASS LIBs, in contrast, presents an entirely different set of interfacial dynamics. Although the intrinsic cathodic electrochemical stability of SEs is still not optimum, unlike OLEs, actual cell performance is not strongly affected by this relatively poor electrochemical stability of SE. This difference is a consequence of irregular contact between solid components and the sluggish decomposition kinetics of SEs. In addition, an external pressure constantly applied to ASS batteries for better material contact also limits the exfoliation of graphite, suppressing the formation of new and reactive surfaces. Thus, direct comparison between OLE and SE batteries demonstrates that there is a far more negligible irreversible capacity in the 1st cycle when SE is used. More specifically, the study was carried out by comparing two electrolytes, 1 M LiPF6 in ethylene carbonate : dimethyl carbonate (LP30) and LiI-Li3PS4 (LPSI), and two different graphite composite electrodes for each electrolyte, one with graphite : poly(vinylidene fluoride) : super P carbon 80 : 10 : 10 and the other with graphite : LPSI : carbon nanofiber 60 : 35 : 05.28 The cells with LP30, a liquid configuration, exhibited a high specific 1st cycle charge capacity (above 300 mAh g−1) at C/10, but were accompanied by a high irreversible capacity of 122.0 mAh g−1. When LPSI was used as the electrolyte, the C-rate had to be decreased to C/50 to achieve the same level of capacity values, but the irreversible capacity was only 4.4 mAh g−1, resulting in 98.5% coulombic efficiency from the 1st cycle. According to cyclic voltammetry recorded on the Li|LPSI|graphite cells, cathodic currents were observed at 1.8 V vs. Li+/Li and intensified at around 0.7 V vs. Li+/Li, which can be attributed to the formation of SEI containing sulfur-based and phosphorous-based components such as Li2S, Li3P and LiI. Although the reduction of SE was confirmed, considering the small 1st cycle irreversible capacity of ASS LIB with the graphite anode, the reduction decomposition of LPSI is very limited owing to the suppressed exfoliation of graphite.

584

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes

To further avoid the decomposition of SEs, coating formation on graphite is also an effective means. Yang et al. used LiI as a coating material, which is stable at the lithium insertion potential, and deposited it directly onto the graphite surface.29 An uncoated graphite sample, examined in Li metal cells with Li3PS4, showed a discharge capacity of 248 mA h g−1 with coulombic efficiency of 90% at 0.13 mA cm−2, while the presence of 5 wt% LiI coating improved these values to 348 mAh g−1 and coulombic efficiency >96%. Impedance spectroscopy recorded during the first charge process exhibited increasing cell resistance in the case of uncoated materials, while it remained constant for the LiI-coated graphite. This can be explained by the thermodynamic stability of LiI at the potential at which Li insertion and de-insertion occur, preventing SEs from decomposing. To confirm their findings, the authors also performed X-ray absorption spectroscopy (XAS) of coated and uncoated graphite before and after the first charge process. After the first cycling, the presence of Li3P and Li2S was confirmed, indicating that the sulfide-based SE underwent reductive decomposition. These results suggest that LiI can suppress decomposition of sulfide SEs, leading to better electrochemical performance. Contact between SE and graphite must be improved for enhancement in the rate capability even though this could lead to a risk of decomposition at the EEI. Kim et al. reported a new scalable fabrication protocol of a graphite anode for ASS LIBs30 in which a homogeneous SE solution, specifically Li6PS5Cl (LPSCl) in ethanol, was infiltrated into the graphite and LiCoO2 (LCO) electrodes and dried at 180  C in an Ar atmosphere. Energy dispersive X-ray spectroscopy (EDXS) of the LPSCl-infiltrated LCO exhibited SEs occupying the spaces between active particles, confirming the excellent penetration of liquefied SE into electrodes. The LPSCl-infiltrated graphite and LCO electrodes exhibited high reversible capacities of 364 mAh g−1 and 141 mAh g−1, respectively, at 0.14 mA cm−2 (equivalent to 0.1C) at 30  C, which are similar to those of LIBs with OLEs. The infiltration of SE liquid dispersion into the electrodes was found to be an efficient means of improving electrochemical performance through the formation of EEI available for electronic and ionic conductions. Although the presence of SE in the anode supports ion conduction, the increase in the amount of inactive material in the electrode inevitably reduces the energy density of ASS LIBs. In place of typical composite graphite anodes using SEs as a binder, diffusion-dependent graphite anodes composed of mostly active materials are more promising in terms of the energy density.31 In contrast to the Li+ migration seen in composite anodes due to the Li+ migration through conduction pathways formed by SEs, Li+ transport in diffusion-dependent graphite anodes occur within the electrode only by diffusion, enabled by seamless interfaces between active materials. The structural difference between conventional and diffusion-dependent graphite can be seen also in the impedance spectroscopy. The diffusion-dependent electrodes are thus able to achieve a very high loading level 15.2 mg cm−2, showing 643 mAh cm−3 at 0.1C at 60  C. By increasing the C rate to 0.2C, the capacity decreased to approximately 440 mAh cm−3. To improve the diffusion process, another cell was cycled at 90  C and in this case, the cell was able to retain a high volumetric capacity above 600 mAh cm−3 even at 0.3C. For oxide-based materials, the anode development becomes complicated due to their high crystallinity. One of the effective methods to produce a composite anode with a reduced interfacial resistance is spark-plasma-sintering process.32 Another effective means is the preparation of a successive SE matrix and then integration with graphite. Oh et al. prepared interconnected Li7La3Zr2O12 (LLZO) networks by using cellulose as a template.33 A slurry based on synthesized LLZO, natural graphite, and poly(vinylidene fluoride) was prepared using N-methyl-2-pyrrolidone as a solvent. A coated and dried composite electrode was examined in Li metal ASS cells using Li6PS5Cl as the electrolyte. The cell achieved a higher capacity of 279 mAh g−1, which is 2.3 times higher than the cell prepared with conventional LLZO at a 1C-rate. In addition to the development of these graphite anodes, various characterizations of graphite anodes have also been reported.34 Optical microscopy (OM) observation is one of the most effective means of tracking the cycling of graphite because the color of graphite changes from black, through red, to gold during lithiation. OM observation is carried out with respect to the lithiated graphite anodes in Li-metal ASS cells by changing the weight composition of the graphite : Li3PS4 x : 100 − x (x ¼ 50, 60 and 70). When x ¼ 50, the color of graphite particles changes to gold, suggesting full lithiation; the gold color was not homogeneous for x ¼ 60 and was not achieved for x ¼ 70. These results suggest that the presence of SE is important in increasing the surface of graphite available for full lithiation. Cross-sectional images taken during operando confocal microscopy have also been reported. From these images, it has been confirmed that there are dynamic changes in the Li+ ion conduction pathway during cycling. After the first charge, most of the graphite particles became gold, but as the cycle number increased, the amount of gold-colored graphite decreased drastically and only the graphite particles close to the SE layer were gold in color. This indicates that graphite near the current collector, far from EEI, was not lithiated sufficiently and a Li+ concentration gradient was formed within the graphite layer during cycling. The electrode thickness was found to have increased after cycling and crack formation was also observed from post-mortem SEM analyses. These observations have shown the causes behind disconnection of Li+ conduction pathways and capacity fade.

2.3

Alloying-type anode

Alloying-type anodes, such as Si, Sn, Al, as well as others, are capable of storing more lithium per mass and volume than graphite anodes.27 These anodes undergo 50–300% volume change during lithiation/delithiation, which causes loss of electric contact of the electrode and capacity fade of the battery. In addition, volume changes produce fresh surfaces, which are exposed to the electrolyte, and when OLEs are used, electrolyte decomposition continuously occurs at these fresh surfaces, as discussed above for graphite, but are actually more pronounced for alloying-type anodes.34 Much engineering effort has been made to mitigate this issue and many alternative materials have been proposed, such as nanoscale silicon to reduce mechanical strain on particles, stress-relief buffer matrices based on carbon, and new functional binders. In the case of ASS LIBs, all the components are solid with a certain volume

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes

585

and shape, and in addition, there is a pressure applied from external part of the cell. Thus, system optimization including application of adequate cell pressure also has a noticeable effect on the anode performance. This idea suggests that it may be easier to achieve stable and long-term cycling of alloy anodes in ASS environment without complicated structuring of active materials. Since alloying-type anodes lack conductivity, they are often used alongside carbonaceous additives, which are also added as a buffer to counteract the volume change of the active material. These composite anodes can be prepared both via mortar milling, tumbler ball milling, or high-energy ball milling. In the case of Sn-C composite electrodes synthesized via mortar milling, a synergic effect of the combination was observed.35 ASS Li-metal cells having an electrode based on Sn-C (8:2 w/w mixture) and LPSI exhibited a discharge capacity of 405 mAh g−1, while those having electrodes based on either Sn or C were found to be 672.9 mAh g−1 and 271.3 mAh g−1, respectively. Based on these two capacity values and the composition of the Sn-C electrode, it is possible to estimate the capacity value of the composite mathematically, and it was found to be 347.6 mAh g−1, which is 50 mAh g−1 smaller than the real cell capacity. These values can be attributed to improved charge transfer rate and better kinetics of lithiation/delithiation of the carbon in the presence of Sn. The cell with the composite electrode maintained a capacity value above 320 mAh g−1. Post mortem SEM analysis of the Sn-C electrode showed an increase in the roughness of the electrode surface but not the formation of large cracks that are commonly seen in cells using OLE. This confirms that the Sn-C electrode underwent volume change but that it was not large enough to cause cracking because of the reduced pulverization of Sn aided by the use of solid-state materials. Milling method also affects the performance of batteries.24 When high-energy ball milling is applied to Si, C, and LPSI-based composites, XRD patterns of the composite samples were found to change with the duration of milling.24 The peaks related to graphite disappeared after 2 h of milling, and a weak peak of silicon carbide became visible after 4 h of milling, which is not favorable for anode materials because of its low electrical conductivity. In addition, electrodes prepared by high-energy ball milling exhibited a lower capacity compared to those prepared by simple mortar grinding. Okuno et al. prepared nanoporous Si particles having an average size of 9.4 nm via air-oxidation demagnesiation of Mg2Si and compared its performance with commercially available non-porous Si.36 For the preparation of composite electrodes based on Si, Li3PS4, and a carbon-based conductive additive, either mortar milling or high-energy ball milling was utilized. With respect to non-porous Si, the cells with hand-milled and mechanically milled electrodes exhibited first discharge capacities of 1056 and 782 mAh g−1, respectively (Fig. 4). However, these values decreased sharply during the first 25 cycles and converged to 90 and 201 mAh g−1 in the 150th cycle. In contrast, those of nanoporous Si were 2498 and 1532 mAh g−1 in the first cycle, and 1097 and 1220 mAh g−1 at the 150th cycle. Better capacity retention of nanoporous Si is believed to arise from the presence of nano pores which can serve a function as buffer regions for large volume changes. From EDX mapping images of electrodes, aggregation of micrometer-sized Si grains was observed in the hand-milled anode. The authors explained that the lower first discharge capacity of mechanically milled sample comes from the presence of an oxide layer on Si formed during mechanical milling, which inevitably produces fresh surfaces due to high frequency application. However, the capacity retention was better for the mechanically milled sample and this is due to the homogeneous dispersion of Si. Comparison between the nanoporous and non-porous Si sample demonstrates that nanoporous Si is suitable for long-term cycling. In addition to porosity, the particles size of alloying-type anodes has an important effect on improving capacity and cycling performance. Dunlap et al. synthesized a Si-C anode via pyrolysis of a mixture of coal-tar-pitch and Si powders of various particles diameters, 44 mm, 1–3 mm, and 50 nm,37 which was then mixed with 77.5Li2S-22.5P2S5. Thus-prepared composite anode with 50 nm Si particles outperformed the composites containing the 2 mm-sized Si particles in terms of first cycle capacity, coulombic efficiency, and capacity retention thanks to limited mechanical stress on the Si nano-particles. Using 50 nm Si particles, the authors also prepared samples with different composition ratios between Si-C and SE ranging from 6:4 to 10:0 (w/w). While energy density of the cell decreased by increasing the mass fraction of SE, the electrode with the largest amount of SE (Si-C:SE ¼ 6:4 in weight) improved capacity retention with cycling. These values are trade-offs and are important metrics to consider when shifting to full-cell configurations. Cross section SEM images using a Focused Ion Beam (FIB) after 100 cycles revealed that there were less cracks and voids in samples with larger amounts of SE.

Fig. 4 Cycle performances of Li||Si cells with either nanoporous or non-porous Si, mixed via mortar milling or high-energy ball milling.36

586

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes

Integration of the alloying-type anode, among them Si and Sn, with carbon additives, is a common strategy to obtain an anode material that has both a higher capacity and longer cycle life. However, it has been found that the presence of carbon in composite electrodes accelerates the degradation of sulfide-based electrolytes. Tan et al., prepared a mSi anode with and without 20 wt% carbon additive and assembled a full cell with mSi||NCM811. In a XRD pattern of the electrode based only on Si and Li6PS5Cl, signals related to SE and Si were observed. Although a formation of SEI was expected, the low amount formed at the interface was most likely not detectable with this bulk technique. In contrast, in electrodes based on Si, carbon, and SE, most of the diffraction signals of SE became less discernible, indicating severe SE decomposition; instead, the signals of Li2S became visible, which is a major decomposition product of Li6PS5Cl. Moreover, by means of XPS analysis, a greater extent of SE decomposition in the presence of carbon was confirmed. Due to the decomposition accelerated by carbon, the authors prepared 99.9 wt% mSi with 0.1% PTFE by a wet mixing procedure and coated it on a copper current collector. By cycling in a full-cell configuration at 1.2 mA cm−2, the mSi anode was found to deliver reversible capacities of more than 11 mAh cm−2 (>2890 mAh g−1).

3

Conclusion

For the development of anode materials suitable for ASS batteries, “homogeneity” is an essential requirement in addition to the intrinsic properties of the anode, in terms of forming seamless SE and anodes and creating extended close contact between them. However, since both SE and anode materials are solids, they can only make point-to-point contact within the materials themselves and at their interfaces, which cause inhomogeneous current distribution and electrode utilization, ultimately leading to battery failure. To address this issue, in the case of Li metal anodes, organic materials such as polymers and ILs are often combined with SE to form composites or interlayers; these significantly improve the wettability of the interface. Anode materials alternative to Li metal are also reported such as insertion-type and alloying-type anodes. For any kinds of anode materials, the pressure applied to the batteries plays an important role to keep the homogeneities. The pressure needs to be adjusted carefully, not to deform the electrodes but to reduce or counteract volume changes of electrode materials. In the case of powder-based electrodes, compositional study becomes also important to form homogeneous pathways for ion and electronic conductions. Furthermore, the development of the anode materials needs to be combined with those of cathodes and cannot be completed on its own because the volume change of anode strongly depends on the battery capacity. Although the solid nature of SEs makes their development difficult, there are possibilities that can be realized because of their solid state, e.g., SE sandwiched by quasi-solid-state interlayers. This widens the possibility of materials development and, hence, increasing advancement in the design of anode materials is anticipated.

References 1. Banerjee, A.; Wang, X.; Fang, C.; Wu, E. A.; Meng, Y. S. Interfaces and Interphases in All-Solid-State Batteries with Inorganic Solid Electrolytes. Chem. Rev. 2020, 120, 6878–6933. https://doi.org/10.1021/acs.chemrev.0c00101. 2. Sun, F.; Dong, K.; Osenberg, M.; Hilger, A.; Risse, S.; Lu, Y.; Kamm, P. H.; Klaus, M.; Markötter, H.; García-Moreno, F.; Arlt, T.; Manke, I. Visualizing the Morphological and Compositional Evolution of the Interface of InLi-Anode|Thio-LISION Electrolyte in an all-Solid-State Li–S Cell by in Operando Synchrotron X-Ray Tomography and Energy Dispersive Diffraction. J. Mater. Chem. A 2018, 6, 22489–22496. https://doi.org/10.1039/c8ta08821g. 3. Krauskopf, T.; Richter, F. H.; Zeier, W. G.; Janek, J. Physicochemical Concepts of the Lithium Metal Anode in Solid-State Batteries. Chem. Rev. 2020, 120, 7745–7794. https:// doi.org/10.1021/acs.chemrev.0c00431. 4. Zhang, L.; Yang, T.; Du, C.; Liu, Q.; Tang, Y.; Zhao, J.; Wang, B.; Chen, T.; Sun, Y.; Jia, P.; Li, H.; Geng, L.; Chen, J.; Ye, H.; Wang, Z.; Li, Y.; Sun, H.; Li, X.; Dai, Q.; Tang, Y.; Peng, Q.; Shen, T.; Zhang, S.; Zhu, T.; Huang, J. Lithium Whisker Growth and Stress Generation in an In Situ Atomic Force Microscope-Environmental Transmission Electron Microscope Set-up. Nat. Nanotechnol. 2020, 15, 94–98. https://doi.org/10.1038/s41565-019-0604-x. 5. Famprikis, T.; Canepa, P.; Dawson, J. A.; Islam, M. S.; Masquelier, C. Fundamentals of Inorganic Solid-State Electrolytes for Batteries. Nat. Mater. 2019, 18, 1278–1291. https:// doi.org/10.1038/s41563-019-0431-3. 6. Wenzel, S.; Leichtweiss, T.; Krüger, D.; Sann, J.; Janek, J. Interphase Formation on Lithium Solid Electrolytes—An In Situ Approach to Study Interfacial Reactions by Photoelectron Spectroscopy. Solid State Ionics 2015, 278, 98–105. https://doi.org/10.1016/j.ssi.2015.06.001. 7. Hartmann, P.; Leichtweiss, T.; Busche, M. R.; Schneider, M.; Reich, M.; Sann, J.; Adelhelm, P.; Janek, J. Degradation of NASICON-Type Materials in Contact with Lithium Metal: Formation of Mixed Conducting Interphases (MCI) on Solid Electrolytes. J. Phys. Chem. C 2013, 117, 21064–21074. https://doi.org/10.1021/jp4051275. 8. Mazzapioda, L.; Tsurumaki, A.; Di Donato, G.; Adenusi, H.; Navarra, M. A.; Passerini, S. Quasi-Solid-State Electrolytes - Strategy Towards Stabilising Li|Inorganic Solid Electrolyte Interfaces in Solid-State Li Metal Batteries. Energy Mater. 2023. https://doi.org/10.20517/energymater.2023.03. 9. Ma, C.; Cheng, Y.; Yin, K.; Luo, J.; Sharafi, A.; Sakamoto, J.; Li, J.; More, K. L.; Dudney, N. J.; Chi, M. Interfacial Stability of Li Metal-Solid Electrolyte Elucidated via In Situ Electron Microscopy. Nano Lett. 2016, 16, 7030–7036. https://doi.org/10.1021/acs.nanolett.6b03223. 10. Schwöbel, A.; Hausbrand, R.; Jaegermann, W. Interface Reactions Between LiPON and Lithium Studied by In-Situ X-Ray Photoemission. Solid State Ionics 2015, 273, 51–54. https://doi.org/10.1016/j.ssi.2014.10.017. 11. Wenzel, S.; Sedlmaier, S. J.; Dietrich, C.; Zeier, W. G.; Janek, J. Interfacial Reactivity and Interphase Growth of Argyrodite Solid Electrolytes at Lithium Metal Electrodes. Solid State Ionics 2018, 318, 102–112. https://doi.org/10.1016/j.ssi.2017.07.005. 12. Diallo, M. S.; Shi, T.; Zhang, Y.; Peng, X.; Shozib, I.; Wang, Y.; Miara, L. J.; Scott, M. C.; Tu, Q. H.; Ceder, G. Effect of Solid-Electrolyte Pellet Density on Failure of Solid-State Batteries. Nat. Commun. 2024, 15, 858. https://doi.org/10.1038/s41467-024-45030-7. 13. Ham, S.-Y.; Yang, H.; Nunez-cuacuas, O.; Tan, D. H. S.; Chen, Y.-T.; Deysher, G.; Cronk, A.; Ridley, P.; Doux, J.-M.; Wu, E. A.; Jang, J.; Meng, Y. S. Assessing the Critical Current Density of all-Solid-State Li Metal Symmetric and Full Cells. Energy Storage Mater. 2023, 55, 455–462. https://doi.org/10.1016/j.ensm.2022.12.013. 14. Luo, S.; Wang, Z.; Li, X.; Liu, X.; Wang, H.; Ma, W.; Zhang, L.; Zhu, L.; Zhang, X. Growth of Lithium-Indium Dendrites in all-Solid-State lithium-Based Batteries with Sulfide Electrolytes. Nat. Commun. 2021, 12, 6968. https://doi.org/10.1038/s41467-021-27311-7. 15. Cao, D.; Sun, X.; Li, Q.; Natan, A.; Xiang, P.; Zhu, H. Lithium Dendrite in all-Solid-State Batteries: Growth Mechanisms, Suppression Strategies, and Characterizations. Matter 2020, 3, 57–94. https://doi.org/10.1016/j.matt.2020.03.015.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes

587

16. Yang, C.; Xie, H.; Ping, W.; Fu, K.; Liu, B.; Rao, J.; Dai, J.; Wang, C.; Pastel, G.; Hu, L. An Electron/Ion Dual-Conductive Alloy Framework for High-Rate and High-Capacity Solid-State Lithium-Metal Batteries. Adv. Mater. 2019, 31, e1804815. https://doi.org/10.1002/adma.201804815. 17. Huo, H.; Chen, Y.; Luo, J.; Yang, X.; Guo, X.; Sun, X. Rational Design of Hierarchical “Ceramic-in-Polymer” and “Polymer-in-Ceramic” Electrolytes for Dendrite-Free Solid-State Batteries. Adv. Energy Mater. 2019, 9. https://doi.org/10.1002/aenm.201804004. 18. Hu, Y.; Zhong, Y.; Qi, L.; Wang, H. Inorganic/Polymer Hybrid Layer Stabilizing Anode/Electrolyte Interfaces in Solid-State Li Metal Batteries. Nano Res. 2020, 13, 3230–3234. https://doi.org/10.1007/s12274-020-2993-4. 19. Wu, F.; Fang, S.; Kuenzel, M.; Diemant, T.; Kim, J.-K.; Bresser, D.; Kim, G.-T.; Passerini, S. Bilayer Solid Electrolyte Enabling Quasi-Solid-State Lithium-Metal Batteries. J. Power Sources 2023, 557, 232514. https://doi.org/10.1016/j.jpowsour.2022.232514. 20. Gao, Y.; Wang, D.; Li, Y. C.; Yu, Z.; Mallouk, T. E.; Wang, D. Salt-Based Organic-Inorganic Nanocomposites: Towards a Stable Lithium Metal/Li(10) GeP(2) S(12) Solid Electrolyte Interface. Angew. Chem. 2018, 57, 13608–13612. https://doi.org/10.1002/anie.201807304. 21. Xiong, S.; Liu, Y.; Jankowski, P.; Liu, Q.; Nitze, F.; Xie, K.; Song, J.; Matic, A. Design of a Multifunctional Interlayer for NASCION-Based Solid-State Li Metal Batteries. Adv. Funct. Mater. 2020, 30, 2001444. https://doi.org/10.1002/adfm.202001444. 22. Reisacher, E.; Kaya, P.; Knoblauch, V. Percolation Behavior of a Sulfide Electrolyte–Carbon Additive Matrix for Composite Cathodes in All-Solid-State Batteries. Batteries 2023, 9, 595. https://doi.org/10.3390/batteries9120595. 23. Shi, T.; Tu, Q.; Tian, Y.; Xiao, Y.; Miara, L. J.; Kononova, O.; Ceder, G. High Active Material Loading in All-Solid-State Battery Electrode via Particle Size Optimization. Adv. Energy Mater. 2019, 10. https://doi.org/10.1002/aenm.201902881. 24. Branchi, M.; Maresca, G.; Tsurumaki, A.; Suzuki, N.; Croce, F.; Panero, S.; Voje, J.; Aihara, Y.; Navarra, M. A. Silicon-Based Composite Anodes for All-Solid-State Lithium-Ion Batteries Conceived by a Mixture Design Approach. ChemSusChem 2023, 16, e202202235. https://doi.org/10.1002/cssc.202202235. 25. Nam, Y. J.; Oh, D. Y.; Jung, S. H.; Jung, Y. S. Toward Practical All-Solid-State Lithium-Ion Batteries with High Energy Density and Safety: Comparative Study for Electrodes Fabricated by Dry- and Slurry-Mixing Processes. J. Power Sources 2018, 375, 93–101. https://doi.org/10.1016/j.jpowsour.2017.11.031. 26. Asenbauer, J.; Eisenmann, T.; Kuenzel, M.; Kazzazi, A.; Chen, Z.; Bresser, D. The Success Story of Graphite as a Lithium-Ion Anode Material – Fundamentals, Remaining Challenges, and Recent Developments Including Silicon (Oxide) Composites. Sustainable Energy Fuels 2020, 4, 5387–5416. https://doi.org/10.1039/d0se00175a. 27. Lewis, J. A.; Cavallaro, K. A.; Liu, Y.; McDowell, M. T. The Promise of Alloy Anodes for Solid-State Batteries. Joule 2022, 6, 1418–1430. https://doi.org/10.1016/j. joule.2022.05.016. 28. Maresca, G.; Tsurumaki, A.; Suzuki, N.; Tsujimura, T.; Aihara, Y.; Navarra, M. A. Improvement of Graphite Interfacial Stability in All-Solid-State Cells Adopting Sulfide Glassy Electrolytes. ChemElectroChem 2021, 8, 689–696. https://doi.org/10.1002/celc.202001291. 29. Yang, S.; Yamamoto, K.; Mei, X.; Sakuda, A.; Uchiyama, T.; Watanabe, T.; Takami, T.; Hayashi, A.; Tatsumisago, M.; Uchimoto, Y. High Rate Capability from a Graphite Anode Through Surface Modification with Lithium Iodide for All-Solid-State Batteries. ACS Appl. Energy Mater. 2021, 5, 667–673. https://doi.org/10.1021/acsaem.1c03166. 30. Kim, D. H.; Oh, D. Y.; Park, K. H.; Choi, Y. E.; Nam, Y. J.; Lee, H. A.; Lee, S. M.; Jung, Y. S. Infiltration of Solution-Processable Solid Electrolytes into Conventional Li-Ion-Battery Electrodes for All-Solid-State Li-Ion Batteries. Nano Lett. 2017, 17, 3013–3020. https://doi.org/10.1021/acs.nanolett.7b00330. 31. Kim, J. Y.; Park, J.; Lee, M. J.; Kang, S. H.; Shin, D. O.; Oh, J.; Kim, J.; Kim, K. M.; Lee, Y.-G.; Lee, Y. M. Diffusion-Dependent Graphite Electrode for All-Solid-State Batteries with Extremely High Energy Density. ACS Energy Lett. 2020, 5, 2995–3004. https://doi.org/10.1021/acsenergylett.0c01628. 32. Wu, Z.; Li, X.; Zheng, C.; Fan, Z.; Zhang, W.; Huang, H.; Gan, Y.; Xia, Y.; He, X.; Tao, X.; Zhang, J. Interfaces in Sulfide Solid Electrolyte-Based All-Solid-State Lithium Batteries: Characterization, Mechanism and Strategy. Electrochem. Energy Rev. 2023, 6. https://doi.org/10.1007/s41918-022-00176-0. 33. Cui, C.; Ye, Q.; Zeng, C.; Wang, S.; Xu, X.; Zhai, T.; Li, H. One-Step Fabrication of Garnet Solid Electrolyte with Integrated Lithiophilic Surface. Energy Storage Mater. 2022, 45, 814–820. https://doi.org/10.1016/j.ensm.2021.12.027. 34. Oh, P.; Yun, J.; Choi, J. H.; Saqib, K. S.; Embleton, T. J.; Park, S.; Lee, C.; Ali, J.; Ko, K.; Cho, J. Development of High-Energy Anodes for All-Solid-State Lithium Batteries Based on Sulfide Electrolytes. Angew. Chem. 2022, 61, e202201249. https://doi.org/10.1002/anie.202201249. 35. Maresca, G.; Tsurumaki, A.; Suzuki, N.; Yoshida, K.; Panero, S.; Aihara, Y.; Navarra, M. A. Sn/C Composite Anodes for Bulk-Type All-Solid-State Batteries. Electrochim. Acta 2021, 395, 139104. https://doi.org/10.1016/j.electacta.2021.139104. 36. Okuno, R.; Yamamoto, M.; Kato, A.; Takahashi, M. Stable Cyclability Caused by Highly Dispersed Nanoporous Si Composite Anodes with Sulfide-Based Solid Electrolyte. J. Electrochem. Soc. 2020, 167, 140522. https://doi.org/10.1149/1945-7111/abc3ff. 37. Dunlap, N. A.; Kim, S.; Jeong, J. J.; Oh, K. H.; Lee, S.-H. Simple and Inexpensive Coal-Tar-Pitch Derived Si-C Anode Composite for All-Solid-State Li-Ion Batteries. Solid State Ionics 2018, 324, 207–217. https://doi.org/10.1016/j.ssi.2018.07.013.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free” Till Fuchsa,b, Burak Aktekina,b, Felix Hartmanna,b , Simon Burkhardtc, and Jürgen Janeka,b,d , aInstitute of Physical Chemistry, Justus Liebig University Giessen, Giessen, Germany; bCenter for Materials Research (ZfM/LaMa), Justus Liebig University Giessen, Giessen, Germany; cVARTA Microbattery GmbH, Ellwangen, Germany; dBattery and Electrochemistry Laboratory (BELLA), Karlsruhe Institute of Technology (KIT), Eggenstein-Leopoldshafen, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 Introduction 2 General considerations of “anode-free” cell concepts 2.1 Energy density estimation 2.2 Manufacturing advantages and challenges 2.3 Electrochemical stability of solid electrolytes against lithium metal 2.3.1 The impact of electrolyte reduction reactions 2.3.2 Lithium inventory loss in “anode-free” cells 2.3.3 Overcoming electrolyte instability 3 Possible realizations of the “anode-free” cell design 3.1 Planar metal current collector|solid electrolyte interfaces 3.2 Carbonaceous seed interlayers 3.3 Three-dimensional concepts of current collector|electrolyte interfaces 4 Summary and outlook Acknowledgments References

589 590 591 591 592 592 593 593 594 594 595 595 597 597 597

Abstract In this chapter, the role of metal anodes in future solid-state batteries is discussed, which have the potential to significantly advance short-term energy storage by increasing energy density, safety, and manufacturability. Further, the concept of “anodefree” secondary lithium metal batteries is analyzed in detail regarding potential advantages and challenges. This design allows a cell to be assembled without the need for a lithium metal reservoir, as the anode active material is deposited onto the current collector within the initial charging step. Different strategies to achieve “anode-free” cell operation are discussed with regard to their individual merits. By examining these concepts and their interconnectedness, further research is needed to pave the way for more sustainable and effective energy storage solutions.

Glossary

“Dead” lithium Lithium that is no longer active or functional in a battery, typically due to an irreversible disconnection of lithium to the anode or chemical changes. “Anode-free” Refers to a type of battery that is assembled without anode-active material present, as this is deposited within the first formation/charging step. 3D host layer A three-dimensional structure or material that serves as a host for lithium deposition. Carbonaceous interlayer A layer composed of carbon-based materials placed between two layers or components to guide lithium deposition morphology. Coulometric titration time analysis (CTTA) Novel developed technique able to quantify the degree of degradation of solid lithium|electrolyte interfaces. Current collector (CC) The current collector used to contact the electrode present within the battery, which is especially important in “anode-free” designs for the growth of the active material. Electrochemical impedance spectroscopy (EIS) A measurement technique used to analyze the electrical impedance of a system in response to a small, alternating current, often used for studying the behavior of electrochemical systems. Mixed conducting interphase (MCI) Abbreviation for “Mixed Conducting Interphase,” describing a type of non-protecting interphase capable of conducting both ions and electrons. Nucleation The initial process of lithium deposition at the interface between an electrolyte and current collector. Porous trilayer A structure composed of three layers with open spaces or pores between them, with a dense layer separating the porous layers. Seed metal A small amount of metal or material used as a starting point for the nucleation and growth of lithium within “anode-free” batteries.

588

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00169-9

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

589

Solid electrolyte interphase (SEI) A degradation layer that forms on the surface of the anode, influencing the overall cell resistance and evolution of a battery. Whisker A thin, filament-like structure that can grow from a metal through the current collector during lithium growth, unlike a dendrite which usually grows into a solid electrolyte.

Key points

• • • •

1

Cell concepts for secondary alkali metal batteries, in which the alkali metal is only present in a charged state (so-called “anode-free”), are described. The advantages of such a cell design in terms of battery performance and costs are elaborated. The electrolyte degradation in contact with lithium is discussed with a focus on “anode-free” cell design. The role of a variety of negative electrode current collector designs is discussed.

Introduction

Alkali metal-ion-based secondary batteries are high-energy electrochemical power sources. A well-established version of such a battery is the lithium-ion battery (LIB), which is used, for example, in consumer electronics, power tools, electric vehicles, and stationary energy storage systems. Another less common but emerging example of an alkali metal-ion-based secondary battery is the sodium-ion battery (SIB). All batteries of this type store energy by transferring alkali metal ions and storing these as alkali metal ions in materials that allow the reversible insertion and extraction thereof. The success of the LIB in changing the daily lives of mankind around the world undoubtedly justifies the relevance and omnipresence of the alkali metal-ion-based secondary battery. However, its energy density is approaching a physicochemical limit that conflicts with the high demand for improved battery performance and for new applications that are powered by secondary batteries. To meet these demands, changes are required in the way energy is stored in such batteries.1,2 One approach for such a change is the transition from storing alkali metal ions in host materials, such as graphite or silicon in the case of LIBs, to in situ depositing the corresponding alkali metal as such during the charging process. The gain in energy density that can be achieved by this transition at the material level is remarkable. The theoretical specific capacity of graphite is 372 mA h g−1, which is only 10 % of the theoretical specific capacity of lithium metal (3860 mA h g−1). For this reason, the concept of using an alkali metal as an anode active material has been pursued in both primary and secondary lithium metal batteries (LMBs). While primary LMBs are a mass product today, the commercialization of secondary LMBs has experienced fluctuations over the past 50 years: After research in the 1970s and the commercialization of electronic devices with secondary LMBs in the 1980s, hazardous incidents banned secondary LMBs from the market. In the 2010s, electric vehicles with polymer-based secondary LMBs were introduced to the market, but to date, elevated temperature as required operating conditions for these batteries prevent them from becoming more widespread in terms of other applications and other vehicle manufacturers. It can be expected that this approach will undergo further years of research and development to ensure ultimate stability of the LMB under other operating conditions and to improve the battery performance even further. The example of the commercialized polymer-based LMBs illustrates an intermediate step in the development of secondary alkali metal batteries. Their production and function rely on the usage of lithium metal foil as anode active material. There is considerable interest in reducing the amount of lithium metal foil in the battery production process to enhance the energy density of the cells and to reduce production costs as discussed in Section 2.2 of this chapter. Therefore, research is focused on a more advanced and more challenging development for secondary alkali metal batteries, so-called “anode-free” alkali metal batteries. The term “anode-free” is frequently used but can create a misleading impression of the actual cell design, as any electrochemical cell always has two electrodes. The negative electrode (i.e. anode during discharge) in a conventional LIB contains active materials such as graphite and silicon, and passive materials, such as binders, conductive additives, and the anode current collector (CC). The term “anode-free” is derived from the idea that lithium is deposited in situ as lithium metal on a CC and not intercalated into an anode active material. In such a case, where no anode active material is present in the discharged state, the cell could be referred to as “anode active material-free,” or short: “anode-free.” However, once the cell is charged and lithium metal is deposited on the CC, the anode active material of the cell is lithium metal, making it an LMB. For the sake of simplicity, readability, and comprehensibility, and despite the misleading impression mentioned above, the abbreviated term “anode-free” is nevertheless used in the battery community and also in this chapter. In this chapter, the cell concept of an “anode-free” alkali metal battery is explained. Lithium and sodium metal batteries serve as specific examples to describe the key aspects and challenges. The relevance of this cell design is illustrated by comparing the energy density of different cell concepts and the impact of avoiding lithium metal foil on cell manufacturing. The challenge of depositing a homogeneous lithium metal layer during the charging process is also discussed before the chapter proceeds with three practical examples of “anode-free” LMBs.

590

2

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

General considerations of “anode-free” cell concepts

The working principle of “anode-free” alkali metal secondary batteries is schematically illustrated for lithium-based cells in Fig. 1. In contrast to commercial LIBs, where graphite serves as the anode active material (Fig. 1A), LMBs with a lithium metal anode (Fig. 1B) and ultimately with “anode-free” designs (Fig. 1C) offer certain advantages, especially in terms of the energy density, but also present some challenges. The main characteristics of an “anode-free” alkali metal secondary battery is that the negative electrode does not contain any active material in the pristine (discharged) state that can be used to operate the cell. It therefore relies on being combined with cathode active materials (CAMs) that contain alkali metal. Both components are separated by a corresponding liquid electrolyte and a separator or by a solid electrolyte (SE) separator. Different strategies for constructing “anodefree” cells are pursued in the existing literature. These are discussed in more detail in Section 3, where the focus is on cell designs that involve planar metal CCs in direct contact with SE separators (Fig. 1D and Section 3.1), the use of carbonaceous seed interlayers (Fig. 1E and Section 3.2), and the application of three-dimensional framework SE separators (Fig. 1F and Section 3.3). The challenges that are associated with combining an ether- or carbonate-based liquid electrolyte with a lithium metal anode are well-known in the scientific literature: During cycling of the cell, the solid electrolyte interphase (SEI) builds up and cracks are formed repeatedly, resulting in a non-uniform SEI. In addition, lithium is deposited inhomogeneously, leading to dendritic lithium growth during charging. Upon discharging, lithium dendrites that have not yet caused a short circuit can be dissolved close to the CC, detach from it, and become electronically insulated lithium deposits (“dead lithium”). Both processes lead to continuous consumption of lithium, which conflicts with zero-excess lithium in “anode-free” LMBs.3 Research into new liquid electrolytes that better meet the requirements, especially for LMBs, has led to significant progress in the past years and will continue so in the future. A comprehensive overview of the development of LMBs and electrolytes for LMBs can be found in the literature.4 This is mentioned to emphasize that all the challenges faced by alkali metal batteries with liquid electrolytes present even greater challenges for “anode-free” alkali metal batteries. Particularly over the past decade, there has been significant increase in R&D focused on alkali metal anodes in secondary solid-state batteries (SSBs) as well. One key driving force is the hope to achieve more reliable and safer cell operation with alkali metal anodes in SSBs compared to similar batteries utilizing liquid electrolytes.5–7 Also, and at the latest since Lee et al. showcased

Fig. 1 Comparison of different anode concepts in lithium-based battery cells. (A) A conventional lithium-ion battery (LIB) operates with a graphite anode and a certain cathode active material (CAM). A lithium metal battery (LMB) can contain (B) a lithium metal anode or (C) have an “anode-free” design. Three “anode-free” cell concepts are illustrated: (D) a planar metal current collector (CC) in direct contact with a separator, (E) an additional carbonaceous seed interlayer between the CC and separator, and (F) a three-dimensional separator framework, e.g., a porous trilayer design with a solid electrolyte (SE) framework combined with a dense separator.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

591

superior SSB cell performance with unsurpassed energy densities at elevated temperatures by employing an engineered anode CC with Ag-C composite layers, the SSB research efforts for “anode-free” cell concepts were intensified.8 In the following, we will emphasize “anode-free” lithium metal SSBs because the challenges associated with lithium metal are comparable to the ones associated with other alkali metals. However, there is more scientific literature available on lithium metal in “anode-free” cells than on “anode-free” cells with other alkali metals. Before going into practical implementation examples and discussing the challenges on the electrochemical stability of SEs with lithium, lithium dendrite formation and morphology/ microstructure control of the plated lithium, the advantages of “anode-free” cell designs are rendered plausible in the following with respect to the energy density and manufacturing considerations.

2.1

Energy density estimation

CAMs frequently used in LIBs such as LiNixCoyMnzO2 (NCM), or LiNixCoyAlzO2 (NCA) contain lithium in their discharged state. During the charging process of an LMB, lithium from the cathode is deposited as lithium metal on the lithium metal foil used as anode. The limit of reversibly stored lithium, i.e., the total amount of charge that can be used for cycling, is therefore determined by the amount of charge that can be extracted/inserted from/into the CAM. The lithium metal foil serves as a substrate for lithium deposition in an LMB but does not contribute to the electrochemical energy storage process, thus reducing the energy density of an LMB cell. Albeit being formally a passive substrate, this foil often fulfills the function of a reservoir to mitigate inevitable and irreversible lithium consumption/loss processes, e.g., during formation of interphases, electrochemical side reactions, and detachment of “dead lithium.” Since any additional (passive) lithium metal is omitted in “anode-free” cell setups, careful control of the lithium inventory is paramount to not only attain but also, being more challenging, to sustain very high energy densities during cycling. The benefits in terms of the specific energy and energy density of “anode-free” LMB cells over LMB with a lithium foil and conventional LIBs with graphite anodes were calculated by Heubner et al. and are shown in Table 1 considering liquid and solid electrolytes.3 It is worth noting that these values can be surpassed with even more “extreme” cell configurations. For instance, the group of Hong Li recently showcased the operation of lithium battery cells with wgrav > 700 Wh kg−1 and wvol > 1600 Wh l−1, essentially by removing most passive components, employing very thin CCs (dCu ¼ 6 mm and dAl ¼ 9 mm), thin Li metal anodes (e.g., dLi metal ¼ 20 mm), and a lithium-rich NCM cathode (Li1.2Ni0.13Co0.13Mn0.54O2) as well as by increasing the operating voltage window to 1.25 V–4.8 V.9 Whilst the experimental data in this report are insufficient to provide a comprehensive understanding of cell kinetics and long-term stability, the given estimates about the energy densities illustrate that the values in Table 1 do not represent the upper limit. Nonetheless, to adequately assess the feasibility of a battery technology for mass markets, all essential performance metrics, such as rate capability, cycle life, large-scale manufacturability, cost, and sustainability, must be taken into consideration. Inherently, the energy densities of sodium-based cells are lower compared to their lithium-based counterparts due to the larger ionic radius and less negative standard reduction potential of sodium. For example, wgrav and wvol of most-promising SIBs are estimated to be in the range of 90–160 Wh kg−1 and 250–375 Wh l−1, respectively, roughly half the values for state-of-the-art LIBs.10 Research of “anode-free” sodium metal batteries is, even more than for LMBs, in the infantry but values up to 350 Wh kg−1 and 700 Wh l−1 seem conceivable.10,11 However, previous experiments on laboratory scale demonstrated that it is possible to sustain wgrav up to 400 Wh kg−1 during cycling of an “anode-free” cell, using a carbonated Al CC and a presodiated pyrite CAM (the energy density was calculated based on the mass of a Na1.5FeS2 and a carbon nucleation interlayer).12

2.2

Manufacturing advantages and challenges

Avoiding lithium metal foil or other lithium reservoirs at the anode during battery production leads also to a cost reduction in several aspects. The most trivial factor for the cost reduction is that expenses associated with purchasing lithium foil are avoided. Moreover, the reactivity of lithium metal toward the surrounding atmosphere and the materials required for processing the foil is comparably high. A second contribution to cost savings therefore results from the lower requirements for storage and production environments as well as for all materials and tools in contact with the foil. This is also advantageous in terms of safety considerations during manufacturing. Table 1 Comparison of specific energies wgrav and energy densities wvol calculated by Heubner et al. for different lithium-based battery cell configurations (materials basis) in the discharged state: Lithium ion-battery (LIB), solid-state battery (SSB), and lithium metal battery (LMB) with and without a lithium metal foil as anode. “Anode-free” LMB

LIB

Li-SSB

LMB

Electrolyte

Liquid

Solid

Liquid

Solid

Liquid

Solid

Anode Anode thickness/mm wgrav/Wh kg−1 wvol/Wh l−1

Graphite 130 277 751

Graphite 130 266 751

Li metal 36 406 1197

Li metal 36 393 1197

– 0 423 1514

– 0 408 1514

For these calculations, only parameters in the electrode stacks (CC to CC, without cell housing), a cathode with 96 wt% NCM811 (dcat ¼ 100 mm), a positive Al CC and negative Cu CC (dAl ¼ dCu ¼ 10 mm), and different separators (dsep ¼ 15 mm) were included. Further details about these estimations can be found in Ref. 3.

592

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

Commercial lithium metal foils for battery application are inherently covered by native passivation layers, in particular resulting from reactions of the freshly produced foil with residual humidity and carbon dioxide during transport, processing, and storage – even if stored in very clean surroundings like a glovebox atmosphere.13 Although studies have shown that these native layers can enhance the overall cell performance of LMBs, the resulting performance depends on the specific composition, thickness, morphology, and homogeneity of these layers.14 Without pre-manufacturing artificial passivation layers, which come at higher costs, their control can be challenging. Lithium metal, which is plated in situ during charging of an “anode-free” cell, is on the contrary highly reactive. This must be considered in terms of electrolyte stability, SEI formation, and side reactions. Nonetheless, it is important to highlight that, as discussed in detail in the following sections, host structures and lithophilic CC coatings are frequently utilized to ensure stable cycling of “anode-free” cells. This may reduce some of the above-mentioned fundamental manufacturing and cost advantages of “anode-free” cells, and such trade-offs should be considered in forthcoming and more comprehensive cost evaluations.3 In the case of “anode-free” SSB cells, the cost estimations are currently even more uncertain due to the early stage of SSB research, with practical insights mostly confined to laboratory experiments. Numerous approaches for enhancing lithium metal anodes and “anode-free” cells configurations have proven successful on laboratory scale. This research primarily concentrates on adjusting the interfaces/interphases as well as electrode morphology/ structure and is instrumental in addressing the issue from a mechanistic point of view. Nevertheless, the feasibility of upscaling and implementing such strategies into large-scale battery production is often neglected in the literature and generally under-studied, which calls for an intensification of application-oriented and rational R&D.

2.3

Electrochemical stability of solid electrolytes against lithium metal

SEs are often considered as better candidates for LMBs due to their beneficial mechanical properties and higher thermal stability, offering a potential advantage against dendritic lithium growth and associated safety risks.15 However, most SE materials are known to be thermodynamically unstable upon contact with lithium metal,16 and as a result, they show parasitic electrolyte side reactions. For instance, in ceramic-type SEs, some promising ionic conductors such as NASICON-type Li1.5Al0.5Ti1.5(PO4)3 (LATP) and Li1.5Al0.5Ge1.5(PO4)3 (LAGP) are known to be unstable with lithium anodes,17 while other materials such as lithium phosphorus oxynitride (LiPON) decomposes upon lithium contact but enables a kinetic stabilization by forming reaction products that are themselves stable and of low electronic conductivity (e.g., Li3P, Li2O, and Li3N).15 The garnet-type oxide material Li7La3Zr2O12 (LLZO) is one of the few exceptional SEs with high reduction stability. It has a low intrinsic reduction limit at around 0.05–0.07 V vs. Li+/Li according to density functional theory calculations, which is very close to the electrochemical potential of lithium metal.18–20 The predicted threshold for electrolyte reduction is rather low (advantageously), however, it is still slightly above the potential of lithium metal. This small potential difference is probably not sufficient to overcome the nucleation barrier for side reaction products, thereby resulting in a kinetically stabilized interface.18 On the other hand, it has been suggested that reduction of LLZO by formation of a very thin lithiated21 or oxygen-deficient22,23 LLZO interphase could also result in kinetic passivation. In any case, it is apparent from experimental studies that LLZO is rather stable upon contact with lithium metal since no considerable time-dependent changes are observable in the impedance of a Li|LLZO interface,24,25 and no significant charge is accumulated in coulometric titration time analysis (CTTA) experiments.26 The high reduction stability of LLZO makes it a promising SE candidate for cells with lithium metal anodes, however, the rigid mechanical characteristics make the Li|LLZO interface more prone to contact issues – which can easily lead to higher interfacial resistance and heterogeneous lithium plating during operation.27 These issues can eventually lead to dendritic lithium growth due to microstructural defects and formation of microcracks, thereby posing a risk of short-circuit in LLZO-based cells.28,29 Additionally, transforming such oxide particles into a SE separator sheet necessitates high-temperature sintering,30 which leads to practical challenges and further chemical stability concerns during production. There have been significant research efforts in addressing these issues, however, given their importance, alternative SE materials have also attracted great interest. Among the alternatives, sulfide-based SEs are quite promising due to their high ionic conductivities and room temperature plasticity/sinterability.31 The thermodynamic stability window of sulfide-based SEs is comparably narrow, and they are not electrochemically stable upon contact with lithium metal.18 For instance, the superionic conductor Li10GeP2S12 (LGPS) has a very high ionic conductivity around 12 mS cm−1 and can be sintered at room temperature,32 but suffers from severe electrolyte side reactions when in contact with lithium metal due to the presence of electronically conductive components among the reaction products (e.g., Ge or LixGe alloys).33 The formation of such decomposition layers with sufficiently high electronic conductivity, i.e., mixed conducting interphase (MCI), would lead to continuous growth of these layers and result in fast cell failure, particularly when the lithium inventory of the cell is limited as in the case of “anode-free” cells. In the case of another important class of sulfide-based SEs, e.g., lithium argyrodites (Li6PS5X, X ¼ Cl, Br, I), the degree of side reactions is relatively low due to low electronic conductivity of decomposition products. The growth of decomposition layers (i.e., the SEI) is kinetically limited and follows a parabolic growth behavior due to diffusion-controlled reactions occurring at the Li|sulfide SE interface.

2.3.1 The impact of electrolyte reduction reactions In principle, it is not desirable to have irreversible electrolyte side reactions – in any electrochemical cell – since such reactions decrease the Coulomb efficiency and, in most cases, result in accelerated capacity fading. In the exemplary case of the LGPS SE, continuous reactions at the Li|SE interface would cause the formation of a comparably thick MCI that increases the cell resistance significantly. The estimates made via electrochemical impedance spectroscopy (EIS) measurements predict a resistance contribution

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

593

of RMCI  4.3 kO cm2 after one year of aging at the Li|LGPS interface.33 This value is rather high and would result in significant capacity fading of cells with lithium metal and LGPS SE. In the case of SEI-forming LPSCl SE, electrolyte reduction products are Li2S, LiCl, and Li3P.34–36 Due to the formation of significantly thinner layers of decomposition products, the resistance contribution is also notably lower, e.g., RSEI  0.3–0.4 kO cm2 (after one year of aging) for the lithium argyrodites Li6PS5Cl and Li6PS5Br.34 Even though the extent of side reactions is significantly reduced as compared to MCI-forming LGPS SE, it is still an obstacle for practical battery systems demanding long operational times. This shows a clear need for further research in order to optimize SEI properties for sulfide-based SEs. In most studies investigating the SEI growth, the experimental setup typically consists of a planar lithium metal anode and an SE pellet. Consequently, a uniform and planar interphase growth is often assumed. Even though such studies are quite useful in providing a fundamental understanding of reactivity between lithium metal and SE, lithium metal is plated in situ on the CC surface in “anodefree” cells. Electrolyte side reactions occurring spontaneously during lithium plating can exacerbate the non-uniform lithium plating anticipated in this cell configuration, and thereby result in even more severe side reactions. Due to factors such as nucleation barriers at the interface, imperfect contact between the CC and SE pellet, presence of contaminants, etc., dendritic lithium growth can be favored and further complicate the SEI-growth dynamics. Therefore, careful handling of these issues becomes even more vital in “anode-free” cells, necessitating a thorough understanding of side reaction mechanisms and SEI-passivation characteristics.

2.3.2 Lithium inventory loss in “anode-free” cells In academic research, it is reasonable to employ a simplified experimental setup to minimize the influence of secondary parameters, i.e., those not central to the experiment. Thus, if the primary objective is to study one particular electrode, in this case an “anode-free” current collector as the working electrode, it makes sense to use counter electrodes (CEs) that can reliably operate at a constant potential and with an excess of lithium. Thick lithium metal anodes, or even more preferably In/InLi alloy anodes, are suitable candidates for this purpose. Thus, it is not surprising that this experimental configuration is commonly employed in the literature. However, in such a cell configuration with excess lithium inventory, lithium losses due to electrolyte side reactions (or dead lithium formation) will be masked/compensated in each cycle from the excess lithium provided by the CE. In practice, the lithium inventory will be limited by the lithium content of the CAM in commercial cells, and thus, lithium compensation from the CE will not be possible. Therefore, it is important to account for such lithium losses – which are known to be quite substantial for both liquid and solid electrolyte-based “anode-free” cells.37–39 For these reasons, it is critical to quantitatively evaluate the reactivity of SEs with respect to lithium metal in “anode-free” cells. Indirect quantitative estimations of side reactions (i.e., SEI growth) can be made by EIS analysis of the Li|SE interface.34 However, these estimates are based on assumptions for the effective ionic conductivities of SEI films, which cannot be readily determined accurately. Direct quantifications are possible via electrochemical measurements performed in “anode-free” cell configuration such as the recently demonstrated CTTA technique,26 or post-mortem characterization approaches such as timeof-flight secondary ion mass spectrometry (TOF-SIMS) combined with atomic force microscopy (AFM),40 titration gas chromatography,41,42 or mass spectrometry titration.43,44 Experiments performed without lithium cycling (i.e., static aging experiments) reveal mainly the intrinsic reactivity of the electrolyte while dynamic experiments such as Coulomb efficiency tests can provide information on the overall effect of lithium loss in SEI growth, “dead lithium” formation, and kinetic losses. It should be noted that lithium inventory losses on the negative electrode can also be masked to some degree in “anode-free” full cells as a result of kinetic discharge capacity loss of the positive electrode,45 or electrolyte oxidation reactions.46 Therefore, it is important to complement galvanostatic cycling experiments with other electrochemical and analytical characterization methods.

2.3.3 Overcoming electrolyte instability The most straightforward approach to address the issues caused by electrolyte instability is to protect the surface of active electrode material with an electrochemically/chemically stable coating, or with a coating that forms an artificial SEI. Ideally, this coating should be thin, conformal, and possess low electronic conductivity while maintaining high ionic conductivity. In the case of “anode-free” cells, since the active material (i.e., lithium metal) is plated in situ during the operation, implementing a coating approach is not so straightforward. Instead, a suitable material can be used to coat the surface of the CC prior to cell assembly. However, the protective coating must withstand the significant volume change caused by lithium plating. There is also a risk of lithium penetration through the coating film if dendritic lithium growth occurs. Therefore, it is important to use coating films that not only protect against side reactions but are also mechanically durable against local volume fluctuations and related local stress. In previous studies, different coating approaches have been studied for “anode-free” cells, such as coatings with polyethylene oxide (PEO),47 polyvinylidene fluoride PVDF,48,49 or metal fluorides.50 An alternative approach is to modify the SE composition by doping with specific elements and design an electrolyte that can form an SEI film with optimized passivation characteristics. As discussed in previous sections, in the case of SEs with metal elements other than lithium, the stability with lithium metal is usually low due to MCI formation. The presence of these metals can be beneficial for ion transport kinetics, but usually at the expense of accelerated side reactions with lithium metal. Instead, the choice of metal cations such as Ca2+ and La3+ as dopants can increase the stability since such metal cations are more resistant against reduction. Also, increasing the fraction of cations such as P5+ and H+ or increasing the degree of hybridization by anion substitution may also be a strategy to mitigate the side reactions.18,51,52 The complex morphology of lithium plating/stripping, combined with the significant volume changes and related local stresses, results only in partial success of such approaches in “anode-free” cells. Further research is necessary to optimize such cells – likely involving multiple strategies.

594

3 3.1

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

Possible realizations of the “anode-free” cell design Planar metal current collector|solid electrolyte interfaces

The most basic design strategy to realize the negative electrode in an “anode-free” cell configuration solely consists of a CC sheet contacted to the SE as depicted in Fig. 1D.53,54 As CC metal, copper is frequently used due to its straightforward handling both as a foil and in PVD techniques, e.g., in combination with LLZO or LIPON. Using non-alloying metals such as copper (which dissolves lithium, but very slowly) or nickel furthermore has the benefit that irreversible lithium loss due to incorporation into the CC is strongly reduced. Thereby the volumetric energy density of the cell is maximized, as the full charge stored in the CAM can be electrodeposited.53 One challenge associated with using a simple planar configuration is the control of the lithium morphology during plating. Frequent issues include the formation of lithium whiskers, dendrites, and growth of isolated islands at the interface.37,55,56 Whisker formation happens, when a lithium island is growing and penetrating the CC, thereby reducing the rate capability in further cycles and facilitating the growth of dendrites. All of the mentioned effects may lead to a current focusing at the interface and therefore further dendrite growth, ultimately followed by a hard short-circuit of the cell. Several strategies exist to control the microstructure of plated lithium, such as applying stack pressure or pulsed plating protocols. To understand these strategies, first the general lithium formation needs to be understood, which can be split into two periods (see Fig. 2). At the beginning, lithium nuclei need to form, which usually requires to overcome a nucleation overpotential of several tens of mV. After the formation of nuclei at the interface, they continue to grow in the second period without further nucleation. In the best-case scenario, a high number of spatially dispersed nuclei is formed, which then continue to grow and thereby form a lithium film. One method to control this process therefore is to change the lithium nucleation density, which frequently is employed by increasing the current density either initially or throughout the plating as pulses. Thereby, the nucleation overpotential is increased, leading to more lithium nuclei with a smaller size as shown by several works on the Cu|LLZO as well as Cu|LIPON interface.37,57 However, this strategy simultaneously introduces a higher risk of dendrite formation during cycling at high current densities, which is why it needs to carefully balance the benefits of a more homogeneous nucleation and the issues of facilitated dendrite growth. Another way of changing the nucleation density is by a deliberate change of temperature. For example, Motoyama et al. were able to show a strong nucleation density increase at lower temperatures due to a higher overvoltage for nucleation at the Cu|LIPON interface.57 A third strategy works by employing artificial seed layers to increase the nucleation density, as recently investigated by Haslam et al. by using Au particles at the Cu|LLZO interface.58 Other strategies, however, rather target the lithium growth period instead of the lithium nucleation period to form lithium films. For this, a rather thick CC is needed to suppress the penetration thereof, as shown for thin CC.37,55 However, the use of a thick CC is also challenging because of the worse contact compared to evaporated thin Cu films. As discussed in the next chapter, pressure can be a suitable tool to guide the lithium growth in film-like morphology during the growth phase by plastic deformation of the plated whiskers.53,59 One challenge associated with any “anode-free” cell concept is the high volume change during cycling, as usually no host for the lithium is present at the anode side. Assuming a practical application requires 5 mA h cm−2, at least 25 mm are plated and stripped in each charge-discharge cycle,60,61 which could amount to height changes in the mm-range of a multilayered cell stack. An uneven pressure distribution both spatially and in time may follow, thereby creating favorable conditions for dendrite initiation and growth.

Fig. 2 Typical voltage profile during plating at a planar CC|SE interface with a non-alloying CC. Two phases can be identified being the nucleation as characterized by a distinct voltage minimum Nuc and the growth phase, which can be identified by a flat voltage plateau with Plat. The schematics below indicate that during period 1, the nucleation of dispersed lithium particles occur, which can then grow into a film-like morphology during period 2. Yet completely unknown is the microstructure, e.g., grain size and orientation, of the plated lithium.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

595

Another concern when using the concept of a layered, planar arrangement of a CC at an SE is that plated lithium anode can bring issues during the subsequent stripping. Pore formation due to insufficient vacancy diffusion coefficients within the lithium metal so far limit the available discharge capacity to only around 1.0 mA h cm−2 even for current densities as low as 0.1 mA cm−2.25,62–64 It is yet unknown if plated lithium metal may enable higher stripping capacities, e.g., because of a vastly different microstructure. As the microstructure can have a profound impact on the discharge performance,65–67 it is of high interest to analyze the plated lithium and possible parameters controlling the microstructure in future studies. The most promising lithium plating performance at planar CC|SE combinations so far all involve the use of a thick (>20 mm) metal foil,53,59,68 despite the difficulty of achieving a good contact to the SE. It seems to be more important to have a mechanically robust CC that mitigates a mechanical failure thereof. Achieving a good contact between malleable thiophosphate SEs is possible when using high formation pressures. For example, Sandoval et al. used 15 MPa both for the preparation and the lithium deposition in Cu|Li6PS5Cl|Li cells to obtain conformal interfaces.68 This enabled the deposition of 10 mm of lithium with a film-like morphology at 0.25 mA cm−2, which was not possible when using thin CC films on oxide SEs. Depositing a film-like morphology on oxide LLZO-type SEs is made possible by laminating a Cu CC onto the pellet at around 900  C and 50 MPa as showcased by the groups of Sakamoto and Dasgupta.53,59 Here, a good conformal contact to the SE can be combined with a mechanical robust CC. Interfaces prepared in this manner can be used to deposit several tens of microns of lithium in film-like morphology at >0.5 mA cm−2. A similar concept is supposedly used by QuantumScape company, thereby enabling high-rate lithium cycling up to several mA cm−2.69 However, it is unclear how the CC is prepared in detail and how the surface of the oxide separator was modified.

3.2

Carbonaceous seed interlayers

An alternative design strategy to enable the film-like lithium growth in “anode-free” cells is pioneered by Lee et al. and is schematically depicted in Fig. 1E, which uses an engineered CC based on a carbon-matrix and incorporated silver nanoparticles.8 While the exact working function of its individual components is still debated in literature, several clear differences can be discussed when compared to using planar non-alloying CC. One advantage is the easy fabrication of cells with an Ag-C CC sheet, which is softer than traditional metal foils. This means that no high temperature/high pressure preparation steps are needed, as it is currently the case for Cu|LLZO alternative designs. Additionally, it seems that these engineered CC homogenize the lithium morphology during plating to be film-like, without the formation of whiskers, dendrites or other effects typically present for planar metal|SE interfaces. On the other hand, the energy density of the cell is lowered due to the employment of an additional layer at the CC side of the cell. Nevertheless, Lee et al. were able to successfully cycle a prototype pouch cell with 0.6 Ah over 1000 cycles with a high-Ni cathode active material, exhibiting an energy density of >900 Wh l−1 at 0.5  C and 60  C. The exact working principle of the Ag-C interlayer concept is, however, still highly debated in literature. An immediate question that arises when examining the concept tackles the question of why the lithium is deposited after passing the Ag-C composite layer, although this layer exhibits a high electronic conductivity. At first glance, a charge transfer and therefore lithium plating would be expected at the Ag-C|SE interface. The cross-sectional analysis still undoubtedly reveals lithium being plated after passing through the nanocomposite. A follow-up study by Kim et al. discussed this question with regards to the different interfacial adhesion energies between the interlayer and the SE or the metal substrate.70 The two cases are defined with lithium either plating before passing the interlayer or afterwards and the respective interfacial energies are calculated. Their analysis showed that lithium will pass through an amorphous carbon layer before depositing, while graphite will lead to a plating before passing. Additionally, the interfacial energies can be tuned via changing the temperature. Using a temperature of 100  C, it was possible to direct the deposition of lithium after passing through a graphite interlayer due to changes in the diffusion properties as well as interfacial energies. Furthermore, the role of the Ag nanoparticles in the initial concept is discussed by Spencer-Jolly et al. to be the reduction of the nucleation overpotential as well as mitigation of dendrite formation during plating.71 Their operando X-ray diffraction analysis shows that graphite is first lithiated during charging, which is followed by a chemical reaction with the Ag nanoparticles to form Ag-Li alloys. However, if the charging rate is high enough so that the diffusivity inside the graphite is insufficient, lithium will directly nucleate unevenly. This conclusion suggests that the Ag nanoparticles are only useful in combination with a carbon matrix. On the other hand, yet to be peer-reviewed analyses by the group of Ceder question the necessity of a carbon matrix.72 They instead propose that the silver particles are first lithiated, followed by a deposition of lithium at the CC and subsequent dissolution of the silver into the lithium, thereby homogenizing the morphology through the volume expansion by alloy formation. The carbon matrix only serves as a support for the silver particles and potentially to guide the nucleation. However, experiments by Haslam et al. further show that homogeneous lithium plating is also possible by using Au particles at the interface between Cu and LLZO without the use of a carbon matrix.58 Follow-up analyses are required to completely elucidate the mechanism of lithium plating at these carbon-based matrices filled with some form of metal particles.

3.3

Three-dimensional concepts of current collector|electrolyte interfaces

In a typical “anode-free” cell configuration, the role of the CC is to serve as a planar substrate for lithium deposition. If the morphological irregularities of the plated lithium are neglected, then the negative electrode (i.e., freshly deposited lithium metal

596

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

film) can be considered to be in planar geometry during the operation of the cell. Current density is widely recognized as a critical factor influencing the tendency toward dendritic lithium growth,73 and extensive research efforts focus on preventing its occurrence at high current densities. On the other hand, an alternative strategy would be to reduce the effective current density by increasing the electrode-electrolyte contact surface area, while maintaining the absolute current constant. In this direction, the implementation of 3D-structured CC as depicted in Fig. 1F has been shown to improve the cycling performance at reasonably high current densities.27,73 In such strategies, to achieve higher power capabilities, it is essential for the SE to adequately infiltrate the CC framework – ensuring an efficient ionic transport pathway. The volume change occurring during lithium plating and stripping is another critical parameter affecting the cycling performance. Such changes can compromise the contact between the electrolyte and electrode and result in an increased interfacial resistance. It is therefore crucial to accommodate the volume expansion during lithium plating. The use of 3D-structured porous electrode frameworks (with dedicated empty space for lithium plating) is a sensible design approach to accommodate volume changes. In such designs, however, the use of a framework that allows good ionic and electronic conduction is advantageous to ensure uniform lithium plating throughout the entire structure, and to reach high Coulomb efficiency. This approach was adopted previously for alloy-based negative electrodes as in the case of partially dealloyed Li-Mg alloy on a garnet electrolyte, and resulted in significant improvement in cycling performance since this alloy can act as an electron and ion conductive framework during cell cycling.74 Alternatively, an ionically conductive and mechanically rigid framework can subsequently be coated to make its surface electronically conductive.75 It should be noted that such cell architectures also provide significant advantages in terms of cell stack pressure requirements, but at the expense of energy density. In a recent study, Alexander et al. proposed the use of a porous ‘single-phase’ mixed ion/electron conducting (MIEC) garnet as a CC framework for lithium metal anodes (see Fig. 3).76 As demonstrated in this figure, this MIEC framework ensures more uniform lithium plating throughout the pores since lithium plating is not limited to regions close to the CC. In this study, the single-phase MIEC garnet was prepared by replacing La with Pr, and further doping the material with Ga and Ce in order to maintain the structure and ensure high electronic/ionic conductivity.76 The cycling performance of symmetrical cells (prepared by infiltrating liquid metal into the ZnO-coated pores) has demonstrated the highest reported values for critical current density of 100 mA cm−2, and galvanostatic cycling currents and capacities (60 mA cm−2 and 30 mA h cm−2, respectively). These results are very promising and make this cell architecture an extremely interesting candidate also for “anode-free” cells as this can help to increase energy density of such cell architectures.

Fig. 3 Schematic demonstration of lithium plating and stripping behavior inside the porous framework (A) without the MIEC garnet and (B) with the MIEC garnet. Reproduced with permission from Springer Nature Alexander, G.V; Shi, C.; O’Neill, J.; Wachsman, E.D. Extreme Lithium-Metal Cycling Enabled by a Mixed Ion- and Electron-Conducting Garnet Three-Dimensional Architecture. Nat. Mater. 2023. https://doi.org/10.1038/s41563-023-01627-9.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

4

597

Summary and outlook

The alkali metal anode marks a cornerstone in the strategy for future rechargeable batteries with superior specific energy and energy density. It has been investigated since a long time in cells based on liquid electrolytes, was early commercialized, but after critical failures there is yet no new commercial approach apparent. Thus, it is an open question whether any surface modification or thin surface layer can achieve stable operation of alkali metal anodes at high enough current densities under both plating and stripping conditions. For this reason, cell concepts with SE separators or all-solid-state cell concepts are meanwhile considered as necessary for the successful implementation of the metal anode. In the early stage of this development, the mechanical rigidity of the SE was considered a major factor for the stability of the lithium metal anode and the suppression of dendrite growth. Meanwhile, the understanding of the metal anode has advanced considerably, and it became clear that the kinetics of the alkali metal anode is far more complicated than originally expected. Commercial cells with metal anodes are yet only available in the case of lithium with polymer electrolyte operating at elevated temperature. The elevated temperature is required to achieve a sufficiently high electrolyte conductivity, yet it also results in softening of the polymer which then limits the current density during charging and does not allow fast charging. Serious efforts in research and development of cell concepts with inorganic SE separators, either in combination with liquid or gel polymer catholytes or in all-solid-state cells are widely seen today, attempting to overcome the limitations of lithium metal-polymer batteries. The high conductivity and improved chemical stability of lithium metal that can be achieved with advanced inorganic SEs is the basis for these developments. Meanwhile it is obvious that several effects and specific properties of mostly crystalline inorganic SEs are critical hurdles on the way to reversible high-rate alkali metal anodes with minor loss of lithium in side reactions. Firstly, the most promising SEs are polycrystalline, and their grain boundaries can be favorable regions for growth of metal filaments or for mechanical failure and subsequent dendrite growth. Thus, the study of grain boundaries and their design for improved resistance against any kind of metal penetration or cracking is of prime interest in the future. Even if single crystalline or glassy electrolytes would be employed, dislocations and structural inhomogeneities appear to open pathways for dendrite initiation and growth. Secondly, most inorganic SEs are reduced by their corresponding alkali metal, which is meanwhile well documented in the case of lithium and sodium. This leads in favorable cases to the limited growth of an SEI, that allows stable operation, however, which also consumes a certain amount of lithium and affects the lithium inventory. To overcome this, thin protecting layers need to be applied which would complicate the production process, or intrinsically stable SEs need to be utilized. This is the reason for the strong interest and research activity in the development of lithium garnet-based separators. These materials are practically stable against lithium metal and may offer stable operation without loss of lithium. Whether this stable operation can be achieved reliably at high operation currents for sufficiently thick metal layers is one of the most pressing questions today. If it could be achieved, the “anode-free” concept would offer superior specific energies and energy densities, as well as fast kinetics. This would indeed be a major step forward about 30 years after the commercialization of the LIB. Currently, three design strategies are pursued for “anode-free” cell concepts, of which two appear as promising. Either an effective seed layer is being used that allows homogeneous plating and stripping, or a 3D host layer is being used. While the seed layer concept maintains the advantage of a reduced volume, the 3D host layer concept may be more reliable and high-rate compatible – at the cost of losing the volume advantage. In view of all available reports, “anode-free” cell concepts appear as promising and useful, and there is no physicochemical reason why the reversible cycling of alkali metal anodes with desired thickness (in the range of a few 10 mm) should not be successful at the end. Probably the major problem will be to upscale successful lab-scale solutions to the industrial scale.

Acknowledgments We acknowledge financial support by the German Federal Ministry of Education and Research (BMBF) under the projects FB2-Koord (03XP0431), FB2-Char (03XP0433D), and LiSI2 (03XP0509B).

References 1. Janek, J.; Zeier, W. G. A Solid Future for Battery Development. Nat. Energy 2016, 1 (9), 16141. https://doi.org/10.1038/nenergy.2016.141. 2. Xu, J.; Cai, X.; Cai, S.; Shao, Y.; Hu, C.; Lu, S.; Ding, S. High-Energy Lithium-Ion Batteries: Recent Progress and a Promising Future in Applications. Energy Environ. Mater. 2023, 6 (5), 1–26. https://doi.org/10.1002/eem2.12450. 3. Heubner, C.; Maletti, S.; Auer, H.; Hüttl, J.; Voigt, K.; Lohrberg, O.; Nikolowski, K.; Partsch, M.; Michaelis, A. From Lithium-Metal toward Anode-Free Solid-State Batteries: Current Developments, Issues, and Challenges. Adv. Funct. Mater. 2021, 31 (51). https://doi.org/10.1002/adfm.202106608. 4. Wang, H.; Yu, Z.; Kong, X.; Kim, S. C.; Boyle, D. T.; Qin, J.; Bao, Z.; Cui, Y. Liquid Electrolyte: The Nexus of Practical Lithium Metal Batteries. Joule 2022, 6 (3), 588–616. https://doi.org/10.1016/j.joule.2021.12.018. 5. Schmaltz, T.; Hartmann, F.; Wicke, T.; Weymann, L.; Neef, C.; Janek, J. A Roadmap for Solid-State Batteries. Adv. Energy Mater. 2023, 2301886. https://doi.org/ 10.1002/aenm.202301886.

598

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

6. Bates, A. M.; Preger, Y.; Torres-Castro, L.; Harrison, K. L.; Harris, S. J.; Hewson, J. Are Solid-State Batteries Safer than Lithium-Ion Batteries? Joule 2022, 6 (4), 742–755. https://doi.org/10.1016/j.joule.2022.02.007. 7. Cheng, X.-B.; Zhao, C.-Z.; Yao, Y.-X.; Liu, H.; Zhang, Q. Recent Advances in Energy Chemistry between Solid-State Electrolyte and Safe Lithium-Metal Anodes. Chem 2019, 5 (1), 74–96. https://doi.org/10.1016/j.chempr.2018.12.002. 8. Lee, Y.; Fujiki, S.; Jung, C.; Suzuki, N.; Yashiro, N.; Omoda, R.; Ko, D.; Shiratsuchi, T.; Sugimoto, T.; Ryu, S.; Ku, J. H.; Watanabe, T.; Park, Y.; Aihara, Y.; Im, D.; Han, I. T. High-Energy Long-Cycling All-Solid-State Lithium Metal Batteries Enabled by Silver-Carbon Composite Anodes. Nat. Energy 2020, 5, 299–308. https://doi.org/10.1038/ s41560-020-0575-z. 9. Li, Q.; Yang, Y.; Yu, X.; Li, H. A 700 Whkg −1 Rechargeable Pouch Type Lithium Battery. Chinese Phys. Lett. 2023, 40 (4), 048201. https://doi.org/10.1088/0256-307X/40/4/ 048201. 10. Yang, T.; Luo, D.; Liu, Y.; Yu, A.; Chen, Z. Anode-Free Sodium Metal Batteries as Rising Stars for Lithium-Ion Alternatives. iScience 2023, 26 (3), 105982. https://doi.org/ 10.1016/j.isci.2023.105982. 11. Tian, Y.; An, Y.; Wei, C.; Jiang, H.; Xiong, S.; Feng, J.; Qian, Y. Recently Advances and Perspectives of Anode-Free Rechargeable Batteries. Nano Energy 2020, 78 (April), 105344. https://doi.org/10.1016/j.nanoen.2020.105344. 12. Cohn, A. P.; Muralidharan, N.; Carter, R.; Share, K.; Pint, C. L. Anode-Free Sodium Battery through in Situ Plating of Sodium Metal. Nano Lett. 2017, 17 (2), 1296–1301. https://doi.org/10.1021/acs.nanolett.6b05174. 13. Otto, S. K.; Moryson, Y.; Krauskopf, T.; Peppler, K.; Sann, J.; Janek, J.; Henss, A. In-Depth Characterization of Lithium-Metal Surfaces with XPS and ToF-SIMS: Toward Better Understanding of the Passivation Layer. Chem. Mater. 2021, 33 (3), 859–867. https://doi.org/10.1021/acs.chemmater.0c03518. 14. Otto, S. K.; Fuchs, T.; Moryson, Y.; Lerch, C.; Mogwitz, B.; Sann, J.; Janek, J.; Henss, A. Storage of Lithium Metal: The Role of the Native Passivation Layer for the Anode Interface Resistance in Solid State Batteries. ACS Appl. Energy Mater. 2021, 4 (11), 12798–12807. https://doi.org/10.1021/acsaem.1c02481. 15. Cheng, D.; Wynn, T. A.; Wang, X.; Wang, S.; Zhang, M.; Shimizu, R.; Bai, S.; Nguyen, H.; Fang, C.; Kim, M.; Li, W.; Lu, B.; Kim, S. J.; Meng, Y. S. Unveiling the Stable Nature of the Solid Electrolyte Interphase between Lithium Metal and LiPON via Cryogenic Electron Microscopy. Joule 2020, 4 (11), 2484–2500. https://doi.org/10.1016/j. joule.2020.08.013. 16. Hatzell, K. B.; Chen, X. C.; Cobb, C. L.; Dasgupta, N. P.; Dixit, M. B.; Marbella, L. E.; McDowell, M. T.; Mukherjee, P. P.; Verma, A.; Viswanathan, V.; Westover, A. S.; Zeier, W. G. Challenges in Lithium Metal Anodes for Solid-State Batteries. ACS Energy Lett. 2020, 5 (3), 922–934. https://doi.org/10.1021/acsenergylett.9b02668. 17. Hartmann, P.; Leichtweiss, T.; Busche, M. R.; Schneider, M.; Reich, M.; Sann, J.; Adelhelm, P.; Janek, J. Degradation of NASICON-Type Materials in Contact with Lithium Metal: Formation of Mixed Conducting Interphases (MCI) on Solid Electrolytes. J. Phys. Chem. C 2013, 117 (41), 21064–21074. https://doi.org/10.1021/jp4051275. 18. Xiao, Y.; Wang, Y.; Bo, S.-H.; Kim, J. C.; Miara, L. J.; Ceder, G. Understanding Interface Stability in Solid-State Batteries. Nat. Rev. Mater. 2019, 5 (2), 105–126. https://doi.org/ 10.1038/s41578-019-0157-5. 19. Richards, W. D.; Miara, L. J.; Wang, Y.; Kim, J. C.; Ceder, G. Interface Stability in Solid-State Batteries. Chem. Mater. 2016, 28 (1), 266–273. https://doi.org/10.1021/acs. chemmater.5b04082. 20. Zhu, Y.; He, X.; Mo, Y. Origin of Outstanding Stability in the Lithium Solid Electrolyte Materials: Insights from Thermodynamic Analyses Based on First-Principles Calculations. ACS Appl. Mater. Interfaces 2015, 7 (42), 23685–23693. https://doi.org/10.1021/acsami.5b07517. 21. Ma, C.; Cheng, Y.; Yin, K.; Luo, J.; Sharafi, A.; Sakamoto, J.; Li, J.; More, K. L.; Dudney, N. J.; Chi, M. Interfacial Stability of Li Metal-Solid Electrolyte Elucidated via in Situ Electron Microscopy. Nano Lett. 2016, 16 (11), 7030–7036. https://doi.org/10.1021/acs.nanolett.6b03223. 22. Zhu, Y.; Connell, J. G.; Tepavcevic, S.; Zapol, P.; Garcia-Mendez, R.; Taylor, N. J.; Sakamoto, J.; Ingram, B. J.; Curtiss, L. A.; Freeland, J. W.; Fong, D. D.; Markovic, N. M. Dopant-Dependent Stability of Garnet Solid Electrolyte Interfaces with Lithium Metal. Adv. Energy Mater. 2019, 9 (12), 1803440. https://doi.org/10.1002/aenm.201803440. 23. Connell, J. G.; Fuchs, T.; Hartmann, H.; Krauskopf, T.; Zhu, Y.; Sann, J.; Garcia-Mendez, R.; Sakamoto, J.; Tepavcevic, S.; Janek, J. Kinetic versus Thermodynamic Stability of LLZO in Contact with Lithium Metal. Chem. Mater. 2020, 32 (23), 10207–10215. https://doi.org/10.1021/acs.chemmater.0c03869. 24. Kim, Y.; Yoo, A.; Schmidt, R.; Sharafi, A.; Lee, H.; Wolfenstine, J.; Sakamoto, J. Electrochemical Stability of Li6.5La3Zr1.5M0.5O12 (M ¼ Nb or Ta) against Metallic Lithium. Front. Energy Res. 2016, 4 (MAY), 1–7. https://doi.org/10.3389/fenrg.2016.00020. 25. Krauskopf, T.; Hartmann, H.; Zeier, W. G.; Janek, J. Toward a Fundamental Understanding of the Lithium Metal Anode in Solid-State Batteries - An Electrochemo-Mechanical Study on the Garnet-Type Solid Electrolyte Li6.25Al0.25La3Zr2O12. ACS Appl. Mater. Interfaces 2019, 11 (15), 14463–14477. https://doi.org/10.1021/acsami.9b02537. 26. Aktekin, B.; Riegger, L. M.; Otto, S.-K.; Fuchs, T.; Henss, A.; Janek, J. SEI Growth on Lithium Metal Anodes in Solid-State Batteries Quantified with Coulometric Titration Time Analysis. Nat. Commun. 2023, 14 (1), 6946. https://doi.org/10.1038/s41467-023-42512-y. 27. Wang, C.; Fu, K.; Kammampata, S. P.; McOwen, D. W.; Samson, A. J.; Zhang, L.; Hitz, G. T.; Nolan, A. M.; Wachsman, E. D.; Mo, Y.; Thangadurai, V.; Hu, L. Garnet-Type Solid-State Electrolytes: Materials, Interfaces, and Batteries. Chem. Rev. 2020, 120 (10), 4257–4300. https://doi.org/10.1021/acs.chemrev.9b00427. 28. Ren, Y.; Shen, Y.; Lin, Y.; Nan, C.-W. Direct Observation of Lithium Dendrites inside Garnet-Type Lithium-Ion Solid Electrolyte. Electrochem. commun. 2015, 57, 27–30. https://doi.org/10.1016/j.elecom.2015.05.001. 29. Shen, F.; Dixit, M. B.; Xiao, X.; Hatzell, K. B. Effect of Pore Connectivity on Li Dendrite Propagation within LLZO Electrolytes Observed with Synchrotron X-Ray Tomography. ACS Energy Lett. 2018, 3 (4), 1056–1061. https://doi.org/10.1021/acsenergylett.8b00249. 30. Wu, T.-T.; Guo, S.; Li, B.; Shen, C.-Y.; Liu, X.-H.; Cao, A.-M. Garnet-Type Solid-State Electrolytes: Crystal Structure, Interfacial Challenges and Controlling Strategies. Rare Met. 2023, 42 (10), 3177–3200. https://doi.org/10.1007/s12598-023-02323-y. 31. Sakuda, A.; Hayashi, A.; Tatsumisago, M. Sulfide Solid Electrolyte with Favorable Mechanical Property for All-Solid-State Lithium Battery. Sci. Rep. 2013, 3 (1), 2261. https://doi. org/10.1038/srep02261. 32. Kamaya, N.; Homma, K.; Yamakawa, Y.; Hirayama, M.; Kanno, R.; Yonemura, M.; Kamiyama, T.; Kato, Y.; Hama, S.; Kawamoto, K.; Mitsui, A. A Lithium Superionic Conductor. Nat. Mater. 2011, 10 (9), 682–686. https://doi.org/10.1038/nmat3066. 33. Wenzel, S.; Randau, S.; Leichtweiß, T.; Weber, D. A.; Sann, J.; Zeier, W. G.; Janek, J. Direct Observation of the Interfacial Instability of the Fast Ionic Conductor Li 10 GeP 2 S 12 at the Lithium Metal Anode. Chem. Mater. 2016, 28 (7), 2400–2407. https://doi.org/10.1021/acs.chemmater.6b00610. 34. Wenzel, S.; Sedlmaier, S. J.; Dietrich, C.; Zeier, W. G.; Janek, J. Interfacial Reactivity and Interphase Growth of Argyrodite Solid Electrolytes at Lithium Metal Electrodes. Solid State Ionics 2018, 318, 102–112. https://doi.org/10.1016/j.ssi.2017.07.005. 35. Schwietert, T. K.; Arszelewska, V. A.; Wang, C.; Yu, C.; Vasileiadis, A.; de Klerk, N. J. J.; Hageman, J.; Hupfer, T.; Kerkamm, I.; Xu, Y.; van der Maas, E.; Kelder, E. M.; Ganapathy, S.; Wagemaker, M. Clarifying the Relationship between Redox Activity and Electrochemical Stability in Solid Electrolytes. Nat. Mater. 2020, 19 (4), 428–435. https://doi.org/10.1038/s41563-019-0576-0. 36. Tan, D. H. S.; Wu, E. A.; Nguyen, H.; Chen, Z.; Marple, M. A. T.; Doux, J. M.; Wang, X.; Yang, H.; Banerjee, A.; Meng, Y. S. Elucidating Reversible Electrochemical Redox of Li6PS5Cl Solid Electrolyte. ACS Energy Lett. 2019, 2418–2427. https://doi.org/10.1021/acsenergylett.9b01693. 37. Fuchs, T.; Becker, J.; Haslam, C. G.; Lerch, C.; Sakamoto, J.; Richter, F. H.; Janek, J. Current-Dependent Lithium Metal Growth Modes in “Anode-Free” Solid-State Batteries at the Cu|LLZO Interface. Adv. Energy Mater. 2023, 13 (1), 2203174. https://doi.org/10.1002/aenm.202203174. 38. Krauskopf, T.; Mogwitz, B.; Hartmann, H.; Singh, D. K.; Zeier, W. G.; Janek, J. The Fast Charge Transfer Kinetics of the Lithium Metal Anode on the Garnet-Type Solid Electrolyte Li6.25Al0.25La3Zr2O12. Adv. Energy Mater. 2020, 10 (27), 2000945. https://doi.org/10.1002/aenm.202000945. 39. Huang, W.; Zhao, C.; Wu, P.; Yuan, H.; Feng, W.; Liu, Z.; Lu, Y.; Sun, S.; Fu, Z.; Hu, J.; Yang, S.; Huang, J.; Zhang, Q. Anode-Free Solid-State Lithium Batteries: A Review. Adv. Energy Mater. 2022, 12 (26), 1–16. https://doi.org/10.1002/aenm.202201044. 40. Otto, S.; Riegger, L. M.; Fuchs, T.; Kayser, S.; Schweitzer, P.; Burkhardt, S.; Henss, A.; Janek, J. In Situ Investigation of Lithium Metal–Solid Electrolyte Anode Interfaces with ToF-SIMS. Adv. Mater. Interfaces 2022, 9 (13), 2102387. https://doi.org/10.1002/admi.202102387. 41. Fang, C.; Li, J.; Zhang, M.; Zhang, Y.; Yang, F.; Lee, J. Z.; Lee, M.-H.; Alvarado, J.; Schroeder, M. A.; Yang, Y.; Lu, B.; Williams, N.; Ceja, M.; Yang, L.; Cai, M.; Gu, J.; Xu, K.; Wang, X.; Meng, Y. S. Quantifying Inactive Lithium in Lithium Metal Batteries. Nature 2019, 572 (7770), 511–515. https://doi.org/10.1038/s41586-019-1481-z.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Negative Electrodes - “Anode Free”

599

42. Hobold, G. M.; Gallant, B. M. Quantifying Capacity Loss Mechanisms of Li Metal Anodes beyond Inactive Li0. ACS Energy Lett. 2022, 7 (10), 3458–3466. https://doi.org/ 10.1021/acsenergylett.2c01845. 43. Tao, M.; Xiang, Y.; Zhao, D.; Shan, P.; Yang, Y. Protocol for Quantifying Inactive Lithium in Anode-Free Lithium Batteries by Mass Spectrometry Titration. Commun. Mater. 2022, 3 (1), 50. https://doi.org/10.1038/s43246-022-00273-z. 44. Xiang, Y.; Tao, M.; Zhong, G.; Liang, Z.; Zheng, G.; Huang, X.; Liu, X.; Jin, Y.; Xu, N.; Armand, M.; Zhang, J. G.; Xu, K.; Fu, R.; Yang, Y. Quantitatively Analyzing the Failure Processes of Rechargeable Li Metal Batteries. Sci. Adv. 2021, 7 (46), 1–12. https://doi.org/10.1126/sciadv.abj3423. 45. Nanda, S.; Gupta, A.; Manthiram, A. Anode-Free Full Cells: A Pathway to High-Energy Density Lithium-Metal Batteries. Adv. Energy Mater. 2021, 11 (2), 1–18. https://doi.org/ 10.1002/aenm.202000804. 46. Aktekin, B.; Lacey, M. J.; Nordh, T.; Younesi, R.; Tengstedt, C.; Zipprich, W.; Brandell, D.; Edström, K. Understanding the Capacity Loss in LiNi0.5Mn1.5O4-Li4Ti5O12 Lithium-Ion Cells at Ambient and Elevated Temperatures. J. Phys. Chem. C 2018, 122 (21), 11234–11248. https://doi.org/10.1021/acs.jpcc.8b02204. 47. Assegie, A. A.; Cheng, J. H.; Kuo, L. M.; Su, W. N.; Hwang, B. J. Polyethylene Oxide Film Coating Enhances Lithium Cycling Efficiency of an Anode-Free Lithium-Metal Battery. Nanoscale 2018, 10 (13), 6125–6138. https://doi.org/10.1039/c7nr09058g. 48. Tamwattana, O.; Park, H.; Kim, J.; Hwang, I.; Yoon, G.; Hwang, T. H.; Kang, Y. S.; Park, J.; Meethong, N.; Kang, K. High-Dielectric Polymer Coating for Uniform Lithium Deposition in Anode-Free Lithium Batteries. ACS Energy Lett. 2021, 6 (12), 4416–4425. https://doi.org/10.1021/acsenergylett.1c02224. 49. Abrha, L. H.; Nikodimos, Y.; Weldeyohannes, H. H.; Hagos, T. T.; Wang, D. Y.; Huang, C. J.; Jiang, S. K.; Wu, S. H.; Su, W. N.; Tsai, M. C.; Hwang, B. J. Effects of a Thermally Electrochemically Activated b-PVDF Fiber on Suppression of Li Dendrite Growth for Anode-Free Batteries. ACS Appl. Energy Mater. 2021, 4 (4), 3240–3248. https://doi.org/ 10.1021/acsaem.0c03015. 50. Lee, J.; Choi, S. H.; Im, G.; Lee, K. J.; Lee, T.; Oh, J.; Lee, N.; Kim, H.; Kim, Y.; Lee, S.; Choi, J. W. Room-Temperature Anode-Less All-Solid-State Batteries via the Conversion Reaction of Metal Fluorides. Adv. Mater. 2022, 34 (40), 1–9. https://doi.org/10.1002/adma.202203580. 51. Nolan, A. M.; Zhu, Y.; He, X.; Bai, Q.; Mo, Y. Computation-Accelerated Design of Materials and Interfaces for All-Solid-State Lithium-Ion Batteries. Joule 2018, 2 (10), 2016–2046. https://doi.org/10.1016/j.joule.2018.08.017. 52. Tian, Y.; Sun, Y.; Hannah, D. C.; Xiao, Y.; Liu, H.; Chapman, K. W.; Bo, S. H.; Ceder, G. Reactivity-Guided Interface Design in Na Metal Solid-State Batteries. Joule 2019, 3 (4), 1037–1050. https://doi.org/10.1016/j.joule.2018.12.019. 53. Wang, M. J.; Carmona, E.; Gupta, A.; Albertus, P.; Sakamoto, J. Enabling “Lithium-Free” Manufacturing of Pure Lithium Metal Solid-State Batteries through in Situ Plating. Nat. Commun. 2020, 11 (1), 5201. https://doi.org/10.1038/s41467-020-19004-4. 54. Neudecker, B. J.; Dudney, N. J.; Bates, J. B. “Lithium-Free” Thin-Film Battery with In Situ Plated Li Anode. J. Electrochem. Soc. 2000, 147 (2), 517–523. https://doi.org/ 10.1149/1.1393226. 55. Krauskopf, T.; Dippel, R.; Hartmann, H.; Peppler, K.; Mogwitz, B.; Richter, F. H.; Zeier, W. G.; Janek, J. Lithium-Metal Growth Kinetics on LLZO Garnet-Type Solid Electrolytes. Joule 2019, 3 (8), 2030–2049. https://doi.org/10.1016/j.joule.2019.06.013. 56. Ning, Z.; Jolly, D. S.; Li, G.; De Meyere, R.; Pu, S. D.; Chen, Y.; Kasemchainan, J.; Ihli, J.; Gong, C.; Liu, B.; Melvin, D. L. R.; Bonnin, A.; Magdysyuk, O.; Adamson, P.; Hartley, G. O.; Monroe, C. W.; Marrow, T. J.; Bruce, P. G. Visualizing Plating-Induced Cracking in Lithium-Anode Solid-Electrolyte Cells. Nat. Mater. 2021, 20 (8), 1121–1129. https://doi.org/10.1038/s41563-021-00967-8. 57. Motoyama, M.; Hirota, M.; Yamamoto, T.; Iriyama, Y. Temperature Effects on Li Nucleation at Cu/LiPON Interfaces. ACS Appl. Mater. Interfaces 2020, 12 (34), 38045–38053. https://doi.org/10.1021/acsami.0c02354. 58. Haslam, C.; Sakamoto, J. Stable Lithium Plating in “Lithium Metal-Free” Solid-State Batteries Enabled by Seeded Lithium Nucleation. J. Electrochem. Soc. 2023, 170 (4), 040524. https://doi.org/10.1149/1945-7111/accab4. 59. Kazyak, E.; Wang, M. J.; Lee, K.; Yadavalli, S.; Sanchez, A. J.; Thouless, M. D.; Sakamoto, J.; Dasgupta, N. P. Understanding the Electro-Chemo-Mechanics of Li Plating in Anode-Free Solid-State Batteries with Operando 3D Microscopy Understanding the Electro-Chemo-Mechanics of Li Plating in Anode-Free Solid-State Batteries with Operando 3D Microscopy. Matter 2022, 5, 3912–3934. https://doi.org/10.1016/j.matt.2022.07.020. 60. Randau, S.; Weber, D. A.; Kötz, O.; Koerver, R.; Braun, P.; Weber, A.; Ivers-Tiffée, E.; Adermann, T.; Kulisch, J.; Zeier, W. G.; Richter, F. H.; Janek, J. Benchmarking the Performance of All-Solid-State Lithium Batteries. Nat. Energy 2020, 5 (3), 259–270. https://doi.org/10.1038/s41560-020-0565-1. 61. Albertus, P.; Babinec, S.; Litzelman, S.; Newman, A. Status and Challenges in Enabling the Lithium Metal Electrode for High-Energy and Low-Cost Rechargeable Batteries. Nat. Energy 2018, 3 (1), 16–21. https://doi.org/10.1038/s41560-017-0047-2. 62. Krauskopf, T.; Richter, F. H.; Zeier, W. G.; Janek, J. Physicochemical Concepts of the Lithium Metal Anode in Solid-State Batteries. Chem. Rev. 2020, 120 (15), 7745–7794. https://doi.org/10.1021/acs.chemrev.0c00431. 63. Wang, M. J.; Choudhury, R.; Sakamoto, J. Characterizing the Li-Solid-Electrolyte Interface Dynamics as a Function of Stack Pressure and Current Density. Joule 2019, 3 (9), 2165–2178. https://doi.org/10.1016/j.joule.2019.06.017. 64. Lee, K.; Kazyak, E.; Wang, M. J.; Dasgupta, N. P.; Sakamoto, J. Analyzing Void Formation and Rewetting of Thin in Situ-Formed Li Anodes on LLZO. Joule 2022, 6 (11), 2547–2565. https://doi.org/10.1016/j.joule.2022.09.009. 65. Sandoval, S. E.; McDowell, M. T. Lithium Metal Anodes in Solid-State Batteries: Metal Microstructure Matters. Matter 2023, 6 (7), 2101–2102. https://doi.org/10.1016/j. matt.2023.05.017. 66. Singh, D. K.; Fuchs, T.; Krempaszky, C.; Schweitzer, P.; Lerch, C.; Richter, F. H.; Janek, J. Origin of the Lithium Metal Anode Instability in Solid-State Batteries during Discharge. Matter 2023, 6 (5), 1463–1483. https://doi.org/10.1016/j.matt.2023.02.008. 67. Singh, D. K.; Fuchs, T.; Krempaszky, C.; Mogwitz, B.; Burkhardt, S.; Richter, F. H.; Janek, J. Overcoming Anode Instability in Solid-State Batteries through Control of the Lithium Metal Microstructure. Adv. Funct. Mater. 2023, 33 (1), 2211067. https://doi.org/10.1002/adfm.202211067. 68. Sandoval, S. E.; Lewis, J. A.; Vishnugopi, B. S.; Nelson, D. L.; Schneider, M. M.; Cortes, F. J. Q.; Matthews, C. M.; Watt, J.; Tian, M.; Shevchenko, P.; Mukherjee, P. P.; McDowell, M. T. Structural and Electrochemical Evolution of Alloy Interfacial Layers in Anode-Free Solid-State Batteries. Joule 2023, 7 (9), 2054–2073. https://doi.org/10.1016/ j.joule.2023.07.022. 69. Holme, T. A Discussion of QuantumScape’s Battery Technology Performance Results, 2021; 14th of January 2021. https://www.quantumscape.com/resources/blog/adiscussion-of-quantumscapes-battery-technology-performance-results/. 70. Kim, S.; Yoon, G.; Jung, S. K.; Park, S.; Kim, J. S.; Yoon, K.; Lee, S.; Kang, K. High-Power Hybrid Solid-State Lithium-Metal Batteries Enabled by Preferred Directional Lithium Growth Mechanism. ACS Energy Lett. 2023, 8, 9–20. https://doi.org/10.1021/acsenergylett.2c02150. 71. Spencer-Jolly, D.; Agarwal, V.; Doerrer, C.; Hu, B.; Zhang, S.; Melvin, D. L. R.; Gao, H.; Gao, X.; Adamson, P.; Magdysyuk, O. V.; Grant, P. S.; House, R. A.; Bruce, P. G. Structural Changes in the Silver-Carbon Composite Anode Interlayer of Solid-State Batteries. Joule 2023, 7 (3), 503–514. https://doi.org/10.1016/j.joule.2023.02.001. 72. Ceder, G.; Tu, H. The Mechanism by Which a Ag-Carbon Buffer Layers Improves Li Plating in Solid-State Batteries. MRS Spring Meeting & Exhibit Abstracts 2023, EN06.06.05. 73. Xu, R.; Liu, F.; Ye, Y.; Chen, H.; Yang, R. R.; Ma, Y.; Huang, W.; Wan, J.; Cui, Y. A Morphologically Stable Li/Electrolyte Interface for All-Solid-State Batteries Enabled by 3D-Micropatterned Garnet. Adv. Mater. 2021, 33 (49), 1–10. https://doi.org/10.1002/adma.202104009. 74. Yang, C.; Xie, H.; Ping, W.; Fu, K.; Liu, B.; Rao, J.; Dai, J.; Wang, C.; Pastel, G.; Hu, L. An Electron/Ion Dual-Conductive Alloy Framework for High-Rate and High-Capacity Solid-State Lithium-Metal Batteries. Adv. Mater. 2019, 31 (3), 1–7. https://doi.org/10.1002/adma.201804815. 75. Xu, S.; McOwen, D. W.; Wang, C.; Zhang, L.; Luo, W.; Chen, C.; Li, Y.; Gong, Y.; Dai, J.; Kuang, Y.; Yang, C.; Hamann, T. R.; Wachsman, E. D.; Hu, L. Three-Dimensional, Solid-State Mixed Electron-Ion Conductive Framework for Lithium Metal Anode. Nano Lett. 2018, 18 (6), 3926–3933. https://doi.org/10.1021/acs.nanolett.8b01295. 76. Alexander, G. V.; Shi, C.; O’Neill, J.; Wachsman, E. D. Extreme Lithium-Metal Cycling Enabled by a Mixed Ion- and Electron-Conducting Garnet Three-Dimensional Architecture. Nat. Mater. 2023,. https://doi.org/10.1038/s41563-023-01627-9.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Thin-Film Type Cells Chuangjie Guo and Yaoyu Ren, School of Materials Science and Engineering, State Key Lab of New Ceramics and Fine Processing, Tsinghua University, Beijing, China © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 3 4 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 5 6 6.1 6.1.1 6.1.2 6.1.3 6.2 6.2.1 6.2.2 6.2.3 6.2.4 6.3 6.4 6.4.1 6.4.2 6.4.3 7 8 9 10 References

Introduction Characteristics of configuration Performance evaluation Fabrication techniques PVD CVD ALD Electrochemical deposition Sol-gel deposition Printing Tape casting Aerosol deposition Initial development of TFBs Lithium-based TFBs Solid electrolytes LiPON Garnet-like structure Polymer-based Cathodes Li(Co1-xMx)O2 (M ¼ Ni, Mn) layered structure Li(NixMn2-x)O4 spinel structure Transition metal oxides (Li)TiS2 Anodes Other components Current collectors Substrates Encapsulation materials 3D TFBs Flexible TFBs Cell development and commercialization Outlook

601 602 603 603 603 603 603 603 604 604 604 604 604 605 605 605 606 606 606 607 607 608 608 608 608 608 609 609 609 609 611 611 612

Abstract The chapter is titled “Thin-Film Type Cells” to maintain consistency throughout the book. However, in the main text, it will be referred to as Thin Film Type Batteries (TFBs) to distinguish it from other cells, as it is often used in the context of solar cells. The development of TFBs has followed four main trends: the shift from primary to secondary batteries, the move from low to high operating voltage batteries, the transition from non-lithium TFBs to exclusively lithium-based TFBs, and the progression from planar to 3D TFBs. This chapter provides an introduction to the characteristics, fabrication techniques and historical development of TFBs with a focus on lithium-based TFBs. It also explores the development of three-dimensional TFBs to enhance capacity and power capability, as well as flexible TFBs for use in wearables and flexible devices. Recent efforts on the commercialization of TFBs are also discussed, along with an outlook for future TFB development.

Glossary Area enhancement factor (in three-dimensional thin film batteries) The area enhancement factor is defined by the ratio between the effective surface area of the three-dimensional structure and its footprint (or geometric) area. Typically, a higher area enhancement factor indicates a greater capacity of a battery per footprint that can be attained.

600

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00218-8

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Thin-Film Type Cells

601

Areal specific capacity/energy (of a battery) Areal specific capacity/energy refers to the amount of electrical energy or capacity stored per unit footprint area of a battery. Areal specific capacity is typically expressed in units of milliampere hour per square centimeter (mAh cm−2). Areal specific energy is typically expressed in units of watt hour per square centimeter (Wh cm−2). Electrochemical stability windows (of solid electrolytes) Electrochemical stability windows refer to the range of voltages in which a material (e.g., a solid electrolyte) can operate without undergoing any electrochemical reactions or degradation. These voltage limits are important in determining the suitability and stability of a material for use in various electrochemical systems, such as batteries, fuel cells, and electrolyzers. Energy density (of a battery) Energy density is a measure of how much energy is available within a unit volume or mass of the battery. It is typically expressed in units of watt hour per litter (Wh L−1) or watt hour per kilogram (Wh kg−1). Li-rich cathode Li-rich cathode refers to a type of cathode material used in lithium-ion batteries that contains a higher proportion of lithium ions compared to other cathode materials. Li-rich cathodes are often written as Li1+xTM1-xO2 (TM: transition metal), composed of a combination of a layered structure, LiTMO2 and a lithium-excess structure, Li2TMO3. The higher lithium content in these cathodes allows for a higher capacity for lithium-ion intercalation and de-intercalation and therefore can provide a greater amount of energy storage capacity. Lithiated/delithiated Lithiated/delithiated refers to the process of adding or removing lithium ions to or from a material or compound, typically a lithium-ion battery electrode. Primary battery A primary battery, also known as a disposable battery, is a type of electrochemical cell that is designed to be used once and then discarded after the energy is depleted. It cannot be recharged or reused. Primary batteries typically provide relatively high and stable voltage output but have a limited capacity. Common examples of primary batteries include alkaline batteries, zinc-carbon batteries, and lithium batteries. They are commonly used in devices such as flashlights, remote controls, and portable electronic devices. Rate capability (of a lithium battery) The rate capability of a lithium battery refers to its ability to discharge or charge at a high rate of power. It is a measure of how quickly a lithium battery can deliver or accept energy. A higher rate capability means that the battery can deliver or absorb power more rapidly. It is often measured in terms of C-rate, where 1C represents the capacity of the battery being discharged or charged over a period of 1 h. Secondary battery A secondary battery, also known as a rechargeable battery or accumulator, is a type of battery that can be recharged and used multiple times. Unlike primary batteries, which are disposable and cannot be recharged, secondary batteries can be connected to an external power source to replenish the stored energy and be reused. Common examples of secondary batteries include lithium-ion batteries, nickel-cadmium batteries, and lead-acid batteries. These batteries find extensive use in electronic devices, electric vehicles, and renewable energy storage systems. Solid-state battery A solid-state battery is a type of battery that uses a solid electrolyte instead of a liquid or gel electrolyte found in traditional batteries. In the technical context of the battery industry, solid-state batteries. Transference number The transference number, also known as the ionic transference number, is a measure of the ability of an ion to move relative to other ions in electrolytic solutions. It is used to describe the fraction of the total electric current carried by a specific ion in the solution. The transference number is usually represented by the symbol “t+” or “t−,” indicating the transference number for cations or anions, respectively. It characterizes the mobility of the ion, with higher transference numbers indicating greater ability to move and carry electric current in the solution.

Key points

• • • • •

1

Thin film batteries feature thin films for all their essential components, typically with thicknesses in the range of a few microns. TFBs have seen significant development in response to the demand for long-lasting and low-capacity miniaturized energy storage devices in microelectronics and flexible electronic devices Presently, TFBs exhibit high safety and long lifespan, while also striving for higher areal specific energy density and power density. Initial TFBs were non-lithium while nowadays’ TFBs are exclusively lithium-based. Recent developments of TFBs are toward being 3D and flexible.

Introduction

A thin film Type battery (TFB) is a type of battery that incorporates thin film materials for its construction. It is a compact and lightweight power source that is designed to be extremely thin, typically 100  C), LIPB must show sufficient stable performance even under a high-temperature overcharged condition. Unfortunately, currently, there are no suitable materials in GPE to solve this problem completely. A flat cell design using an aluminum-laminated film in LIPB contributes to the improvement of safety because of its good heat release properties. The relative surface area per cell capacity is larger in the flat-type cell than that of cylindrical or prismatic cells. In addition, the accumulated gas in the cell is easily released from the thermally sealed area of the aluminum-laminated film, whereas metal cans used in cylindrical or prismatic housings require a rapture vent to release the pressure inside. The choice of liquid free SPE is one way to improve safety at abuse condition. Perea et al. reported the difference of the thermal stability between SPE and liquid electrolyte with an accelerating rate calorimeter (ARC).10 In case of LiFePO4 cathode and Li metal anode battery, the onset temperature of the liquid-type cell at 100% SOC starts at 90  C compared to 247  C for the allsolid-state cell. The addition of fire-retardant material is one of the intrinsic ways to improve safety. Cui et al. reported fire-proof polymer electrolyte using porous mechanic enforcer (polyimide, PI), a fire-retardant additive (decabromodiphenyl ethane, DBDPE), and an ionic conductive polymer electrolyte (PEO/LiTFSI).11 They compared abused cell working test under flame test at fully charged condition. The selected cathode and anode were LiFePO4 and Li4Ti5O12, respectively. In case of liquid electrolyte/conventional polymer separator (EC/DEC/polyethylene (PE)) cell, and conventional solid polymer electrolyte (PEO/LiTFSI) cell could not light the LED bulbs after ignition for 18 and 24 s, respectively. However, proposed PI/DBDPE/PEO/LiTFSI cell at 24 s was still as bright as that before ignition. These findings encourage us the way to improve safety of LIPBs in the future.

3 3.1

Solvent-free lithium-ion polymer battery Solvent-free solid polymer electrolyte

To improve the intrinsic safety of the battery, a nonvolatile electrolyte has been applied in LIPB for a long time. A solvent-free genuine SPE is a suitable material for the design of a safe LIPB. The design of the LPB was proposed by M. B. Armand in 1979.12 The specific ionic conductivity of the first-generation PEO was 10−8 S cm−1 at room temperature. Then, the conductivity was improved to the order of 10−4 S cm−1 at room temperature by a suitable polymer matrix design, the addition of filler, the use of lithium salt, and so on. However, the conductivity is still 1 or 2 orders lower than that of liquid electrolyte or GPE. Therefore, the battery using SPE requires a heating system to maintain a suitable temperature (60–90  C) to enhance the conductivity, so such a battery is difficult for portable applications but is developed for large energy storage systems such as those for electric vehicle and stationary use. Hydro Québec developed a 10 Wh class LPB in 1995.13 EDF also developed 40 Ah class LPB in 1997.14 Bolloré Blue Solutions commercialized real scale LPB for EV in 2012.15 However, the negative electrode used was lithium metal. Thus, the battery was not included in the family of LIPBs.

3.2

Lithium-ion acceptable negative electrodes for solvent-free solid polymer electrolyte

There were two main barriers for realizing LIPB using SPE. One was a combination of negative electrodes, which showed reversible intercalation properties of lithium ions. The other was the insufficient reversibility of 4 V class positive electrodes such as LiMO2 (M ¼ Mn, Co, Ni, etc.). A lithium-ion intercalation property observed in a graphite structure with SPE was first reported by Yazami et al., in 1994.16 However, the irreversible capacity at the first intercalation was extremely higher than the reversible one. Lithium-ion in LIPB is supplied from a positive electrode; thus, a large irreversible capacity, in other words, extra lithium consumption at a negative electrode due to the formation of solid electrolyte interphase (SEI), decreases the reversible capacity at the positive electrode. Since then, the combination of a carbon-related negative electrode and SPE was believed to be difficult for obtaining good reversibility. The origin of irreversibility between graphite anode and SPE was investigated by Xu et al., using XPS.17 They found water residue in SPE was a key reactant for generation of large amounts of LiOH on the surface of the graphite. Li4Ti5O12 (LTO) is another promising material for lithium-ion acceptable negative electrolyte. Perea et al. reported high reversibility (155–157 mAh g−1) with high 1st coulombic efficiency (96–97%) using LTO and SPE.10 LTO shows higher redox potential (1.5 V vs. Li/Li+) than that of graphite (0.1 V vs. Li/Li+) so that irreversible Li loss due to SEI formation is smaller than that of graphite. In 2007, Imanishi et al. reported a 300 mAh g−1 reversible capacity using surface-modified MCMB with solvent-free polyether-based SPE.18 Their approach involves an active formation of low-crystalline carbon using polyvinyl chloride to decrease the residual irreversible reaction at the surface. The reported first coulombic efficiency was 65%. Furthermore, Kobayashi et al. also reported a 360 mAh g−1 reversible specific capacity without the surface modification of graphite in 2008.19 They focused on the electronic-conductive additives in a negative electrode. The conventionally used acetylene black (AB) had a large surface area (50–100 m2 g−1), resulting in a large irreversible capacity. They used a vapor-grown carbon fiber (VGCFs, surface area: 15 m2 g−1) instead of AB to decrease the surface area in the electrode and obtained a coulombic efficiency of 79% at the first cycle. The capacity retention was 75% at the 250th cycle at C/8, which was comparable to that of a liquid electrolyte system. Silicon-related materials are other promising high-capacity negative electrodes. Pure silicon exhibits a large volume change during charging and discharging (400% vs. initial); thus, it is not suitable for SPE, which shows a restricted shape flexibility at the interface. Silicon monoxide (SiO) shows a small volume change (200% vs. initial). Liu et al., first reported the combination of SiO and SPE in 2005.20 They used nanosized SiO and Li2.6Co0.4N and obtained a reversible specific capacity of 500 mAh g−1. This result

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolyte Cells

619

indicates the possibility of fabricating LIPB using SPE without using lithium metal. Furthermore, Kobayashi et al. reported a reversibility of 1000 mAh g−1 by using micron-sized SiO with SPE.19 They applied polyimide as a binder in a negative electrode instead of conventional binders, such as PVdF, to improve the supporting properties of electrode particles. The first coulombic efficiency of SiO with SPE was 78%, which is comparable to that of a liquid electrolyte system (79%). These results suggest that the irreversible capacity loss at the first cycle is mainly due to the irreversible reaction inside the electrode particles. In other words, the irreversible reaction slightly depends on electrolyte species. Further improvement of reversibility was expected using carbon and silicon-related materials with SPE by optimizing their surface modification, the use of additives, the preparation procedure of electrodes, and so on.

4 4.1

Advanced solid polymer electrolyte design for high voltage cathode Polyether-based SPEs

As mentioned above, another difficulty in realizing LIPB was the insufficient reversibility of 4 V class positive electrodes with SPE. The EO unit commonly used in SPE was assumed to be oxidized at >4 V vs. Li/Li+. A cyclic voltammetry test using a platinum electrode also supported this assumption. Cathodic current was observed at a voltage higher than 4 V vs. Li/Li+ using an EO-based SPE. Therefore, lithium-ion phosphate (LiFePO4) was considered the most suitable positive electrode for LIPB, because its redox potential was lower (3.4 V vs. Li/Li+) than that of other oxide-based positive materials (3.7–4.2 V vs. Li/Li+). Certainly, the cycle performance of LiFePO4 with SPE was stable without a significant decrease in capacity, and already commercialized by Bolloré Blue Solutions.15 The combination of a 4 V class positive electrode (LiCoO2) and SPE was investigated by Matsui et al.21 The capacity retention was 63% at the 200th cycle vs. initial.

4.2

Ceramic/polymer composite using polyether-based SPE

The improvement of reversibility using LiCoO2 with SPE was achieved by the surface modification of positive electrodes. Kobayashi et al. prepared Li3PO4-coated LiCoO2, and the coated LiCoO2 showed a reversible specific capacity (200 mAh g−1) up to 4.6 V with SPE.22 On the contrary, uncoated LiCoO2 could not charge to 4.6 V because of the degradation at the interface between SPE and the positive electrode. They determined the change in positive electrode/SPE interface impedance by electrochemical impedance spectroscopy at constant potentials and found a significant improvement of impedance growth by the surface modification of the positive electrode. The coating species protects the SPE from oxidation and/or forms a stable surface on the positive electrode, which is restricted neither to only Li3PO4, nor to Al2O3, LiAl0.5Ge1.5(PO4)3, (LATP) and so on. The positive electrode is also not restricted to LiCoO2. Miyashiro et al. reported the reversibility of a 5 V class positive electrode (LiNi0.5Mn1.5O4) with SPE obtained by surface modification using Li3PO4.23

4.3

Ceramic/polymer composite using other SPEs

Ionic conductivity of oxide-based electrolyte is higher than that of SPE based one. Then, LATP was sandwiched by heterogeneous SPEs to extend electrochemical window. Duan et al., proposed an Ni rich layered cathode/poly(acrylonitrile)(PAN)/LATP/polyethylene glycol diacrylate/Li cell.24 In the material design, oxidation-resistance PAN is in contact with the cathode while reduction tolerant polyethylene glycol diacrylate contacts with Li metal anode. In case of LiNi0.6Co0.2Mn0.2O2 (NCM622), a capacity retention of 81.5% with an average Coulombic efficiency of 99.8% is achieved after 270 cycles.

4.4

Plastic crystal-based SPEs for high-voltage positive electrodes

Succinonitrile (SN) is one of the promising lithium-ion conductive plastic crystal above room temperature and reported by Alarco et al.25 Ionic conductivity of SN with 5 mol% LiTFSI reached above 3  10−3 Scm−1 at room temperature with high oxidation potential (6 V vs. Li/Li+). Though the poor mechanical strength limited their application, mixing of flexible polymer etc. were investigated. Zhou et al. reported ternary lithium-salt organic ionic plastic crystal polymer composite electrolyte and demonstrated high reversibility (80% capacity retention with 4.6 V cutoff voltage vs. Li/Li+) using LiNi1/3Mn1/3Co1/3O2 cathode.26 The addition of SN is helpful also to the high loading cathode preparation in solid-state lithium batteries. Zhang et al., reported high loading (10.5 mg cm−2) LiNi0.5Co0.2Mn0.3O2 composite cathode with SN and LiTFSI as additives is used to fabricate solid-state batteries as shown in Fig. 3, which exhibited a high initial discharge capacity of 146.9 mAh g−1 and good capacity retention of 86.89% after 150 cycles at 0.1  C under 25  C.27

4.5

Polymer/ceramic/polymer sandwich electrolyte for dendrite-free lithium anode

Oxide-based inorganic electrolyte such as LATP and Li7La3Zr2O12 (LLZO) are well known good Li-ion conductor with sintered pellet shape. Zhou et al. proposed to introduce a cross-linked polymer electrolyte between ceramic pellet and electrodes.28 In the sandwich design, the double-layer electric field at the Li/polymer interface is reduced due to the blocked salt anion transfer, which induced high Coulombic efficiency of 99.8–100% over 640 cycles.

620

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolyte Cells

Fig. 3 Schematic diagram of solid-state battery. Ceramic-based composite electrolyte is consisted of Li6.75La3Zr1.75Ta0.25O12 and PVdF/HFP. Composite cathode was prepared by mixing LiNi0.5Co0.2Mn0.3O2, PVdF, electronic conductive additive (Super P), lithium bis(trifluoromethylsulfonyl) imide (LiTFSI) and Succinonitrile (SN). Adapted from Zhang, B.; Chen, L.; Hu, J.; Liu Y.; Liu, Y.; Feng, Q.; Zhu, G.; Fan L-Z. Solid-State lithium Metal Batteries Enabled with High Loading Composite Cathode Materials and Ceramic-Based Composite Electrolytes. J. Power Sources 2019, 442, 227230, with copyright from permission from Elsevier.

4.6

Current status of solvent-free lithium-ion polymer battery

Solvent-free LIPBs have no flammable liquid; thus, an intrinsic improvement of safety is expected. The 1st solvent-free lithium-ion polymer battery was reported by Prosini et al. using LiMn2O4 cathode and LTO anode.29 Through the obtained active material utilization was about 50%, they proved the possibility to realize solvent-free lithium-ion battery. In 2008, Kobayashi et al. reported better cycle performance solvent-free lithium-ion dry polymer battery using LiFePO4 cathode polyether-based SPE, and graphite anode.19 The 1st reversible capacity based on LiFePO4 was 128 mAh g−1 (approximately 80% vs. reversible capacity of LiFePO4), which suggested the capacity loss of the battery was derived from the irreversible loss of graphite due to SEI formation. The capacity retention at the 80th cycle was 64% vs. 1st cycle so that further improvement in performance was required. Shono et al. prepared 4 V class LiNi1/3Co1/3Mn1/3O2/polyether-based SPE/graphite cell with pseudo-Li reference electrode, and capacity fade mechanism was investigated.30 They confirmed the loss of active Li was derived from the continuous SEI formation at graphite anode as shown in Fig. 4. Therefore, the suppression of Li loss is required for long life battery design. On the other hand, formed SEI at graphite anode is not main contribution of impedance growth of the cell, but the main impedance increase is due to the cathode/SPE interface. Based on the investigation of weak points, Kobayashi et al. achieved long life (capacity retention 60%

Fig. 4 Variations in AC impedance with number of cycles at the end of charge in (a) [Graphite (Gr) | solid polymer electrolyte (SPE) | Li]-[Li | SPE| LiNi1/3Mn1/3Co1/3O2 (NMC)], (b) [Li | SPE | MMC], (c) [Li | SPE | Gr]. Adapted from Shono, K.; Kobayashi, T.; Tabuchi, M.; Ohno, Y.; Miyashiro, H.; Kobayashi, Y. Proposal of Simple and Novel Method of Capacity Fading Analysis Using pseudo-Reference Electrode in lithium-Ion Cells: Application to Solvent-Free lithium-Ion Polymer Batteries. J. Power Sources 2014, 247, 1026–1032, with copyright from permission from Elsevier.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolyte Cells

LiBF4

LiBF4

+

621

LiTFSI LiTFSI

LiBOB

Solid polymer electrolyte

LiNi1/3Mn1/3Co1/3O2

Surface modified graphite

Fig. 5 Cycle performance of lithium-ion dry polymer battery with [LiNi1/3Mn1/3Co1/3O2 (NMC)| solid polymer electrolyte (SPE) | Graphite (Gr)]. Composite cathode is consisted with NMC, SPE, lithium tetrafluoroborate (LiBF4), and antioxidant additive. SPE is consisted with SPE, LiBF4 and lithium bis(oxalate)borate (LiBOB) at cathode interface, and lithium bis(trifluoromethylsulfonyl) imide (LiTFSI) at anode interface. Composite anode is consisted with surface modified Gr, vapor grown carbon fiber (VGCF©), SPE, and LiTFSI. Adapted from Kobayashi, Y.; Shono, K.; Kobayashi, T.; Ohno, Y.; Tabuchi, M.; Oka, Y.; Nakamura, T.; Miyashiro, H. A Long Life 4 V Class lithium-Ion Polymer Battery with Liquid-Free Polymer Electrolyte. J. Power Sources 2017, 341, 257–263, with copyright from permission from Elsevier.

after 5400 cycles at 50  C, Fig. 5) lithium-ion dry polymer battery using ether-based SPE.31 The introduced boron-based lithium salt prohibits further oxidation of SPE at the cathode interface. In addition, the surface modification of graphite by annealing of polyvinyl chloride mostly prohibits the continuous consumption of lithium at the graphite anode. The improvement of safety is the most critical interest for solvent-free LIPB. To verify the event quantitatively, Kobayashi et al. designed a battery safety evaluation system and compared forced internal short circuit event in various LIBs and LIPB32. In the evaluation system, tested battery was placed in pressure-tight enclosure, charged to 120% SOC, and heated at 70  C. They prepared 13 Wh LiNi1/3Co1/3Mn1/3O2/graphite cells with polyether-based SPE or liquid electrolyte, and compared maximum outgas pressure (DPe) with forced internal short circuit with a nail penetration. They also tested commercial LiFePO4/liquid electrolyte/graphite and LiCoO2/gel electrolyte/graphite cells and compared the safety event. As the capacity of commercial cells were different from prepared cells, outgas pressure value was compared based on Wh (DPe/Wh) as shown in Fig. 6. The order of DPe/Wh was LiCoO2/gel > LiNi1/3Co1/3Mn1/3O2/liquid > LiFePO4/liquid > LiNi1/3Co1/3Mn1/3O2/SPE. The LIPB cell after the event showed no drastic shape change as shown in Fig. 7. They confirmed intrinsic safety improvement of SPE by using real-scale quantitative analysis. The issue of polyether-based electrolyte is insufficient high-rate performance. Choi et al. selected plastic crystal and demonstrated high-rate cycle operation using LiCoO2 cathode and Li4Ti5O12 anode.33 Here, they designed thin, deformable, and safety-reinforced plastic polymer electrolyte by combining a plastic polymer electrolyte with a porous polyethylene terephthalate (PET) nonwoven. The reported rate performance (70% at 2  C) is promising for the practical use.

Fig. 6 Maximum outgas pressure based on capacity (DPe/Wh) of various Lithium-ion batteries and Lithium-ion polymer battery with forced internal short circuit by nail penetration at 120% state of charge at 70  C. LFP: LiFePO4, LCO: LiCoO2, NMC: LiNi1/3Mn1/3Co1/3O2, Gr: Graphite, Liquid: Liquid electrolyte, Gel: Gel electrolyte, SPE: Solid Polymer Electrolyte. Adapted from Kobayashi, Y.; Shono, K.; Miyashiro, H. Stability Evaluation of Lithium-Ion Batteries by Overcharge and Simulated Internal Short-Circuit Test. CRIEPI Report 2016, Q15006, with copyright from permission from CRIEPI.

622

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolyte Cells

Fig. 7 Schematic of 13 Wh [LiNi1/3Mn1/3Co1/3O2 (NMC)| solid polymer electrolyte (SPE) | Graphite (Gr)] battery after nail penetration at 120% state of charge at 70  C. Adapted from Kobayashi, Y.; Shono, K.; Miyashiro, H. Stability Evaluation of Lithium-Ion Batteries by Overcharge and Simulated Internal Short-Circuit Test. CRIEPI Report 2016, Q15006, with copyright from permission from CRIEPI.

5

Conclusion

Lithium-ion polymer battery has been developed as a derivative of LIB by the partial substitution of gel polymer in an electrolyte. Hence, the battery performances (energy density, cycle, and rate) of LIPB are now comparable to those of LIB. Most of the materials (positive and negative electrodes, separator, liquid electrolyte, exterior package, and so on) in LIPB are compatible with those in LIB. This trend (material sharing) is expected to reduce the costs of LIB and LIPB. Although flat-type battery using an aluminum-laminated package is an intrinsic characteristic of LIPB, LIB without a polymer also has employed the same package. In addition, although the safety issue is expected to improve using GPE, no essential improvement of safety has been reported for LIPB. The development of LIPB will be required not only the good bland image of a ‘polymer’ but also intrinsic safety in the future. Recent research trend is changing to safe material choices such as dry polymer, and various mixture of inorganic electrolyte, plastic crystal with polymer materials. The role of the polymer is changing from “ionic conductor” or “liquid impetration material” to “flexible buffer”. Such material design will provide high-energy density, long life, and safe lithium-ion polymer battery.

References 1. Aihara, Y.; Kodama, M.; Nakahara, K.; Okise, H.; Murata, K. Characteristics of a Thin Film lithium-Ion Battery Using Plasticized Solid Polymer Electrolyte. J. Power Sources 1997, 65, 143–147. 2. Croce, F.; Appetecchi, G. B.; Persi, L.; Scrosati, B. Nanocomposite Polymer Electrolytes for Lithium Batteries. Nature 1998, 394, 456–458. 3. Kono, M.; Nishiura, M.; Ishiko, E. Behavior of Interfacial Resistance at lithium Electrode for Gel Electrolyte Using Novel Three-Dimensional Network Polymer Host. J. Power Sources 1999, 81-82, 748–751. 4. Guerfi, A.; Duchesne, S.; Kobayashi, Y.; Vijh, A.; Zaghib, K. LiFePO4 and Graphite Electrodes with Ionic Liquids Based on Bis(Fluorosulfonyl)Imide (FSI)− for Li-Ion Batteries. J. Power Sources 2008, 175, 866–873. 5. Akashi, H.; Sekai, K.; Tanaka, K. A Novel Fire-Retardant Polyacrylonitrile-Based Gel Electrolyte for lithium Batteries. Electrochim. Acta 1998, 43, 1193–1197. 6. Feuillade, G.; Perch, P. Ion-Conductive Macromolecular Gels and Membranes for Solid lithium Cells. J. Appl. Electrochem. 1975, 5, 63–69. 7. Tarascon, J.-M.; Gozdz, A. S.; Schmutz, C.; Shokoohi, F.; Warren, P. C. Performance of Bellcore’s Plastic Rechargeable Li-Ion Batteries. Solid State Ion. 1996, 86-88, 49–54. 8. Yamamoto, T.; Hara, T.; Segawa, K.; Honda, K.; Akashi, H. 4.4 V Lithium-Ion Polymer Batteries with a Chemical Stable Gel Electrolyte. J. Power Sources 2007, 174, 1036–1040. 9. Hu, P.; Zhao, J.; Wang, T.; Shang, C.; Zhang, J.; Qin, B.; Liu, Z.; Xiong, J.; Cui, G. A Composite Gel Polymer Electrolyte with High Voltage Cyclability for Ni-Rich Cathode of lithium-Ion Battery. Electrochem. Commun. 2015, 61, 32–35. 10. Perea, A.; Dontigny, M.; Zaghib, K. Safety of Solid-State Li Metal Battery: Solid Polymer Versus Liquid Electrolyte. J. Power Sources 2017, 359, 182–185. 11. Cui, Y.; Wan, J.; Ye, Y.; Liu, K.; Chou, L.-Y.; Cui, Y. A Fireproof, Lightweight Polymer-Polymer Solid-State Electrolyte for Safe Lithium Batteries. Nano Lett. 2020, 20, 1686–1692. 12. Armand, M. B.; Chabagno, J. M.; Duclot, M. J. In Fast Ion Transport in Solids; Vashishta, P., Mundy, J.-N., Shenoy, G. K., Eds.; Elsevier: New York, 1979; pp. 131–136. 13. Gauthier, M.; Bélanger, A.; Bouchard, P. B.; Kapfer, B.; Ricard, S.; Vassort, G.; Armand, M.; Sanchez, J. Y.; Krause, L. Large lithium Polymer Battery Development: The Immobile Solvent Concept. J. Power Sources 1995, 54, 163–169. 14. Braudry, P.; Lascaud, S.; Majastre, H.; Bloch, D. Lithium Polymer Battery Development for Electric Vehicle Application. J. Power Sources 1997, 69, 432–435. 15. Vervaeke, M.; Calabrese, G. Prospective Design in the Automotive Sector and the Trajectory of the Bluecar Project: An Electric Car Sharing System. Int. J. Veh. Des. 2015, 68, 245–264. 16. Yazami, R.; Zaghib, K.; Deschamps, M. Carbon Fibres and Natural Graphite as Negative Electrodes for lithium Ion-Type Batteries. J. Power Sources 1994, 52, 55–59. 17. Xu, C.; Sun, B.; Gustafsson, T.; Edstrom, K.; Brandell, D.; Hahlin, M. Interface Layer Formation in Solid Polymer Electrolyte lithium Batteries: An XPS Study. J. Mater. Chem. A 2014, 2, 7256–7264. 18. Imanishi, N.; Ono, Y.; Hanai, K.; Uchiyama, R.; Liu, Y.; Hirano, A.; Takeda, Y.; Yamamoto, O. Surface-Modified Meso-Carbon Microbeads Anode for Dry Polymer lithium-Ion Batteries. J. Power Sources 2008, 178, 744–750.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Solid Polymer Electrolyte Cells

623

19. Kobayashi, Y.; Seki, S.; Mita, Y.; Ohno, Y.; Miyashiro, H.; Charest, P.; Guerfi, A.; Zaghib, K. High Reversible Capacities of Graphite and SiO/Graphite with Solvent-Free Solid Electrolyte for lithium-Ion Batteries. J. Power Sources 2008, 185, 542–548. 20. Liu, Y.; Yang, J.; Imanishi, N.; Hirano, A.; Takeda, Y.; Yamamoto, O. Composite Anode Containing Nano-SiO1.1 and Li2.6Co0.4N with Solid PEO Electrolytes for lithium-Ion Batteries. J. Power Sources 2005, 146, 376–379. 21. Matsui, S.; Muranaga, T.; Higobashi, H.; Sakai, T. Liquid-Free Rechargeable Li Polymer Battery. J. Power Sources 2001, 97–98, 772–774. 22. Kobayashi, Y.; Seki, S.; Yamanaka, A.; Miyashiro, H.; Mita, Y.; Iwahori, T. Development of High-Voltage and High-Capacity all-Solid-State lithium Secondary Batteries. J. Power Sources 2005, 146, 719–722. 23. Miyashiro, H.; Seki, S.; Kobayashi, Y.; Ohno, Y.; Mita, Y.; Usami, A. All-Solid-State lithium Polymer Secondary Battery with LiNi0.5Mn1.5O4 by Mixing of Li3PO4. Electrochem. Commun. 2005, 7, 1083–1086. 24. Duan, H.; Fan, M.; Chen, W.-P.; Li, J.-Y.; Wang, P.-F.; Wang, W.-P.; Shi, J.-L.; Yin, Y.-X.; Wan, L.-J.; Guo, Y.-G. Extended Electrochemical Window of Solid Electrolytes Via Heterogeneous Multilayered Structure for High-Voltage lithium Metal Batteries. Adv. Mater. 2019, 31, 1807789. 25. Alarco, P.-J.; Abu-Lebdeh, Y.; Abouimrane, A.; Armand, M. The Plastic-Crystalline Phase Succinonitrile as a Universal Matrix for Solid-State Ionic Conductors. Nat. Mater. 2004, 3, 476–481. 26. Zhou, Y.; Wang, X.; Zhu, H.; Armand, M.; Forsyth, M.; Greene, G. W.; Pringle, J. M.; Howlett, P. C. Ternary lithium-Salt Organic Ionic Plastic Crystal Polymer Composite Electrolytes for High Voltage, all-Solid-State Batteries. Energy Storage Mater. 2018, 15, 407–414. 27. Zhang, B.; Chen, L.; Hu, J.; Liu, Y.; Liu, Y.; Feng, Q.; Zhu, G.; Fan, L.-Z. Solid-State lithium Metal Batteries Enabled with High Loading Composite Cathode Materials and Ceramic-Based Composite Electrolytes. J. Power Sources 2019, 442, 227230. 28. Zhou, W.; Wang, S.; Li, Y.; Xin, S.; Manthiram, A.; Goodenough, J. B. Plating a Dendrite-Free Lithium Anode with a Polymer/Ceramic/Polymer Sandwich Electrolyte. J. Am. Chem. Soc. 2016, 138, 9385–9388. 29. Prosini, P. P.; Mancini, R.; Petrucci, L.; Contini, V.; Villano, P. Li4Ti5O12 as Anode in all-Solid-State, Plastic, lithium-Ion Batteries for Low-Power Applications. Solid State Ion. 2001, 144, 185–192. 30. Shono, K.; Kobayashi, T.; Tabuchi, M.; Ohno, Y.; Miyashiro, H.; Kobayashi, Y. Proposal of Simple and Novel Method of Capacity Fading Analysis Using pseudo-Reference Electrode in lithium-Ion Cells: Application to Solvent-Free lithium-Ion Polymer Batteries. J. Power Sources 2014, 247, 1026–1032. 31. Kobayashi, Y.; Shono, K.; Kobayashi, T.; Ohno, Y.; Tabuchi, M.; Oka, Y.; Nakamura, T.; Miyashiro, H. A Long Life 4 V Class lithium-Ion Polymer Battery with Liquid-Free Polymer Electrolyte. J. Power Sources 2017, 341, 257–263. 32. Kobayashi, Y.; Shono, K.; Miyashiro, H. Stability Evaluation of Lithium-Ion Batteries by Overcharge and Simulated Internal Short-Circuit Test; CRIEPI Report, 2016; p. Q15006. 33. Choi, K.-H.; Cho, S.-J.; Kim, S.-H.; Kwon, Y. H.; Kim, J. Y.; Kee, S.-Y. Thin, Deformable, and Safety-Reinforced Plastic Crystal Polymer Electrolytes for High-Performance Flexible Lithium-Ion Batteries. Adv. Funct. Mater. 2014, 24, 44–52.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells Felix Hippauf and Sahin Cangaz, Fraunhofer IWS Dresden, Dresden, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 1.1 1.2 2 2.1 2.2 2.3 3 3.1 3.1.1 3.1.2 3.1.3 3.2 4 References

Introduction What is the motivation behind developing solid-state batteries From semi- to all solid-state Processing of large-size cells Lithium metal anode Silicon anode Composite electrodes + solid electrolyte films Practical inorganic solid-electrolyte cells Sulfide-based bulk cells Toyota Samsung Solid Power Oxide-based micro-batteries Conclusion

625 626 628 628 629 630 631 633 634 634 635 637 638 639 640

Abstract The chapter is dealing with practical solid-state batteries that already made the leap into application or at least into a working prototype. The first of its kind was a primary solid-state Li-iodide battery in the late 1960s. However, we will focus on secondary battery types. Initially, an overview is given about the different types of solid-state batteries and what is the motivation behind their development. State of the art lithium ion cells are already advanced in their technology and production. However, solid-state batteries enable the chance to drastically increase safety and energy density. Subsequently, the challenges in scaling-up the production of solid-state batteries are highlighted. In the subsequent chapter, focus is laid on selected practical cells based on different inorganic solid electrolytes. Sulfide electrolytes gained much attention due to their promising properties. Therefore, different companies developed prototype cells for potential use in electric vehicles. Already commercially available, but attracted less public attention, are oxide-based MLCC solid-state batteries. Their potential field of application lies in IoT and wearable devices.

Glossary ASSB All solid-state battery BEV Battery electric vehicle DBE Dry battery electrode EVs Electric vehicles IoT Internet of things IP Intellectual property LGPS Li10GeP2S12 LiSiPSC Li9.54Si1.74P1.44S11.7Cl0.3 MLCC Multi-layer ceramic capacitor PTFE Poly(tetrafluoroethylene) PVDF Poly(vinylidene fluoride) SE Solid electrolyte SEI Solid electrolyte interphase SMDs Surface-mounted devices SoA State-of-the-art SOC State-of-charge SSBs Solid-state batteries SSEs Sulfidic solid electrolytes TFB Thin film battery TRL Technology readiness level

624

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00318-2

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

625

Key points

• • • •

1

Highlighting advantages of solid-state cells Categorize solid-state batteries Emphasize challenges during production Describe selected bulk cells

Introduction

The following chapter is dealing with practical solid-state batteries that already made the leap into application or at least into a working prototype. The first of its kind was a primary solid-state Li-iodide battery in the late 1960s. For a matter of fact, this was the first practical application of Li-batteries. However, the range of application was strictly limited due to its high internal resistance, which was caused by the poor solid electrolyte (SE) LiI. Between 1984 and 1989, more than 80,000 Li-I2 batteries were produced in the soviet part of Germany and used for pacemakers, which had the longest service life at that time in all over the world. The leakageand short circuit-free design of this primary battery made it the perfect candidate for this medical application. The aspect of higher safety is still something that automatically is implied when talking about solid-state batteries, especially when possible applications are in the mobility sector. Images of burning electric vehicles (EVs) are going around the world, despite the fact that the total numbers are by far lower than for their combustion counterparts. However, today’s applications often have higher demands regarding energy and power density as well as long cycle life. Hence, other solid-state batteries (SSBs) were developed mainly categorized by their electrolyte type: oxide, sulfide, or polymer electrolyte. The number of scientifically related publications has increased tremendously over the last years. Especially, after the discovery of a solid electrolyte with ionic conductivities comparable to liquid electrolytes in 2001,1 sulfide-based solid electrolytes have initialized a true boom for solid-state batteries leading to an exponential growth in publication numbers and start-ups (Fig. 1). Solid electrolytes are enablers to rethink interfaces and cell chemistry fundamentally leading to a playground of new possibilities for scientists. A wide range of promising material developments and analytical approaches have been reported at the level of small cells, including the stable utilization of high energy lithium and silicon anodes that boost the specific energy and energy density of the cells to levels beyond state-of-the-art (SoA) Lithium-ion batteries.2 The commercial application of lithium metal anodes within conventional liquid electrolyte secondary lithium-ion batteries literally went up in flames (Moli Energy).3 The solid electrolyte is expected to be ultimate solution for this, since it physically blocks lithium dendrites from reaching the cathode. According to the literature, the volumetric energy density of an solid-state battery can be increased by up to 70% compared to its liquid counterpart by implementing a lithium metal anode, which would lead to ranges of electric vehicles >1000 km/charge (an often considered milestone for consumers; please note that the range of a vehicle is not directly related to the energy density of the battery pack).4 So much for the theory. However, reality is more complicated and with solid interfaces there come also numerous challenges to translate the promising results from small lab cells to larger prototypes. Depending on the type of electrolyte, many process steps need to be re-evaluated and adapted since the binder (to prepare electrode sheets) and the formation of an interface between the

Fig. 1 Number of publications on solid-state batteries, 1949–2024. (Data obtained from publication search in ScopusW data base by Elsevier.)

626

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

solid electrolyte and active material have a huge influence on the performance. Furthermore, solid-state batteries are quite sensitive toward their cycling conditions, e.g. temperature, stack pressure and so on.

1.1

What is the motivation behind developing solid-state batteries

Currently, state-of-the-art lithium-ion batteries are the ultimate solution for many applications. They increasingly replace lead-acid and nickel metal hydride systems. Their performance has improved constantly since their introduction to the market.5 Today’s mass-produced commercial cells deliver specific energies and energy densities
270 Wh/kg and 650 Wh/l with cycle lives exceeding 1000 cycles and more. A sharp decline in battery pack prices was observed reaching values below 100 $/kWh 2023.6,7 This was mainly achieved due to scaling effect and the latest decrease in raw material prices by adopting cheap existing positive electrode chemistries, such as lithium iron phosphate (LFP). LFP cells are intrinsically safer compared to its high energy alternative NMC and, hence, can be assembled in a battery pack using fewer safety measures, which usually reduce the energy density of the final device (so-called cell-to-pack configurations).8 The dilemma occurs because LIBs are under development now for nearly 40 years and show great stability, high energy density, and decreasing cost. Just recently, Mercedes demonstrated a 1000 km test drive with a single charge using common LIB by drastically increasing the energy efficiency of its prototype EV.9 An honest question would be: are solid-state batteries even required given the high technology level of SoA cells? Standard cells are not even at their end of development. By improving the anode and cathode chemistry step by step, small incremental improvements are made year by year. The answer is not clear but by enabling fundamentally new electrode chemistries like lithium metal, pure silicon or conversion-based cathode, SSBs would not just improve by small increments but make a big jump by up to 70% increase in energy density and 40% in specific energy.4 So there is a clear motivation from a scientific point of view. However, this only holds as long as promising results from the lab can be transferred to realistic prototype cells and mass production. These are two separate achievements and requirements in final applications are more complex. First, most results on SSBs in literature are based on unrealistic conditions like mm-thick separator pellets, high electrolyte contents and unrealistic high stack pressures (the situation is rather similar for liquid electrolyte cells based on coin cells). Giving into account that some results are obtained at pressures >50 MPa, it needs to be said that this can only be applied by heavy springs and steel frames, which would be unrealistic to apply in a real application.10,11 Fortunately, there is more and more literature about prototype cells being published and we need to learn more about how to cycle these cells under real conditions. On the one hand, electrolyte content and separator thickness of lab cells must be drastically reduced to obtain a competitive over-all energy density. On the other hand, it is required to add inactive weight and volume (e.g. electrolyte, additives, binder, current collectors, packaging) to construct a functioning cell. Later, cells must be assembled into a pack with additional structural elements (e.g. pressure plates, safety elements or further auxiliary systems like cooling and voltage control). Now, the claim of solid-state batteries is that they can be cycled at even higher temperatures than common LIBs due to the absence of volatile liquids.12 For example, Hitachi Zosen demonstrated that their solid-state 140 mAh prototype cell can be operated from −40 to 100  C at environmental pressures of 0.01 Pa.13 This would eliminate the need for complex cooling systems and pressure elements. Furthermore, solid-state batteries are suitable for a simplified bipolar stacking. Toyota being one of the pioneers in the field of sulfide-based solid-state batteries was working on such a setup. This would further reduce the need for complex auxiliary systems and safe weight on inactive components. What is automatically implied when talking about solid-state batteries, is the increased safety and stability. This might come from the image of a solid interface being stable and impregnable against dendrites etc. For sure, a solid electrolyte has a negligible chance to evaporate and exit the cell casing to form a flammable gas. It is expected that the risk of thermal events can be lowered substantially. However, it is safe to say that solid electrolytes are everything else than impregnable against dendrites and most solid electrolytes (or metallic lithium) are at least combustible (sulfide or polymers) when reacting with oxygen or form flammable and toxic gases (e.g. H2S) in contact with water, representing a clear safety risk. Comprehensive safety testing is essential to prove the higher safety of solid-state cells. However, it is impossible using small lab cells. In case of any safety event, the heat is completely absorbed by the massive cell casing shutting down following and more critical events. It is most critical that safety tests are being done with real prototype cells that store a considerably large amount of energy. So far, the literature about that is more than thin. There is proof that in the case of a charged high-nickel NMC, the cathode can certainly release oxygen at elevated temperatures, which reacts with the sulfide-based electrolyte similar to with a liquid electrolyte.14 To evaluate how this observation translates into standardized safety tests (nail-penetration, external short circuit, etc.), large prototype or battery packs are required, and this is rarely represented in peer-reviewed publications and more in press-releases of start-ups or companies. Samsung SDI published their zero-excess cell concept based on sulfide solid electrolytes. The cell will be described more in detail later, but the performed also safety tests on relatively large 1.2 Ah prototypes (Fig. 2).15 The cell was charged to 4.25 V and heated up to 210  C for more than an hour. While the 1.4 Ah LIB counterpart showed all signs of a classic thermal runaway, the SSB prototype remained its cell voltage and did not show any gas formation or internal heat formation despite the fact that the lithium metal at the anode side was already molten at this point. This result highlights the potential for EV battery development, however it is very specific by the type of electrolyte and active materials. It is crucial to provide profound safety and performance testing of realistic prototypes not only at the beginning but also at the end-of-life and different state-of-charge (SOC). This is mandatory to reach higher technology readiness level (TRL), since batteries always bear a certain chance for failures. If SSBs can hold their promise for higher safety, this would again mean that larger packs can

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

627

Fig. 2 Heating test of a lithium ion battery using liquid (a) and solid (b) electrolyte. Temperature was increased from 25 to 210  C at 5  C/min. From Lee, Y.-G.; Fujiki, S.; Jung, C.; Suzuki, N.; Yashiro, N.; Omoda, R.; Ko, D.-S.; Shiratsuchi, T.; Sugimoto, T.; Ryu, S.; Ku, J. H.; Watanabe, T.; Park, Y.; Aihara, Y.; Im, D.; Han, I. T. High-Energy Long-Cycling All-Solid-State Lithium Metal Batteries Enabled by Silver–Carbon Composite Anodes. Nat. Energy 2020, 5, 299–308. DOI:10.1038/ s41560-020-0575-z. Reproduced with permission of Springer Nature Limited.

Fig. 3 Technology readiness level scale for EV battery applications. From Frith, J. T.; Lacey, M. J.; Ulissi, U. A Non-Academic Perspective on the Future of Lithium-Based Batteries. Nat. Commun. 2023, 14 (1), 420. https://doi.org/10.1038/s41467-023-35933-2. Licensed under CC BY.

be manufactured using less safety measures (inactive material) required for safe usage, which then leads to higher cell-to-pack efficiencies. Second, to have any chance of getting into a larger scale application like EV market, the different SSB technologies must reach higher TRL levels. While the previous chapter was about reaching a TRL of 4 (to get a true estimation about the potential and risk of the respective technology), the following chapter will deal with different aspects of going beyond TRL 5 and 6 (Fig. 3). Scientists operating at TRL 1–4 are often not confronted with end-user requirements.16 However, a successful market entry of a new cell type heavily depends on satisfying the needs of end-consumers and on the complex connections between the requirements of the application and the cell itself. Since 1/3 of global battery production goes into the electric vehicle market, the public debate is mostly influenced by that. Based on the fact that the value of the battery pack makes up for up to 50% of the vehicle cost, one

628

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

constant motivation for cell development is always the reducing the cost ($/kWh). In some cases this becomes even more important than achieving higher energy densities, which can be seen at the latest hype about sodium ion batteries. Of course, by drastically increasing the storage capacity of active materials, the material costs per kWh should fall. In the lab-stage, a true cost estimation is not possible. Partially, this is due to material costs itself. Although, solid electrolytes contain more lithium than liquid electrolytes, prices that are paid at the moment (7–35 $/g) are purely made up and do not reflect actual raw material prices at the market. Once supply chains for new materials (active materials and electrolytes) are established, the prices will go down. Material costs are only one part of the final product. Manufacturing costs likely will be higher than 50% during prototype production. By optimization and scaling up the production, it is expected to cut this part down to less than 25%.17 A plant for solid-state batteries will likely be completely different from a plant that produces SoA LIBs.7 Introducing the solid electrolyte into mass production requires the development of completely new and disruptive processes and components for cell casing etc. The details of that will be highlighted later. However, there is no standard solid-state battery so far. Hence, there is no widespread standard for production machines and invest costs will likely be higher. Furthermore, most solid electrolytes or cell components are sensitive for moisture making it more complex and expensive to handle these materials at large scale. It takes several years to commission from capital expenditure to when production starts, and it is likely that new production processes will have a lower yield. Then, the cost forecast can result in unrealistic expectations of the competitiveness of the final cell and the final product will be more expensive than the existing state of the art.7

1.2

From semi- to all solid-state

The challenges in processing and scaling are enormous and only a limited number of companies are capable of producing such cells. The current news may suggest otherwise, since many start-ups promise that they have already established their technology in large scale. However, it is necessary to take a closer look and put everything in the right perspective, since it is a long road to establish mass production for the electric vehicle market. First, not every solid-state battery is a true all solid-state battery (ASSB). To minimize the gap between theoretical and practical performance, so-called hybrid systems have been proposed. A soft material like a polymer or liquid can help to bridge grain boundaries between two solid particles. Recent studies have shown that a liquid electrolyte contributes toward an improved electrochemical performance.18 It is common that such batteries with a substantial fraction of liquid electrolyte (polymer gel electrolytes with solvent additives; liquid fraction 10–15 wt%) are still considered as Semi-SSBs (Fig. 4).19 For simplicity, they are often just called solid-state batteries, which is misleading. When the majority of charge transfer is governed by the ions in a liquid electrolyte, the solid electrolyte particle simply acts as a filler, a quite expensive filler, and the whole system is merely a modification of current LIBs. They can be processed with well-established slurry coating routines. Hence, required adaptions of existing factories and production lines are negligible, which reduces the potential production cost and results in an easier scale-up. This is a good argument for market entry, since the battery pack makes up a major fraction of cost for an electric vehicle. However, this cell concept might just be a marketing strategy and not a real advantage of SSB over LIB. Companies like Factorial Energy and WeLion manufacture already prototype cells with capacities 100 Ah and strive toward mass production. Just recently, NIO placed a 150 kWh Semi-Solid-state battery pack inside its ET7 and demonstrated a 1000 km driving range with only a single charge.20 The cell consists of Si-C composite anode and a high nickel cathode delivering an energy density of 360 Wh/kg at rather high solvent content. However, a drastic increase in safety will only be achieved when the fraction of liquid electrolyte becomes much smaller. To move from semi-SSBs to All-solid-state batteries (ASSBs), the liquid fraction needs to be reduced completely or almost. In some cases, a certain wetting of interfaces with liquids or polymers might help to bridge grain boundaries or increase the mechanical stability of battery components. However, in the following, we will focus solely on “all-solid” concepts.

2

Processing of large-size cells

The successful commercialization of solid-state batteries is critically linked to the up-scaled production of large cells. Otherwise critical evaluations such as safety tests or cost assumption won’t have any meaning. To compete with existing SoA production processes, there are some basic requirements that approaches from the literature do often ignore. A slow production speed is often seen as a critical showstopper. Current slurry coaters can operate at a production speed of 100 m/min but require vast amounts of energy for drying and solvent recovery. A lower production speed might be compensated if the system has a lower footprint and consumes less energy, but the decision is very specific. Another elemental requirement is that the double-sided coating of electrodes must be possible. Otherwise, it becomes impossible to prepare prototype cells with practical capacity values. Since solid-state batteries fundamentally depend on new electrode systems, new mixing and coating techniques need to be established, which strongly depends on the type of electrolyte (polymer, sulfide, or oxide) and active material (Fig. 5).21 For LIBs, composite electrodes always consist of particle-based active material that are casted via a slurry process with minor adaptions. For solid-state batteries, additional steps like consolidation or sintering are required. Processes for composite electrodes can be adapted with some effort (new solvent and binder), while the fabrication of solid electrolyte separators differs completely from existing

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

629

Fig. 4 Comparison of various cell concepts with different fractions of liquid components. From Huo, H.; Janek, J. Solid-State Batteries: From ‘All-Solid’ to ‘AlmostSolid’. Natl. Sci. Rev. 2023, 10 (6), nwad098. https://doi.org/10.1093/nsr/nwad098. Licensed under CC BY.

processes for polyolefin separators. Also, the integration of bulk anodes like lithium or silicon will require completely new processes. Since, there is no standard SSB, every solid-state battery factory will have its own special design.

2.1

Lithium metal anode

Although the lithium metal anode is considered as the holy grail for battery scientists, the up scaled manufacturing of lithium metal is a great challenge that needs to be solved in order enable this high-energy anode for the broad market. Conventional top-down lithium foil production is carried out under inert atmosphere and includes an extrusion process. The product is a foil with a minimum thickness of 100 mm (20 mAh/cm2), which displays a large excess when combined with standard cathodes.22 Roll pressing is commonly employed to reduce the thickness further. Currently, 20 mm Li-foil is commercially available. However, as the thickness decreases, the width of foil becomes also narrower due to fraying out of the edges and the surface of lithium is covered with lubricants.7 Free-standing lithium foil is very fragile to handle due to its poor mechanical properties. Hence, it needs to be laminated on a current collector foil like copper or stainless steel.23 All these steps are not trivial to scale-up for production of EV-batteries.

630

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

Fig. 5 Process chains for solid electrolyte separator fabrication. From Schnell, J.; Günther, T.; Knoche, T.; Vieider, C.; Köhler, L.; Just, A.; Keller, M.; Passerini, S.; Reinhart, G. All-Solid-State Lithium-Ion and Lithium Metal Batteries – Paving the Way to Large-Scale Production. J. Power Sources 2018, 382, 160–175. DOI:10.1016/j.jpowsour.2018.02.062. Reproduced with permission of Elsevier.

Bottom-up approaches include PVD24 or melt-depositions25 and are suitable to produce thin layers ( 30 m (Fig. 7). Second, binder and solvents need to be adapted. Standard binder for LIB are based on polar solvents and polymers (e.g. PVDF). Sulfide SE exhibit a certain instability against polar or protic solvents. They dissolve or at least deteriorate in most solvents or in contact with residual moisture. Moving to non-polar solvents limits the selection of available polymer binders that can still be dissolved and show a good interaction with the active material or

Fig. 7 Schematic illustration of a slurry-based battery electrode coating line with slurry preparation, coating, drying, calendering, rewinding and solvent-recovery. ©Fraunhofer IWS.

632

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

SE particle. One solution is found in co-polymers like NBR, HNBR, SBR or acrylates,15,34,35 which contain polar segments that interact with the composite and non-polar segments that lead to a good solubility in solvents like xylene or toluene. The used solvents must be recaptured after evaporation and drying, purified by distillation and recovered for reuse in the process. Although, composite electrodes can be prepared this way, the binder itself is a limiting factor in terms of performance. After removing the solvent, the polymer binder covers parts of the surface area of active materials and acts as an electric and ionic isolator. In solid-state batteries, the surface coverage of active materials with the electrolyte is lower than for liquid electrolytes. Further reducing the surface leads to a bottleneck for ion transport.36,37 To compensate for that, cells are often cycled at 60  C or higher, which leads to a decreased energy efficiency. An alternative to that is the so-called dry battery electrode (DBE), which omits any kind of solvent during its production and, therefore also any kind of negative interaction between solvent and SE. There are different approaches described in the literature that are suitable for bulk batteries.38,39 Dry spraying is one of them where the blended particles are sprayed on the current collector. An electrostatic field can help to improve the efficiency of the deposition. A subsequent hot rolling then melts the binder and increases the adhesion.40 Dry spraying deposition can be used with most types of particles and the choice of binder is relatively simple. However, controlling mass loading, thickness and homogeneity might be difficult. Hot pressing and melt extrusion are well established techniques in the polymer industry and are based on the thermoplasticity of polymers above the crystallization temperature. They work well for solid-state polymer electrolytes and are easily scalable to produce large sheet-type electrodes. However, it becomes challenging to manufacture electrodes with low or no polymer fraction (e.g. sulfide-based SSBs), where particles can easily be damaged or agglomerate under the influence of too high shear forces. Fibrillation processes depict the most advanced method to prepare dry battery electrode (DBE) and attracted world-wide attention after Tesla bought Maxwell and its attached intellectual property (IP).41 For sulfide-based solid-state batteries, it was already demonstrated in 2019 that dry processing with the Poly(tetrafluoroethylene) (PTFE) binder can be realized at extremely low binder amounts down to 0.1 wt% (Fig. 8).42,43 In addition to that, the binder does not cover the surface of active materials but forms a spider-web like structure entangling the active material particle and the electrolyte. Hence, ionic pathways are not altered and the influence of the binder on the cathode impedance can be neglected. While slurry coatings often have their problems with high loadings due to crack formation, loadings up to 10 mAh/cm2 have been realized already using PTFE as a binder.44,45 The limiting factor here is not the binder anymore but the morphology of the cathode itself defined by particle packing of active material and solid electrolyte. This approach can also be used to prepare electrolyte separators with competitive thicknesses 8 mS/cm are possible (depending on the used SE).46 The stability of PTFE against the lithium is a problem, since it can facilitate dendrite growth. Nevertheless, it was shown that stable operation is possible against anodes like silicon or as long as the lithium it protected.47 This type of DBE production has the potential to preserve the performance of the cathode without compromise and avoids issues regarding the chemical stability of the SE. However, limitations in terms of binder choice and scalability are present. Under certain conditions, the use of alternative binders is possible that might bypass implications with the anode.48 The question of scalability raised since many results in the literature are based on manual kneading and film formation with low production speed. Calender-based dry electrode formation enables continuous production of electrode layers and has been used industrially for different applications like NiMH batteries, fuel cells and super capacitors for the last decades. The active material is mixed with conductive additives and PTFE. Under impact of shear forces, the PTFE form fibrils that are several tens of micrometers long and entangle the particles in a network-like structure. In a subsequent step, this composite is pressed between to calender rolls producing a film of several hundred micrometers in thickness. In further rolling devices, the film is rolled out to the desired target thickness. The free-standing film is finally laminated onto the substrate foil to form the finished electrode tape. Such a multi-state film

Fig. 8 Schematic diagram of dry electrode fabrication based on a dry premixing of NCM, C, SE, and PTFE binder followed by shearing force induced film formation. From Hippauf, F.; Schumm, B.; Doerfler, S.; Althues, H.; Fujiki, S.; Shiratsushi, T.; Tsujimura, T.; Aihara, Y.; Kaskel, S. Overcoming Binder Limitations of Sheet-Type Solid-State Cathodes Using a Solvent-Free Dry-Film Approach. Energy Storage Mater. 2019, 21, 390. DOI:10.1016/j.ensm.2019.05.033. Reproduced with permission of Elsevier.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

633

Fig. 9 Illustration of DRYtraecW-based electrode coating. Left: Scheme of a production equipment for Roll-to-Roll DBE coating with relatively low footprint. Right: Scheme of tandem film formation and coating process step based on different roller speeds, film formation on the faster roller and transfer to the current collector foil. ©Fraunhofer IWS.

formation process based in free-standing tapes induce several risks. The free-standing film requires a certain mechanical tensile strength. If the binder content is extremely low, the risk of film cracking is extremely high, especially for thin films. This would cause an entire production stop. To prevent such risks, a roll-supported process route during the formation of the DBE was proposed (DRYtraec®).49,50 In this variant of the DBE-formation, the pre-treated mixture is placed directly in a calender nip. Different rotational speeds of the rolls apply a shear force directly at the center of the nip leading to the formation of a film on the faster roll. The intermediate of a free-standing film is avoided until the film is transferred from the roll toward the substrate foil. A major advantage of the system is that the target loading is directly adjusted in the first nip by adjusting the roller distance and speeds. Thus, further rolling out can be avoided reducing the footprint of such a system. The minimum size of such a coating device consists of a 3-roll calender. If a fourth roll is implemented, DRYtraec® can carry out simultaneous double-side coatings, which is complicated for slurries (Fig. 9) The production speed of calender-based routes is usually limited to the powder feed and the calender hardware. With, DRYtraec, coating speeds of 5 m/min were demonstrated, which is lower than for slurry coatings. Since there is no need to place a 100 m dryer under inert atmosphere, several calender systems could operate in parallel to compensate for the lower production speed. In combination with a superior rate performance, DBE is a unique match for solid-state batteries (maybe all-dry pouch cell). In summary, dry battery electrode coating displays enormous chances and advantages for future green production, e.g. lower energy demand and compatibility for next-generation batteries, lower equipment footprint. It is clear that this technology must play an important role in realizing the ambitious growth rates for battery production and will gradually complement the existing slurry technology. Dry coating will thus provide an essential basis for a sustainable and resource-efficient electrode fabrication. It is also clear that further material, process and equipment innovations will be needed along this path over the next years to help the technology achieve a breakthrough toward higher TRL levels.

3

Practical inorganic solid-electrolyte cells

Due to the previously mentioned characteristics and promises that the solid-state technology holds, current trends in research and development of all solid-state batteries can be divided into two types. One is the thin film battery (TFB) with capacities in the range of mAh to several mAh. The production of such small-capacity ASSBs is well established, and a niche market has been created without much of public attention. This type of cells is often linked to oxide-type solid electrolytes. The main reason is the high grain boundary resistance between SE and active material, which limits the active material loading. Oxides are usually much harder than sulfide SE and they cannot be cold-sintered at room or moderate temperatures. Since higher temperatures often promote side reactions between active material and SE, fabrication techniques are limited to thin film deposition techniques. A novelty display hybrid SSBs, where an oxide separator can be fabricated at high temperatures separately from active materials. Companies like ProLogium and QuantumScape are following this route. Promising results were also presented by Wachsman et al.51–54 The market for solid-state secondary batteries actually began with so-called surface mounted devices (SMDs) based on TFBs. Cymbet developed such TFBs for wearables and small medical sensors since it was founded in 2000. TDK also successfully has developed their multi-layer ceramic capacitor (MLCC)-type solid-state battery. Typical applications are wearables, Internet of Things (IoT) gadgets, or SMDs. Assuming no critical requirements in energy density and current density those type of ASSBs suggest further possibility of development for efficient smart devices because they are thin and small in shape, and extremely reliable even under harsh conditions. The other trend is more toward bulk cells with considerable high capacities (1–100 Ah) caused by the strong market driving force of xEV applications. Due to their outstanding ionic conductivity, process ability at room temperature and promising results in literature, sulfide-based electrolytes are at the center of latest developments. For example, Toyota has recently announced their

634

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

commitment to start commercial use of their ASSBs by 2027/28. They have already demonstrated different electric-powered vehicles using their sulfide-based ASSBs,55 among them fully license plate registered cars that conducted test drives to collect performance data.56 Samsung also reported the fabrication of large-format laminated cells based on sulfides combining their anode-free cell design with innovative solvent electrode fabrication.15 In the following section, we will explore selected cell setups that seem close to commercialization. However, there is no claim of being complete, since new statements about solid-state batteries are emerging on a daily basis.

3.1

Sulfide-based bulk cells

Sulfide electrolytes gain the highest attention from the EV-sector at the moment. Most prominent players in the field are Samsung SDI, Solid Power and Toyota as well as a bunch of smaller Start-ups. Toyota is one of the pioneers and holds the most patents in this technology. Together with the Tokyo Institute of Technology (Tokyo Tech), they published a significant paper in 2011 on a new solid electrolyte Li10GeP2S12 (LGPS) outperforming liquid electrolytes in terms of ionic conductivity.57

3.1.1 Toyota With its total number of 1331 registered patents related with solid-state battery technology, Toyota is the leading company on the development of all solid-state batteries using sulfidic solid electrolytes (SSEs).58 In particular, innovation of novel SSEs gained a significant pace after Toyota founded its battery research division in 2008.59 The collaborative work of the division with the Tokyo tech in 2016 leaded to the development of the SSE Li9.54Si1.74P1.44S11.7Cl0.3 (LiSiPSCl) with an outstanding ionic conductivity >25 mS/cm.60 Thereby, they have demonstrated a LiSiPSCl-based 0.7 mAh ASSB cell that can be charged and discharged as fast as 3 min for >1000 cycles at 25  C (C-rate: 18C). Even though the cell properties were far away from the realistic conditions (low loading, thick separator pellet etc.), the thermal stability (up to 100  C) and power capability of the cell at elevated temperatures (C-Rate: up to 1500C) are extremely promising. In 2017, with the aim of boosting development of novel SSE chemistries, Toyota research institute has also announced $35 million investment on computer technologies like machine learning and artificial intelligence.61 Only 3 years after that, the company introduced its first prototype battery electric vehicle (BEV) at the Tokyo Olympic games,56 which is fully powered by all-solid-state batteries during its drive. Yet, there was no concrete disclosure on the utilized cell chemistry for the prototype battery pack of the BEV (Fig. 10).62 Chirro Yada, the head of Toyota motor Europe, has stated in an interview that the cathode material relies on the lithium transition metal oxides with a typical voltage of 4 V.63 Probably, the battery pack was not optimized in terms of energy density since the company made no statement on the driving range of the BEV back at the time. In the late 2023, Toyota took a serious step after signing a cooperation together with Idemitsu Kosan co. Ltd., who has decades of experience on the development/production of SSEs and holds the third highest number of registered patents on ASSBs (Fig. 11). The close partnership between the companies dates back to 2013, where they initiated a joint research project on the development of crack-resistant solid electrolytes, resulting in a successful innovation on SSEs.64 Idemitsu is originally a petroleum company and utilizes oil wastes to produce compounds like Li2S, which is a core ingredient required for manufacturing sulfidic solid electrolytes. The company holds sufficient raw materials and a pilot production facility for SSEs. According to the joint announcement in the press conference, a 3-phase route map is planned to realize the commercialization of SSBs particularly for BEVs with a driving range > 1000 km and 10-min charging time (SoC ¼ 10–80%) by 2027/2028.64,65 By that, both parties aim not only to enable mass production of SSEs but also to ensure sustainable material supply for future BEVs.

Fig. 10 Toyota’s prototype battery module based on all solid-state battery cell stacks.62 This photo was obtained from Toyota under terms of their download agreement.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

635

Fig. 11 Koji Sato (left), the president of the Toyota Motor Corporation and Shunichi Kito (right), the president of Idemitsu Kosan shake hands at a press conference in Tokyo on October 12th in 2023.64 This photo was obtained from Toyota under terms of their download agreement.

3.1.2 Samsung Next to Toyota, one of the pioneers in the area of sulfide-based solid-state batteries is Samsung. At least from the beginning of their research, there are publications available from their Japan-based research branch SRJ and first publications date back to 2013. Next to that, Samsung also holds many patents in that area. Already in 2014, they published a practical relevant rocking chair type allsolid-state lithium-ion battery. They focused on important bottle necks in the battery like a buffer coating at the cathode material that limits side reactions with the electrolyte and developed slurry-processes to fabricate sheet electrodes and separators. In comparison to many literature results from that time (based on mm thick pellets), Samsung already demonstrated a 1 Ah class demonstrator cell with a thin separator of 200 mm (Fig. 12).66 The cathode was based on zirconate coated NCA coated by a xylene-based slurry with NBR binder. The anode consisted only of graphite, which was surprising that the ionic conductivity of graphite alone is sufficient to operate the anode. This enabled the choice of different binder and solvents and simplifies the anode fabrication. The Poly(vinylidene fluoride) (PVDF) binder does not dissolve in the xylene slurry. Hence, it was possible to cast the separator directly on top of the anode and assemble it together with the cathode after drying. If the binder dissolves again during a coating on top the electrode, the coated slurry may intrude deeply in the layer beneath. They prepared a 125 mAh cell that showed

Fig. 12 (a) The 1 Ah class demonstration cell. The cell consists of three parallel stacks of the single cells (coated on both sides); (b) 0.1C Discharge curves of (A) a single cell (single side coated), (B) a single cell (both sides coated) and (c) stacked battery (three parallel stacks of (B)) at 25C. From Ito, S.; Fujiki, S.; Yamada, T.; Aihara, Y.; Park, Y.; Kim, T. Y.; Baek, S.-W.; Lee, J.-M.; Doo, S.; Machida, N. A Rocking Chair Type All-Solid-State Lithium Ion Battery Adopting Li2O–ZrO2 Coated LiNi0.8Co0.15Al0.05O2 and a Sulfide Based Electrolyte. J. Power Sources 2014, 248, 943–950. DOI:10.1016/j.jpowsour.2013.10.005. Reproduced with permission of Elsevier.

636

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

an 85% capacity retention after 100 cycles at 60  C without any artificial external pressure. Finally, a multi-layer pouch cell with 1 Ah was demonstrated at practical loading of 2.5 mAh/cm2. Although the binder amount was optimized, even a small amount of binder can significantly influence the cell impedance since the binder cannot be swollen by liquid electrolyte. The polymer restricts the reaction surface area and hinders charge transfer between active material and SE. The process and the binder material must be carefully optimized to develop a realistic prototype cell. In order to improve their cell performance, they worked on solvent free electrode fabrication. The utilized PTFE binder forms a network of fibrils that entangle the active material and do not interfere with ionic pathways through the electrode. This enables supreme rate performance even at high loadings >6 mAh/cm2, which is relevant to achieve practical energy densities.42 However, the best cathode is useless without a suitable anode that enables the uptake of lithium at relevant rates. The holy grail of lithium ion batteries is lithium metal and it becomes even better, if it does not exist in its elemental form at the initial state. Sometimes these cell concepts are called anode-free although there still is an anode.26 The lithium metal forms directly during charging at the current collector as lithium ions from the cathode are reduced. Since, there is no additional lithium incorporated except for the lithium inventory of the cathode, these anodes should be called zero-excess anodes and have the maximum theoretical energy density possible. The problem is that the naked current collector foil is never perfectly contacting the solid electrolyte and lithium metal that form in-situ tends to grow in larger crystals instead as a uniform layer. Hence, dendrites can easily form and penetrate the separator. Samsung approached this problem by developing a soft interlayer that bridges the gaps between the current collector and the electrolyte.15,67 They used a thin slurry coating of carbon enriched with silver nanoparticles, which enhance the ionic and electric conductivity of the layer. Such a cell stack can be compressed uniformly at ultra-high pressures without lithium metal creeping, which simplifies cell assembly. Compared to their previous prototype, areal loading and specific capacity of the NMC were drastically increased to >6.8 mAh/cm2 and > 210 mAh/g, respectively (Fig. 13). Furthermore, they could minimize the thickness of the separator to 30 mm, which is close to commercial polyolefin separators of classic LIB. In most liquid electrolytes, such a cell concept would fade relatively rapid as the coulomb efficiency is low due to constant SEI formation at the anode.68 The crux about solid electrolytes is that the anode interface stabilizes quickly. Hence, Samsung could demonstrate a small 0.6 Ah bi-layer prototype cell that cycled for more than 1000 cycles with a capacity retention >80%. Therefore, two anodes were placed around a double-sided NMC cathode. Superior rate performance was demonstrated even at temperatures below 0  C due to the minimized cell impedance of the dry cathode. A 10-layer pouch cell (5.8 Ah) was assembled as well, that could be successfully charged and discharge and achieved a record breaking energy density of >900 Wh/L. After that, it went relatively silent in terms of publishing scientific results except for a few announcements. In late 2023, they announced that they have currently completed their pilot line for solid-state batteries in Suwon.69 At the beginning of 2024, further details followed.70 They confirmed that they will use warm isostatic pressing in order to compact their solid-state batteries. The method was already described in their earlier publication. However, it was unclear if this method could be transferred to an industrial process. Within the interBattery 2024 in Korea, they finally announced that Samsung SDI will start its mass production of solid-state batteries by 2027 (Fig. 14).72 The company refers to an all solid-state concept achieving a volumetric energy density of 900 Wh/L. This indicates that the concept is based on their zero-excess anode, which was published in 2020, with some minor adaptions. The pilot plant in Suwon is currently delivering prototype cells and the company stated that every aspect of its plan for mass production will be addressed until 2027 from development, production line, project launch to supply chain management.

Fig. 13 Schematic of an ASSB composed of a NMC cathode with a high areal capacity (>6.8 mAh cm−2), SSE and a Ag-C nanocomposite anode layer that does not require excess Li. Al and SUS foil were used as current collectors for the cathode and anode, respectively. From Lee, Y.-G.; Fujiki, S.; Jung, C.; Suzuki, N.; Yashiro, N.; Omoda, R.; Ko, D.-S.; Shiratsuchi, T.; Sugimoto, T.; Ryu, S.; Ku, J. H.; Watanabe, T.; Park, Y.; Aihara, Y.; Im, D.; Han, I. T. High-Energy Long-Cycling AllSolid-State Lithium Metal Batteries Enabled by Silver–Carbon Composite Anodes. Nat. Energy 2020, 5, 299–308. DOI:10.1038/s41560-020-0575-z. Reproduced with permission of Springer Nature.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

637

Fig. 14 Employees of Samsung SDI presenting actual all-solid-state prototype cells at InterBattery 2024 in Seoul, Korea.71 This photo was obtained from Samsung SDI.

3.1.3 Solid Power Solid Power was founded in 2011 by Doug Campbell, Conrad Stoldt, and Sehee Lee as a spin-off from the University of Colorado Boulder. Since then, they raised more than 700 million USD of funding from investors.73 Their core technology is based on their own proprietary sulfide-based solid electrolyte produced in their own pilot electrolyte production facility since early 2023. The targeted production capacity is 30–60 metric tons per year by 2028, which will enough to power 800,000 electrified vehicles. They partnered with SK Innovations (probably about cathode active materials) as well as leading OEMs like BMW and Ford. The main strategy in their early development was to leverage existing lithium-ion manufacturing equipment to ensure minimal CapEx investments. Their cell manufacturing processes are already used globally for a high volume of traditional LIBs, which will enable possible manufacturer of their licensed SSB technology to meet volume and cost requirements of OEMs. This means they focused on standard slurry-based coating technologies for their separator, anode, and cathode sheets. After slitting and calendaring, electrode sheets are stacked to form a multi-layer pouch cell. However, it is not clear how they consolidate the cell stack. Their first generation of solid-state batteries is based on a high-content silicon anode, which enables high charge rates (1000 cycles, which outperforms existing SoA lithium-ion batteries. Replacing their silicon anode with thin lithium metal increases the specific energy up to 440 Wh/kg. In general, the cell chemistry works, and Solid Power produced 20 Ah prototype cells on their production equipment already in 2020. However, their cells at that time exhibited “only” a specific energy of 320 Wh/kg suggesting a lower loading or thicker separator. As described before, handling lithium metal at large scale leads to numerous challenges. By replacing the NMC cathode with a conversion-type FeS2 cathode material, they aim to supply a cell chemistry with an even higher specific energy of 560 Wh/kg. By getting rid of nickel and cobalt based active materials, they expect a 15–35% cost advantage over existing lithium-ion batteries at the pack level. Solid Power continued their cooperation with OEMs and they delivered 20 Ah prototype cells in 2022 and first 60 Ah A−1 EV cells to BMW in late 2023 (Fig. 16).75

Fig. 15 Schematic of Solid Power’s different all solid-state battery technologies including design targets for cell performance properties.74 This photo was obtained from Solid Power.

638

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

Fig. 16 An employee of Solid Power presents two different sized all-solid-state prototype cells.76 This photo was obtained from Solid Power.

3.2

Oxide-based micro-batteries

In the shadow of the hype around sulfide-based solid-state batteries, small oxide-based solid-state batteries based on multi-layer ceramic capacitors (MLCC) have already been commercialized by different companies.77,78 The advantage of using the MLCC technique is that is already established for capacitors and that stacking of more than 100 layers is possible (Fig. 17). This enhances the capacity of such batteries compared to classic LIPON-based thin film batteries. The MLCC manufacturing process is the following: Starting from a suspension of raw materials and binder, the so-called green sheets are casted. A pair of two electrodes and an SE layer (often NASICON) is stacked together with the current collector precursors. The electrodes are stacked with a slight offset relatively to each other. Hence, cathode and anode can be contacted later from different sites of the stack. High precision is required to stack more than 100 layers. After the binder is burned out, the full stack is sintered at temperatures above 1000  C to obtain the multilayer battery. Finally, the cell stack is contacted by soldering from opposite sites and the battery can be mounted. TDK was one of the first companies to release such a multilayer micro-battery. The size of their CeraCharge™ product is 4.4  3.0  1.1 mm. The capacity is 0.1 mAh at a nominal cell voltage of 1.5 V (Fig. 18). The constant discharge current can be up to 1 mA. However, the cell endures also short pulse currents of up to 5 mA. Taiyo Yuden presented a MLCC battery in 2018.81 Their cell comes in dimensions of 8x8x4 mm and consists of a multilayer stack of 200 layers. An alternative but larger cell measures 32  20 mm and 30 layers. Compared to TDK they offer higher cell voltages of 3.2 and 2.4 V with 1 mAh and 0.7 mAh capacity respectively. FDK followed a similar approach by using their own high voltage cathode material Li2CoP2O7. Against lithium, it offers a nominal discharge voltage of 4.7 V resulting in a final cell voltage of 3 V (Fig. 19). Due to optimization of inner structure and layer thickness (and higher cell voltage), their 0.5 mAh cell shows an increased energy density of 65 Wh/L compared to conventional products. The company Maxell is also successful in this field. Their cell voltage is moderate compared to other cells (2.3 V; probably LCO vs. LTO) (Table 1). However, they specialized in preparing relatively large MLCC batteries with capacities of 8 and 16 mAh. Furthermore, they developed a prototype with 200 mAh, which is outstanding for this type of battery. So far, the highest energy density among MLCC solid-state batteries was reported by Murata.79 They used a high-capacity cathode material Li3V2(PO4)3 delivering a cell voltage of 3.8 V83 with capacities of up to 25 mAh. This results in an extremely high energy density of 158 Wh/L.

Fig. 17 Schematic of MLCC solid-state battery.79 This photo was obtained from Murata.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

639

Fig. 18 Typical discharge characteristic of TDK’s CeraCharge™, Constant current discharge with 20 mA (0.2C), 100 mA (1C), 200 mA (2C), 500 mA (5C), 1000 mA (10C) to 0 V.80 This photo was supplied by TDK with publishing permission.

Fig. 19 Left: Discharge characteristics of Li2CoP2O7 versus Lithium, right: FDK’s Small all-solid-state SMD battery.82 This photo was supplied by FDK with publishing permission.

Table 1

Comparison of different commercially available MLCC-type solid-state batteries.

Dimensions/mm Capacity/mAh Cathode Cell Voltage/V

TDK

FDK

Taiyo Yuden

Murata

Maxell

4.4  3.0  1.1 0.1 LiVPO4 1.5

4.5  3.2  1.6 0.5 Li2CoP2O7 3

884 1 LiCoPO4 3.2

5  5  2/10  10  6 25 Li3V2PO4 3.8

10.5  10.5  4 8–200 LCO? 2.3

Such micro-batteries are expected to be mounted on circuit boards for IoT and wearable devices. All suppliers announced extreme cycle lifes of >1000 cycles and stability in a broad temperature window with low self-discharge even under harsh conditions. The calendar aging of these batteries is negligible making them a perfect solution for applications with long service life (>20 years) and good reliability (e.g. energy harvesting etc.). These are all niche applications with special needs but the mentioned energy densities are far away from projected values of sulfide-based bulk batteries. At least right now, production capacities are far away from several TWh per year, which is the projected global demand for batteries in 2030.84 In 2020, Murata announced that they will be able to produce 100,000 MLCC batteries per month, which scales to an annual production of 114 kWh.83

4

Conclusion

The most promising applications for true all solid-state batteries are EVs and most car companies as well as battery suppliers are currently putting great efforts in their development. Since CO2 emission regulations are becoming stricter every year all over the

640

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

world and prices for fossil fuels are rising, a large percentage of cars with combustion engines will be replaced by environmentally friendly cars in the future. It is not sure if battery-powered electric vehicles will dominate the market of the future. Fuel cell vehicles also depict a very promising technology. However, even they require a small battery and the very high energy efficiency of a battery-power train is a great advantage. In every scenario, this will induce a large demand for secondary batteries with highest safety and energy density. As mentioned above, solid-state electrolytes promise a higher safety than liquid ones, which is particularly important for large-sized EV batteries. Additionally, the energy density of the ASSB battery pack is expected to increase not only because of new high energy cell chemistries, but also because ASSB can omit different safety and cooling systems that contribute to the mass of the battery pack. In recent years, there has been a vast development in the market by start-ups and companies with many very promising results. Some of them are really advanced and on their way to mass production delivering prototype cells to car manufacturers, where they are implemented in first vehicles. If they can truly compete with cheap liquid electrolyte cells and manage a broad market entry, time will tell.

References 1. Kanno, R.; Murayama, M. Lithium Ionic Conductor Thio-LISICON: The Li2 S - GeS2 - P 2 S 5 System. J. Electrochem. Soc. 2001, 148 (7), A742–A746. https://doi.org/ 10.1149/1.1379028. 2. Passerini, S., Bresser, D., Moretti, A., Varzi, A., Eds. Encyclopedia of Electrochemistry: Batterries. Solid-State Batteries with Inorganic Electrolytes; Wiley, 2020. 3. Xie, J.; Lu, Y.-C. A Retrospective on Lithium-Ion Batteries. Nat. Commun. 2020, 11, 2499. https://doi.org/10.1038/s41467-020-16259-9. 4. Janek, J.; Zeier, W. G. A Solid Future for Battery Development. Nat. Energy 2016, 1 (9), 1167. https://doi.org/10.1038/nenergy.2016.141. 5. Raccichini, R.; Qiao, M.; Dongare, P. The Journey of an Electrifying (R)evolution. Nat. Commun. 2021, 12. https://doi.org/10.1038/s41467-021-24410-3. 6. Ziegler, M. S.; Trancik, J. E. Re-Examining Rates of Lithium-Ion Battery Technology Improvement and Cost Decline. Energy Environ. Sci. 2021, 4, 1635–1651. https://doi.org/ 10.1039/D0EE02681F. 7. Frith, J. T.; Lacey, M. J.; Ulissi, U. A Non-Academic Perspective on the Future of Lithium-Based Batteries. Nat. Commun. 2023, 14 (1), 420. https://doi.org/10.1038/s41467023-35933-2. 8. Yang, X.-G.; Liu, T.; Wang, C.-Y. Thermally Modulated Lithium Iron Phosphate Batteries for Mass-Market Electric Vehicles. Nat. Energy 2021, 6 (2), 176–185. https://doi.org/ 10.1038/s41560-020-00757-7. 9. Holzer, H. Mercedes Vision EQXX Rekord: 1000 Kilometer mit einer Akkufüllung; 2022. https://www.adac.de/rund-ums-fahrzeug/autokatalog/marken-modelle/mercedes-benz/ mercedes-eqxx/ (Accessed 28 May 2024). 10. Chen, Y.-T.; Jang, J.; Oh, J. A. S.; Ham, S.-Y.; Yang, H.; Lee, D.-J.; Vicencio, M.; Lee, J. B.; Tan, D. H. S.; Chouchane, M.; Cronk, A.; Song, M.-S.; Yin, Y.; Qian, J.; Chen, Z.; Meng, Y. S. Enabling Uniform and Accurate Control of Cycling Pressure for All-Solid-State Batteries; 2023. 11. Tan, D. H. S.; Chen, Y.-T.; Yang, H.; Bao, W.; Sreenarayanan, B.; Doux, J.-M.; Li, W.; Lu, B.; Ham, S.-Y.; Sayahpour, B.; Scharf, J.; Wu, E. A.; Deysher, G.; Han, H. E.; Hah, H. J.; Jeong, H.; Lee, J. B.; Chen, Z.; Meng, Y. S. Carbon-Free High-Loading Silicon Anodes Enabled by Sulfide Solid Electrolytes. Science 2021, 373 (6562), 1494–1499. https://doi. org/10.1126/science.abg7217. 12. Horowitz, Y.; Schmidt, C.; Yoon, D.; Riegger, L. M.; Katzenmeier, L.; Bosch, G. M.; Noked, M.; Ein-Eli, Y.; Janek, J.; Zeier, W. G.; Diesendruck, C. E.; Golodnitsky, D. Between Liquid and All Solid: A Prospect on Electrolyte Future in Lithium-Ion Batteries for Electric Vehicles. Energ. Technol. 2020, 8 (11), 2000580. https://doi.org/10.1002/ ente.202000580. 13. Anon. All-Solid-State Lithium-Ion Batteries. https://www.hitachizosen.co.jp/english/business/field/functional/as-lib.html (Accessed 28 May 2024). 14. Kim, T.; Kim, K.; Lee, S.; Song, G.; Jung, M. S.; Lee, K. T. Thermal Runaway Behavior of Li6PS5Cl Solid Electrolytes for LiNi0.8Co0.1Mn0.1O2 and LiFePO4 in All-Solid-State Batteries. Chem. Mater. 2022, 34 (20), 9159–9171. https://doi.org/10.1021/acs.chemmater.2c02106. 15. Lee, Y.-G.; Fujiki, S.; Jung, C.; Suzuki, N.; Yashiro, N.; Omoda, R.; Ko, D.-S.; Shiratsuchi, T.; Sugimoto, T.; Ryu, S.; Ku, J. H.; Watanabe, T.; Park, Y.; Aihara, Y.; Im, D.; Han, I. T. High-Energy Long-Cycling All-Solid-State Lithium Metal Batteries Enabled by Silver–Carbon Composite Anodes. Nat. Energy 2020, 5, 299–308. https://doi.org/10.1038/ s41560-020-0575-z. 16. Liu, J.; et al. Pathways for Practical High-Energy Long-Cycling Lithium Metal Batteries. Nat. Energy 2019, 3, 180–186. https://doi.org/10.1038/s41560-019-0338-x. In press. 17. Duffner, F.; Mauler, L.; Wentker, M.; Leker, J.; Winter, M. Large-Scale Automotive Battery Cell Manufacturing: Analyzing Strategic and Operational Effects on Manufacturing Costs. Int. J. Prod. Econ. 2021, 232, 107982. https://doi.org/10.1016/j.ijpe.2020.107982. 18. Weiss, M.; Simon, F. J.; Busche, M. R.; Nakamura, T.; Schröder, D.; Richter, F. H.; Janek, J. From Liquid- to Solid-State Batteries: Ion Transfer Kinetics of Heteroionic Interfaces. Electrochem. Energ. Rev. 2020, 3 (2), 221–238. https://doi.org/10.1007/s41918-020-00062-7. 19. Huo, H.; Janek, J. Solid-State Batteries: From ‘All-Solid’ to ’Almost-Solid’. Natl. Sci. Rev. 2023, 10 (6). https://doi.org/10.1093/nsr/nwad098. nwad098. 20. Geiger, T.; Rudschies, W. Nio ET7: Der Tesla-Jäger im ADAC Test; 2023. https://www.adac.de/rund-ums-fahrzeug/autokatalog/marken-modelle/nio/nio-et7/ (Accessed 28 May 2024). 21. Schnell, J.; Günther, T.; Knoche, T.; Vieider, C.; Köhler, L.; Just, A.; Keller, M.; Passerini, S.; Reinhart, G. All-Solid-State Lithium-Ion and Lithium Metal Batteries – Paving the Way to Large-Scale Production. J. Power Sources 2018, 382, 160–175. https://doi.org/10.1016/j.jpowsour.2018.02.062. 22. Duffner, F.; Kronemeyer, N.; Tübke, J.; Leker, J.; Winter, M.; Schmuch, R. Post-Lithium-Ion Battery Cell Production and Its Compatibility with Lithium-Ion Cell Production Infrastructure. Nat. Energy 2021, 6 (2), 123–134. https://doi.org/10.1038/s41560-020-00748-8. 23. Albertus, P.; Babinec, S.; Litzelman, S.; Newman, A. Status and Challenges in Enabling the Lithium Metal Electrode for High-Energy and Low-Cost Rechargeable Batteries. Nat. Energy 2018, 3 (1), 16–21. https://doi.org/10.1038/s41560-017-0047-2. 24. Vanleeuw, D.; Sapundjiev, D.; Sibbens, G.; Oberstedt, S.; Salvador Castiñeira, P. Physical Vapour Deposition of Metallic Lithium. J. Radioanal. Nucl. Chem. 2014, 299 (2), 1113–1120. https://doi.org/10.1007/s10967-013-2669-6. 25. Schönherr, K.; Schumm, B.; Hippauf, F.; Lissy, R.; Althues, H.; Leyens, C.; Kaskel, S. Liquid Lithium Metal Processing into Ultrathin Metal Anodes for Solid State Batteries. Chem. Eng. J. Adv. 2022, 9, 100218. https://doi.org/10.1016/j.ceja.2021.100218. 26. Lohrberg, O.; Voigt, K.; Maletti, S.; Auer, H.; Nikolowski, K.; Heubner, C.; Michaelis, A. Benchmarking and Critical Design Considerations of Zero-Excess Li-Metal Batteries. Adv. Funct. Mater. 2023, 2214891. 27. Huo, H.; Janek, J. Silicon as Emerging Anode in Solid-State Batteries. ACS Energy Lett. 2022, 7 (11), 4005–4016. https://doi.org/10.1021/acsenergylett.2c01950. 28. Huo, H.; Jiang, M.; Bai, Y.; Ahmed, S.; Volz, K.; Hartmann, H.; Henss, A.; Singh, C. V.; Raabe, D.; Janek, J. Chemo-Mechanical Failure Mechanisms of the Silicon Anode in Solid-State Batteries. Nat. Mater. 2024, 23, 543–551. 29. Cangaz, S.; Hippauf, F.; Reuter, F. S.; Doerfler, S.; Abendroth, T.; Althues, H.; Kaskel, S. Enabling High-Energy Solid-State Batteries with Stable Anode Interphase by the Use of Columnar Silicon Anodes. Adv. Energy Mater. 2020, 10 (34), 2001320. https://doi.org/10.1002/aenm.202001320. 30. Saager, S.; Decker, L.; Kopte, T.; Scheffel, B.; Zimmermann, B. High-Performance Anodes Made of Metallic Lithium Layers and Lithiated Silicon Layers Prepared by Vacuum Technologies. Batteries 2023, 9 (2), 75. https://doi.org/10.3390/batteries9020075.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

641

31. Grandjean, M.; Perrey, M.; Randrema, X.; Laurier, J.; Chenevier, P.; Haon, C.; Liatard, S. Low Pressure Cycling of Solid State Li-Ion Pouch Cells Based on NMC – Sulfide – Nanosilicon Chemistry. J. Power Sources 2023, 585, 233646. https://doi.org/10.1016/j.jpowsour.2023.233646. 32. Singer, C.; Töpper, H.-C.; Kutsch, T.; Schuster, R.; Koerver, R.; Daub, R. Hydrolysis of Argyrodite Sulfide-Based Separator Sheets for Industrial All-Solid-State Battery Production. ACS Appl. Mater. Interfaces 2022, 14 (21), 24245–24254. https://doi.org/10.1021/acsami.2c01099. 33. Kim, K. T.; Woo, J.; Kim, Y.-S.; Sung, S.; Park, C.; Lee, C.; Park, Y. J.; Lee, H.-W.; Park, K.; Jung, Y. S. Ultrathin Superhydrophobic Coatings for Air-Stable Inorganic Solid Electrolytes: Toward Dry Room Application for All-Solid-State Batteries. Adv. Energy Mater. 2023, 13 (43). https://doi.org/10.1002/aenm.202301600. 34. Lee, J.; Choi, J. W. Block Copolymer Binders with Hard and Soft Segments for Scalable Fabrication of Sulfide-Based All-Solid-State Batteries. EcoMat 2022, 4 (4), e12193. 35. Riphaus, N.; Strobl, P.; Stiaszny, B.; Zinkevich, T.; Yavuz, M.; Schnell, J.; Indris, S.; Gasteiger, H. A.; Sedlmaier, S. J. Slurry-Based Processing of Solid Electrolytes: A Comparative Binder Study. J. Electrochem. Soc. 2018, 165 (16), A3993–A3999. https://doi.org/10.1149/2.0961816jes. 36. Nam, Y. J.; Oh, D. Y.; Jung, S. H.; Jung, Y. S. Toward Practical All-Solid-State Lithium-Ion Batteries with High Energy Density and Safety: Comparative Study for Electrodes Fabricated by Dry- and Slurry-Mixing Processes. J. Power Sources 2018, 375, 93–101. https://doi.org/10.1016/j.jpowsour.2017.11.031. 37. Takada, K. Solid-State Lithium Battery with Graphite Anode. Solid State Ion. 2003, 158 (3-4), 269–274. https://doi.org/10.1016/S0167-2738(02)00823-8. 38. Li, Y.; Wu, Y.; Wang, Z.; Xu, J.; Ma, T.; Chen, L.; Li, H.; Wu, F. Progress in Solvent-Free Dry-Film Technology for Batteries and Supercapacitors. Mater. Today 2022, 55, 92–109. https://doi.org/10.1016/j.mattod.2022.04.008. 39. Lu, Y.; Zhao, C.-Z.; Yuan, H.; Hu, J.-K.; Huang, J.-Q.; Zhang, Q. Dry Electrode Technology, the Rising Star in Solid-State Battery Industrialization. Matter 2022, 5 (3), 876–898. https://doi.org/10.1016/j.matt.2022.01.011. 40. Ludwig, B.; Zheng, Z.; Shou, W.; Wang, Y.; Pan, H. Solvent-Free Manufacturing of Electrodes for Lithium-ion Batteries. Sci. Rep. 2016, 6, 23150. https://doi.org/10.1038/ srep23150. 41. Schumm, B.; Kaskel, S. Dry Battery Electrode Processing, What’s Next? Next Energy 2023, 1, 100009. https://doi.org/10.1016/j.nxener.2023.100009. 42. Hippauf, F.; Schumm, B.; Doerfler, S.; Althues, H.; Fujiki, S.; Shiratsushi, T.; Tsujimura, T.; Aihara, Y.; Kaskel, S. Overcoming Binder Limitations of Sheet-Type Solid-State Cathodes Using a Solvent-Free Dry-Film Approach. Energy Storage Mater. 2019, 21, 390. https://doi.org/10.1016/j.ensm.2019.05.033. 43. Fiedler, M.; Cangaz, S.; Hippauf, F.; Dörfler, S.; Abendroth, T.; Althues, H.; Kaskel, S. Mechanistic Insights into the Cycling Behavior of Sulfur Dry-Film Cathodes. Adv. Sustain. Syst. 2023, 7 (4). https://doi.org/10.1002/adsu.202200439. 44. Kim, J.; Park, K.; Kim, M.; Lee, H.; Choi, J.; Park, H. B.; Kim, H.; Jang, J.; Kim, Y.-H.; Song, T.; Paik, U. 10 mAh cm−2 Cathode by Roll-to-Roll Process for Low Cost and High Energy Density Li-Ion Batteries. Adv. Energy Mater. 2024, 14 (10), 2303455. 45. Yao, W.; Chouchane, M.; Li, W.; Bai, S.; Liu, Z.; Li, L.; Chen, A. X.; Sayahpour, B.; Shimizu, R.; Raghavendran, G.; Schroeder, M. A.; Chen, Y.-T.; Tan, D. H. S.; Sreenarayanan, B.; Waters, C. K.; Sichler, A.; Gould, B.; Kountz, D. J.; Lipomi, D. J.; Zhang, M.; Meng, Y. S. A 5 V-Class Cobalt-Free Battery Cathode with High Loading Enabled by Dry Coating. Energ. Environ. Sci. 2023, 16 (4), 1620–1630. https://doi.org/10.1039/D2EE03840D. 46. Zhang, Z.; Wu, L.; Zhou, D.; Weng, W.; Yao, X. Flexible Sulfide Electrolyte Thin Membrane with Ultrahigh Ionic Conductivity for All-Solid-State Lithium Batteries. Nano Lett. 2021, 21, 5233–5239. https://doi.org/10.1021/acs.nanolett.1c01344. 47. Lee, D. J.; Jang, J.; Lee, J.-P.; Wu, J.; Chen, Y.-T.; Holoubek, J.; Yu, K.; Ham, S.-Y.; Jeon, Y.; Kim, T.-H.; Lee, J. B.; Song, M.-S.; Meng, Y. S.; Chen, Z. Physio-Electrochemically Durable Dry-Processed Solid-State Electrolyte Films for All-Solid-State Batteries. Adv. Funct. Mater. 2023, 33 (28), 2301341. 48. Gyulai, A.; Bauer, W.; Ehrenberg, H. Dry Electrode Manufacturing in a Calender: The Role of Powder Premixing for Electrode Quality and Electrochemical Performance. ACS Appl. Energy Mater. 2023, 6 (10), 5122–5134. https://doi.org/10.1021/acsaem.2c03755. 49. Schumm, B.; Kaskel, S. Dry Battery Electrode Processing, What’s Next? Next Energy 2023, 1 (2), 100009. https://doi.org/10.1016/j.nxener.2023.100009. 50. Tschöcke, S.; Althues, H.; Schumm, B.; Kaskel, S.; SCHULT, C.; FRITSCHE, D.; Schönherr, K. Process for Producing a Dry Film and Dry Film and Dry Film Coated Substrate; 2017. DE 10 2017 208 220 A1 2018.11.22, May 16, 2017. 51. Fu, K.; Gong, Y.; Hitz, G. T.; McOwen, D. W.; Li, Y.; Xu, S.; Wen, Y.; Zhang, L.; Wang, C.; Pastel, G.; Dai, J.; Liu, B.; Xie, H.; Yao, Y.; Wachsman, E. D.; Hu, L. Three-Dimensional Bilayer Garnet Solid Electrolyte Based High Energy Density Lithium Metal–Sulfur Batteries. Energ. Environ. Sci. 2017, 10 (7), 1568–1575. https://doi.org/10.1039/C7EE01004D. 52. Fu, K. K.; Gong, Y.; Dai, J.; Gong, A.; Han, X.; Yao, Y.; Wang, C.; Wang, Y.; Chen, Y.; Yan, C.; Li, Y.; Wachsman, E. D.; Hu, L. Flexible, Solid-State, Ion-Conducting Membrane with 3D Garnet Nanofiber Networks for Lithium Batteries. Proc. Natl. Acad. Sci. U. S. A. 2016, 113 (26), 7094–7099. https://doi.org/10.1073/pnas.1600422113. 53. Han, X.; Gong, Y.; Fu, K. K.; He, X.; Hitz, G. T.; Dai, J.; Pearse, A.; Liu, B.; Wang, H.; Rubloff, G.; Mo, Y.; Thangadurai, V.; Wachsman, E. D.; Hu, L. Negating Interfacial Impedance in Garnet-Based Solid-State Li Metal Batteries. Nat. Mater. 2017, 16 (5), 572–579. https://doi.org/10.1038/nmat4821. 54. Luo, W.; Gong, Y.; Zhu, Y.; Fu, K. K.; Dai, J.; Lacey, S. D.; Wang, C.; Liu, B.; Han, X.; Mo, Y.; Wachsman, E. D.; Hu, L. Transition from Superlithiophobicity to Superlithiophilicity of Garnet Solid-State Electrolyte. J. Am. Chem. Soc. 2016, 138 (37), 12258–12262. https://doi.org/10.1021/jacs.6b06777. 55. Yada, C.; Brasse, C. Better Batteries with Solid-State Instead of Liquid-Based Electrolytes. ATZ Elektron Worldw 2014, 9 (3), 10–15. https://doi.org/10.1365/s38314-014-0244-8. 56. Day, L. Toyota Is Road Testing a Prototype Solid State Battery EV; 2021. https://www.thedrive.com/tech/42287/toyota-is-road-testing-a-prototype-solid-state-battery-ev. 57. Kamaya, N.; Homma, K.; Yamakawa, Y.; Hirayama, M.; Kanno, R.; Yonemura, M.; Kamiyama, T.; Kato, Y.; Hama, S.; Kawamoto, K.; Mitsui, A. A Lithium Superionic Conductor. Nat. Mater. 2011, 10 (9), 682–686. https://doi.org/10.1038/NMAT3066. 58. Yumae, S.; Akama, K. Toyota Secures Huge Lead in Solid-State Battery Patents; 2022. https://asia.nikkei.com/Business/Technology/Toyota-secures-huge-lead-in-solid-statebattery-patents. 59. Toyota UK media site. Toyota Motor Corporation Unveils Full Global Battery Electric Line-Up; 2021. https://media.toyota.co.uk/toyota-motor-corporation-unveils-full-globalbattery-electric-line-up/. 60. Kato, Y.; Hori, S.; Saito, T.; Suzuki, K.; Hirayama, M.; Mitsui, A.; Yonemura, M.; Iba, H.; Kanno, R. High-Power All-Solid-State Batteries Using Sulfide Superionic Conductors. Nat. Energy 2016, 1 (4), 16030. https://doi.org/10.1038/nenergy.2016.30. 61. Scott, A. Toyota Hunts for Novel Battery Materials. C&EN Global Enterp 2017, 95 (15), 12. https://doi.org/10.1021/cen-09515-notw9. 62. Toyota Group., 2023. https://global.toyota/en/album/images/39865919/. 63. Robinson, A. L.; Janek, J. Solid-State Batteries Enter EV Fray. MRS Bull. 2014, 39 (12), 1046–1047. https://doi.org/10.1557/mrs.2014.285. 64. Toyota Times. Idemitsu & Toyota Team Up to Create Global Standard for All-Solid-State Batteries; 2023. https://toyotatimes.jp/en/toyota_news/1046.html?padid=ag478_from_ newsroom. 65. Toyota Group. Electrified Technologies - Batteries, Fundamental Technologies to Innovate BEV; 2023. https://global.toyota/en/newsroom/corporate/39330500.html. 66. Ito, S.; Fujiki, S.; Yamada, T.; Aihara, Y.; Park, Y.; Kim, T. Y.; Baek, S.-W.; Lee, J.-M.; Doo, S.; Machida, N. A Rocking Chair Type All-Solid-State Lithium Ion Battery Adopting Li2O–ZrO2 Coated LiNi0.8Co0.15Al0.05O2 and a Sulfide Based Electrolyte. J. Power Sources 2014, 248, 943–950. https://doi.org/10.1016/j.jpowsour.2013.10.005. 67. Suzuki, N.; Yashiro, N.; Fujiki, S.; Omoda, R.; Shiratsuchi, T.; Watanabe, T.; Aihara, Y. Highly Cyclable All-Solid-State Battery with Deposition-Type Lithium Metal Anode Based on Thin Carbon Black Layer. Adv. Energy Sustain. Res. 2021, 192, 2100066. https://doi.org/10.1002/aesr.202100066. 68. Heubner, C.; Maletti, S.; Auer, H.; Hüttl, J.; Voigt, K.; Lohrberg, O.; Nikolowski, K.; Partsch, M.; Michaelis, A. From Lithium-Metal toward Anode-Free Solid-State Batteries: Current Developments, Issues, and Challenges. Adv. Funct. Mater. 2021, 31 (51), 2106608. https://doi.org/10.1002/adfm.202106608. 69. Choi, J. Samsung SDI Unveils Solid State Battery Road Map; 2023. https://www.businesskorea.co.kr/news/articleView.html?idxno=203912. 70. Lee, S. Samsung SDI Upgrading Way It Manufactures Solid-State Battery; 2024. https://thelec.net/news/articleView.html?idxno=4710. 71. SDI News. Samsung SDI to Present Essence of Super-Gap Battery Technology at InterBattery; 2024. https://www.samsungsdi.com/sdi-news/3522.html?pageIndex=1& idx=3522&searchCondition=0&searchKeyword=. 72. Randall, C. Samsung SDI to Start Mass Production of Solid-State Batteries in 2027; 2024. https://www.electrive.com/2024/03/05/samsung-sdi-to-start-mass-production-ofsolid-state-batteries-in-2027/.

642

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Inorganic Solid-Electrolyte Cells

73. Technology for a New Breed of Battery: Investor Deck, 2023. https://www.solidpowerbattery.com/files/doc_presentation/2023/10/SLDP-Investor-Presentation-updatedtemplate-Sept-2023-FINAL.pdf (accessed 10.2023). 74. All-Solid-State Battery Cell Technology, 2024 https://www.solidpowerbattery.com/all-solid-state-batteries/default.aspx (accessed April 12, 2024). 75. BMW Group Pressclub USA. An Example of a Solid Power 60Ah All Solid State Prototype; 2023. https://www.press.bmwgroup.com/usa/photo/detail/P90534022/An-example-ofa-Solid-Power-60Ah-all-solid-state-prototype-cell-shown-at-a-media-and-stakeholder. 76. Solid Power Inc. Banner; 2024. https://www.solidpowerbattery.com/files/design/banner/battery-banner.jpg (accessed April 12, 2024). 77. Nikkei Electronics, n.d. All-Solid-State Battery Mounted on a Board: IoT Terminal Is Highly Functional. https://xtech.nikkei.com/atcl/nxt/mag/ne/18/00036/. 78. Nikkei Electronics. All-Solid-State Batteries: To EV/IoT Now, 2018; accessed https://xtech.nikkei.com/dm/atcl/mag/15/318381/201801/. 79. Anon. Murata’s Oxide-based Solid-state Batteries for Expanding the Range of Applications for IoT Devices and Supporting More Advanced Wearables; 2020. https://article. murata.com/en-global/article/solid-state-battery-that-supports-wearables-1 (Accessed 27 May 2024). 80. TDK Electronics AG. Rechargeable Multilayer Ceramic Chip Battery; 2022. https://www.tdk-electronics.tdk.com/inf/75/ds/CeraCharge_BCT1812M101AG.pdf. 81. Taiyo Yuden. Growing with Society toward new IoT Era: Integrated Report; 2018. https://www.yuden.co.jp/en/ir/2018ar/download/pdf/Yuden_AR18-E_all.pdf. 82. FDK Corporation., 2019. https://www.fdk.com/whatsnew-e/release20190509-e.html. 83. Murata Manufacturing Co., Ltd. Murata Develops Solid-State Battery with Industry’s Highest Energy Density. For Wearables Applications, Oxide Ceramic Electrolyte Solution Provides Reliability and Durability; 2019. https://www.murata.com/en-global/news/batteries/solid_state/2019/0626?intcid1=mar_art_xxx_tao_xxx_tao-link. 84. Thielmann, A.; Wietschel, M.; Funke, S.; Grimm, A.; Hettesheimer, T.; Langkau, S.; Loibl, A.; Moll, C.; Neef, C.; Plötz, P.; Sievers, L.; Espinoza, L. T.; Edler, J. Batterien für Elektroautos: Faktencheck und Handlungsbedarf: Sind Batterien für Elektroautos der Schlüssel für eine nachhaltige Mobilität der Zukunft?; 2020. https://www.isi.fraunhofer.de/ content/dam/isi/dokumente/cct/2020/Faktencheck-Batterien-fuer-E-Autos.pdf.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Production Mareike Partscha, Benedikt Stumperb, Jonas Dhomb, and Julian Schwenzelc, aFraunhofer Institute for Ceramic Technologies and Systems IKTS, Dresden, Germany; bFraunhofer Institute for Casting, Composite and Processing Technology IGCV, Augsburg, Germany; c Fraunhofer Institute for Manufacturing Technology and Advanced Materials IFAM, Bremen, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 3 3.1 3.2 3.3 3.4 4 4.1 4.2 4.3 4.4 5 5.1 5.1.1 5.2 6 7 References

Introduction Manufacturing processes Polymer-based solid-state batteries Specific properties of polymer electrolyte materials Free-standing polymer electrolyte separators and composite cathodes Matrix-supported polymer electrolyte separator Cathode-supported polymer electrolyte based cell Sulfide-based solid-state batteries Specific properties of sulfide electrolyte materials Free-standing sulfide-based electrolyte separator membranes Free-standing sulfide-based composite cathodes Lamination of complete sulfide-based cells Oxide-based solid-state batteries Specific properties of oxide electrolyte materials Free-standing separator-supported oxide cells Free-standing cathode-supported oxide cells Thin-film solid-state batteries Summary and brief outlook

644 645 647 647 647 648 648 649 649 650 650 651 652 652 653 653 654 654 655

Abstract The replacement of the liquid electrolyte by a solid electrolyte leads to changed requirements for the solid/solid interfaces, especially in terms of (electro)chemical and mechanical stability, but also requires adjustments in the manufacturing process of the batteries themselves. In addition, adjustments are required depending on the type of electrolyte selected—polymer, sulfide, oxide or thin film. This results in a wide range of process combinations and cell concepts, the central element of which is a multilayer structure consisting of current collectors, (composite) electrodes and the electrolyte-separator, with the interface between the individual components being of central importance.

Glossary Battery energy density Energy density is the measure of how much energy a battery contains in proportion to its weight or volume. It is measured in units of Wh/kg or Wh/L. Bipolar battery Bipolar batteries are battery cells that consist of stacked, serially connected electrodes. Calendaring Calendaring is the compression of dried electrodes resulting from the coating and drying of electrode slurry to reduce porosity and improve particle contacts Cation transference number The transference number describes the charge transport and accordingly the current transport of a specific ion. Doctor blading Doctor blading/tape casting is a casting process used to manufacture electrodes from a slurry that is cast in a thin layer onto a collector foil and then dried. The process is adapted from ceramic tape manufacturing. Formation and aging Formation is the initial charge and discharge of the manufactured lithium-ion battery with a small current. Aging generally refers to placing the battery at room temperature or high temperature after initial charging. Lithium dendrites Lithium dendrites are formed by the heterogeneous deposition of lithium on the current collector. They cause the risk of short circuits and loss of capacity in batteries. Pouch cell Pouch cells are a type of lithium-ion battery that features a flexible, flat pouch-shaped design. Round cell A round cell is a single unit of a cylindrical battery containing positive and negative electrodes, separator and electrolyte. Sintering aids Sintering aids typically assist in lowering the sintering temperature and eliminating pores from ceramic materials, either through the presence of a liquid phase or through other mechanisms that promote densification of ceramics, such as introducing dislocations and enhancing diffusion during thermal treatments.

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00236-X

643

644

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Production

Sintering Sintering is the process of compacting and shaping a solid mass of material by heating it to a temperature below its melting point and is usually accompanied by grain growth, as grain boundary reduction is energetically favorable. It is commonly used in the manufacture of ceramics, metals, and other materials. Slot die coating Slot die coating is a coating technique for applying electrode slurry or extruded thin films to collector foils. The process was first developed for the industrial production of photographic paper. tape casting Tape casting/doctor blading is a casting process used to manufacture electrodes from a slurry that is cast in a thin layer onto a collector foil and then dried. The process is adapted from ceramic tape manufacturing.

Key points

• • • • • •

1

Challenges in the manufacturing of solid-state batteries. Manufacturing of polymer-based batteries. Manufacturing of sulfide-based batteries. Manufacturing of oxide-based batteries. Manufacturing of thin-film batteries. Conclusion.

Introduction

Solid-state batteries are seen as the next promising development step to meet the growing demands on battery storage systems. The demands of electromobility, the fastest growing application for batteries, are certainly at the heart of this. Expectations are very high, especially with regard to increasing the energy density and intrinsic safety of the cells. Solid-state batteries can make a significant contribution in the coming years. Solid electrolytes, such as polymers, sulfides and oxides, and others, were initially considered for battery applications primarily because of their (electrochemical) material properties, such as ionic conductivity. With increasing experience in battery cell design, additional aspects such as chemical and mechanical stability and processability were added. Considerations regarding the large-scale production of these batteries and, in the future, their recycling, motivate further research. The requirements for advanced solid electrolyte batteries are diverse and include cell performance, manufacturing efficiency, and competitive material costs compared to lithium-ion batteries. The overall capital and operating costs are mainly determined by the reliability and complexity of the production processes, the required environmental conditions such as dry rooms, and the efficiency of material consumption during production. As improvements in lithium-ion batteries continue, the benchmarks for solid-state batteries are constantly evolving. In addition, solid-state batteries are more of a portfolio of cell chemistries and include a wide range of solid electrolytes (polymer, sulfide, oxide and hybrid concepts) with different processing characteristics and associated cell concepts.1 The substitution of liquid electrolytes leads to some fundamental changes in the cell structure compared to conventional lithium-ion batteries, which are reflected in the manufacturing process. In today’s manufacturing processes for the mass production of lithium-ion batteries, individual electrode foils for the cathode and anode are produced by wet coating aluminum or copper current collector foils with a slurry based on the storage material, the so-called active material. Together with the very thin separator membrane, a stack or jelly roll is formed. This electrode-separator assembly is then contacted and placed in the cell housing. Finally, the liquid electrolyte is filled into the cell and the unit is sealed. In this process, the electrolyte diffuses into the open porosity of the electrodes and separator, forming the liquid/solid interface between the electrode and electrolyte—activating the electrochemical cell. This last step must be performed in a dry atmosphere as the electrolyte is extremely sensitive to moisture. Solid electrolyte batteries, on the other hand, require charge transfer across a solid/solid interface that is formed as a cohesive lamination during the manufacture of individual components. This interface is critical to the formation of interfacial resistances that affect the performance of the cell not only after manufacture, but throughout its lifetime. The chemical stability of the materials and the electrochemical compatibility at the electrode/electrolyte interface are critical factors. Since many of the currently known solid electrolytes have a limited electrochemical window—either toward oxidation or reduction—additional measures, such as the introduction of protective layers or the combination of different electrolytes, are required. In addition, solid electrolytes have different elasticity and stiffness properties that respond to volume changes in the electrode materials and thermo-mechanical stress. All of these aspects are incorporated into the cell design and must be considered in the strategic selection of process operations. It should also be noted that there are still no standards for the selection of materials for solid-state batteries, and more classes of materials are under development. Therefore, the development of suitable process workflows remains a fundamental task, requiring tailor-made adaptations for each cell concept.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Production

2

645

Manufacturing processes

Due to the complexity of the topic, this chapter will provide a brief overview of selected relevant manufacturing processes and routes to illustrate common challenges. Solid electrolyte batteries typically consist of multilayer structures formed by the current collectors, cathode, anode, and electrolyte-separator. The current collectors are typically made of aluminum or copper, analogous to conventional cells. Lithium is particularly attractive as an anode due to its high energy density. In addition, composite electrodes are used, especially for the cathodes. As with conventional lithium-ion batteries, the composite cathode consists of the active material, conductivity additives, binders, and the electrolyte. In conventional electrodes, the liquid electrolyte is in the pores of the electrodes. In solid-state batteries, the solid electrolytes are either infiltrated into the pores (soluble polymer electrolytes) or mixed with the electrode slurry. The typical active materials are the same as those used in lithium-ion batteries (metal oxides for cathodes). In order to adequately compensate for the high capacity of the lithium anode, very high loadings on the cathode side and cathode’s electrolytes with high ionic conductivity are usually required. Electrolyte separators consist of the electrolytes themselves, which provide the ionic conductivity. Polymer binders provide mechanical stability and processability. Polymer electrolytes in particular are stabilized by spacers such as nonwovens. Since the electrolyte must be integrated into the manufacturing process of the electrode and separator, it is necessary to manufacture the components under dry room conditions due to the hydrolysis sensitivity of the materials. This can significantly increase operating costs compared to the use of liquid electrolytes. The electrochemical cell itself is realized either by laminating the individual components or by sequentially depositing individual layers on a supporting component. In contrast to conventional batteries, where the current collector is the supporting structural element in the cell, the electrolyte separator or cathode in solid-state batteries can perform this function. This is referred to as electrolyte-supported or cathode-supported cell concepts. Wet and dry thick film deposition technologies are the primary methods used to manufacture composite electrodes and electrolyte-separators. Thick film technologies include roll-to-roll doctor blade or slot die coating, screen printing, dry powder coating, yielding coatings, or stand-alone films ranging from a few microns to several hundred microns. Intermediate and protective layers, typically a few nanometers to a few micrometers thick, are produced by thin film processes such as physical and chemical vapor deposition or spin coating. The manufacturing process of solid electrolyte batteries often follows the following sequence: in a first manufacturing step, the material components of the supporting component—electrolyte-separator or composite cathode—are mixed to form a slurry and a foil is prepared. The second component is then applied by coating or laminating to form a cathode-separator laminate, which is completed by the anode. Various technology chains are described to produce the individual components or laminates. Wet chemical processes (e.g., tape casting) are based on mixtures of the active material and conductive additives for composite electrodes, as well as solid electrolyte materials, binders and organic solvents. These require slurries produced in an upstream dispersion process with low viscosity and low solids content. After the coating process, the high solvent content must be evaporated by drying, which results in additional porosity in the electrode layer. After drying, this porosity is reduced by densification, e.g., by calendering for polymer or sulfide electrolytes or thermal sintering for oxide electrolytes. Highly viscous composite mixtures are achieved by compounding and subsequent extrusion of films, especially of polymer electrolyte materials. This has the advantage of significantly reducing the amount of solvent required, thereby reducing the energy required to dry the coating and the length of the drying section. However, powders and binders are subjected to very high shear rates and mechanical stress during extrusion, which is not ideal for all materials. Depending on the material and the desired layer morphology, a much more moderate compression can then be applied. For some classes of materials, so-called dry coating processes are also possible, ranging from processing the active materials together with thermoplastic polymer electrolytes to various thin-film processes. Depending on the process sequence, lamination and/or thermal processes are integrated to form the multilayer cells. Since the interfaces between the electrolyte, electrodes and collectors are formed via the solid phase, the quality of the interface and the interfacial resistance, which are crucial for the performance of the battery, are subject to quality control. A suitable combination of materials and the choice of appropriate manufacturing processes are crucial. Both have to be adapted in terms of (electro)chemical and (thermo)mechanical boundary conditions as well as processability. During the manufacturing process, challenges arise in particular from the fact that solid electrolyte materials are processed right from the start of electrode manufacture. The solid electrolyte materials are intensively mixed with the active material to form particle-particle interfaces and the subsequent composite cathode. Since most solid electrolytes are highly sensitive to moisture and even air, all manufacturing steps must be performed in a dry room environment under a protective atmosphere. Fig. 1 provides an overview of the general process design for the manufacture of solid-state batteries.2 In current developments, lithium is the preferred anode material because it has a very high specific capacity (3862 mAh g−1), allowing cells with high energy density to be realized. The lithium is typically introduced as a foil produced by extrusion. To ensure that these foils can still be handled, they are thicker than actually required for the electrochemical reaction itself. The future goal is to reduce the amount of metallic lithium in the cell to the electrochemical minimum. This may require the application of very thin

646

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Production

Fig. 1 Process options for solid-state battery manufacturing. Adapted from Schnell, J.; Günther, T.; Knoche, T.; Vieider, C.; Köhler, L.; Just, A.; Keller, M.; Passerini, S.; Reinhart, G. All-Solid-State lithium-Ion and lithium Metal Batteries – Paving the Way to Large-Scale Production. J. Power Sources 2018, 382, 160–175. https://doi.org/10.1016/j.jpowsour.2018.02.062

Table 1

Solid-state battery (SSB) manufacturing processes. Polymer SSB

Composite cathode, separator Anode Cell assembly

Sulfide SSB

Oxide SSB

Extrusion or wet Wet processing: Wet processing: processing: slurry mixing, coating, slurry mixing, coating, drying, slurry mixing, coating, drying, calendering lamination, sintering drying, calendering Extrusion process (Li foil) Extrusion, calendering, lamination Wet processing (Si anode) Electrode stacking and stack pressing, packaging, formation and aging

Optimization potential Dry processes or green solvent-based processes Melt or vapor-based process (Li foil), host-structure for Li-free anodes Omit formation and aging

Adapted from Schmaltz, T.; Hartmann, F.; Wicke, T.; Weymann, L.; Neef, C.; Janek, J. A Roadmap for Solid-State Batteries. Adv. Energy Mater. 2023, 13, 2301886. doi:https://doi. org/10.1002/aenm.202301886

lithium layers or even the application of host layers for lithium deposition in so-called anode-less concepts, where the lithium is deposited directly from the cathode onto the current collector host structure. One approach to avoid dendrite growth is to use a silicon anode (3579 mAh g−1) as the lithium host. However, the very large volume change during charging and discharging requires consideration of the cell concept as well as specific interface engineering to avoid loss of contact between the anode and electrolyte over the life of the cell. Due to dendrite growth during lithium deposition in the charging process and the volume changes of the electrodes, additional functional layers are often applied to the interface between the lithium metal anode and the separator to suppress dendrite growth and ensure a homogeneous field distribution. The assembly of solid-state battery cells is very similar to that of conventional stacked cells, as laminated single cells are packaged as a cell stack in a pouch bag. In the future, other cell formats such as round cells may become relevant. Special requirements for cell assembly arise from the high reactivity of the lithium anode and the adhesive properties of the lithium foils, which require special attention and process control with consistent exclusion of moisture. The extent to which aging and the formation of stable electrode/electrolyte interfaces are required for solid-state batteries is still not well understood. These two processes are major cost drivers in the production of conventional lithium batteries. It would be very advantageous if these steps could be omitted in solid-state batteries due to the higher stability of the electrolytes. However, there is currently little experience in this regard. Table 1 summarizes typical processes for each electrolyte and cell component. Since there is no mass production of solid-state batteries yet, some of the considerations presented here are based on experimental work, theoretical considerations or considerations based on analogous processes for other products.

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Production

3 3.1

647

Polymer-based solid-state batteries Specific properties of polymer electrolyte materials

Among solid-state batteries, polymer-based solid-state batteries are the most extensively studied solid electrolytes. Compared to inorganic (sulfide or oxide) solid-state batteries, polymer solid-state batteries offer advantages such as high flexibility (Fig. 2), good processability, and significantly lower production costs. In addition, polymer solid electrolytes have excellent contact with the electrodes, resulting in low interfacial resistance. Compared to lithium-ion batteries, high energy density can be achieved by combining the solid electrolyte and polymer-based composite cathode with a lithium metal anode. Since moisture-sensitive lithium metal anode and electrolytes are used, the process must be carried out in a dry room or even in an inert gas atmosphere to avoid unwanted reactions. With the polymer-based composite cathode, the solid electrolyte can be added directly during mixing or infiltrated into the pores later. Compaction of the composite cathode is aimed at significantly reducing porosity to achieve high ionic conductivity, since any pore is an obstacle for the lithium ions. By using a solid electrolyte, the individual components can be connected in series (bipolar battery concept), which allows the packaging design to be optimized. A major disadvantage is the low ionic conductivity at room temperature. Therefore, polymer cells are often operated at higher temperatures to increase the mobility of the polymer chains.3 Solid polymer electrolytes are typically prepared and processed by dissolving the polymer matrix and conducting salt in solvents. Common materials for the polymer matrix and conducting salts are listed in Table 2. In addition, additives are often used to increase the ionic conductivity.5 Several manufacturing concepts have been described for polymer solid-state batteries, as shown in Fig. 3. The free-standing separator and composite electrodes (a) and the matrix-supported separator (c) can be manufactured independently, resulting in production advantages. In the cathode-supported concept (b), the solid electrolyte is applied directly to the composite cathode. The different process routes are described below.

3.2

Free-standing polymer electrolyte separators and composite cathodes

In the manufacture of the free-standing composite electrodes and solid electrolyte, the slurry is applied to a low-adhesion substrate (e.g., polytetrafluoroethylene, PTFE). The coating is thereafter dried to evaporate the solvent, as shown in Fig. 4. The subsequent calendering step compacts the free-standing components. After the composite cathode is produced, the free-standing solid

Fig. 2 Polymer-based free-standing electrolyte-separator (Fraunhofer IFAM).

Table 2

Components of the solid polymer electrolyte.4

Component

Type

Abbreviation

Solvent

Acetonitrile N-Methyl-2-pyrrolidone Poly(ethylene oxide) Poly(methyl methacrylate) Polyacrylonitrile Poly(vinylidene difluoride) Lithium bis(trifluoromethanesulphonyl)imide Lithium bis(oxalatoborate) Lithium bis(fluorosulfonyl)imide Butylene nitrile

MeCN NMP PEO PMMA PAN PVDF LiTFSI LiBOB LiFSI SN

Polymer matrix

Conducting salt Additives

648

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Production

Fig. 3 Manufacturing processes for polymer-based solid electrolyte components: (a) Fabrication of a free-standing composite electrode or separator with subsequent lamination, (b) Coating of a separator on a supporting composite cathode, (c) Matrix-supported separator to be laminated with a free-standing composite cathode.

Fig. 4 Process diagram for laminating free-standing components.

Fig. 5 Process diagram for laminating matrix-supported polymer electrolyte-separators.

electrolyte can be laminated onto the composite cathode. After this process, the substrate foil can be removed from the free-standing solid electrolyte. The free-standing component lamination offers several advantages, including a simple and precisely adjustable coating process, good scalability to an industrial scale, and the ability to produce all components independently from each other, which simplifies quality management significantly.6 One disadvantage of this concept is that the free-standing components are mechanically fragile and structurally unstable. This makes further handling more difficult and requires a substrate that must be removed during processing. Additionally, the electrolyte-separator must have a thickness of around 100 mm in order to be processable, which often exceeds the required dimension and energy density of the cell.

3.3

Matrix-supported polymer electrolyte separator

Another way to fabricate solid electrolytes is to coat and infiltrate a porous support matrix with a solid electrolyte slurry, as shown in Fig. 5. The use of the support matrix defines the thickness and allows the production of thin separators, increasing the energy density of the cell. The mechanical stability and flexibility of the separator results in production advantages. Conventional separators, which are also used in conventional lithium-ion batteries, are often used as the support matrix. This provides additional mechanical stability and prevents the separator from being penetrated by dendrite growth. Like the free-standing solid electrolyte separator, the matrix-supported solid electrolyte separator can be manufactured separately from the polymer-based composite cathode, resulting in process advantages.3 The components are assembled by lamination as described above.

3.4

Cathode-supported polymer electrolyte based cell

For optimal energy density, a highly capacitive cathode is essential, which can have layer thicknesses of around 100 mm. This component is therefore ideally suited for use as a supporting element in the cell structure, and the solid electrolyte slurry can be applied directly to the cathode, as shown in Fig. 3b. This approach is referred to as the cathode-supported cell concept. In this context, analogous to the self-supporting solid electrolytes mentioned above, a wet solid electrolyte slurry is formulated and applied to the cathode using a doctor blade or slot die, with the cathode serving directly as the supporting substrate, as shown in Fig. 6. The polymer-based composite cathode can be calendered separately or together with the solid electrolyte separator.3

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Production

649

Fig. 6 Process diagram for the cathode-supported cell concept.

Challenges in this process sequence include coordinating the formulations of the cathode and separator—for example, ensuring that the solvents used do not damage the other component. There are also surface quality requirements for the cathode, as its roughness affects the uniform thickness of the electrolyte-separator.

4 4.1

Sulfide-based solid-state batteries Specific properties of sulfide electrolyte materials

Sulfide-based solid electrolytes (Fig. 7) have received great interest due to their excellent lithium-ion conductivity and mechanical softness, making them promising candidates for processing in all solid-state batteries. There are several electrolyte materials available such as LPS (Li2S–P2S5) derivates, thio-LISICON ceramics and lithium argyrodites. A brief overview is given in Table 3.7 All the mentioned compounds have an intrinsically high cation transference number close to 1, indicating a high mobility of the lithium ions in the lattice. A critical aspect of these materials is their reactivity with moisture. Due to their low thermodynamic stability, these materials are highly hygroscopic and form highly toxic and flammable H2S. The rate of formation of the toxic gas depends on the material and the relative humidity. A systematic study of LPS glasses showed values of 0.5–2.0 cm3
g−1
min−1 at 50% relative humidity and 25  C.8 Consequently, all manufacturing processes must be carried out under controlled atmospheric conditions to protect the materials and workers.

Fig. 7 Sulfide-based free-standing electrolyte-separator (Fraunhofer IFAM).

Table 3

Selected sulfide electrolytes.

Type

Formula

Ionic conductivity/mS
cm−1

LPS (crystalline) LPS (glass-ceramic) thio-LISICON thio-LISICON thio-LISICON Argyrodite Argyrodite Argyrodite

b-Li3PS4 Li7P3S11 Li10GeP2S12 Li10SnP2S12 Li10SiP2S12 Li6PS5Cl Li6PS5Br Li5.5PS4.5Cl1.5

0.1 5.2 12.0 3.2 2.3 1.1 1.9 9.4

650

Lithium Batteries – Lithium Secondary Batteries – Lithium All-Solid State Battery | Production

The density of sulfide-based electrolytes ( 1.8–2.0 g
cm−3) is twice that of conventional liquid electrolytes ( 1.0 g
cm−3). Therefore, thin separators and cathodes with surface capacities in the range of 3–10 mAh
cm−2 are required to achieve energy densities comparable to or higher than those of state-of-the-art lithium-ion batteries.9 This requires new and specialized manufacturing developments to produce thin, multilayer, dense systems with appropriate interfaces, as conventional processing approaches are not suitable.

4.2

Free-standing sulfide-based electrolyte separator membranes

Scalable manufacturing processes for solid electrolyte separators are critical for the large-scale production of sulfide-based solid-state batteries. The ionic conductivity is inversely proportional to the thickness of the electrolyte-separator, which requires the development of thin separator membranes for the realization of high energy density batteries. However, the separators are also responsible for the electrical separation of the cathode and anode to prevent contact and short circuits. Therefore, a balance must be found between minimizing the thickness and maintaining the mechanical integrity of solid-state electrolyte-separators. Two general strategies are used to produce thin membranes: In recent years, solvent-based processing of electrolytes in separators has gained interest through the use of tape casting technologies to coat membranes less than 100 mm thick. This requires a prior slurry process in which the electrolyte powder is mixed with suitable binders and solvents. In this context, the stability of sulfide materials against solvents is very important. It has been found that only non-polar solvents with low donor numbers, such as toluene, xylene or aliphatic hydrocarbons, are chemically stable to the sulfide electrolytes. Accordingly, only a limited group of polymers can be used as binders, especially non-polar rubbers such as styrene-butadiene rubber (SBR)10 or (hydrogenated) nitrile-butadiene rubber ((H)NBR). In addition to adjusting the rheological properties, the amount of binder determines the performance of the final separator membrane. Increasing the amount of binder improves mechanical stability while reducing ionic conductivity. The challenge is to optimize the binder/sulfide ratio as a function of material and particle size to obtain suitable separator membranes. After deposition and drying, densification is necessary to obtain a dense layer without any porosity. For densification, uniaxial pressing, isostatic pressing or calendering have been applied. Compared to uniaxial and isostatic pressing, calendaring technology is particularly promising because it can be implemented in a continuous roll-to-roll manufacturing process. For the production of free-standing separators, the bendability and flexibility of the components are crucial. Therefore, the type and amount of binder, density and layer thickness of the separator must be optimized. The best densification is achieved by isostatic pressing at elevated temperatures up to 150  C.11 However, isostatic pressing as a batch process is only possible as a post-process of the complete battery cell in the final cell manufacturing process. To achieve a good conductive layer, the substrate used is also important. Polymer films must be wettable by the slurry while providing a stable mechanical backbone for the layer. In addition, the deposition of thin (1 for the fuel-rich region and equivalence ratio < 1 for the fuel-poor region). The self-ignition temperature of the venting gas is calculated by chemical kinetic modeling with the different equivalence ratio when the ignition delay time is set to 60s (Fig. 10 a and b). Interestingly, at different equivalence ratios, high-nickel batteries always (e.g. NCM811 and NCM955 batteries) exhibited a higher self-ignition temperature of the venting gas than that of LiFePO4 battery, resulting in mild external combustion. This is because high-nickel NCM batteries emit less combustible gases than LiFePO4 batteries due to their strong oxidizability of high-nickel cathodes. Moreover, the self-ignition temperature decreases with the increase of the equivalence ratio for all batteries, indicating the more drastic gas combustion and explosion when the fuel is rich. Fig. 10c compares the venting

666

Battery Types – Lithium Batteries – Lithium Battery Safety | Overview

Fig. 10 (a) Venting gas composition when battery goes into thermal runaway.22 (b) Self-ignition temperature of venting gas from batteries with different cathode when the ignition delay time is 60s. (c) Self-ignition temperature and gas temperature of venting gas from batteries with different cathodes when the ignition delay time is 60s and equivalence ratio is 1. (d) Laminar flame speed of venting gas for batteries with different cathodes during thermal runaway.11 From Zhao, L.; Hou, J.; Feng, X.; Xu, C.; Wang, H.; Liu, H.; Hou, B.; Rui, X.; Lu, L.; Bao, C.; Ouyang, M. The Trade-Off Characteristic between Battery Thermal Runaway and Combustion. Energy Storage Mater. 2024, 69, 103380.

gas temperature and self-ignition temperature of different batteries. The high-nickel batteries have a higher exhaust temperature of venting gas because of the violent chemical reactions within the battery during thermal runaway. As the exhaust temperature of LiFePO4 battery is lower than the self-ignition temperature, they may achieve thermal runaway with a large amount of gas, but not combustion. Once an ignition source is present during thermal runaway, the LiFePO4 battery will cause a violent combustion and explosion. Fig. 10d shows the calculation result of laminar flame speeds for the combustible gases based on the premixed laminar flame speed calculation (PLFC) model.23,24 For the PLFC model parameters, the temperature of the unreacted gas and the ambient temperature were set to 400 K, the pressure was set to 1 atm, and the inlet flow rate was set to 40 cm s−1. LiFePO4 battery was found to exhibit a higher laminar flame speed (LFS) than NCM batteries, indicating that LiFePO4 batteries are more flammable during venting. Furthermore, the LFS increases with the increase of the equivalence ratio, and reaches its maximum value when the equivalence ratio reaches 1.1. As the equivalence ratio increases, the LFS decreases, and the gas production of the NCM batteries follows a similar trend to that of the LiFePO4 battery. As the content of H2 and C2H4 in the venting gas increases, it appears that combustion and explosion risk increases, as well as poorer ignition and explosion characteristics.22 It has been demonstrated that the LiFePO4 battery venting gas explosion limit is lower, the explosion overpressure is greater, and the explosion index is higher. Consequently, the cathode oxidizability is stronger in the high-nickel battery, resulting in a more intense thermal runaway reaction but mitigated external combustion and explosion. On the contrary, the LiFePO4 battery released the most flammable gases and posed the greatest risk of external combustion and explosion.11

5

Trade-off characters between internal reaction and external combustion

A recent study illustrates that there is a trade-off between thermal runaway within the battery and external combustion (Fig. 11). Cathode oxidation is linearly correlated with the intensity of thermochemical reactions within battery components. The nickel-rich

Battery Types – Lithium Batteries – Lithium Battery Safety | Overview

667

Fig. 11 Trade-off relationship between internal thermal runaway and external combustion in lithium-ion batteries.11 From Zhao, L.; Hou, J.; Feng, X.; Xu, C.; Wang, H.; Liu, H.; Hou, B.; Rui, X.; Lu, L.; Bao, C.; Ouyang, M. The Trade-Off Characteristic between Battery Thermal Runaway and Combustion. Energy Storage Mater. 2024, 69, 103380.

NCM battery exhibits the most intense exothermic reactions and energy release due to the strong oxidation of its cathode, leading to the highest thermal runaway temperature of 862.4  C within the battery. The LiFePO4 cathode with the weak oxidation shows the lowest temperature of 297.4  C within the battery, but with the highest laminar flame speed of 55 cm s−1 outside the battery. The reason for this is that the LiFePO4 cathode rarely oxidizes the electrolyte, instead, the electrolyte reacts with the anode primarily, thereby generating substantial reductive gases and accumulating the explosion risks. For LiFePO4 battery, the cathode is only weakly oxidizing and seldom reacts with other compounds. It is proposed that most of the electrolyte will react with the highly-reductive lithiated anode and generates flammable gases, causing severe combustion and even explosion in LiFePO4 battery. However, for NCM batteries, the cathodes pose the strong oxidation and part of electrolyte can be oxidized, which is proposed that more CO and CO2 gases generated and emitted, leading to a lower explosion risk than LiFePO4 battery. LiFePO4 battery exhibits the higher laminar flame speed outside and the lower maximum temperature inside during battery thermal runaway, while NCM batteries possess intense internal thermal runaway and weak external combustion hazards. Therefore, within the battery, thermal runaway intensity has an order of NCM811 battery > NCM622 battery > NCM523 battery > NCM333 battery > LiFePO4 battery. The battery with NCM811 cathode shows the most intensive chemical reactions among the battery components inside the cell. Outside the battery, on the contrary, the LiFePO4 battery released the most flammable gases and posed the greatest risk of external combustion and explosion. The relationship between internal thermal runaway and external combustion is that when one is safer, the other is more dangerous.11

6

Electrolyte design for high-safety battery

Thermal runaway is driven by a series of chemical chain reactions, commencing from the thermal decomposition of SEI at temperatures of 70  C  120  C inside the batteries, which leads to the final catastrophic events. Generally, the SEI layer acts as a barrier between the electrode and the electrolyte for stable cycling, but it may also serve as a fuel for the thermal runaway process. Under thermal abuse conditions, the inorganic-rich SEI derived from conventional ester-based electrolytes will decompose and dissolve, resulting in the generation of heat and in turn increasing the battery’s temperature. When the SEI layer breaks down, the highly-reactive lithiated anode is exposed, causing the intensive reduction reactions of the lithium salt and solvents, thereby further increasing the battery temperature. At the site of SEI cracking, carbonate solvents undergo reduction with generation of reductive gases such as H2, CH4, C2H4, etc. These reductive gases then migrate to the cathode side, attacking the cathode structure and accelerate cathode degradation with reactive oxygen release, then the battery goes into an uncontrollable state. When it comes to high-nickel and ultrahigh-nickel cathodes, they are more vulnerable to the anode-releasing reductive gasses, resulting in deteriorated safety performance. Therefore, SEI degradation signals the possibility that the above chain exothermic reactions may be taking place within the battery. As a result of the high thermal stability of inorganic matter, the inorganic-rich SEI is proposed to postpone thermal decomposition and mitigate unfavorable side effects. Developing electrolytes that can form inorganic-rich interface and are resistant to reduction is essential for enhancing the intrinsic thermal safety of batteries. To determine the relationship between chemical composition and thermal stability of SEI, we used temperature-dependent XPS to detect the SEI composition changes upon heating (Fig. 12). In base electrolyte of 1 M conventional electrolyte, the P/F-containing species have changed significantly when heating

668

Battery Types – Lithium Batteries – Lithium Battery Safety | Overview

Fig. 12 High-safety 1.2 Ah NMC955|Gr battery achieved by thermally stable SEI. (a) Temperature-dependent XPS of anodes collected from BE-based battery; (b) Temperature-dependent XPS of anodes collected from SE-based battery; (c) DSC results of An+electrolyte; (d) hot box test of 1.2 Ah NMC955|Gr battery with BE and SE, respectively; (e) 1.2 Ah battery wreckage after thermal runaway.

from 25  C to 200  C, indicating that the SEI is unstable under thermal abuse and decomposes in practical batteries. Comparatively, the SEI containing inorganic-rich (LiF, Li3N and Li2SO3 etc.) remains almost stable till 200  C. The thermally stable and inorganic-rich SEI that signals the strong resistance to high temperatures, which suppresses the strong side reactions between highly-reactive charged anode and electrolyte. Then, the heat evolution behavior of the mixture between electrolyte and fully-charged powder was investigated by DSC (Fig. 12). Upon mixing, the mixture sample of anode and conventional electrolyte generated heat earlier, with a peak at 99.5  C. In contrast, charged graphite with safe electrolyte is thermally stable up to 124.6  C and exhibits a low reaction rate with a smaller heat release. This initial release of heat was primarily caused by the decomposition of SEI and the reaction between lithiated graphite and electrolyte. This indicates that the inorganic-rich SEI derived from safe electrolyte could enhance the interface stability and mitigate the reaction between anode and electrolyte, thus improving the thermal failure onset temperature and lowering the heat release rate, thereby enhancing battery safety. Thermal safety can be enhanced by reducing the high-reactively agents in the electrolyte. The solvent of ethylene carbonate (EC) was demonstrated to plays a triggering role in thermal runaway of NMC811|Gr cell. The electrolyte with lower EC concentration is proposed to mitigate thermal runaway, especially enhancing the trigger point of battery thermal runaway. When the EC concentration was artificially reduced from 31% (normal) to 6% and 0% by weight, and the heat generation of Ca+Ely decreased

Battery Types – Lithium Batteries – Lithium Battery Safety | Overview

669

obviously. If EC was entirely removed from the electrolyte, the heat generation of Ca+Ely (0% EC) was comparable to that of pure Ca. The EC-free electrolyte significantly suppressed the self-heating rate with improved the tipping point of thermal runaway, further demonstrating the critical role of EC during thermal runaway of the LIBs.15 Fluorinated solvents or electrolytes possess good antioxidant capacity that provides high compatibility to high-voltage or high-nickel cathode and flame retardance; thus, they are considered as a promising solution for advanced lithium-ion batteries carrying both high-energy density and high safety. Moreover, the fluorinated electrolytes are widely used to form stable electrolyte interphase, due to their chemical reactivity with lithiated graphite or lithium. It was demonstrated that the flame-retardant fluorinated electrolytes help to reduce the flammability when battery goes into thermal runaway. The fluorinated solvent of 1,1,2,2-tetrafluoroethyl-2,2,3,3-tetrafluoropropylether (D2 or TTE from DAIKIN company) in the electrolyte lowered the heat release rate and total heat release of NMC811 battery, which is supposed to be promising in reducing the external fire hazard of lithium-ion batteries.14

7

Summary and outlook

This study demonstrates a trade-off relationship between internal thermal chemical reactions and external combustion for batteries during thermal runaway. Inside the batteries, the intensity of internal thermal runaway is linearly related to cathode oxidizability. The intensity of internal chemical reactions has an order of NCM955 battery > NCM811 battery > NCM622 battery > NCM523 battery > NCM333 battery > LiFePO4 battery. With an increase in cathode oxidizability, the internal chemical reaction of battery materials becomes more intense with tremendous energy release within the battery. Outside the battery, the weaker the oxidizability of cathode, the stronger the external combustion. The combustion intensity has an order of LiFePO4 battery > NCM333 battery  NCM523 battery  NCM622 battery > NCM811 battery  NCM955 battery. LiFePO4 battery emits the highest concentration of H2 and C2H4, which increases the risk of explosion. By using air and N2 dilution, NCM batteries can achieve only a thermal runaway but without combustion. A trade-off exists between internal thermal runaway and external combustion. The stronger the energy release of chemical reactions inside, the weaker the fire and explosion outside. Increasing nickel content in a battery (e.g. NCM811, NCM955) leads to an intensified internal thermal runaway due to a decrease in cathode thermal stability. In contrast, the LiFePO4 batteries and low-nickel batteries are more likely to exhibit electrolyte reduction reactions, resulting in the release of more flammable gases, thereby increasing the risk of explosions. LiFePO4 batteries have the good balance of energy, power, longevity, and cost, occupying the largest market share with the help of blade batteries, cell-to-pack batteries, etc.25 However, LiFePO4 batteries with a capacity of MWh to GWh require further investigation due to their high failure rate and large number of cells.13 The high-energy NCM batteries are struggling in battery safety to compete with the LiFePO4 batteries in vehicle application scenarios. As long as NCM batteries are intrinsically safe, they could still outperform LiFePO4 batteries, as the latter has reached the limit of their energy density. The intrinsic safety of NCM batteries can be improved by electrolyte design, electrode modification,26 separator enhancement, and cell design guided by the reaction map proposed.27 Solid-state batteries and batteries with other cell chemistries must first solve energy, power, longevity, and cost problems, then safety issues may arise. It is hoped that this review will provide new insights and ideas for tackling the problems of high-energy lithium and lithium-ion batteries failing.

References 1. Feng, X.; Fang, M.; He, X.; Ouyang, M.; Lu, L.; Wang, H.; Zhang, M. Thermal Runaway Features of Large Format Prismatic Lithium Ion Battery Using Extended Volume Accelerating Rate Calorimetry. J. Power Sources 2014, 255, 294–301. 2. Myung, S.; Maglia, F.; Park, K.; Yoon, C. S.; Lamp, P.; Kim, S.; Sun, Y. Nickel-Rich Layered Cathode Materials for Automotive Lithium-Ion Batteries: Achievements and Perspectives. ACS Energy Lett. 2017, 2, 196–223. 3. Li, M.; Lu, J.; Chen, Z.; Amine, K. 30 Years of Lithium-Ion Batteries. Adv. Mater. 2018, 30. 4. Liu, L.; Xu, J.; Wang, S.; Wu, F.; Li, H.; Chen, L. Practical Evaluation of Energy Densities for Sulfide Solid-State Batteries. eTransportation 2019, 1, 100010. 5. Ren, D.; Hsu, H.; Li, R.; Feng, X.; Guo, D.; Han, X.; Lu, L.; He, X.; Gao, S.; Hou, J.; Li, Y.; Wang, Y.; Ouyang, M. A Comparative Investigation of Aging Effects on Thermal Runaway Behavior of Lithium-Ion Batteries. eTransportation 2019, 2, 100034. 6. Hasan, M. K.; Mahmud, M.; Ahasan Habib, A. K. M.; Motakabber, S. M. A.; Islam, S. Review of Electric Vehicle Energy Storage and Management System: Standards, Issues, and Challenges. J. Energy Storage 2021, 41, 102940. 7. Zhai, H.; Li, H.; Ping, P.; Huang, Z.; Wang, Q. An Experimental-Based Domino Prediction Model of Thermal Runaway Propagation in 18,650 Lithium-Ion Battery Modules. Int. J. Heat Mass Transfer. 2021, 181, 122024. 8. Feng, X.; Ouyang, M.; Liu, X.; Lu, L.; Xia, Y.; He, X. Thermal Runaway Mechanism of Lithium Ion Battery for Electric Vehicles: A Review. Energy Storage Mater. 2018, 10, 246–267. 9. Feng, X.; Zheng, S.; Ren, D.; He, X.; Wang, L.; Cui, H.; Liu, X.; Jin, C.; Zhang, F.; Xu, C.; Hsu, H.; Gao, S.; Chen, T.; Li, Y.; Wang, T.; Wang, H.; Li, M.; Ouyang, M. Investigating the Thermal Runaway Mechanisms of Lithium-Ion Batteries Based on Thermal Analysis Database. Appl. Energy 2019, 246, 53–64. 10. Feng, X.; Ren, D.; He, X.; Ouyang, M. Mitigating Thermal Runaway of Lithium-Ion Batteries. Joule 2020, 4, 743–770. 11. Zhao, L.; Hou, J.; Feng, X.; Xu, C.; Wang, H.; Liu, H.; Hou, B.; Rui, X.; Lu, L.; Bao, C.; Ouyang, M. The Trade-Off Characteristic between Battery Thermal Runaway and Combustion. Energy Storage Mater. 2024, 69, 103380. 12. Liu, X.; Ren, D.; Hsu, H.; Feng, X.; Xu, G.; Zhuang, M.; Gao, H.; Lu, L.; Han, X.; Chu, Z.; Li, J.; He, X.; Amine, K.; Ouyang, M. Thermal Runaway of Lithium-Ion Batteries without Internal Short Circuit. Joule 2018, 2, 2047–2064. 13. Feng, X. N.; Ren, D. S.; Ouyang, M. G. Safety of Lithium Battery Materials Chemistry. J. Mater. Chem. A 2023, 11, 25236–25246.

670

Battery Types – Lithium Batteries – Lithium Battery Safety | Overview

14. Hou, J.; Wang, L.; Feng, X.; Terada, J.; Lu, L.; Yamazaki, S.; Su, A.; Kuwajima, Y.; Chen, Y.; Hidaka, T.; He, X.; Wang, H.; Ouyang, M. Thermal Runaway of Lithium-Ion Batteries Employing Flame-Retardant Fluorinated Electrolytes. Energy Environ. Mater. 2023, 6, e12297. 15. Hou, J.; Feng, X.; Wang, L.; Liu, X.; Ohma, A.; Lu, L.; Ren, D.; Huang, W.; Li, Y.; Yi, M.; Wang, Y.; Ren, J.; Meng, Z.; Chu, Z.; Xu, G.-L.; Amine, K.; He, X.; Wang, H.; Nitta, Y.; Ouyang, M. Unlocking the Self-Supported Thermal Runaway of High-Energy Lithium-Ion Batteries. Energy Storage Mater. 2021, 39, 395–402. 16. Xu, G.; Huang, L.; Lu, C.; Zhou, X.; Cui, G. Revealing the Multilevel Thermal Safety of Lithium Batteries. Energy Storage Mater. 2020, 31, 72–86. 17. Tomaszewska, A.; Chu, Z.; Feng, X.; O’Kane, S.; Liu, X.; Chen, J.; Ji, C.; Endler, E.; Li, R.; Liu, L.; Li, Y.; Zheng, S.; Vetterlein, S.; Gao, M.; Du, J.; Parkes, M.; Ouyang, M.; Marinescu, M.; Offer, G.; Wu, B. Lithium-Ion Battery Fast Charging: A review. eTransportation 2019, 1, 100011. 18. Li, Y.; Liu, X.; Wang, L.; Feng, X.; Ren, D.; Wu, Y.; Xu, G.; Lu, L.; Hou, J.; Zhang, W.; Wang, Y.; Xu, W.; Ren, Y.; Wang, Z.; Huang, J.; Meng, X.; Han, X.; Wang, H.; He, X.; Chen, Z.; Amine, K.; Ouyang, M. Thermal Runaway Mechanism of Lithium-Ion Battery with LiNi0.8Mn0.1Co0.1O2 Cathode Materials. Nano Energy 2021, 85. 19. Feng, X.; Zheng, S.; He, X.; Wang, L.; Wang, Y.; Ren, D.; Ouyang, M. Time Sequence Map for Interpreting the Thermal Runaway Mechanism of Lithium-Ion Batteries with LiNixCoyMnzO2 Cathode. Front. Energy Res. 2018, 6. 20. Zheng, Y.; Shi, Z.; Ren, D.; Chen, J.; Liu, X.; Feng, X.; Wang, L.; Han, X.; Lu, L.; He, X.; Ouyang, M. In-Depth Investigation of the Exothermic Reactions between Lithiated Graphite and Electrolyte in Lithium-Ion Battery. J. Energy Chem. 2022, 69, 593–600. 21. Wang, Y.; Feng, X.; Peng, Y.; Zhang, F.; Ren, D.; Liu, X.; Lu, L.; Nitta, Y.; Wang, L.; Ouyang, M. Reductive Gas Manipulation at Early Self-Heating Stage Enables Controllable Battery Thermal Failure. Joule 2022, 6, 2810–2820. 22. Wang, H.; Xu, H.; Zhang, Z.; Wang, Q.; Jin, C.; Wu, C.; Xu, C.; Hao, J.; Sun, L.; Du, Z.; Li, Y.; Sun, J.; Feng, X. Fire and Explosion Characteristics of Vent Gas From Lithium-Ion Batteries after Thermal Runaway: A Comparative Study. eTransportation 2022, 13, 100190. 23. Sieradzka, M.; Rajca, P.; Zajemska, M.; Mlonka-Me˛ drala, A.; Magdziarz, A. Prediction of Gaseous Products from Refuse Derived Fuel Pyrolysis Using Chemical Modelling Software - Ansys Chemkin-Pro. J. Clean. Prod. 2020, 248, 119277. 24. An, S.; Jung, J. C. Kinetic Modeling of Thermal Reactor in Claus Process Using CHEMKIN-PRO Software. Case Stud. Therm. Eng. 2020, 21, 100694. 25. Zhang, J.; Wang, Y.; Jiang, B.; He, H.; Huang, S.; Wang, C.; Zhang, Y.; Han, X.; Guo, D.; He, G.; Ouyang, M. Realistic Fault Detection of Li-Ion Battery Via Dynamical Deep Learning. Nat. Commun. 2023, 14, 5940. 26. Wu, Y.; Ren, D.; Liu, X.; Xu, G.-L.; Feng, X.; Zheng, Y.; Li, Y.; Yang, M.; Peng, Y.; Han, X.; Wang, L.; Chen, Z.; Ren, Y.; Lu, L.; He, X.; Chen, J.; Amine, K.; Ouyang, M. High-Voltage and High-Safety Practical Lithium Batteries with Ethylene Carbonate-Free Electrolyte. Adv. Energy Mater. 2021, 11, 2102299. 27. Hong, Y.; Zhang, Y.; Li, C.; Zhang, F.; Gao, F.; Sheng, J.; Su, S.; Chen, S.; Xu, C.; Jin, C.; Wang, H.; Zheng, Y.; Wang, H.; Feng, X.; Ouyang, M. High-Security Prismatic Battery with Cover Filled Agent. J. Energy Storage 2023, 64, 107133.

Battery Types – Lithium Batteries – Lithium Battery Safety | Chemical Hazards Changyong Jin, School of Vehicle and Mobility, Tsinghua University, Beijing, China © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 2.1 2.1.1 2.1.2 2.1.3 2.2 2.2.1 2.2.2 2.2.3 2.2.4 3 3.1 3.1.1 3.1.2 3.2 3.3 3.3.1 3.3.2 4 References

Introduction Standard of chemical hazard Hazard classification Health hazard Physical hazard Environmental hazard Standards for classification of chemical hazards Globally harmonized system of classification and labelling of chemicals, GHS Regulation (EC) No 1272/2008 on Classification, Labelling and Packaging of substances and mixtures, CLP CGA P-20-2017: Standard for classification of toxic gas mixtures Occupational safety and health administration’s hazard communication standard, HCS Li-battery material chemical hazard Cathode material Element Metal oxides Anode material Electrolyte Solvent Lithium hexafluorophosphate Conclusion

672 672 672 672 672 672 673 673 673 673 673 674 674 674 674 674 674 674 675 675 676

Abstract Lithium-ion batteries are widely used in modern technology, but their chemical properties pose significant health and environmental risks. This introduction explores the chemical hazards of lithium-ion batteries, focusing on their potential impacts. From a health perspective, the primary concerns are chemical leakage and fires, which can cause severe skin, respiratory, and other health issues. Environmentally, lithium-ion batteries can harm the environment during production, use, and improper disposal, leading to pollution and ecological disruption. Although lithium-ion batteries provide great convenience, their chemical characteristics also bring non-negligible risks. Thoroughly understanding these risks and implementing appropriate preventive measures are crucial for protecting human health and the environment.

Key points

• • •

Classifications of health, physical, and environmental hazards for chemical substances. Globally harmonized system (GHS), CLP Regulation, and other standards for chemical hazard. Detailed analysis of the toxicity and reactivity of cathode metals, anode graphite, and electrolyte solvents/salts.

Abbreviations ACGIH CLP CO DMC DME EC EMC GHS HCS HF

American Conference of Governmental Industrial Hygienists Classification, labelling and packaging Carbon monoxide Dimethyl carbonate Dimethyl ether Ethylene carbonate Ethyl methyl carbonate Globally harmonized system Hazard communication standard Hydrogen fluoride

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00317-0

671

672

Battery Types – Lithium Batteries – Lithium Battery Safety | Chemical Hazards

LCO LFP LMO LTO OSHA PC

1

Lithium cobalt oxide Lithium iron phosphate Lithium manganese oxide Lithium titanium oxide Occupational safety and health administration Propylene carbonate

Introduction

Lithium-ion batteries, as a crucial component of modern technology, are widely used in a variety of devices, ranging from portable electronic products to electric vehicles. However, the chemical properties of these batteries not only bring immense convenience but also raise serious concerns about health and environmental risks. This introduction aims to explore the chemical hazards of lithium-ion batteries, particularly focusing on their potential impacts on human health and the environment. From a health perspective, the primary hazards of lithium-ion batteries include chemical leakage and fires. Chemical substances inside the batteries, such as lithium, cobalt, and nickel, can cause severe skin and respiratory irritation if leaked, and even more serious health problems. Additionally, overheating or damage to the batteries can lead to fires, releasing toxic gases that pose a threat to human health. In terms of environmental risks, lithium-ion batteries can cause environmental harm during their production, usage, and disposal phases. The manufacturing process generates harmful waste and emissions, and improperly disposed of batteries can contaminate soil and water sources, disrupting ecosystems. Moreover, the mining of lithium also significantly impacts the environment, such as groundwater pollution and ecological destruction. Overall, while lithium-ion batteries provide great convenience in modern life, their chemical characteristics also bring non-negligible health and environmental risks. Therefore, understanding these risks thoroughly and taking appropriate preventive and mitigative measures are crucial for protecting human health and the environment. The key points covered in this chapter include: 1. Classifications of health, physical, and environmental hazards for chemical substances; 2. Globally harmonized system (GHS), CLP Regulation, and other standards for chemical hazard; 3. Detailed analysis of the toxicity and reactivity of cathode metals, anode graphite, and electrolyte solvents/salts;

2 2.1

Standard of chemical hazard Hazard classification

2.1.1 Health hazard Chemicals are classified as health hazards if they pose risks such as acute or chronic toxicity, skin corrosion/irritation, serious eye damage/irritation, respiratory or skin sensitization, germ cell mutagenicity, carcinogenicity, reproductive toxicity, specific target organ toxicity (whether from single or repeated exposure), and aspiration hazard. For example, during thermal runaway in batteries, a large amount of mixed toxic gases, such as carbon monoxide (CO) and hydrogen fluoride (HF), are produced. Inhalation of these gases can lead to severe respiratory tract injuries and trigger a range of health issues, including headaches, vomiting, fatigue, confusion, loss of appetite, chest pain, rapid breathing, increased heart rate, loss of consciousness, and even cardiac arrest.

2.1.2 Physical hazard Physical hazards refer to the dangers arising from the physical properties of chemicals, rather than their chemical composition. These hazards include flammable liquids and solids, self-heating substances, explosive materials, oxidizers and organic peroxides, pressurized gas containers, corrosive gases, hygroscopic and deliquescent materials, pyrophoric and high-temperature substances, cryogenic gases and liquids, as well as magnetic materials. These substances may cause fires, explosions, or other forms of harm under certain conditions. The physical hazards present in lithium batteries mainly include thermal runaway, fire, and explosion risks. Lithium batteries can experience thermal runaway under abuse conditions such as overcharging, deep discharging, dropping, or overheating. Thermal runaway is a phenomenon where the chemical reactions inside the battery become uncontrollable, leading to a rapid increase in internal temperature. This can potentially lead to fires and, in extreme cases, explosions.

2.1.3 Environmental hazard Environmental hazards refer to the potential negative impacts of certain substances or actions on the natural environment, including pollution of air, water, and soil. These hazards can stem from the leakage of chemical substances, improper disposal of waste, industrial emissions, excessive use of pesticides and fertilizers, and other various pollution sources.

Battery Types – Lithium Batteries – Lithium Battery Safety | Chemical Hazards

673

The environmental hazards associated with lithium batteries primarily involve their production, use, and disposal processes. During production, the manufacturing of lithium batteries can lead to the release of harmful chemical substances, affecting air and water quality. During use, damage to batteries or improper handling can cause the leakage of harmful substances, polluting soil and water sources. If disposed of improperly, the heavy metals and toxic chemicals contained in spent lithium batteries can leach into the environment, causing long-term impacts on ecosystems.

2.2

Standards for classification of chemical hazards

2.2.1 Globally harmonized system of classification and labelling of chemicals, GHS The Globally Harmonized System (GHS) is an internationally recognized standard managed by the United Nations, aimed at replacing various classification and labeling schemes for hazardous materials previously used around the world. It provides a unified standard for the classification of substances and mixtures according to health hazards, physical hazards, and environmental hazards. The system includes unified elements for hazard communication, including requirements for labeling and safety data sheets. Health hazards include classifications related to acute toxicity, skin corrosion/irritation, serious eye damage/irritation, respiratory or skin sensitization, reproductive toxicity, germ cell mutagenicity, carcinogenicity, specific target organ toxicity (from single or repeated exposure), and aspiration hazard caused by chemicals. Environmental hazards focus on the acute and chronic toxic impacts of chemicals on the aquatic environment, including toxicity to aquatic life, persistence, bioaccumulation, and potential damage to the ozone layer.

2.2.2 Regulation (EC) No 1272/2008 on Classification, Labelling and Packaging of substances and mixtures, CLP Regulation (European Commission, EC) No 1272/2008, also known as the CLP Regulation, pertaining to the Classification, Labelling, and Packaging of substances and mixtures, encompasses comprehensive criteria for health and environmental hazards. Health hazards under this regulation include various categories like acute toxicity, skin and eye irritation or damage, respiratory or skin sensitization, mutagenicity, carcinogenicity, reproductive toxicity, and specific target organ toxicity. It emphasizes clear criteria for classifying substances and mixtures that pose such risks to human health. Regarding environmental hazards, the CLP Regulation focuses on the classification of substances and mixtures that are hazardous to the aquatic environment, addressing acute and chronic aquatic toxicity and bio accumulative potential. The regulation ensures that hazards are clearly communicated through standard labeling and packaging, thus safeguarding human health and minimizing environmental impacts.

2.2.3 CGA P-20-2017: Standard for classification of toxic gas mixtures This standard provides guidelines and criteria for the classification of toxic gas mixtures based on their health hazards, which is based on the LC50 rat 1 h value (lethal concentration 50 for rats over a period of 1 h) toxicity threshold criterion, the definition of LC50 see Eq. (1). LC50 ¼ P i

1

Ci LC50i

(1)

In the formula, LC50 refers to the lethal concentration 50% (volume fraction) of the mixed gas, specifically indicating the concentration of gas in the air that can cause half of the tested adult male and female albino rats to die within 14 days when continuously inhaled for 1 h. Ci represents the concentration (volume fraction) of the ith toxic gas in the mixed gas, and LC50i is the lethal concentration 50 (volume fraction) of the ith toxic component in the mixed gas. This standard categorizes toxic gases or gas mixtures into four hazard zones: Hazard zone A (LC50 rat 1 h value 200 ppm), Hazard zone B (200 ppm < LC50 rat 1 h value 1000 ppm), Hazard zone C (1000 ppm < LC50 rat 1 h value 3000 ppm), and Hazard zone D (3000 ppm < LC50 rat 1 h value 5000 ppm). The United Nations’ GHS (Globally Harmonized System) also has similar classification standards, which, based on LC50 rat 4 h values, divide gas mixtures into four categories: Acute toxicity category 1 (LC50 rat 4 h value 100 ppm), Acute toxicity category 2 (100 ppm < LC50 rat 4 h values 500 ppm), Acute toxicity category 3 (500 ppm < LC50 rat 4 h values 2500 ppm), and Acute toxicity category 4 (2500 ppm < LC50 rat 4 h values 20,000 ppm).

2.2.4 Occupational safety and health administration’s hazard communication standard, HCS The Occupational Safety and Health Administration’s (OSHA) Hazard Communication Standard (HCS) focuses on ensuring that information about the chemical hazards and associated protective measures is communicated to workers. Regarding health hazards, the HCS includes criteria for classifying a chemical as a health hazard if it poses risks such as carcinogenicity, reproductive toxicity, respiratory or skin sensitization, or specific target organ toxicity. It requires that these hazards be clearly communicated to workers through labels and safety data sheets. The standard, however, primarily concentrates on workplace health hazards and does not extensively cover environmental hazards, which are typically addressed under the regulations of environmental protection agencies. Nevertheless, HCS ensures that workers are informed about the potential health risks associated with chemicals they might encounter in their workplace, enabling them to take necessary precautions to safeguard their health.

674

3 3.1

Battery Types – Lithium Batteries – Lithium Battery Safety | Chemical Hazards

Li-battery material chemical hazard Cathode material

3.1.1 Element Lithium batteries’ raw materials include a large number of metal elements such as lithium, copper, nickel, cobalt, manganese, and aluminum. Other metals detected at very low levels include barium, chromium, silver, thallium, vanadium, zinc, and lead.1 Lithium, the core element of lithium-ion batteries, is reported in literature2 as not having bio accumulative properties, and its toxicity to humans and the environment is relatively low. Lithium intake, whether from food and water or occupational exposure, is not considered to pose toxicological hazards. Copper exposure can potentially cause liver damage, brain injury, and skeletal harm. Nickel exposure may lead to dermatitis, chronic asthma, and an increased risk of cancer.3 Aluminum, also a significant component of lithium batteries, is generally considered non-toxic or having very slow-acting toxicity. However, in work environments with high concentrations of aluminum dust, there is an increased risk of developing pulmonary fibrosis. Additionally, frequent use of aluminum products has been suggested to potentially lead to chronic neurological damage, including diseases such as Alzheimer’s.4 Cobalt exposure can lead to a complex clinical syndrome, marked by a range of systemic health effects. Key areas affected include neurological systems, where exposure may result in hearing and visual impairments; cardiovascular systems, potentially leading to heart-related issues; and the endocrine system, affecting hormonal balances. It’s also important to note that chronic exposure to cobalt at levels deemed acceptable is not expected to pose significant health hazards.5 Excessive intake of manganese would damage the nervous system.1

3.1.2 Metal oxides The cathode materials of batteries primarily consist of metal oxides like ternary materials (LiNixCOyMn1-x-yO2), lithium iron phosphate (LiFePO4), lithium manganese oxide (LiMn2O4), lithium cobalt oxide (LiCoO₂) and lithium titanate (Li2TiO3). In a study,6 researchers compared the health conditions of mice after inhaling three different cathode materials: LCO (Lithium Cobalt Oxide), LFP (Lithium iron phosphate), and LTO (Lithium titanium oxide). They found that LTO induced mild acute inflammation; both LFP and LCO induced acute and sub-chronic inflammation, but only LCO led to a fibrotic response. It was observed that exposure to LCO material significantly increased the chances of acute and sub-chronic lung inflammation in mice, with cobalt playing a crucial role in LCO lung toxicity. Additionally, cobalt compounds were found to induce cytotoxicity, apoptosis, inflammatory responses, and genotoxicity in vitro.7 LiMn2O4 itself is not considered a toxic substance. However, it is important to note that during the handling of batteries, especially when they are failing or being processed, harmful substances may be released. For instance, discarded LMO (lithium manganese oxide) batteries can be a potential source of pollution due to the presence of manganese, organic solvents, and other toxic materials. Unprocessed discharge of these materials can cause severe contamination to soil and water bodies. Additionally, with the depletion of high-grade manganese ore resources over the past few decades, there has been an increased use of low-grade manganese minerals, leading to a rise in solid waste production.8

3.2

Anode material

According to literature,9 both synthetic and natural graphite meet the standards set by the American Conference of Governmental Industrial Hygienists (ACGIH) for harmful dust. Prolonged exposure to environments with graphite dust can lead to lung diseases. There have been studies analyzing harmful components in graphite powder from discarded lithium batteries, focusing on the potential health risks associated with exposure to these materials. Research has been conducted on the hazardous components in the graphite powder from spent lithium batteries. The results indicated that all spent graphite samples should be classified as hazardous waste. This classification is primarily due to the excessive levels of nickel (Ni), and in some cases, cadmium (Cd) and fluoride (F) present in the samples. Additionally, the life cycle impact assessment results revealed that the main contributors to the hazardous nature were Ni, Co, and Al, rather than the graphite itself.10

3.3

Electrolyte

The electrolyte is a key component of lithium-ion batteries, responsible for the transfer of lithium ions between the anode and cathode, facilitating the battery’s charging and discharging process. The electrolyte typically comprises components such as solvents, electrolyte salts, and additives. Below is a toxicity analysis of these components, considering the electrolyte solvent and electrolyte salt.

3.3.1 Solvent Currently, the solvents primarily used in lithium-ion battery electrolytes include Dimethyl Carbonate (DMC), Ethylene Carbonate (EC), Ethyl Methyl Carbonate (EMC), Propylene Carbonate (PC), and Dimethyl ether (DME). DMC is a highly flammable solvent that is non-irritating and non-sensitizing. According to literature,11 studies using a mouse model have assessed the allergenic potential and immunotoxicity caused by skin exposure to DMC. Although significant data gaps

Battery Types – Lithium Batteries – Lithium Battery Safety | Chemical Hazards

675

still exist in the complete toxicological evaluation of this chemical, these results suggest that DMC is an immunotoxic chemical when assessed in a mouse model. At the same time, DMC is a well-established solvent and is also a green reagent of considerable interest. It is a non-polar, non-proton solvent with good miscibility with water, biodegradable in the atmosphere, and non-toxic. According to solvent selection guidelines, it can serve as a potential substitute for Methyl Ethyl Ketone, Ethyl Acetate, Methyl Isobutyl Ketone, and most other ketones.12 EC is acutely toxic if ingested, and prolonged or repeated exposure can cause damage to organs. Additionally, it can cause severe irritation to the eyes and cause skin irritation.13 EMC is a flammable solvent with relatively low acute toxicity to humans. However, contact with skin, inhalation, or ingestion can lead to damage to body organs.14 PC is considered an environmentally friendly material and a solvent with relatively low toxicity. However, it can cause serious eye damage or irritation and is classified as Category 2 under the OSHA’s HCS. DME is a colorless, odorless liquid with low toxicity, but it is highly flammable and should be used with caution. It may cause irritation to the skin and eyes and is classified as having Specific Target Organ Toxicity (single exposure) Category 3. DME decomposes slowly in the environment and may have some impact on water bodies and soil.

3.3.2 Lithium hexafluorophosphate Lithium Hexafluorophosphate (LiPF6) is one of the most commonly used electrolyte salts due to its excellent electrical conductivity and chemical stability. However, LiPF6 can decompose into harmful substances when exposed to high temperatures or water. When a lithium battery is damaged, and LiPF6 comes into contact with water, it decomposes into LiF, Phosphoryl trifluoride (POF3), HF, as shown in Eq. (2). LiPF6 itself decomposes into LiF and PF5, as shown in Eq. (3).15 PF5 further reacts with water in the air, producing POF3 and HF,16 as shown in Eq. (4). LiPF6 + H2 O ! LiF + POF3 + 2HF

(2)

LiPF6 ! LiF + PF5

(3)

PF5 + H2 O ! POF3 + 2HF

(4)

LiPF6 can cause irritation to the eyes, skin, and respiratory tract. Contact with LiPF6 may result in pain, redness, and itching, and it has the potential to damage vital organs like the central nervous system, liver, and kidneys.17 PF5 is a colorless, irritating gas and a highly reactive and corrosive chemical substance. The health hazards from PF5 primarily arise through inhalation and skin contact. Inhalation of PF5 can lead to severe respiratory irritation and damage, including bronchitis and pulmonary edema, and can be fatal in extreme cases. Skin contact with PF5 can cause severe burns and corrosive injuries.18,19 HF can cause health hazards through inhalation, ingestion, and skin absorption.20–23 It is classified as Category 1 under OSHA for acute toxicity - dermal, irritation, and eye irritation. Specifically, HF has a strong corrosive effect on the skin, capable of penetrating deep layers, leading to necrosis and ulcers that are difficult to heal. Eye contact with high concentrations of hydrogen fluoride can cause corneal perforation. Exposure to its vapors can lead to bronchitis, pneumonia, and other respiratory ailments. Long-term exposure may cause chronic inflammation of the respiratory tract, leading to periodontitis and fluorosis. The fluoride ions in hydrogen fluoride dehydrate and corrode tissues, as fluorine is one of the most reactive non-metal elements. Upon skin contact with HF, fluoride ions continuously dissociate and penetrate deep tissues, dissolving cell membranes and causing liquefactive necrosis of the epidermis, dermis, subcutaneous tissue, and even muscle layers. Fluoride ions can also inhibit the activity of enolase, reducing the oxygen uptake capacity of skin cells. Inhalation of high concentrations of hydrogen fluoride mist can cause bronchitis and hemorrhagic pulmonary edema. HF can also be absorbed through the skin, causing severe poisoning. HF is also considered to lead to the formation of various hydrofluorocarbons (HFCs), which are generally insoluble in water and considered to be of low toxicity to humans. Regarding the ecotoxicity of these HFCs, they were also revealed to be not very toxic to aquatic organisms (such as algae, water fleas, and fish) and terrestrial plants.24 The PVDF binder material may decompose to HF when reaching 400  C, while the PVDF not classified as hazardous.

4

Conclusion

Lithium-ion batteries have become ubiquitous in modern technology, but their chemical compositions also introduce significant health and environmental risks that must be thoroughly understood and managed. This document has explored the diverse chemical hazards associated with the key materials used in lithium-ion batteries, including metals, metal oxides, solvents, and salts, which can pose severe consequences through leakage, thermal runaway, or improper disposal - ranging from skin/respiratory irritation and organ damage to chronic systemic effects. Similarly, the environmental impacts stemming from lithium-ion battery manufacturing, usage, and waste management pose serious concerns regarding pollution, ecosystem disruption, and resource depletion. To address these risks, it is crucial that comprehensive safety standards and regulations are strictly followed, and that manufacturers, users, and waste handlers adopt robust safety protocols and mitigation strategies throughout the entire lithium-ion battery lifecycle. As demand for this technology continues to grow, ongoing vigilance and collaborative efforts among all stakeholders are essential to ensuring the safe and sustainable deployment of lithium-ion batteries.

676

Battery Types – Lithium Batteries – Lithium Battery Safety | Chemical Hazards

References 1. Kang, D. H. P.; Chen, M.; Ogunseitan, O. A. Potential Environmental and Human Health Impacts of Rechargeable Lithium Batteries in Electronic Waste. Environ. Sci. Technol. 2013, 47 (10), 5495–5503. 2. Aral, H.; Vecchio-Sadus, A. Toxicity of Lithium to Humans and the Environment—A Literature Review. Ecotoxicol. Environ. Saf. 2008, 70 (3), 349–356. 3. Parmar, M.; Thakur, L. S. Heavy Metal Cu, Ni and Zn: Toxicity, Health Hazards and their Removal Techniques by Low Cost Adsorbents: A Short Overview. Int. J. Plant Animal Environ. Sci. 2013, 3 (3), 143–157. 4. Sorenson, J. R.; et al. Aluminum in the Environment and Human Health. Environ. Health Perspect. 1974, 8, 3–95. 5. Leyssens, L.; et al. Cobalt Toxicity in Humans—A Review of the Potential Sources and Systemic Health Effects. Toxicology 2017, 387, 43–56. 6. Sironval, V.; et al. Respiratory Hazard of Li-Ion Battery Components: Elective Toxicity of lithium Cobalt Oxide (LiCoO2) Particles in a Mouse Bioassay. Arch. Toxicol. 2018, 92, 1673–1684. 7. Simonsen, L. O.; Harbak, H.; Bennekou, P. Cobalt Metabolism and Toxicology—A Brief Update. Sci. Total Environ. 2012, 432, 210–215. 8. He, M.; et al. Sustainable and Facile Process for Li2CO3 and Mn2O3 Recovery from Spent LiMn2O4 Batteries Via Selective Sulfation with Waste Copperas. J. Environ. Chem. Eng. 2023, 11 (3), 110222. 9. Bergmann, J. D.; Hilaaki, R. J. Comparative Inhalation Hazards Of Titanium Dioxide, Synthetic And Natural Graphite. In Proceedings of the VIIth International Pneumoconioses Conference, Pittsburgh, Pennsylvania, USA, August 23–26, 1988; US Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Institute for Occupational Safety and Health, 1990. 10. Zhang, Z.; et al. Potential Environmental and Human Health Menace of Spent Graphite in lithium-Ion Batteries. Environ. Res. 2023, 117967. 11. Anderson, S. E.; et al. Immunotoxicity and Allergic Potential Induced by Topical Application of Dimethyl Carbonate (DMC) in a Murine Model. J. Immunotoxicol. 2013, 10 (1), 59–66. 12. Pyo, S.-H.; et al. Dimethyl Carbonate as a Green Chemical. Curr. Opin. Green Sustainable Chem. 2017, 5, 61–66. 13. Husár, B.; Liska, R. Vinyl Carbonates, Vinyl Carbamates, and Related Monomers: Synthesis, Polymerization, and Application. Chem. Soc. Rev. 2012, 41 (6), 2395–2405. 14. Lebedeva, N. P.; Boon-Brett, L. Considerations on the Chemical Toxicity of Contemporary li-Ion Battery Electrolytes and their Components. J. Electrochem. Soc. 2016, 163 (6), A821. 15. Teng, X.-G.; et al. Study on Thermal Decomposition of Lithium Hexafluorophosphate by TG–FT-IR Coupling Method. Thermochim. Acta 2005, 436 (1–2), 30–34. 16. Solchenbach, S.; et al. Quantification of PF5 and POF3 from Side Reactions of LiPF6 in li-Ion Batteries. J. Electrochem. Soc. 2018, 165 (13), A3022. 17. Aroca, R.; et al. Vibrational Spectra and Ion-Pair Properties of lithium Hexafluorophosphate in Ethylene Carbonate Based Mixed-Solvent Systems for lithium Batteries. J. Solution Chem. 2000, 29, 1047–1060. 18. Zackrisson, M.; Schellenberger, S. Toxicity of Lithium Ion Battery Chemicals-Overview with Focus on Recycling; Borås, Sweden: Rise, 2020. 19. Sankara Reddy, N.; et al. Bio-Control Efficiency of Pseudomonas Fluorescens Against Stem Rot of Tuberose Caused by Sclerotium Rolfsii. Plant Arch. 2021, 21, 1. 09725210. 20. Alexeeff, G. V.; Lewis, D. C.; Ragle, N. L. Estimation of Potential Health Effects from Acute Exposure to Hydrogen Fluoride Using a “Benchmark Dose” Approach. Risk Anal. 1993, 13 (1), 63–69. 21. Meldrum, M. Toxicology of Hydrogen Fluoride in Relation to Major Accident Hazards. Regul. Toxicol. Pharmacol. 1999, 30 (2), 110–116. 22. Sheet, S. D. Hydrogen Fluoride; Airgas: Radnor, PA, 2017. 23. Das, P. Fluorine: Risk Assessment, Environmental, and Health Hazard. Hazardous Gases 2021, 153–167. 24. Tsai, W.-T. An Overview of Environmental Hazards and Exposure Risk of Hydrofluorocarbons (HFCs). Chemosphere 2005, 61 (11), 1539–1547.

Battery Types – Lithium Batteries – Lithium Battery Safety | Thermal Hazards Yu Wang, Beijing Key Laboratory of Green Chemical Reaction Engineering and Technology, Department of Chemical Engineering, Tsinghua University, Beijing, China © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

1 2 2.1 2.2 2.2.1 2.2.2 2.2.3 2.2.4 2.2.5 2.3 2.3.1 2.3.2 2.4 2.4.1 2.4.2 2.4.3 3 3.1 3.1.1 3.1.2 3.1.3 3.2 3.2.1 3.2.2 4 References

Introduction Battery thermal runaway The characteristics of battery thermal runaway Thermal runaway mechanisms Reaction sequence and reaction mechanism analysis Anode-related reactions Cathode-related reactions Electrolyte, separator, and current collector related reactions Chemical crosstalks Smoke, fire and explosion Smoke Fire and explosion Mitigating battery thermal runaway Thermally stable materials Functional materials and structures Crosstalk mitigations Thermal runaway propagation Thermal runaway propagation mechanisms TR propagation driven by heat transfer TR propagation driven by gas and fire Shift from heat propagation to fire propagation Mitigating thermal runaway propagation Mitigating the heat propagation mode Avoiding the shift from heat propagation to fire propagation Outlooks

678 679 679 681 681 682 682 683 683 683 683 683 684 684 685 685 685 685 685 686 686 686 686 687 687 687

Abstract Batteries have played an important role in promoting green energy initiatives and applications. However, the thermal hazards associated with batteries, including high temperatures, fires, and explosions, can cause severe damage and have raised broad safety concerns. Battery thermal hazards originate from battery thermal runaway (TR). In real applications, the thermal runaway of one battery can propagate to adjacent batteries and the whole battery pack, leading to catastrophic thermal runaway propagation. This chapter will provide a comprehensive review of battery thermal hazards, covering both TR and TR propagation, including the most advanced research on TR and propagation behaviors, mechanisms, and mitigating technologies.

Glossary Accelerating rate calorimeter A calorimetry that provides adiabatic environment, commonly used to test battery thermal runaway characteristics. Battery module An assembly of individual batteries integrated together to provide a desired voltage and capacity for powering electronic devices. Battery pack An assembly of individual battery modules. Cathode electrolyte interface The interface layer between cathode and electrolyte in batteries. Current collector A conductive metal sheet that facilitates the transfer of electrons between the electrode material and the external circuit. Diethylene carbonate A carbonate solvent that commonly adopted in commercial batteries. Differential scanning calorimetry A calorimetry with high resolution that commonly used to test material thermal features. Dimethyl carbonate A carbonate solvent that commonly adopted in commercial batteries. Electrolyte additive A chemical added to the electrolyte to enhance performance, longevity, or safety, usually in small amounts of

Discharge

VO+2 + 2H+ + e − Eo ¼ 1:00 V

(I)

Eo being the electrode potential. At the negative electrode: V 3+ + e −

Charge

>

Discharge

V 2+ Eo ¼ −0:26 V

(II)

Batteries – Battery Types – Redox-Flow Batteries | Overview

3

Fig. 1 Principle of a redox flow battery (RFB), showing recirculation of an electrolyte through a cell compartment-tank loop with cell divided by a cation-exchange membrane. The cell is shown under charge. From Watt-Smith, M.J.; Wills, R.G.A.; Walsh, F.C. Secondary Batteries - Flow systems Overview. In Encyclopedia of Electrochemical Power Sources, 1st edn., Garche, J.; Dyer, C.; Moseley, P.; Ogumi, Z.; Rand, D.; Scrosati, B. (eds.), Elsevier, 2009, pp. 438–442.

Cell reaction: V 3+ + VO2+ + H2 O

Charge

>

Discharge

V 2+ + VO+2 + 2H+ V ocell ¼ 1:26 V

(III)

Vcell being the voltage. Each electrode reaction must be reversible with fast kinetics and favorable thermodynamics for the forward and reverse reactions.4,5 In order to maintain a high cell voltage, the positive and negative electrode reactions should be separated by a large potential difference with low overpotentials associated with each reaction. Ohmic voltage losses must be low in the electrolyte, electrode, and membrane phases. An RFB generally contains an ion-exchange membrane (either cationic or anionic) that serves two purposes. First, the membrane allows transport of the anionic or cationic species required for the necessary redox reactions. Second, it separates the two electrolyte solutions. In a generic RFB, species are oxidized at the positive electrode (anode) and reduced at the negative electrode (cathode) during charge. The reverse procedure occurs during discharge and the original redox species are reformed. This allows RFBs to serve as a versatile means of energy storage for which the power output and energy storage can be designed separately (see below). Systems are being developed for both isolated power networks and wholesale distribution markets. The batteries provide an energy storage facility that can respond rapidly (in the order of seconds) between fully charging and fully discharging operation and that is highly scalable (kW to MW units). In the simplest case, RFBs involve soluble species for all reactants and products. In practice, it has become common to include batteries involving one (or both) half-cell reaction that involves a liquid–solid phase transformation; the case of a negative electrode involving metal deposition and dissolution is particularly common. It is also possible to include positive half-cell reactions that involve the deposition and dissolution of metal oxides.6 There are particular cases that involve a gas phase, as hydrogen7 or oxygen in the negative or positive electrode respectively.

2

Principles of operation and plant design

All redox flow systems operate on the same principle: the combination of two redox-active half-cells. Each full-cell is analogous to that within a conventional battery, except that the active materials in RFBs are highly soluble (and chemically stable) in the electrolyte. During operation, the electrolyte is circulated through the cell. As a result, the power and energy storage of the flow battery can be separated to an extent. A larger volume of electrolyte increases the storage capacity, whereas the number, configuration, and size (electrode area) of the cells largely dictate the available power. Considerable materials and engineering advantages can be gained with RFBs compared with scaling-up conventional battery technologies. It is possible to connect multiple redox flow cells into a stack arrangement through the use of bipolar plates, that is, in a manner similar to that employed in many fuel cells. Electrolyte flow is distributed through the stacks by means of internal manifolds. In addition, scale-up of the system is possible by increasing the size of the electrodes, adding more electrodes to each stack, or by

4

Batteries – Battery Types – Redox-Flow Batteries | Overview

Fig. 2 Divided redox flow stack with four cells and three bipolar electrodes, showing the electrolyte flow through positive and negative electrode compartments. From Watt-Smith, M.J.; Wills, R.G.A.; Walsh, F.C. Secondary Batteries - Flow systems Overview. In Encyclopedia of Electrochemical Power Sources, 1st edn., Garche, J.; Dyer, C.; Moseley, P.; Ogumi, Z.; Rand, D.; Scrosati, B. (eds.), Elsevier, 2009, pp. 438–442.

assembling series or parallel configurations of stacks. A modular stack of four cells connected in electrical series as a bipolar stack is shown schematically in Fig. 2. Electrolytes for the negative and the positive electrode compartments are stored externally and are pumped through the cells. Several features of RFBs have to be taken into account during operation. These include the following.8 Leakage (shunt or bypass) currents. These occur when the conductivity of the electrolyte is high and there is significant ohmic drop through the electrodes and electrical circuit. Current flows through the electrolyte rather than through the bipolar electrodes or intercell electrical circuit so that reaction only takes place at the end electrodes of the stack. Leakage currents can be reduced by increasing the length of the cell inlet and/or outlet manifolds or by reducing the cross-sectional area of the ports. The former approach tends to place greater pressure on the pumps and can increase cost and complicate cell design. Mixing of positive-side and negative-side electrolytes. In systems where separate electrolytes are required for the negative and the positive electrode reactions, a membrane is needed in order to separate the two half-cells and prevent electrolyte mixing. If the membrane or compartment seals fail or unwanted transport of active species through the membrane occurs, numerous problems can result, for example, loss of cell efficiency and parasitic side reactions. If mixing occurs, careful electrolyte management is normally required to rebalance the composition of each electrolyte. In the worst-case scenario, chemical precipitation, dangerous side reactions, or cell failure can take place. Reactant back-mixing. This problem can occur when partially depleted reactants from the cell mix with more concentrated electrolyte in their respective tanks. Accordingly, an imbalance in the concentration of species entering the cell could arise, thereby causing a drop in cell voltage. Flow distribution within the stack. A constant linear flow velocity of electrolyte (0.1–10 cm s−1) should pass along or through the electrodes. Variations in flow through the stack can cause maldistributions of current density and potential (and thereby energy) losses.

3

Examples of redox flow batteries

Various redox systems have been investigated over the past few decades.9 A selection of redox species and their respective redox potentials versus the standard hydrogen electrode (SHE) are presented in Fig. 3. Those systems are divided according to the pH of the electrolyte revealing two clear tendencies; inorganics mostly operated in acidic pH in contrast with organics that do so in neutral-alkaline media. Systems that utilize these redox couples are described in more detail below.

4

Inorganic active materials

Initial studies on RFB were based on the use of metal redox couples as active materials. The high stability of metal atoms at different oxidation states and the good redox kinetics stand as key features of those materials. Metals such as vanadium, chromium or iron were proposed back in 1970s and 1980s together with some non-metallic inorganic compounds, e.g., halides. The interest for those materials have permeated scientific forum, reaching companies and market. While some redox systems are still under study, attention has been shifted to non-toxic, highly abundant materials as iron (see Fig. 3).

Batteries – Battery Types – Redox-Flow Batteries | Overview

5

1.2

Br -/Br3-; 1.04V V(IV)/V(V); 1V BQ; 0.85V Fe(II)/Fe(III); 0.77V

0.8

TEMPO

0.6 Fe(CN)64 - 0.36V to 0.49V

Ferrocene

0.4 0.2

AQDS; 0.19 to 0.22V

0.0 V(II)/V(III); -0.26V Fe(II)/Fe(0); -0.45V Cr(II)/Cr(III); -0.42V

-0.2 -0.4

Viologens

Antraquinone

-0.6 Zn/Zn(II); -0.76V

-0.8

Viologens -0.45V to -0.35V

Potential / V vs. SHE

1.0

Phenazine

Quinoxaline -0.86V to 0.59V

Ce(III)/Ce(IV); 1.44V

Quinoids -0.68V to 0.95V

1.4

Fe TEMPO 0.39V to 0.61 V 0.8V to 1V

1.6

-1.0 Zn/Zn(II); -1.22V

-1.2 -1.4 -1

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

pH Fig. 3 Electrode potentials of selected redox couples used in typical redox flow batteries and their corresponding pH values. Solid black lines represent the thermodynamic potentials for water splitting (oxygen evolution reaction and hydrogen evolution reaction) . SHE, standard hydrogen electrode.

5

Vanadium-vanadium

The vanadium-vanadium redox flow battery (VRB) was largely pioneered by M. Skyllas-Kazacos and coworkers in 1983 at the University of New South Wales, Australia. The technology is now being developed by several organizations across the world, including Invinity, Sumitomo, Dalian Rongke Power, Australian Vanadium, among others, becoming the largest deployment technology among RFBs. A particular feature of the VRB is that it employs the same chemical element in both the anode and the cathode electrolytes. The VRB10 utilizes the four oxidation states of vanadium, and ideally there is one redox couple of vanadium in each half-cell. The V(II)–(III) and V(IV)–(V) couples are used in the negative and positive half-cells, respectively. Typically, the supporting electrolyte is sulfuric acid (2–4 mol dm−3) and the vanadium concentration is in the range of 1–2 mol dm−3. The charge-discharge reactions in the VRB are shown in Eqs. (I)–(III). During operation, the open-circuit voltage is typically 1.4 V at 50% state-of-charge and 1.6 V at 100% state-of-charge. The electrodes used in VRBs are usually carbon felts or other porous, three-dimensional forms of carbon. Batteries of lower power have employed carbon–polymer composite electrodes. A major advantage of the VRB is that the use of the same element in both half-cells helps to avoid problems associated with cross-contamination of the two half-cell electrolytes during long-term usage. The electrolyte has a long lifetime and waste disposal issues are minimized. The VRB also offers high energy efficiency (

Discharge

Br 3− + 2e − Eo ¼ 1:08 V

(IV)

6

Batteries – Battery Types – Redox-Flow Batteries | Overview

Zn2+ + 2e −

Charge

>

Discharge

Zn Eo ¼ −0:76 V

(V)

During discharge, the tribromide ion is reduced to bromide ion at the positive electrode, whereas zinc metal is oxidized to the soluble zinc(II) ion at the negative electrode. Thus, the latter electrode reaction involves the interconversion of solid and solution phases. Owing to their good reversibility and favorable electrode potentials, a number of metal deposition–dissolution reactions are used in flow batteries. In such systems, the energy storage and power of the battery are not fully decoupled as the energy storage capacity will depend on the thickness and morphology of the metallic layer formed. A porous separator is often used between the positive and negative electrodes to avoid the reduction of dissolved bromine during charge. Electrolyte additives (e.g., quaternary ammonium salts)15 can be used to complex any dissolved bromine that has been inadvertently transported through the membrane to the zinc half-cell. There is an increasing interest in the zinc-bromine RFB with a number of running projects in different countries as South Africa, United States and Australia, and a promising forecast for further deployment in the range of gigawatts hour per year.

7

Zinc-cerium

This technology was pioneered by Plurion Systems Inc. (Glenrothes, United Kingdom), a company that developed this system until 2012.While the commercialization activities were stopped, improvement of this battery type is further matter of research. Interestingly, high current densities of up to 100 mA cm−2 and coulombic efficiencies of 70% were claimed. The zinc-cerium system uses methanesulfonic acid as the supporting electrolyte and experiences the same negative electrode reaction as the zinc-bromine system (Eq. V). The reaction at the positive electrode is 2Ce3+ − 2e −

Charge

>

Discharge

2Ce4+ , Eo ¼ 1:44 V

(VI)

The open-circuit voltage of the cell is 2.5 V and falls below 2 V during discharge. Both cerium species remain solvated in the electrolyte during charge and discharge. Despite the benefits of the high cell voltage, a compromise should be accepted between the methanesulfonic acid concentration, the solubility of cerium, the efficiency of the system and the high-cost metal components employed as electrodes.16 Thus, the system has operated with carbon-polymer-based anodes and platinized titanium-mesh cathodes. Use of additives to improve compatibility of the acid cerium electrolyte with zinc, and use of dual membranes has been proposed to improve the system.

8

Zinc-iron

The alkaline zinc—ferrocyanide flow battery was developed by Lockheed Missiles and Space Company in the 1980s, and currently commercialized by ViZn Energy Systems, partnered with the Chinese company, WeView. While Zn—Iron RFB can be operated in a wide range of pH values, best performances have been achieved in alkaline media.17 In basic conditions, the half-cell reactions are as follow: At the positive electrode: 

FeðCN Þ6

4 −

Charge

>

Discharge



FeðCN Þ6

3+

+ e−

Eo ¼ 0:4 −0:5 V

(VII)

At the negative electrode: 

ZnðOHÞ4

2 −

+ 2e −

Charge

>

Discharge

Zn + 4OH − Eo ¼ 1:22 V

(VIII)

The standard cell voltage in alkaline media is 1.58 V, and current densities above 100 mA cm−2 have been achieved. Positive and negative electrolytes are separated by either an anionic membrane or a porous separator, employing carbon fiber or carbon felt as electrode materials. As for other Zn-based RFBs, issues related to the negative half-cell reaction, e.g., Zn dendrite growth, limit the cycle life of zinciron RFB. Development of improved membranes and electrolyte additives have led to prolonged cycle life. In addition, Zn chemistry was extended to the use of air as cathode material18 and demonstrated at megawatt scale by companies as Zinc8 and Sharp. The use of oxygen as cathode material brings the benefits of a metal-air batteries in terms of energy density but also entails a more complex redox process and the triple phase boundary.

Batteries – Battery Types – Redox-Flow Batteries | Overview

9

7

Iron-chromium

The iron-chromium system is one of the earliest types of RFB to be studied and usually operates with an aqueous hydrochloric acid electrolyte that contains the redox-active species. The redox reaction is very simple, which results in low costs, high stability over many cycles, and long life. The National Aeronautics and Space Administration (NASA)-Lewis Research Center conducted much work on this battery during the 1970s and 1980s. As all species are fully soluble, there are no lifetime-limiting factors, such as dendrite formation. The reactions are as follows: At the positive electrode: Fe2+

Charge

>

Discharge

Fe3+ + e − Eo ¼ 0:77 V

(IX)

At the negative electrode: Cr 3+ + e −

Charge

>

Discharge

Cr 2+ Eo ¼ −0:42 V

(X)

The cell operates with either a cationic or an anionic membrane and typically employs carbon fiber or carbon felt as electrode materials. The electrode in the chromium half-cell can be carbon felt with trace amounts of gold or lead. A catalyst is required because the rate of reduction of Cr(III) to Cr(II) is slow.19 In addition, the catalyst must have a high overpotential for hydrogen, as the gas is evolved before chromium is reduced. After initial demonstration at Lion Creek, Ener Vault company offers solutions at megawatt scale. Redox One start-up company continue with the development of the iron-chromium technology.

10

Iron-iron

The iron-iron flow battery was introduced by R. Savinell and L. Hrushka in 1981, and currently commercialized by EES Inc. The main advantage of this system is the abundance and low-cost of the redox-active species: iron.20 The pH value of the electrolytes is maintained below 3 to avoid the precipitation of iron hydroxides. The positive electrode reaction is the same as that for the iron-chromium RFB shown in Eq. (IX). The negative electrode reaction involves iron electroplating and stripping, as shown in Eq. (XI). Fe2+ + 2e −

Charge

>

Discharge

Fe Eo ¼ −0:44 V

(XI)

The standard cell voltage is 1.21 V and the typical operation current density is approximately 50 mA cm−2. Despite cationic membranes can be used, anionic or porous separators are most often employed. The pH value plays a critical role since extreme acidic conditions promotes self-discharge of metallic iron, and values above 3 leads to iron precipitates. Electrolyte additives and moderate operation temperatures have been pointed as solutions used to improve the performances.

11

Bromine-polysulfide

The bromine-polysulfide system21 was developed during the period 1996–2004 under the trade name Regenesys and was publicized as a ‘regenerative fuel cell’, although it is more appropriately described as a redox flow cell. It is the only common RFB system to use anionic inorganic redox-active species for both half-cell reactions. The positive electrode reaction, which involves bromide ion/dissolved bromine conversion, is the same as that for the zinc-bromine RFB shown in Eq. (IV). The negative electrode reaction takes place via the interconversion of sulfide and polysulfide ions and can be simplified to. S4 2 − + 2e −

Charge

>

Discharge

2S2 2 − Eo ¼ 0:47 V

(XII)

The standard cell voltage is approximately 1.55 V. In practice, the open-circuit voltage is approximately 1.5 V; typically, current densities up to 60 mA cm−2 have been used. The two electrolytes are separated by a cationic membrane to prevent reaction between the sulfur and bromine species. Both the bromine and the sulfur ions are present in the electrolytes in the form of their sodium salts. Charge balance during battery operation is achieved by transport of sodium ions through the membrane.

8

Batteries – Battery Types – Redox-Flow Batteries | Overview

12

Soluble lead-acid

The soluble lead flow battery is at an early stage of development.22 It is the only common RFB system to interconvert solid and solution phases at both electrodes. The system operates with an aqueous methanesulfonic acid electrolyte, in which Pb(II) ions are highly soluble. The half-cell reactions are as follows: At the positive electrode: Pb2+ + 2H2 O

Charge

PbO2 + 4H+ + 2e − Eo ¼ 1:46 V

>

Discharge

(XIII)

At the negative electrode: Pb2+ + 2e −

Charge

>

Discharge

Pb Eo ¼ −0:13 V

(XIV)

Cell reaction: 2Pb2+ + 2H2 O

Charge

>

Discharge

Pb + PbO2 + 4H+ V o cell ¼ 1:59 V

(XV)

During charge, layers of lead and lead dioxide are electrodeposited onto the negative and positive electrodes, respectively. During discharge, the lead and lead dioxide redissolve by oxidation and reduction, respectively, to form soluble Pb(II) ion. The operating voltages of this battery are 2.0 and 1.5 V during charge and discharge, respectively, with an open-circuit voltage of 1.7 V.

13

Organic active materials

First reports employing organic active materials in redox flow batteries back in early 2010s were considered as a breakthrough in the area and triggered the gradual growth of examples of aqueous organic redox flow batteries (AORFB).23 The strength of those batteries, besides the intrinsic advantages of aqueous redox flow batteries in terms of safety, relies on the high potential of organic molecules as compounds based on earth abundant elements that could be produced in large scale with a potential low cost.24 Broadening the scope of active materials to carbon-based molecules implies an exponential growth of redox candidates. Indeed, molecular engineering strategies may serve to modulate the redox potential, the solubility and stability of those organic active materials. Quinones, quinoxalines, pyridinium salts, nitroxyl radicals and iron based organometallic complexes are listed as the most relevant chemistries employed in AORFB.25 It should be remarked that all those chemistries match with the simplest case of RFB, which allows to decouple energy and power as the active material remains in solution The bloom of several start-ups and kilowatt scale demonstrators are evidence of initial steps on commercial fruition of those technologies indicating that organic redox species are paving their way to the market. The most relevant cases described with more detail are the quinone-ferrocyanide and viologen-TEMPO systems. It is to say that many variants may fall under this general description with slight changes in the stability, cell voltage and cost of the battery.

14

Anthraquinone-ferrocyanide

The anthraquinone-ferrocyanide system was first reported by M. Aziz and coworkers at University of Harvard based on dihydroxyanthraquinone (DHAQ) (ca. −0.7 V vs. SHE) anolyte and ferrocyanide catholyte operating at alkaline media. Small structural modifications and a change in the operation pH may condition the stability of the system and slightly modify the cell voltage (1.0–1.2 V at open circuit). Thus, substitution of hydroxyl moieties by carboxylate (DBEAQ), phosphonate (DPPEAQ) or pivalate (DPivOHAQ) moieties as hydrophilic moieties have served to boost capacity retention up to 99.9982%/day. As referred to ferrocyanide, the cycling stability of the material is confirmed, yet it presents a low redox potential and low solubility. Linked to the operation pH, a value of 80% round trip efficiency can be reached for (DPivOHAQ)/ferrocyanide system working at pH 14 at 100 mA cm−2. Electrolyte is therefore composed by anthraquinone and ferrocyanide active materials for anolyte and catholyte respectively, and KCl or KOH as supporting electrolytes. A cation exchange membrane is employed as separator between half-cells. At the positive electrode the reaction of ferrocyanide takes places as described in Eq. (VII). The redox reactions of quinone proceeds as follows: At the negative electrode:

Batteries – Battery Types – Redox-Flow Batteries | Overview

9

ðXVIÞ

, E o =-0.4 to -0.7 V During charge the carbonyl groups are reduced to the corresponding aromatic alcohols and depending on the pH of the solution either protons or other cations can be involved in the process. Therefore, the reduction-oxidation of all quinone derivatives is a pH dependent two electron process. In contraposition, at the positive electrode a one electron oxidation of ferrocyanide to ferricyanide takes place. Ferrocyanide is low-cost organometallic complex widely employed in AORFB due to its high stability and low cost, yet its limited solubility conflicts with high energy density targets. Effect of cation has been studied to overcome this limitation. Thus, a relatively low demonstrated energy density (6–8 Wh L−1) as compared to anthraquinone water solubility, and the lack of a more competitive catholyte solution are pointed as current limits for this technology. Efforts have been devoted to identifying and understanding capacity fade mechanisms. In the case of anthraquinones, the exposure of anolyte to air has been identified as a positive strategy to rebalance the electrolyte. Similar to other systems employing different active materials as catholyte and anolyte, maintaining the state of charge balance between both sides is critical for elongating battery lifetime. Following this concept of recovery of active material, US startup Quino Energy, a spinoff of Harvard University, develops AORFB based on DHAQ. Quinone-ferrocyanide chemistry has been already demonstrated at kilowatt scale by Kemiwatt (France). Alternatively, CMBlu has developed flow batteries based on lignin, a bio-based polymer that may work under a similar working principle to quinone based batteries, based on aromatic alcohols and conjugated ketones. Prototypes based on this technology have been developed by CMBlu.

15

Viologen-TEMPO

The viologen-TEMPO system is preferentially operated at neutral pH according to the chemical stability of active materials. Redox reactions of quaternary bipyridines and TEMPO derivatives involve radical species and are not governed by the pH. Therefore, this system is of particular interest for neutral pH batteries. Following initial studies with methyl viologen (MV), a commercial quaternary bipyridinium salt and TEMPOL, continuous development has been focused on tuning of active materials to improve their stability and increase the cell voltage (1.2–1.5 V at open circuit). Thus, including additional hydrophilic groups and bulky substituents, such as trimethylammonium group, serves to stabilize the charged species preventing precipitation or dimerization phenomena and leading to systems with negligible capacity decay over cycling. 1,10-bis[3-(trimethylammonio)propyl]-4,40 -bipyridinium tetrachloride (TMAP-Vi) and 4-trimethylammoniumTEMPO chloride (TMA-TEMPO) are examples of stable viologen and TEMPO derivatives. The redox reactions represented for TEMPOL and MV as archetypical examples of this chemistry are as follows: At the positive electrode:

ðXVIIÞ

, E o =0.8 to 1.0 V At the negative electrode: ðXVIIIÞ , E o =≈ -0.4 V

During charge the pyridinium ring is reduced to the corresponding radical in a one electron process. A second one electron reduction process can be unleashed to increase energy density of the system. However, this is only reversible for certain viologen structures. At the positive electrode a one electron oxidation of nitroxyl radical to the oxoammonium cation takes place. The low concentration of water dissociation products at neutral pH severely affects the ionic conductivity of the electrolyte, therefore salts as NaCl, KCl or NH4Cl are included in the electrolyte formulation. Employing anion exchange membranes practical round trip efficiencies of 70–80% are generally reached at current densities of 40–60 mA cm−2 when anion exchange membranes are employed. German company, Jena Batteries, renamed as CERQ, has demonstrated the Viologen-TEMPO chemistry at kilowatt scale and offers solution for megawatt scale.

10

16

Batteries – Battery Types – Redox-Flow Batteries | Overview

Emerging RFB technologies

Membrane-less RFBs. The need of ion-selective membranes in RFBs increases the battery cost since this element is one of the major contributors to the cost of the stack. Thus, development of systems that do not require the use of any separator is very appealing. However, continuous mixing of catholyte and anolyte become a major issue when the separator is removed. There are two primary strategies to overcome it. On the one hand, deployment of a cell design that allows laminar flow of electrolytes avoids electrolyte mixing. Encouraging results in this direction have been obtained for membraneless H2–Br2 RFB and Vanadium RFB. On the other hand, the use of two immiscible liquids as catholyte and anolyte, respectively, enables spontaneous separation of the active species according to their partition coefficients. Several examples including aqueous biphasic systems have been demonstrated at lab scale. RFBs based on large redox-active species. Instead of expensive ion-selective membranes, porous separators can be used to decrease costs. However, charged species easily cross over to the opposite side leading to self-discharge and reducing energy efficiency. The use of large redox-active species allows confinement of active species by size exclusion using cheap porous separators. Soluble redox-active polymers as well as solid redox-active particles forming a slurry have been explored at lab scale achieving promising results. This approach is based on a conventional cell architecture, in which ion-selective membrane is simply replaced by a porous separator. Redox-mediated RFBs. An emerging approach to tackle the limitations related to the deployment of highly concentrated electrolyte is the redox-mediated flow battery. Solid electroactive materials are confined in the external reservoirs. Spontaneous and reversible charge transfer reaction takes place between redox electrolyte and solid electroactive material, so that dissolved species become redox mediators carrying charges between reactor and reservoir. Thus, energy density of the system is determined by the energy density of the solid electroactive materials, which is very often much larger than that of liquid solution of electroactive materials. Redox-mediated RFBs use conventional cell architecture, while the main differences are found in the external reservoirs in which solid materials are confined. Non-aqueous Flow Batteries. Cell voltage for aqueous flow batteries is limited by the electrochemical stability window of water, which thermodynamically is 1.23 V at room temperature. As for other non-flow battery technologies, the use of non-aqueous solvents in RFBs enables widening the cell voltage above 4 V. A larger cell voltage is pursued not only to increase energy density but also to decrease the cost of the stack. Several examples of non-aqueous flow batteries delivering stable cell voltage of 2.5–3 V have been achieved at lab scale.

17

Concluding remarks

Each redox system possesses both advantages and limitations. These factors are highlighted in Table 1 along with an indication of the current scale of each system. A number of RFB technologies are in the process of being scaled-up and commercialized. The allvanadium and the zinc-bromine technologies have been deployed across the world producing megawatts of electrical power. Other Table 1

General comparison of redox flow battery (RFB) systems.

Redox system

Features

Challenges

Maximum output reported

Vanadiumvanadium

Same element in both electrolytes. High storage efficiency (90%)

Zinc-bromine

100 MW (400 MWh) (Dalian City, Dalian Rongke Power; Dalian Constant Current Energy Storage Power Station Co., Ltd) 2 MWh (192 batteries composed by 12160 kWh modules; California, RedFlow) 250 kW (4 h) (Turlock, California, US, EnerVault)

Iron-iron

Relatively high cell voltage can be achieved. Abundant materials Relatively low-cost chemicals. Low storage costs Low-cost, abundant chemicals

Moderately high costs of efficient membrane. Electrolyte preparation can be difficult Hazardous chemical (bromine and sulfides). High storage costs Low energy density. High Cr electrolyte costs High pH dependence

Brominepolysulfide

Abundant and reasonably low-cost chemicals. High energy efficiencies (70–75%)

Soluble leadacid

Reasonable voltage efficiency (70–80%). High energy efficiency ( N212. The declining rate of discharge capacity exhibits the exact opposite pattern. (ii) Thinner N212 is better suited for ICRFB cycling due to its lower internal resistance, which results in increased EE and electrolyte utilization (shown in Fig. 4a).10 The comprehensive efficiency, or EE, is the most significant metric in the field of energy storage. N212 demonstrates the highest cost performance in ICRFB when considering its lower cost and higher EE. According to the foregoing conclusion, the battery performance of TIHFB with a N212 was reported in Fig. 4b.9 Despite the charging potential being close to 1.26 V for the whole battery cycling, no gassing behavior was detected. The CE rises from 96.4% to 98.6% from 40 to 200 mA cm−2, while the VE drops linearly. The VE is as high as 79.5% even at 200 mA cm−2. The high conductivity of electrolytes based on chloride acids and the fast redox couple kinetics are responsible for the high VE. Additionally, as current density increases, the EE drops in a way comparable to the VE, indicating that the VE plays a significant influence in determining the EE. The EE of TIHFB is higher than the conventional zinc HFB system.18,28 CE achieves 99% throughout the 700 cycles at 200 mA cm−2 between the potential windows of 0.4 V and 1.26 V, demonstrating the excellent reversibility of the electrochemical process. Achieving an average VE and EE of 79% and 78% demonstrates that the single-cell can function steadily over 2 weeks. Despite the penetration of electroactive species causes the discharge capacity to decline with the cycle number, the rate of capacity degradation is just 0.96 per cycle. This cycle performance of the system is significantly superior than that of other HFBs.18,44 In IBA-RFB, it can be essentially determined that the N212 is the one among Nafion series that is most suited for use in IBA-RFBs. Although Nafion has shown outstanding electrochemical performance and chemical stability, its continued use in IBA-RFB is still constrained by their expensive cost (600 USD m−2). Hence, novel membranes must yet be developed in order to increase the commercial growth of IBA-RFB. The low-cost hydrocarbon membrane can be used in place of Nafion. For ICRFB applications, the non-perfluorinated proton exchange membrane (PEM) known as sulfonated poly(ether ether ketone) or SPEEK, has been employed.18 The ICRFB built using the SPEEK membrane obtain superior battery performance as compared to Nafion, and the capital cost of the whole system is further decreased. Nevertheless, it is still considered that the non-perfluorinated PEM will degrade easily in the very acidic and severely oxidative electrolyte environment, leading to short cycle life over the prolonged operation. As shown by Li et al., the protonation of ethereal oxygen atoms in a strong acid media will speed the degradation of SPEEK membrane.45 The ICRFB’s optimal working temperature is 60  C, and it may further worsen and accelerate the degrading process. Further research and observation are needed to ensure the long-term reliability of SPEEK during ICRFB operation. In ZIRFB, Ding et al. reported a similar investigation based on SPEEK.40 The SPEEK for neutral ZIRFB was specifically made in the + K form (SPEEK-K), with the produced SPEEK membrane submerged in a solution of 1 M KOH and 2 M KCl at 80  C. Since the conductive cation ion in ZIRFB is K+, the transfer of pure SPEEK from its H+ form to K+ aims to boost membrane conductivity, which can be verified by infrared spectroscopy and energy dispersive spectrometer. The permeability test demonstrates that there is practically no species crossover over the SPEEK-K membrane, demonstrating it can successfully repel the ion crossover. This is in line with the self-discharge time of ZIRFB with SPEEK-K is 6 h longer than that with Nafion 117. The single-cell constructed using SPEEK-K displays high and steady CE (>95%) and EE (>79%) at 40 mA cm−2. The cost-effective membrane is essential to encourage ZIRFB commercialization since the IEM accounts for the majority of the ZIRFB cost. It should be noted that N117 is approximately 13 times more expensive than SPEEK-based membrane, and it is extremely competitive to use the SPEEK-based membrane in ZIRFB. Since ZIRFB does not need operation at medium to high temperatures and the electrolyte does not display strong acidity, this approach may even be more feasible when considering the potential of the SPEEK membrane degrading. There have been several investigations on using porous membrane/separator in IVRFB as less oxidative Fe2+/Fe3+ and V2+/V3+ redox couples were used. Daramic® holds a dominant position in the lead-acid battery industry because of its low cost ($1–20 m−2), low electrical resistance, superior microporous structure, and designable porosity.18 Wei et al. have reported on the performance of Daramic® as the separator in IVRFB.46 The Daramic® has a porosity of 57% and a median pore diameter (0.15 mm). Consistent efficiency were seen throughout the cycling, which demonstrates exceptional stability. Due to the distinct ion migration process, IVRFB with Daramic® has somewhat lower efficiency than Nafion.18 The linked micropores can offer much wider pathways for charge carriers, and the higher permeability of active species causes lower CE. The increased ohmic resistance of Daramic® results in lower VE since it is significantly thicker than Nafion and does not own hopping mechanism for protons. Daramic® is a viable

34

Batteries – Battery Types – Redox-Flow Batteries | Iron Systems

replacement for expensive Nafion with a fair EE sacrifice (20,000 h) with only a 10% loss in power while meeting other performance and cost-based key performance targets.9

Fig. 2 Polarization curve of a typical PEMFC and a breakdown of losses that contribute to reduction from theoretical cell voltage.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Overview: PEMFC

61

Table 1 EU harmonized automotive reference operating conditions for low-temperature PEMFC single cell testing, assuming air supplied to cathode.8

Cell Anode

Cathode

4 4.1

Parameters

Units

Values

Cell operating temperature Fuel gas inlet temperature Fuel gas inlet humidity Fuel gas inlet pressure Fuel stoichiometry Air inlet temperature Air inlet humidity Air inlet pressure Air stoichiometry



80 85 50 150 1.3 85 30 130 1.5

C C RH % kPag –  C RH % kPag – 

Structure/design System and stack

An outline of the components of a typical PEMFC system for automotive use is shown in Fig. 3. At a high level, the majority of the components are described as a balance of plant and are present to supply gasses at the right conditions to the fuel cell stack, control the temperature and convert the produced electricity into a useful voltage. Components must be chosen to avoid introducing any impurities into the hydrogen. When used in vehicles, fuel cells are often hybridized with batteries of various sizes to smooth the load on the fuel cell and provide power for starting up etc. At the core of a PEMFC system is the stack. As discussed, each fuel cell produces only 0.65 V at maximum load, though a 300 cm2 cell may comfortably produce >500 A at maximum power. It is therefore typical to stack cells together to produce a more useful voltage and power. Hundreds of cells is common for automotive applications. In a stack each cell is connected in series electrically and manifolded together so they are connected in parallel for supply of hydrogen, air, and coolant.

Mixer

Exhaust Injector B (Bypass) Compressed Hydrogen Tank

H2 line Air line LTL Coolant Line HTL Coolant Line

PRD Injector A Hydrogen from Tank H2 Purge Valve

Ejector H2 Recirculation

Exhaust Air to Tail Pipe

Check Valve

Anode Exhaust

Air Filter Over-Pressure Cut-Off Valve

–

Stack Shut-Off Valve

Stack Bypass Valve Voltage Sensor

Differential Pressure Sensor

Air Mass Flow Sensor

Membrane Humidifier

Current Sensor

Cathode Exhaust To High Temperature Coolant Loop

Air Pre-cooler Air Loop Temperature Sensor

Orifice (air bleed for motor cooling)

Motor

Expander

+

FC Stack

Compressor

Demister

Reactant Air

To Low Temperature Coolant Loop Demister

Fig. 3 Schematic of a fuel cell system for a light duty truck adapted from Mass Production Cost Estimation of Direct H2 PEM Fuel Cell Systems for Transportation Applications: 2018 Update.10 PRD: pressure relief device.

62 4.2

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Overview: PEMFC Cell

Each single cell consists of a PEM, sandwiched by catalyst layers, gas diffusion layers with microporous layers (GDLs with MPLs), and current collectors/bipolar plates. Each component is optimized to balance the cell performance, lifetime and cost. The structure of each cell is shown schematically in Fig. 4 and the details of each component are outlined below. The combination of membrane, catalyst layers and GDLs is often referred to as the membrane electrode assembly (MEA).

4.3

Membrane

The membrane serves two principal functions: it prevents mixing of hydrogen and air and allows for the flow of protons while blocking the flow of electrons. Ideally the membrane is as thin and as ionically conductive as possible, but this needs to be balanced against the rate of gas crossover and the mechanical strength. Contemporary membranes are usually 10–20 mm in thickness with thinner membranes also demonstrated. The membrane material must therefore have high proton conductivity, low gas permeability, and excellent mechanical and chemical stability in the operating environment of the fuel cell. The archetypal PEM material is perfluorosulfonic acid (PFSA), a polymer with a perfluorinated backbone with pendant sulfonic acid groups. These materials readily absorb water and form continuous interconnected channels which enable the free protons to transport charge, while providing suitable mechanical and crossover properties, as seen in Fig. 5. Specific examples include Nafion™ and Aquivion™. In modern membranes it is common to have a supporting reinforcement in addition to the conducting phase, as these increase mechanical performance and reduce membrane swelling, increasing lifetime, and enabling thinner membranes. Other additives, such as radical scavengers or recombination catalysts may also be added to improve performance and prolong lifetime. Polymer chemistries other than PFSA, particularly those based on non-fluorinated backbones,11 have also been developed but generally follow the scheme of a polymeric backbone with sulfonate groups. Currently PFSAs dominate deployments for PEMFCs but possible future legislation to limit the use of perfluorinated materials is driving innovation into hydrocarbon materials.

4.4

Catalyst and catalyst layer

The electrochemical reactions inside a PEMFC can only occur at the three-phase interface between (i) electrically connected electrode (ii) proton conductor and (iii) reactant gas. The catalyst and catalyst layers in a PEMFC are designed to provide a high three-phase area and contiguous proton, electron and reactant conducting pathways. A transmission electron microscopy (TEM) image of platinum nanoparticles supported on a high surface area carbon (a typical fuel cell catalyst for anode and cathode) and a schematic of the catalyst layer are shown in Fig. 6. To minimize activation losses and therefore maximize cell performance, the hydrogen oxidation reaction and the oxygen reduction reaction need to occur on an electrode material that is an efficient catalyst for the reaction. Platinum is the most

Fig. 4 Schematic showing different components of a PEMFC single cell.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Overview: PEMFC

63

Fig. 5 Schematic of a typical PFSA membrane showing polymer segmenting into hydrophobic regions of backbone and hydrophilic regions of solvated head groups and liquid water.

Fig. 6 (A) TEM image of Pt/C catalyst powder showing platinum nanoparticles deposited on carbon support and (B) schematic of a catalyst layer showing catalyst particles, ionomer and channels for gas flow.

64

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Overview: PEMFC

commonly used catalyst for the anode’s hydrogen oxidation reaction as it has high stability and activity. It is sometimes alloyed with ruthenium or other metals to improve tolerance to impurities in the hydrogen that may otherwise poison the catalyst. To maximize the surface area, the platinum is used in a nanoparticulate form where spherical particles with a mean diameter of 2–5 nm are the most common. The diameter is chosen to give the largest surface area per gram of expensive catalyst material while avoiding smaller particles that are more unstable and rapidly Ostwald-ripen. To stop nanoparticle agglomeration and aid water, gas, and proton transport, the nanoparticles are supported on a catalyst support. This is most commonly a carbon black or other large surface area highly-graphitic carbon. Though there is a high degree of variability, these are usually solid carbon spherical particles with diameters of 20 nm (primary particles) agglomerated into larger structures of 200 nm diameter (secondary particles) which are then packed together further (see Fig. 6). The geometry of the carbon black particles provides a porous structure that supports gas and water transport. To provide efficient proton conduction through the catalyst layer, a proton conducting material is added. Herein, and commonly in fuel cell literature, this is referred to as ionomer (though this is also a general term for ionically conducting polymers). The ionomer is usually of the same composition as the membrane material and thinly coats the catalyst and catalyst support. Catalyst layers in PEMFCs need to be thin for the resistance to proton, water, gas and electrical flows to be small, keeping ohmic and mass transport losses to a minimum. Typical thicknesses for an anode are on the order of 1–10 mm with catalyst loadings of 0.05 mgPt cm−2. Sometimes additives such as radical scavengers or catalysts are incorporated into the catalyst layer to protect against corrosion in the event of the anode potential increasing during transient events. The structure of the cathode is the same as that of the anode, but the oxygen reduction reaction is much slower than the hydrogen oxidation reaction and therefore the cathode causes most of the activation losses. The high potentials and acidic environment make the cathode environment highly oxidizing which limits the stability of most materials in the catalyst layer, so platinum is still the most commonly-used catalyst for the oxygen reduction reaction. Typical cathode catalyst layer thicknesses are 10 mm with catalyst loadings of 0.2–0.4 mgPt cm−2 where loadings are chosen for each application in a cost vs. performance and durability trade-off. Owing to the cost, large activation losses and relatively high catalyst loadings required, significant work has been performed over the last 30+ years to identify and produce lower cost and more active catalysts for the oxygen reduction reaction. Two classes of catalyst that are used are alloys of platinum with transition metals such as Pt3Ni, Pt3Co, PtPb, Pt71Rh3Ni26,12 which can show high performance and have reduced cost, and doped carbon materials such as Fe-N-C.13

4.5

Gas diffusion layers and microporous layers

The GDL and MPL are layers that are introduced to homogenize properties across the cell. They link the catalyst layer, where the porosity is on the order of nm, with the flow fields which distribute gas and electrons across the cell on the order of cm. These layers need to (i) electrically and thermally connect the catalyst layer and bipolar plates (ii) provide pathways for the efficient transport of reactants and products and (iii) provide mechanical support to the catalyst layer. GDLs are normally a carbon paper or cloth with high electrical conductivity and high porosity. It is critical that GDLs allow removal of water away from the catalyst layer to avoid flooding, while simultaneously allowing reactant gas to reach the electrode. To facilitate this, GDLs are coated with a hydrophobic material, usually polytetrafluoroethylene (PTFE) or similar, to prevent water accumulating in all of the pores in the material, preventing efficient air transport. The distribution in the diameter of the channels also facilitates a two-phase flow as capillary forces prevent condensation in some areas and rapid water flow in others. GDLs are usually also coated with a microporous layer which helps to maintain the right level of humidity inside the catalyst layer while facilitating water transport, minimizing the contact resistance between the GDL and catalyst layer, further homogenizing flow and mechanical support, and preventing the loss of catalyst to the GDL. A typical MPL material is carbon black with a PTFE binder. Carbon is the material of choice for GDLs and MPLs as it offers excellent chemical and mechanical stability and electrical conductivity while PTFE is used due to its excellent stability and high hydrophobicity.

4.6

Bipolar plates, flow fields and gaskets

Bipolar plates act to electrically connect the cells in the stack together while separating the anode and cathode of adjacent cells.14 They also provide mechanical stability to the stack. Key considerations for bipolar plates are their contact resistance, corrosion resistance, cost, weight and processability. Bipolar plates are typically made up of corrosion resistant and lightweight materials. Graphite and graphite polymer composites have been frequently used. Titanium and stainless steel are also popular choices as they can be formed into shape using stamping equipment that is already widely used in mass production. However, these materials require a coating of materials like gold, platinum or carbon to maintain a low contact resistance and in some cases to ensure corrosion resistance. To enable the controlled distribution of gasses into and out of the cell, bipolar plates are designed to have ‘flow fields’ (Fig. 7). These consist of channels allowing gas flow and lands making electrical contact with the GDL. The design of the flow field is critical in providing a uniform distribution of reactants over the entire cell area. In stacks, it is also common for the bipolar plates to be shaped to allow a cooling medium to pass between cells.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Overview: PEMFC

65

Fig. 7 (A) Example of a triple serpentine flow field. (B) A cross section of a cell stack showing the arrangement of flow field land and channels in a typical cell stack configuration.

4.7

Gaskets and end plates

The MEA may incorporate sub-gasket materials around the outside of the active area to provide a surface for gaskets to seal against, reduce the amount of membrane required and allow for easier handing during stack assembly. Gaskets serve to seal the MEA between the bipolar plates. Gasket materials are chosen to provide gas tight seals under operating temperatures and pressure and have a thickness and compressibility such that once the stack is clamped together, enough pressure is applied to each cell to maintain good electrical contact but not so much that compressible materials, such as the GDL, become overly compressed. Another key feature of the materials chosen is that they do not introduce any impurities to the fuel cell which would poison the components. During operation, the thickness of the membrane and other components can change as water is absorbed and as materials expand at higher temperatures. In stacks it is necessary to have a system for applying an appropriate amount of pressure to the cell and accommodating this flex; typically fixed bolts or pneumatic systems are used to apply pressure to single cells used in laboratory testing while spring systems are used to provide pressure to stacks. These usually act on thick, rigid end plates that distribute pressure uniformly across the cell. End plates also incorporate a means to connect gasses and coolant to the stack and may also incorporate sensors, valves, and components for recirculating gas and removing liquid water.

5 5.1

Advantages, disadvantages and key performance indicators Advantages and disadvantages of hydrogen

The advantages and disadvantages of PEMFCs are closely related to those of the hydrogen fuel that they use. The pros and cons of the hydrogen economy are complex and vary with location so are not discussed in detail with only a cursory outline here. Note also that many of the advantages and disadvantages of hydrogen are also shared with other devices operating on it, including various types of fuel cell, combustion engines and boilers. When compared to other fuels, hydrogen’s principal benefit is that it emits nothing but water at the point of use. Emissions generated during the production and distribution of hydrogen are non-zero, but legislation is in place to limit the carbon intensity permitted during the production of ‘clean hydrogen’. This varies by jurisdiction, but in the United Kingdom it is set as 20 gCO2e100 MJLHV−1,15 which is very favorable when compared to the carbon intensity of other fuels.

66

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Overview: PEMFC

When compared to batteries the benefit of hydrogen is that power and energy are decoupled. Hydrogen can be stored and transported comparatively cheaply and quickly, usually as a compressed gas, or liquid. This is particularly true at a ‘grid-scale’ where existing gas infrastructure can store vast amounts of compressed gas between seasons. In vehicles, adding extra range is simply adding extra hydrogen storage and refueling times are very short (50%. This allows PEMFCs to generate a large amount of energy per unit volume. This is particularly important in vehicles where packing into a chassis is often important. They can start-up rapidly, including from sub-zero conditions and have a very dynamic power output from 0 to 100% power in 1 s. PEMFCs also operate without any moving parts or vibrations from the stack itself, decreasing noise and increasing reliability. Unlike some other types of fuel cell, PEMFCs can also be manufactured at scale by adapting well developed mass-production techniques such as screen printing and stamping. Compared to batteries, PEMFCs are safe electrically and chemically once the fuel supply is removed; this is expected to make the recycling of PEMFCs easier than for lithium-ion batteries, though recycling of hydrogen gas cylinders remains a challenge. The two principal advantages that PEMFCs have over hydrogen combustion engines is that they are more energy efficient and produce only water and heat as by-products. Hydrogen combustion engines show fuel efficiencies of 14 g km−1 for IEC vs 10 g km−1 for PEMFC and may emit 20% more NOx than equivalent combustion engines running on gasoline.17

5.3

Disadvantages of PEMFCs

PEMFCs rely on the use of critical raw materials, particularly platinum group metals. While the amount of critical raw materials used has been reduced significantly, they still contribute to the overall cost and result in more brittle supply chains. Many of the components in a PEMFC are susceptible to poisoning, necessitating the use of pure hydrogen with a very low level of some of the impurities. This increases the cost of producing hydrogen and limits the materials and equipment that can be used during its production, transportation, and storage. For some applications that demand very high powers, namely >MW, like intercontinental passenger aircraft and large ships, PEMFCs and the associated hydrogen storage currently have too low a volumetric and gravimetric power density and scale to be suitable replacements for combustion engines.

5.4

Key performance indicators and future development

PEMFC performance, lifetime and cost are sufficient to meet the technical requirements for use in many stationary and transportation applications, yet there is still room for improvement in all areas to support more competitive deployments. Heavy-duty applications such as planes have very demanding requirements that have not yet been fully met. The economic limitation to fuel cell deployment has historically been the capital cost of the fuel cell system. With technological improvements, total cost of ownership is now driven significantly by the hydrogen price with the expectation that hydrogen at a price of €4 kg−1 at the pump provides cost parity with diesel for HGV.18 Technical targets for improvements to the technology are embodied in target key performance indicators - some examples are provided in Table 2. It is important to note that these targets must be met collectively; it is comparatively easy to make a fuel cell last a long time if cost is no issue.

Table 2

European light duty vehicle key performance indicators.19

Property

Unit

State of the art 2017

Target 2030

Durability of fuel cell system until 10% power degradation Hydrogen consumption Availability Maintenance Fuel cell system cost on a 100,000 unit per year basis Areal power density @ 0.66 V Platinum group metal loading Cell volumetric power, excluding endplates

h g km−1 % EUR km−1 EUR kW−1 W cm−2 g kW−1 kW l−1

4000 12 98 0.04 100 1.0 0.4 5.0

7000 10 >99 0.01 40 2.0 0.05 10.0

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Overview: PEMFC

67

Key areas for future research and development include developing new materials and structures for membrane electrode assemblies including thinner, more conductive membranes, more efficient catalyst utilization, and cheaper, more lightweight and durable bipolar plates. Aside from materials challenges, production at increasing scale means that there is an increasing need for activities such as in-line quality control and the development of new manufacturing techniques.

6

Applications

PEMFCs can generally be used in any application where there is a need for electricity, and a supply of air and hydrogen is available. However, there are several applications where hydrogen-fueled PEMFCs have clear benefits over low emission technologies such as lithium-ion batteries and hydrogen combustion engines. Despite being relevant across a wide range of power applications, PEMFCs are not a ‘silver bullet’ that will fully decarbonize every application, but they are likely to be essential in areas which are challenging to electrify by other means. For example, the transport sector has been the fastest growing source of carbon emissions worldwide,20 and contributed to 37% of carbon dioxide emissions from end-use sectors in 2021.21 Within the transport sector however, there are a comparatively smaller number of HGVs and larger vehicles such as trains, ships and planes that both use more energy and are more intensively used, and so have an outsize impact on emissions.

6.1

Transportation

6.1.1 Ground PEMFCs can be used to power vehicles, ranging from small electric cars to trucks. In the case of medium and heavy-duty passenger vehicles, fuel cell electric vehicles (FCEVs) have already been deployed and have comparable performance to battery electric vehicles (BEV) with ranges of approximately 800 km and typical refueling times of around five minutes. Many automotive OEMs have prototype PEMFC vehicles; Toyota and Hyundai both have production models for sale in various regions. In the near to medium term, BEVs are likely to dominate the passenger car market of the UK and Europe but there remains a large amount of uncertainty about the relative role of both technologies as they mature. A key limitation on this segment is the need to reliably supply hydrogen to retail customers across a wide area. The UK has approximately 10,000 petrol stations across the country but there is limited market appetite to invest in adding hydrogen refueling stations as few people have vehicles, while vehicle owners won’t switch until there is sufficient supply. Commercial vehicles such as vans, buses, coaches, and heavy goods vehicles (HGV) tend to operate over longer distances than light duty vehicles. PEMFCs are therefore expected by some to outperform batteries in the medium-term as FCEVs benefit more from the fast-refueling time and higher gravimetric power density of hydrogen with compressed gaseous hydrogen storage. For commercial vehicles which regularly return to a depot, the deployment of refueling infrastructure is less demanding. Industrial vehicles such as forklifts, warehouse robots, mining and agricultural vehicles are more niche applications where PEMFCs are likely to excel. PEMFC-powered forklifts have been commercialized with fast refueling times compared to batteries being a key differentiator. Prototype mining and agricultural vehicles have also been trialed, such as hydrogen-powered tractors, but costs compared to diesel and a lack of infrastructure are barriers to large scale commercialization. Somewhat aptly, notable success stories of PEMFC adoption in industry have been for mining vehicles used in platinum mines. For example, Anglo-American converted a 200-t truck from a diesel-electric powertrain to a 2 MW hydrogen fuel cell capable of hauling 300 t.22 Many countries have only partially-electrified rail systems, with many routes operating using diesel engines. Electrification infrastructure is often prohibitively expensive, disruptive to retrofit and potentially incompatible with the dimensions of existing tunnels and underpasses. Correspondingly, PEMFCs have been deployed for traction in intercity trains with other rail uses being investigated.

6.1.2 Maritime and aeronautical Hydrogen is seen as a key fuel for zero-emission passenger and freight aircraft and there are a number of ways for it to be used. PEMFCs are most likely to be deployed on smaller, regional aircraft where electrically driven propellers are a suitable propulsion system and power requirements are in the 0.1 to 1 MW range. For larger, heavier and faster aeroplanes commonly used in intercontinental travel and where current kerosene powered engines can produce over 50 MW each, hydrogen is more likely to be fed directly to a combustion engine such as a single or twin-aisled aircraft, or indeed a scram jet for propulsion, though PEMFCs may still be used in a hybrid configuration. There are a number of companies developing PEMFC-powered aircraft. PEMFCs have also been deployed for the population of un-manned aerial vehicles where the high gravimetric power density of hydrogen is an advantage. The maritime sector similarly has vessels such as passenger ferries, near-shore, and coastal vessels which operate in the few MW power range. Hydrogen-powered PEMFCs are promising in this area where marine certified systems and prototype vessels featuring PEMFC propulsion have been deployed. At much larger scales, e.g. freight ships, it is likely that systems that can operate directly on methanol, ammonia or other hydrogen carriers would be prioritized. This is also partly a limitation introduced by electric motors. Historically PEMFCs have also been used on military submarines.

68 6.2

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Overview: PEMFC Stationary electricity generation

PEMFCs have been used for stationary power generation, where they can provide electrical energy to residential, commercial, and industrial buildings. PEMFCs are particularly attractive for use in micro-grids, where access to traditional grid electricity is limited and there is a need for small scale seasonal storage that would be prohibitively expensive with batteries. Similarly, they have also been deployed as backup sources of power for critical facilities, such as hospitals and data centers where fast response time, stability when dormant, and safety when not charged with hydrogen are important features. PEMFCs have also been used for a range of portable power generation applications, such as replacements for diesel generators on building sites, to recharge electronic devices and similar. Because of the vast energy storage requirements for matching renewable generation and electricity consumption at the grid scale it has been suggested that fuel cells could be integrated at a large scale into national power grids to provide additional capacity during peak demand periods. While technically feasible, the economic case for this is unclear and there are as yet no plans for this at scale.

6.3

Combined heat and power

PEMFCs have also been deployed as combined heat and power (CHP) systems where as well as providing electrical energy, the heat produced is used for space and water heating. Japan’s ENE-FARM project deployed over 400,000 residential CHP units, the majority of which were PEMFCs capable of supplying 60% of the typical electrical power demand of a normal household. PEMFCs in residential CHP applications have a power around 0.5–10 kWel with a combined efficiency of >90% and a lifetime comparable to normal boilers; 80,000 h, 40,000 on-off cycles and a 10-year maintenance period.

7 7.1

Current and future market status Driving forces

For most use cases where PEMFCs are being considered, the technology works. It is possible, in 2023, to buy a fuel cell car; it refuels in 5 min and drives for >500 km emitting nothing but water from the exhaust. There are hundreds of fuel cell buses running in Europe and small fleets of HGVs, trains, with prototype ships and planes undergoing trials. While technological improvements for reduction in cost and improvements in performance and lifetime are still desirable, PEMFC systems have been successfully trialed in a range of prototype vehicles and stationary power generation applications. Mass production has been demonstrated at the 10,000’s units per year scale and uses well-known manufacturing techniques such as coating, stamping, and welding. The primary driving forces that determine the contemporary market uptake are therefore economic and policy related. For most mass-market use cases PEMFCs are not economically competitive with hydrocarbon fueled combustion engines on a total cost of ownership basis where aggressive carbon-taxes or subsidies are not applied. It is therefore government policies, notably net-zero targets, which require decarbonization of difficult to abate applications, that are the main driver of PEMFC adoption. PEMFCs are also not a technology that can easily be deployed in isolation, rather they are an enabling technology for electrification with hydrogen. The largest barrier to the uptake of PEMFCs is their requirement to operate on hydrogen. The total cost of use of PEMFCs is dominated by the hydrogen price and there is, as yet, fairly limited hydrogen infrastructure for the production and distribution of clean hydrogen such as hydrogen refueling stations or nationwide hydrogen gas grids, though this has been identified and a range of policies are being put in place to address this. The final driver of market penetration is likely to be the ease of access and the price of low carbon hydrogen along with the total cost of ownership. Deployment of PEMFCs may therefore depend on the jurisdiction, with different countries having different hydrogen strategies. For instance, Europe has a strategy to deploy 1780 refueling stations by 2027 supplying 1 million tonnes of hydrogen annually. Projections are that the total cost of ownership of a PEMFC truck is equivalent to a diesel engine with hydrogen prices of €4 kg−1, though such comparisons always rely on several underlying assumptions. The greater system inefficiency for producing hydrogen results in higher well-to-wheel emissions. Overall, PEMFCs operating on clean hydrogen have been projected to have lifetime emissions of 150 g CO2 km−1 compared to 100 g CO2 km−1 for battery electric vehicles and 1000 g CO2 km−1 for diesel engine vehicles.23 Again, this analysis is dependent on a number of uncertain predictions. PEMFCs contain several critical raw materials, principally platinum. The current amount of platinum used in the 2nd generation Toyota Mirai is reported to be > NH3 > H2, indicating that CO is more accessible than H2 to absorb on the Pt surface. The lower adsorption energy of CO on the Pt surface (134 kJ/mol for Pt-CO vs. 87.9 kJ/mol for Pt-H2) supports this observation. CO affects PEMFCs in both low and high current regions, and diluted hydrogen conditions can easily trigger voltage fluctuation. The dominant process varies depending on the operation conditions, and the reactions below displays some typical reactions. CO + Pt $ Pt − COads

(2)

2CO + 2ðPt − Hads Þ $ 2ðPt − COads Þ + H2 Pt − COads + Pt − H2 Oads $ 2Pt + CO2 + 2H + 2e +

(3) −

(4)

CO absorption on the Pt surface can be categorized as either linear-absorbed or bridge-absorbed, and the bonding type on Pt sites is dependent on the CO coverage. Bridge-bonded CO species block more sites than linear-bonded CO adsorption, resulting in higher CO coverage with linear-bonded CO adsorption when sufficient CO is provided. An electrochemical method, reaction (4), effectively eliminates adsorbed CO, but its occurrence is limited to electrode potentials higher than 0.6 V vs. NHE. At potentials lower than 0.6 V vs. NHE, CO-adsorbed Pt sites hinder water adsorption on Pt sites, causing reaction (4) to proceed slowly. Based on this mechanism, optimizing water absorption near Pt sites can efficiently accelerate CO oxidation. The increase in anodic overpotential caused by the accumulation of CO on the surface of Pt catalyst results in CO oxidation and the relief of CO poisoning. Ultimately, this process leads to an increase in the cell voltage. However, the interplay between CO adsorption (poisoning) and CO oxidation (reactivating) causes anodic overpotential oscillation. This phenomenon is observable on the surfaces of Pt/C, PtRu/C, and PdPt/C. The oscillation patterns of voltage have been classified into three types: period-1 with strong relaxation, period-2 with intercalated large and small oscillation amplitudes, and chaos as shown in Fig. 1. It is worth noting that periodic and nonperiodic oscillations are favored differently by varying anode currents and flow rates. High current density tends to cause a chaotic state, whereas high flow rates favor period-1 oscillations. Intermediate values of both current and flow rate tend to favor period-2 oscillations, which exhibit intercalated amplitudes. Table 1 Typical impurities and corresponding maximum concentrations in road vehicle application hydrogen product.1 Impurity

Concentration (ppm)

Water (H2O) Total hydrocarbons except methane Methane (CH4) Carbon monoxide (CO) Carbon dioxide (CO2) Total sulfur compounds Ammonia (NH3) Oxygen (O2) Halogenated compounds

5 2 100 0.2 2 0.004 0.1 5 0.05

From Li, Z.; Wang, Y.; Mu, Y.; Wu, B.; Jiang, Y.; Zeng, L.; Zhao, T. Recent Advances in the Anode Catalyst Layer for Proton Exchange Membrane Fuel Cells. Renew. Sustain. Energy Rev. 2023, 176, 113182.

74

0.6

Cell voltage, V

0.4

Kco / V

0.2

Period 1

0.4 0.2

0.6

Period 2

0.4 0.2

1000 ppm CO/H2

0.5 0.0

1.0 0.8 0.6 0.4 0.2 0.0

f = 200 mL/min

1.0 0.8 0.6 0.4 0.2 0.0

f = 300 mL/min

1.0 0.8 0.6 0.4 0.2 0.0

f = 400 mL/min

200 ppm CO/H2

1.0 0.5 0.0

100 ppm CO/H2

1.0 0.5 0.0 1.0

50 ppm CO/H2

0.5 0.0 0

100 s

100 mA/cm2,f=400mL/min

1.0

f = 100 mL/min

Cell voltage, V

0.2

0.6

200 ppm CO/H2, 100 mA/cm2

1.0 0.8 0.6 0.4 0.2 0.0

0.4

50

100

150

Time / s

200

250

0

50

100

150

200

250

Time / s

Fig. 1 (left) Time series obtained at 50 sccm 100 ppm CO/H2 and 30  C at different cell currents: 527, 437, 327, and 55 mA cm−2 on the PtRu/C surface2; (middle) Cell voltage oscillations at various flow rates (100–400 mL min−1) at 100 mA cm−2 using 200 ppm CO/H2 as an anodic feeding stream; (right) Cell voltage oscillations at various CO concentrations (50–1000 ppm CO) on the PtRu/C anode catalyst layer.3 From Mota, A.; Lopes, P. P.; Ticianelli, E. A.; Gonzalez, E. R.; Varela, H. Complex Oscillatory Response of a PEM Fuel Cell Fed with H2/CO and Oxygen. J. Electrochem. Soc. 2010, 157(9), B1301.; Lu, H.; Rihko-Struckmann, L.; Sundmacher, K. Spontaneous Oscillations of Cell Voltage, Power Density, and Anode Exit CO Concentration in a PEM Fuel Cell. Phys. Chem. Chem. Phys. 2011, 13(40), 18179.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

0.6

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

75

The behavior of CO poisoning can vary depending on different operation conditions and MEA configuration parameters. Any factor that affects CO adsorption and oxidation can change the CO poisoning behavior. Bender and coworkers conducted a study on the performance effect of CO-included hydrogen fuel under various operating conditions, including different temperatures, relative humidities (RH), and current densities. They quantitatively analyzed dry gas concentrations ranging from 1 to 10 ppm. According to the findings, the impact of CO exposure on performance was more pronounced at elevated CO concentrations, current densities, and low RH levels in the ACL. Additionally, gas chromatography analysis indicated that a greater amount of liquid water at 60  C facilitated the conversion of CO to CO2 compared to 80  C. Furthermore, the CO tolerance of MEA was directly linked to the catalyst loading, catalyst supports, and membrane variations. In a study by Papasavva et al., the CO tolerance was compared at platinum loadings of 0.1 mg/cm2 and 0.05 mg/cm2. The results indicated that 0.1 mg/cm2 loading exhibited a smaller voltage drop during poisoning tests. Additionally, Prass et al. discovered that the voltage loss due to CO poisoning accounted for 8%, 41%, and 51% at an ISO concentration of 0.2 ppm when the anodic loading decreased from 50 to 25 and 15 mg/cm2. This suggests that a higher CO tolerance anode is required to reduce the anode catalyst loading without compromising CO poisoning performance. Improving CO tolerance is crucial since the catalyst loading has significantly decreased in the past decade and is expected to continue decreasing in the future. The study also revealed that high graphitized carbon support exhibited lower CO tolerance compared to high surface area carbon support. Interestingly, the impact of carbon graphitization on the robustness of the MEA to CO contamination was even more significant than the amount of catalyst loading. This could be attributed to the added challenge for CO gas to penetrate the nano-sized pores of the high surface carbon and contaminate the Pt catalyst particles. Furthermore, the role of the membrane in CO poisoning cannot be overstated. Reshetenko and coworkers found that when exposed to 2 ppm CO in the hydrogen stream, a membrane with higher oxygen permeability could reduce the impact of CO through catalytic oxidation of adsorbed CO by O2. The performance of the fuel cell was able to fully recover within 1–2 h when pure hydrogen was supplied. Additionally, the researchers compared the oxidation differences when pure O2 and air were introduced in the cathode. In the case of pure O2, there was no pseudo-inductive behavior observed at low frequencies, and the anode overpotential was much lower than the ignition potential for CO electrooxidation (0.085 V). This led to the conclusion that the removal of adsorbed CO was primarily through a chemical pathway facilitated by diffused O2. On the other hand, when air was used as a feed gas, there was clear pseudo-inductance and high anode overpotential (0.280 V), indicating the involvement of both chemical and electrochemical pathways. It is worth noting that the degree of poisoning within a single cell can vary. Reshetenko and coworkers investigated the impact of low concentrations of CO poisoning using a segmented cell and spatial electrochemical impedance spectroscopy (EIS). The study revealed that the inlet segments exhibited a reduced current density due to CO adsorption on the Pt surface, while downstream segments showed an increased current density. The EIS results also showed an increase in high-frequency resistance (HFR), anode, and cathode charge transfer resistances under CO exposure. Typically, the anode impedance is challenging to observe as it is usually covered by the cathode impedance. To directly evaluate the anode impedance and investigate the impedance profile under CO poisoning, Darowicki et al. utilized a dynamic electrochemical impedance spectroscopy (DEIS) technique to identify the anode impedance dynamically. This technique showed great potential for online monitoring of fuel cells. The results indicated that the performance of the ACL was marginally affected by CO at a concentration of 125 ppb. However, a significant impact on the anode impedance and voltage decrease was observed when the concentration reached 250 ppb or higher.

2.3

CO-tolerant catalyst

Developing a CO-tolerant catalyst aims to reduce CO adsorption or enhance the CO oxidation rate. Typically, both chemical and electrochemical oxidation mechanisms are utilized to achieve anti-CO poisoning ability. Recent research on anti-CO poisoning anode for PEMFC was compiled and presented in Table 2 to facilitate comparison. Bimetallic Pt-based catalysts are a promising method for fabricating CO-tolerant Pt-based anodes. This is due to the bifunctional effect, which enables oxidative CO removal at a low applied potential. Moreover, the interaction between the two or three metals can weaken CO bonding and reduce CO surface coverage. To investigate the enhanced CO-tolerance of alloys, spin-polarized density functional theory (DFT) calculations were conducted. The results indicated that the bifunctional effect achieved a lower potential for CO oxidation removal, while the ligand effect resulted in weaker CO bonding and lower surface coverage. Zhang and coworkers conducted an experimental study on a CO-tolerant PtRu core-shell catalyst with Ru enriched in the core and Pt enriched in the shell. The in-house PtRu/C catalyst demonstrated superior HOR catalytic activity compared to commercial Pt/C, with an overpotential of 28.6 mV compared to 45 mV. Moreover, it exhibited better CO-tolerant capability towards HOR than both commercial Pt/C and PtRu/C for CO concentrations ranging from 10 to 500 ppm with a 1-h exposure time. Wang and colleagues conducted a study on the impact of hydrogen dilution on CO poisoning in PtRu/C catalysts through modeling. The results showed that the combined effect of CO-H2-N2 mixtures exacerbates the surface coverage of hydrogen. Lee and colleagues developed a new PiR/C catalyst by replacing Ru with Pt through galvanic replacement to enhance HOR and anti-CO poisoning ability. The introduction of RuO2 altered the electronic structure of Pt in PiR/C, leading to a weaker Pt-CO bond. The oxophilicity of RuO2 enhances CO oxidation by facilitating the transportation of oxygen species. Despite having a much lower Pt content, the HOR activity is comparable to that of commercial Pt/C. Nepel and coworkers investigated the single-cell polarization of Pt/C and PtMo/C anodes at varying concentrations of CO in H2 between 70  C and 105  C. As expected, the CO-tolerance increased with temperature. Pt/C exhibited enhanced tolerance above 80  C due to O2 crossover, while PtMo/C activated MoOx-mediated water gas shift reaction (WGS) at high temperatures, resulting in greater CO tolerance. Under the same anode overpotential, PtMo/C

76 Table 2

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes Recent studies on anti-CO poisoning anode for PEMFC.1

Zhang et al. Lee et al. Nepel et al. Hassan et al. Hassan et al. Wang et al. Zhou et al. Wang et al. Zhang et al. Hassan et al. Hassan et al. Peng et al. Wang et al. Sun et al. Yamazaki et al. Wang et al. Wang et al. Yang et al. Yang et al. Huang et al. Li et al.

Catalyst

Temperature ( C)

CO Concentration

Exposure time

Performance loss

PtRu/C PiR/C PtMo/C PtMo/C PtW/C Pt/Ti0.7W0.3O2 Ru@TiO2 Ru@RuO2/TiO2 Ru@Pt/Ti4O7 PtMo/Mo2C/C Pt/WC/C Pd/W18O49 Pt/2,6-Diacetylpyridine/C PtRu@h-BN/C Rh(DPDS)/C Au/TiO2 ultrathin film IrRuSA-N-C Ir@IrSA-N-C Pt SACs/CrN Pt1@Co1CN Pt-C-AA

r.t. 60 85 85 105 r.t.a r.t.a r.t.a r.t.a 85 105 r.t.a r.t.a 70 60 25 80 95 r.t. r.t. 80

500 ppm 100 ppm 100 ppm 100 ppm 100 ppm 2% 1000 ppm 15% 500 ppm 100 ppm 100 ppm 1000 ppm 100 ppm 300 ppm 100 ppm 0.1% 1000 ppm 1000 ppm 1000 ppm 1000 ppm 10 ppm

1h – 2h 2h 2h – 2000 s 10 h – 2h 2h 10,000 s 5h – – – – – 4000 s 2400 s –

768 mV@1A cm−2 (HC)a 17.09% maximum power density reduction About 300 mV@1A cm−2 About 150 mV@1A cm−2 About 200 mV@1A cm−2 Onset potential 250 mV more positive (HC)a 12.4% loss @ 0.1 V (HC)a 33.3% loss @ 0.1 V (HC)a 179.2 mV overpotential @1A cm−2 About 200 mV@1A cm−2 About 250 mV@1A cm−2 No decrease @0.2 V 13% loss @0.1 V 70 mV loss @ 0.2268 A cm−2 Onset potential less than 0.05 V (HC)a 9% performance loss About 355 mW cm−2 maximum power density loss 987 mW cm−2 maximum power density loss 8.4% loss @ 0.1 V (HC)a 9.4% loss @ 0.1 V (HC)a About 45 mW cm−2 maximum power density loss

a

a HC: half-cell test; r.t: room temperature. From Li, Z.; Wang, Y.; Mu, Y.; Wu, B.; Jiang, Y.; Zeng, L.; Zhao, T. Recent Advances in the Anode Catalyst Layer for Proton Exchange Membrane Fuel Cells. Renew. Sustain. Energy Rev. 2023, 176, 113182.

showed a higher current density and produced more CO2 when exposed to 100 ppm CO in the H2 stream. Hassan and colleagues examined the impact of heat treatment on the activity and stability of CO-tolerant PtMo/C catalysts. The results revealed that thermal treatment improved the catalyst’s ability to tolerate CO and enhanced its stability. The single-cell polarization curves illustrated that the PtMo/C catalyst treated at 600  C exhibited the highest HOR activity and the best potential cycling retention when exposed to 100 ppm CO. However, the study also revealed that the dissolution of Mo species could impede proton conduction in the membrane and oxygen reduction reaction kinetics at the cathode side. The research group also investigated PtW/C as a CO-tolerant catalyst. The electronic effect of WOx in the Pt 5d band was found to reduce CO adsorption. The polarization curves showed that PtW/C outperformed Pt/C under pure hydrogen and 100 ppm CO/H2 gas supply in the anode side at 85  C and 105  C, respectively. An accelerated stress test (AST) was conducted by applying CV from 0.1–0.7 V vs. RHE at 50 mV/s, as depicted in Fig. 2. The significant decrease in PtW/C performance compared to Pt/C was likely due to the dissolution of W species, as evidenced by the presence of WOx peaks in the CV curves of the cathode. Studies have shown that the use of ternary and quaternary materials, such as PtMoFe/C, PtMoRu/C, and PtMoRuFe/C, can improve CO tolerance to some degree. However, PtMo/C still exhibited the best polarization performance when exposed to a 100 ppm CO/H2 stream. One way to improve CO tolerance in fuel cells is by using functional catalyst support. This method is effective because the transfer of electrons between the catalyst and support can induce significant changes in the properties of the catalyst, which is known as strong metal-support interactions (SMSIs). According to the report, a catalyst made of Pt/Ti0.7W0.3O2 and capable of tolerating CO was found to have the lowest onset potential (around 0.05 V vs. RHE) for H2 oxidation when used as an anode catalyst. In contrast, Pt/C and PtRu/C had a much higher potential of over 250 mV in the presence of 2% CO. Furthermore, the Ti0.7W0.3O2 catalyst also demonstrated greater stability in potential cycling over 500 cycles compared to Pt/C and PtRu/C. The commercial PtRu/C catalyst experienced a 30% loss, while the Ti0.7W0.3O2 catalyst only experienced a 5% loss. Zhou and coworkers synthesized a Ru@TiO2 catalyst with lattice confinement, which exhibited high CO tolerance and enhanced HOR activity. The catalyst showed no CO-stripping peak in both acidic and alkaline solutions, indicating weak CO adsorption. The half-cell test results revealed that the catalyst only experienced an 8% loss in limiting current density of HOR activity in the presence of 1000 ppm CO. In a similar fashion, researchers fabricated a Ru@RuO2/TiO2 catalyst that exhibited robust CO tolerance. Unlike the traditional bifunctional mechanism, the Ru@RuO2/TiO2 catalyst achieved high CO tolerance by using a hydrous metal oxide shell to block CO adsorption. This allowed the catalyst to tolerate 1–3% CO and work stably in 1% CO/H2 for 50 h, with 20% of the active sites remaining even in a pure CO environment. Additionally, researchers synthesized a Ru@Pt core-shell catalyst supported on Ti4O7, which showed enhanced CO-tolerance in the half-cell test. However, this performance was only evaluated in a three-electrode cell and the performance at the MEA level was not analyzed. Carbon-supported molybdenum carbide (Mo2C/C) has also been used as a catalyst support with CO-tolerant properties. The Pt/Mo2C/C and PtMo/Mo2C/C catalysts prepared in this manner exhibited greater initial hydrogen oxidation activities and cycling performances in the presence of 100 ppm CO compared to Pt/C and PtMo/C under the same conditions. The aforementioned research group also synthesized carbon-based tungsten carbide (WC/C) and

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

77

Fig. 2 Single-cell polarization curves for PtW/C and Pt/C anodes at 85  C and 105  C with (a)pure hydrogen and (b) H2/100ppm CO and (c and d) after the corresponding AST testing.4 From Hassan, A.; Paganin, V. A.; Ticianelli, E. A. Investigation of Carbon Supported PtW Catalysts as CO Tolerant Anodes at High Temperature in Proton Exchange Membrane Fuel Cell. J. Power Sources 2016, 325, 375–382.

compared its performance to PtW/C. The Pt/WC/C catalyst demonstrated enhanced initial hydrogen oxidation activity, with overpotentials of 364 mV and 398 mV for 1 A/cm2, as well as lower CO oxidation potentials at various temperatures in the presence of 100 ppm CO. In terms of stability, the enhanced stability of Pt/WC/C was attributed to its carbide phase, which is more resistant to corrosion in an acidic medium. Although the W crossover was still detected through CV measurement of the cathode, its negative impact was significantly smaller than that caused by Mo. Likewise, W18O49 was discovered to have synergistic catalytic effects on enhancing HOR performance and CO tolerance when used as a catalyst support. This is due to its ability to transfer electrons and the abundance of exposed active sites. However, full-cell testing results were not presented. Another effective method for mitigating CO adsorption is to use the steric effect and encapsulation effect, which can physically impede the interaction between the catalyst and CO. However, the challenge lies in designing a selective channel for H2 that does not allow CO to pass through. The 2,6-Diacetylpyridine, which could be strongly anchored on the Pt surface through tridentate coordination with a tilted-orientation pyridine ring, was selected as a canopy to isolate H2 with larger CO and H2S. This structure provided a height-limited space and exhibited 100 ppm CO tolerance. Similar research was conducted using 2,6-dihydroxymethyl pyridine, which presented far exceeding performance than commercial Pt/C under 10 ppm CO poisoning due to the steric confinement, as shown in Fig. 3. In a study conducted by Sun and coworkers, commercial PtRu/C catalysts were coated with few-layer graphitic boron nitride (h-BN) shells to create PtRu@h-BN/C core-shell catalysts. These catalysts exhibited comparable hydrogen oxidation reaction (HOR) activity to PtRu/C, but with improved CO-tolerance. Specifically, the PtRu@h-BN/C catalysts experienced a 20% loss in performance with 30 ppm CO and 25% CO2, while PtRu/C lost 60% of its performance under the same conditions.6 The improved CO-tolerance ability and performance of the PtRu@h-BN/C core-shell catalysts were attributed to the

Fig. 3 Polarization performance under CO and H2S poisoning and the corresponding mechanism.5 From Zhang, D.; Liu, W.; Ye, K.; Li, X. High CO and Sulfur Tolerant Proton Exchange Membrane Fuel Cell Anodes Enabled by “Work along Both Lines” Mechanism of 2,6-Dihydroxymethyl Pyridine Molecule Blocking Layer. J. Colloid Interface Sci. 2024, 653, 413–422.

78

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

Fig. 4 Schematic of Au/TiO2 film for anti-poisoning anode.7 From Wang, L.; Zhou, Y.; Yang, Y.; Subramanian, A.; Kisslinger, K.; Zuo, X.; Chuang, Y.-C.; Yin, Y.; Nam, C.-Y.; Rafailovich, M. H. Suppression of Carbon Monoxide Poisoning in Proton Exchange Membrane Fuel Cells via Gold Nanoparticle/Titania Ultrathin Film Heterogeneous Catalysts. ACS Appl. Energy Mater. 2019, 2(5), 3479–3487.

confinement effect of the h-BN shells, which weakened the adsorption of CO on the PtRu surface. Previous studies have also reported that carbon encapsulation can enhance CO tolerance. Therefore, the use of a shell or encapsulation can improve the CO-tolerance of catalysts. Researchers led by Hu discovered that Pt@C/C and Pt alloy Pt-Fe@NC/C, both of which encapsulate carbon layers as physical protective layers, exhibited improved poisoning tolerance by isolating the contact between nanoparticles and toxic molecules. However, the improvement was only evaluated in a half-cell test through CO stripping, and further validation at the MEA level was still needed. Enhancing CO tolerance through the use of additives is a promising approach, although more research is needed. Rh disulfo-deuteroporphyrin has been discovered to reduce the overpotential for CO oxidation, and a carbon-supported Rh deuteroporphyrin 2,4-disulfonic acid [Rh(DPDS)/C] catalyst has been developed that can efficiently oxidize CO below 0.05 V vs. RHE at 60  C. This is because CO has a tendency to coordinate with the Rh atom, followed by a nucleophilic attack by water, which generates RhdCOOH and provides electrons to an electrode to regenerate the free Rh atom. However, it is important to note that the effectiveness of this approach has only been confirmed in the three-electrode test, and further evaluation in the single-cell test is required to determine its practicality. In order to address the issue of CO poisoning, Wang and his team proposed a method of creating an ultrathin film made of Au/TiO2 between the membrane and catalyst layer, as depicted in Fig. 4. The thickness of TiO2 layer was optimized to 0.65 nm, which was thick enough to support CO catalytic activity while ensuring efficient proton transport through Nafion 117. The performance of the MEA that contained the Au/TiO2 layer was tested under 0.1% concentrated CO, and it exhibited a performance loss of less than 10%. In contrast, the control cell that lacked the film showed a 70% decrease in performance, demonstrating the effectiveness of the Au/TiO2 CO oxidation catalyst in enhancing the fuel cell performance against CO poisoning. A research group recently proposed CO-tolerant anode catalysts for PEMFCs derived from MOFs, including single-atom IrRuN-C and Ir nanoparticles@Ir single atom-N-C. They conducted various tests, including half-cell HOR polarization curves, full-cell polarization performance (with and without CO), CO oxidation reaction (COOR), and COdO2 PEMFC cell performance, to demonstrate excellent CO tolerance under a wide range of CO concentrations (10 ppm–1000 ppm) and long-term current retention under high CO concentration (1000 ppm), as illustrated in Fig. 5. The superior CO tolerance was attributed to the synergy between Ir single-atom and Ru single-atom or Ir nanoparticles. In the case of single-atom IrRu-N-C, CO and OH functional groups were likely to absorb on Ir and Ru surfaces, respectively, leading to the formation of COOH and CO2. For IrNP@IrSA-N-C, temperature-programmed desorption (TPD) suggested that the binding on isolated Ir sites was weaker than on metallic Ir(111). Chronoamperometric (CA) tests indicated that IrN4 sites removed the CO adsorbed on neighboring Ir nanoparticles. The synergy between IrN4 and Ir particles was attributed to the water dissociation capability of IrN4, which generated reactive OH and oxidizing CO absorbed on Ir nanoparticles in close proximity. More single-atom-related works are followed, such as Pt1@Co1CN, Pt SACs/ CrN, and carbon-supported Pt single atoms and clusters. The single atoms show great potential in anti-CO poisoning because when the Pt decreases to a single atom level, CO can hardy adsorb on the Pt surface, which is beneficial to increase CO tolerance. 8–10

2.4

Sulfur dioxide (SO2) poisoning behaviors

SO2 is a prevalent air contaminant that is known to occupy active sites, leading to decreased catalytic activities. Moreover, it can diffuse through the membrane from cathode to anode, causing harm to the ACL. The Pt surface adsorbed with SO2 can dissociate into sulfur and sulfur oxides or can be catalytically or electrochemically reduced to sulfur. Therefore, it is crucial to study how SO2 affects the durability and performance of PEMFCs. Tsushima and his team have proposed two types of PEMFC performance

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

79

Fig. 5 (a) HOR polarization curves in H2-saturated 0.1 M HClO4; (b) H2-O2 PEMFC cell performance at 95  C; (c) The metal mass activities of different catalysts at 0.6 V in PEMFC. (d) The polarization curves of PEMFC in H2/1000 ppm CO-saturated 0.1 M HClO4; (e) PEMFC cell performance of IrNP@IrSA-N-C in H2/ (10, 50, 100, 1000) ppm CO at 95  C; (f ) PEMFC maximum peak power density of different anode catalysts in H2/CO at 95  C.11 From Yang, X.; Wang, Y.; Wang, X.; Mei, B.; Luo, E.; Li, Y.; Meng, Q.; Jin, Z.; Jiang, Z.; Liu, C.; Ge, J.; Xing, W. CO-Tolerant PEMFC Anodes Enabled by Synergistic Catalysis between Iridium Single-Atom Sites and Nanoparticles. Angew. Chem. Int. Ed. 2021, 60 (50), 26177–26183. (a)

(b) 0.100

0.100 1st cycle

0.075

8th cycle

0.050

0.050

0.025

0.025

0.000

Current (A)

Current (A)

0.075

–0.025 –0.050 –0.075

8th cycle

1st cycle

0.000 –0.025 –0.050 –0.075

–0.100 –0.100 –0.125 –0.125

–0.150

–0.150

–0.175 0.0

0.2

0.4

0.6

0.8

1.0

Voltage (V) vs. DHE

1.2

1.4

1.6

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Voltage (V) vs. DHE

Fig. 6 CV curves of the anode obtained after the cathode exposure to 15 ppm SO2/air for 2 h at 500 mA cm−2 under different anode RH (a)14%RH, (b)100%RH and constant cathode RH at 100%.12 From Zhai, J.; Hou, M.; Zhang, H.; Zhou, Z.; Fu, J.; Shao, Z.; Yi, B. Study of Sulfur Dioxide Crossover in Proton Exchange Membrane Fuel Cells. J. Power Sources 2011, 196(6), 3172–3177.

degradation due to SO2 contamination: one-stage and two-stage degradation. The one-stage degradation can be partly recovered by supplying neat air, while the two-stage degradation is challenging to recover through neat air supply. It leads to the formation of a monolayer or multilayers of sulfur on the catalyst surface by electrochemical reduction of SO2. The phenomenon of SO2 crossover has been systematically studied by Zhai and colleagues using electrochemical methods under different conditions. It has been observed that the humidity of the CCL is generally higher than that of the ACL due to water generation in the CCL. The presence of water facilitates the migration of SO2 from the cathode to the anode, resulting in severe contamination of the anode catalyst. Even a low concentration of 1 ppm SO2 can cause SO2 crossover, and a greater difference in humidity between the anode and cathode results in more severe SO2 migration, as demonstrated in Fig. 6. Besides, the SO2 coverage of the anode catalyst was 2%, 13%, and

80

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

20% when the SO2 concentration was 15 ppm, 20 ppm, and 30 ppm, respectively, which calculated by the CV scan differences before and after the poisoning, while the gas pressures exhibited no pronounced effect on the SO2 crossover when the cell temperature was kept at 70  C and the RH were 100% for both sides. Meanwhile, the gas pressures did not show a notable impact on SO2 crossover when the cell temperature was kept at 70  C and the RH was 100% for both sides. The impact of cell temperature on SO2 contamination was also studied, and it was discovered that the SO2 crossover from the CCL to the ACL was suppressed at low cell temperatures due to the higher water content and lower gas permeability of the membrane. Additionally, after 120 h of poisoning under 2 ppm SO2 cathode gas supply, the electrochemical active surface area (ECSA) of the ACL decreased by 5% and 11.6% when the current density was 0.2 A/cm2 and 1.0 A/cm2, respectively. Gould and colleagues studied the restorative effects of electrochemical methods on PEMFCs that were exposed to 1 ppm SO2 in the air and RH of 100%/50% (anode/cathode) for 3 h. They achieved this by oxidizing adsorbed sulfur species on the Pt catalysts to sulfate (SO42−) at high potentials and then removing them as water-soluble anions at low potentials. By cycling between 0.09 V and 1.1 V under the mixture gas of H2/N2, they were able to recover 97% of the ECSA. The majority of research has concentrated on the impact of SO2 contamination in the CCL, and there is limited discussion on the effects of poisoning in the ACL.

2.5

SO2-tolerant catalyst

According to reports, graphene has the potential to be utilized as a platinum support, which displays improved SO2 tolerance as compared to the conventional platinum supported on Vulcan XC carbon. The Pt/C and Pt/Graphene depicted 72% and 54% sulfur coverage, respectively. This was credited to the indirect elimination of sulfur ions from Pt through scavenged OH- ions. It was further reported that the incorporation of the Ru element to synthesize Pt6Ru2/C catalyst can boost SO2 tolerance. The Ru element was capable of accelerating the SO2 oxidation while still maintaining higher ECSA after SO2 poisoning, as illustrated in Fig. 7. Based on the DFT study, it was found that the inclusion of Ru element effectively altered the electronic structure of Pt oxidation in the PtRu/C catalyst, resulting in a decreased interaction between Pt and SO2 and a faster rate of SO2ad oxidation.

3 3.1

Cell reversal Introduction

During regular operations, the anode and cathode are supplied with adequate reactants, such as hydrogen and oxygen, to generate electrons and protons. In this scenario, the potential of the cathode is higher than that of the anode. 2H2 + O2 ! 2H2 O E0 ¼ 1:23V ðvs:RHEÞ

(5)

Indeed, the operation of PEMFC is often carried out in challenging conditions such as dynamic cycling, rapid loading, and cold start-up, which can result in reaction obstruction or cell reversal under certain circumstances. Cell reversal refers to the phenomenon where the potential difference between the anode and cathode decreases and even reverses, with the anode exhibiting a higher potential than the cathode. This can be caused by different factors, including insufficient gas supply, uneven distribution of reactants, start-up and shutdown (SU/SD) operations, and rapid changes in loading. To provide a more detailed explanation, cell reversal can be vividly described as the stack current “pushing” a starved cell towards negative voltages. Cell reversal can be categorized into three types: oxidant starvation, dehydration, and fuel starvation. Oxidant starvation occurs when there is a shortage of oxygen, causing protons to recombine on the cathode and producing hydrogen. H+ + 2e− ! H2 E0 ¼ 0V ðvs:RHEÞ

(6)

Fig. 7 Illustration for the Ru introduction to enhance SO2 tolerance in PtRu alloy.13 From Liu, Y.; Du, L.; Kong, F.; Han, G.; Gao, Y.; Du, C.; Zuo, P.; Yin, G. Sulfur Dioxide-Tolerant Bimetallic PtRu Catalyst toward Oxygen Electroreduction. ACS Sustain. Chem. Eng. 2020, 8(2), 1295–1301.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

81

It is also known as the hydrogen pump, occurs when the air stoichiometry is less than 1.0 or when water flooding takes place. It does not cause direct harm to the cell, but it leads to a loss of performance. Dehydration is a condition in which the cell reverses due to a lack of membrane conductivity, resulting in an increase in resistance and a short, rapid oxidation event. The MEA is damaged due to the lack of proton conductivity, which converts energy into heat. Dehydration reversal typically occurs until the membrane stops functioning, although it is rare in practice. The study of fuel starvation is the focus of this section because it is more probable to occur in practical scenarios. Due to its significance, it has been extensively researched.

3.2

Mechanisms and phenomena investigation

When fuel is starved, the carbon oxidation reaction (COR) and oxygen evolution reaction (OER) take place to produce electrons and protons, which are necessary to sustain the external current and internal proton flow. 2H2 O ! H+ + O2 + 4e− E0 ¼ 1:23V ðvs:RHEÞ

(7)

C + 2H2 O ! CO2 + 4H+ + 4e− E0 ¼ 0:21V ðvs:RHEÞ

(8)

−

C + H2 O ! CO + 2H + 2e E ¼ 0:52V ðvs:RHEÞ +

0

(9)

In a moist anode environment, the slow kinetics of the COR result in a preference for the OER, which has a lower thermodynamic potential. The effects of COR are more detrimental than those of OER in the anode, as it can lead to carbon corrosion, detachment of platinum nanoparticles, and collapse of the catalyst layer. Consequently, this increases the ohmic resistance and degrades the performance of the fuel cell. Different perspectives suggest that the rise in ohmic resistance is due to the deformation and contraction of the membrane and the reconfiguration of the ionomer, which could be a consequence of different membrane applications. In situ online mass spectrometry uncovered that the water electrolysis reaction began at a potential range lower than that of carbon oxidation. This finding suggests that extending the duration of the water electrolysis reaction could help reduce carbon corrosion. Thus, research efforts are focused on prolonging the water electrolysis plateau. Nevertheless, it was observed that the fuel cell performance declined during the water electrolysis plateau (WEP), even with an extended WEP time. Furthermore, the longer the WEP time, the greater the performance decline. Despite the presence of performance degradation during the WEP, the degradation rate was noticeably lower than that in the carbon-corrosion dominant region. It is important to note that the AST protocol significantly influenced the cell reversal damage behavior. Pulsed AST resulted in a 2.5-time increase in carbon corrosion rates, a 2.2-time higher loss of ECSA, a 15% greater reduction in anode catalyst layer thickness, and a 65% higher increase in HFR compared to the conventional long-lasting cell reversal test. Based on the measurements taken, it was observed that pulsed AST resulted in increased oxidation of the carbon surface and damage to the ACL. It can be inferred that carbon with a higher surface area or amorphous carbon has a weaker ability to resist corrosion. As a result, it is proposed that amorphous carbon could potentially serve as sacrificial carbon and compensate for the loss of ECSA during the degradation process.

3.3

Anti-reversal catalysts

3.3.1 Catalyst support materials The creation of an anti-reversal catalyst layer heavily relies on the catalyst support. Structural collapse is commonly caused by the carbon support, which is the component with the lowest strength. To improve the reversal tolerance of the catalyst layer, it is a straightforward and effective approach to enhance the corrosion-resistance of carbon support. Research has indicated that increasing the degree of graphitization of the carbon support can significantly reduce the impact of reversal damage. As a matter of fact, higher levels of graphitized carbon were found to be effective in alleviating the damage caused by reversal, with a mere 4% loss in performance when the cell potential was clamped at −0.8 V. Moreover, carbon-free support holds greater promise as it removes the chance of carbon corrosion, despite the need for balancing various factors such as cost, conductivity, and durability. According to reports, WO3, a type of hydrogen spillover material, may act as a hydrogen reservoir to compensate for a temporary proton deficit. As a result, composite electrodes containing WO3 underwent testing using gas diffusion electrodes and pellet electrodes, and were compared to Pt/Vulcan XC-72. The findings revealed that the hydrogen storage charge, measured by the hydrogen adsorption-desorption area, was six times greater than that of Pt/Vulcan XC-72. However, the inadequate conductivity of WO3 poses a challenge for full-cell validation, and further modifications are required to meet the demand of the PEMFC system. A similar study was conducted by Park and coworkers, using Ar plasma surface treatment and atomic layer deposition of catalysts to improve the electrical conductivity of WO3, further increasing the initial performance. In addition, titanium oxide, comprising TiO2 and Ti4O7, has shown significant potential as a reversal tolerant anode support. Specifically, Ti4O7 was utilized as support for both Pt and Ir by Ioroi and colleagues to achieve reversal tolerant anodes. Fig. 8 presents a comparison of BOL/EOL performances and voltage reversal time between Pt/XC-72 based RTA and Pt/graphitized Ketjen black (GKB) based RTA. The findings indicate that the addition of Ir considerably enhances RTA property, with Pt/Ti4O7 RTAs exhibiting a significantly greater effect of Ir, requiring only 1/7 the amount of Ir compared to Pt/C RTAs with similar voltage behavior. Post-mortem analysis of the tested MEA confirms that Pt/Ti4O7 RTA mitigated damage to both the catalyst layer

82

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

Fig. 8 BOL/EOL I-V curves and cell resistance at 80  C, fully humidified H2/O2 (ambient pressure); (a and b) Pt/C-based RTA, (c and d) Pt/GKB-based RTA, and (e and f ) Pt/Ti4O7-based RTA.14 From Ioroi, T.; Yasuda, K. Highly Reversal-Tolerant Anodes Using Ti4O7-Supported Platinum with a Very Small Amount of Water-Splitting Catalyst. J. Power Sources 2020, 450, 227656.

and the microporous layer. Subsequently, Ti4O7 supported IrOx was further investigated as reversal tolerant components, displaying almost no attenuation of cell voltage in less than 100 min of reversal time. Nevertheless, the working mechanism was not explicitly stated, and additional characterizations are necessary.

3.3.2 Oxygen evolution reaction (OER) catalyst The most prevalent catalyst of anode, Pt/C, underwent modification with appropriate protic ionic liquids (PILs) to enhance OER activity and reversal tolerance. The findings indicated a reduction of up to 300 mV in overpotential compared to unmodified

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

83

Pt/C for the OER test and less ECSA loss in the reversal test. The primary concept of this technique is to increase solubility and oxygen diffusivity through PILs, resulting in improved reaction kinetics towards OER. Iridium oxide is the most frequently used OER catalyst, and many different methods have been employed to enhance its anti-reversal capabilities. Roh and colleagues created the monodisperse IrOx deposited on Pt/C, which achieved a reversal tolerance time four times longer than the conventional RTA catalyst (physical mixing of IrOx particles and Pt/C). Interestingly, the monodisperse metallic Ir deposited on Pt/C was also synthesized, but it displayed poor reversal tolerance due to competitive HOR and CO poisoning. To further improve the anti-reversal ability, researchers considered RuO2 because its OER activity is slightly higher than that of IrO2. However, RuO2 demonstrated poor durability and chemical stability in the fuel cell operation environment. Therefore, a RuO2/IrO2 composite was studied to maintain balance. Despite the performance loss caused by the poor conductivity of RuO2 and IrO2, the enhancement of cell reversal tolerance was evident. IrRu is currently a major focus of research as it has the potential to mitigate the effects of cell reversal. In the beginning, the IrRu alloy was applied to Pt-coated 3 M nanostructured thin films (NSTF) using PVD and sputtering to enhance SU/SD adaptability. Researchers, including Kim and colleagues, have investigated the fundamental properties of IrRu-based catalysts and the HOR and OER activities of carbon-supported IrRu alloy under various synthetic conditions. However, there is still controversy surrounding the optimal composition of IrRu for enhancing cell reversal tolerance. While most studies suggest that IrRu4 exhibits better performance, some researchers have also recommended IrRu2 as the better choice. Furthermore, the ternary IrRu4Y0.5 was introduced, showing a 21% improvement in performance and a longer reversal time (64 min) than Pt/C. Nonetheless, research in this area remains limited. It is hoped that more attention will be paid to high-entropy alloys to achieve a balance of HOR, reversal tolerance, and durability.

3.3.3 HOR selective catalyst HOR selective catalysts are a promising solution to mitigate the damage caused by instantaneous potential changes. When air seeps into the anode, oxygen reduction reaction (ORR) ensues, consuming the electrons generated by HOR in an oxygen-rich environment. In 2011, Genorio et al. proposed a selective catalyst design that promotes HOR while suppressing ORR. Similarly, Jung et al. utilized this mechanism to resolve corrosion issues during startup and shutdown events. They created a gas environment-sensitive catalyst by depositing thin platinum layers on hydrogen tungsten bronze (Pt/HxWO3). The catalyst changes into an insulator when exposed to oxygen but regains metallic conductivity and selectively promotes HOR when exposed to hydrogen, as depicted in Fig. 9. Similarly, Pt/TiO2 is also found as HOR selective catalyst since hydrogen spillover occurs and leads to the formation of a conduction pathway on the surface under hydrogen-rich conditions. While the TiO2 surface returns to its insulation nature under oxygen-rich conditions. However, the conductivity loss of these HOR selective catalysts significantly affects the polarization performance, and further efforts are necessary to meet the practical application requirements.15

4 4.1

Low platinum loading anode Introduction

In order to ensure the economic feasibility of PEMFC, it is important to reduce the usage of noble metal platinum to a cost-friendly level. DOE has established a technical target for PGM total content, which should not exceed 0.125 g kW−1 for a 100 kWnet PEMFC stack. This translates to a cathode loading of 0.1 mg cm−2 and an anode loading of 0.025 mg cm−2. Previous studies have indicated that the significant difference between these values suggests the potential for reducing Pt loading. While some studies have achieved ultra-high mass activity in rotating disk electrode (RDE) or low Pt loading in single cell level, it has been theoretically demonstrated that reducing anode loadings from 0.1 to 0.05 mgPt cm−2 has minimal impact on performance. In 2014 and 2020, Toyota, a leading manufacturer of PEMFCs, launched two generations of FC vehicles with an anode Pt loading of 0.05 and 0.025 mgPt cm−2,

Fig. 9 Schematic illustration of HOR selective catalyst while suppressing ORR by tuning the electrical conductivity.16 From Jung, S.-M.; Yun, S.-W.; Kim, J.-H.; You, S.-H.; Park, J.; Lee, S.; Chang, S. H.; Chae, S. C.; Joo, S. H.; Jung, Y.; Lee, J.; Son, J.; Snyder, J.; Stamenkovic, V.; Markovic, N. M.; Kim, Y.-T. Selective Electrocatalysis Imparted by Metal–Insulator Transition for Durability Enhancement of Automotive Fuel Cells. Nat. Catal. 2020, 3 (8), 639–648.

84

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

respectively. However, achieving such low Pt loading is a significant challenge for most PEMFC manufacturers, as it requires extensive optimization of the stack and system. As a result, the actual Pt loading in FC vehicles is often much higher to ensure scalability and durability, as demonstrated by the commercial Gore MEAs with 0.4 mgPt cm−2 cathode and 0.2 mgPt cm−2 anode Pt loading in the DOE durability test. It is important to note that doubling the Pt loading does not necessarily double the output performance due to mass transfer resistance issues. For commercial applications, it is crucial to balance output performance and cost appropriately. While reducing cathode Pt loading has received more attention due to sluggish ORR kinetics and low exchange −2 −2 to 0.025 mgPt cm−2 current density of 10−8 A cm−2 Pt (0.05 A cmPt for anode), a reduction in anode loading from 0.2 mgPt cm results in an 87.5% reduction in Pt, compared to a 75% reduction for cathode loading decreases from 0.4 mgPt cm−2 to 0.1 mgPt cm−2. This advantage is amplified with increased production, highlighting the importance of emphasizing and discussing reliable low platinum loading anodes.

4.2

Recent research

To decrease the Pt loading while maintaining the required output performance, two approaches have been proposed. The first approach involves exposing more activity sites on the Pt surface, while the second approach aims to increase the turnover frequency (TOF) through interactions between platinum and catalyst support, as well as other introduced materials. These methods aim to boost the Pt utilization efficiency by decreasing the Pt loading. To facilitate comparison, recent studies on low anode loading of PEMFC are summarized in Table 3. In the field of low platinum anode research, novel catalyst preparation methods are commonly used to decrease Pt loading while maintaining performance. Billy and colleagues implemented direct liquid injection metal-organic chemical vapor deposition technology to control the Pt loading deposited on the GDL, resulting in improved mass transfer and Pt utilization without varying the thickness of the catalyst layer. The single cell polarization test results indicated a decrease in potential as the anode Pt loading decreased from 0.226 mgPt cm−2 to 0.035 mgPt cm−2, demonstrating the effectiveness of this approach. In addition, the home-made anodes outperformed the commercial electrode with 0.5 mgPt cm−2 Pt loading. This trend was consistent across various conditions, including oxygen or air cathode and different gas pressures ranging from 4 to 1.5 bars, indicating the remarkable universality of the approach. Hsueh and colleagues utilized atomic layer deposition (ALD) to deposit platinum nanoparticles onto nitric acid-treated multiwalled carbon nanotubes, with Pt loading controlled by the number of ALD cycles. Following 100 cycles, the Pt loading was approximately 0.019 mgPt cm−2. Fig. 10 illustrates that the home-made Pt/CNTs prepared by ALD demonstrated superior Table 3

Recent research on the low platinum anode.1 Electrode

Anode Pt loading (mg cm−2)

Cathode Pt loading (mg cm−2)

Maximum power density (mW cm−2)

Pt utilization (kW g−1)

Preparation method

Billy et al.

Pt/GDL

0.035

0.5

600–700

–

Hsueh et al. Martin et al. Dang et al. Breitwieser et al. Fiala et al. Saha et al. Song et al. Ostroverkh et al. Qayyum et al. Mougenot et al.

Pt/CNTs Pt/GDL Pd@Pt/C Pt/C Pt-CeO2 Pt/Mo2C Pt/bonded carbon layers Pt/GDL Pt/GDL Pt/carbon paper

0.019 0.01 0.015 0.029 0.04 0.02 0.01 0.017 0.017 0.01Pd

300–350 700 780 2500 410 400–450 780 950 1100 250

Cooper et al. Liu et al.

0.06 0.0372

572 1001

2.27 30–35 – 88 – 10.4 – – 7.43 250 kW g−1 Pt and 12.5 kW g−1 Pd – Pulse potential deposition – Photo-driven fabrication

Jiang et al.

Pt/a-Carbon Dendritic Pt spherical nanocrowns/photocatalystmodified membrane PtPd/ppy nanowire

0.5 0.01 0.04 0.029 2 0.02 0.4 0.4 0.1 0.01Pd + 0.001Pt 0.5546 0.0533

Direct liquid injection metal organic chemical vapor deposition Atomic layer deposition Electrospray In situ Pulse electrodeposition Direct membrane deposition Magnetron sputtering Atomic layer deposition Atomic layer deposition Magnetron sputtering Pulsed laser deposition Plasma sputtering

0.0846Pt + 0.0528Pd

0.156Pt + 0.0545Pd

762.1

–

Tian et al.

Pt/VACNTs

0.035

0.035

1002

–

Physical vapor deposition + Impregnation reduction Plasma enhanced chemical vapor deposition + Sputtering

From Li, Z.; Wang, Y.; Mu, Y.; Wu, B.; Jiang, Y.; Zeng, L.; Zhao, T. Recent Advances in the Anode Catalyst Layer for Proton Exchange Membrane Fuel Cells. Renew. Sustain. Energy Rev. 2023, 176, 113182.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

85

Fig. 10 (a) Morphological characterization of Pt/CNT prepared by ALD; (b) PEMFC performances for anodes made with commercial material and by the ALD method. The Pt loading of commercial E-Tek and E-Tek electrodes are 0.5 and 0.019 mgPt cm−2, respectively.17 From Hsueh, Y.-C.; Wang, C.-C.; Kei, C.-C.; Lin, Y.-H.; Liu, C.; Perng, T.-P. Fabrication of Catalyst by Atomic Layer Deposition for High Specific Power Density Proton Exchange Membrane Fuel Cells. J. Catal. 2012, 294, 63–68.

Fig. 11 (a) Polarization curves for single-cell performance with different anode compositions and identical commercial cathodes (0.2 mgPt cm−2); (b) Stability test results for the Pd@Pt/C MEA and commercial Pt/C MEA (600 mA cm−2 for 150 h).18 From Dang, D.; Zou, H.; Xiong, Z.; Hou, S.; Shu, T.; Nan, H.; Zeng, X.; Zeng, J.; Liao, S. High-Performance, Ultralow Platinum Membrane Electrode Assembly Fabricated by In Situ Deposition of a Pt Shell Layer on Carbon-Supported Pd Nanoparticles in the Catalyst Layer Using a Facile Pulse Electrodeposition Approach. ACS Catal. 2015, 5 (7), 4318–4324.

performance compared to the commercial anode, with a 96.2% reduction in Pt loading. Martin and his team fabricated electrodes with a remarkably low Pt loading of 0.01 mgPt cm−2 for both the cathode and anode using the electrospray method. These electrodes achieved an impressive output performance of about 600–700 mW cm−2, resulting in an exceptionally high overall platinum utilization rate of 30–35 kW g−1 Pt . Dang and his team discovered that a thin layer of platinum (Pt) atoms deposited on carbon-supported palladium (Pd) nanoparticles (Pd@Pt/C) using in situ pulse electrodeposition resulted in excellent performance in H2/air single fuel cells. The anode and cathode had Pt loadings of 0.015 mgPt cm−2 and 0.04 mgPt cm−2, respectively, outperforming commercial Pt/C MEAs. The detailed test results are presented in Fig. 11. Notably, the Pt atoms only covered the exposed surfaces of the Pd nanoparticles, without any Nafion binder covering the Pt catalysts. The enhancements were attributed to several factors, including the high dispersion of Pt due to the core-shell structure, high Pt utilization due to the lack of Nafion binder coverage, the quantum effect due to the high distribution of Pt, and the interaction between the Pt shell and the Pd core. These factors should be considered to inspire future research directions. Moreover, the Pd@Pt/C exhibited similar durability performance to commercial Pt/C within 150 h, indicating its excellent potential for practical application. Creating innovative catalyst layer structures, such as 3D ordered catalyst layers, is an effective approach to reduce Pt loading. Compared to conventional random catalyst dispersion, an ordered structured catalyst layer maximizes the three-phase boundaries

86

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

Fig. 12 Polarization curves and power density of single PEM fuel cells with Pt/VACNTs ¼ 0.03 mgPt cm−2 and Pt/VACNTs ¼ 0.035 mgPt cm−2 (F-0.02 mgPt cm−2, B-0.015 mg mgPt cm−2) as anodic electrodes.19 From Tian, Z. Q.; Lim, S. H.; Poh, C. K.; Tang, Z.; Xia, Z.; Luo, Z.; Shen, P. K.; Chua, D.; Feng, Y. P.; Shen, Z.; Lin, J. A Highly Order-Structured Membrane Electrode Assembly with Vertically Aligned Carbon Nanotubes for Ultra-Low Pt Loading PEM Fuel Cells. Adv. Energy Mater. 2011, 1 (6), 1205–1214.

and ensures uniform chemical reaction conditions for the Pt catalyst. One example is Pt deposited on vertically aligned carbon nanotubes (VACNTs). As illustrated in Fig. 12, the VACNTs-based catalyst layer with a Pt loading of 0.035 mgPt cm−2 demonstrated comparable polarization performance to the commercial random carbon powder-based catalyst layer with a Pt loading of 0.4 mgPt cm−2, indicating the potential for significant Pt reduction in fuel cells.

5

Perspectives and outlook

In the past years, there have been notable advancements in anode catalysts aimed at improving durability and reducing costs. Firstly, this chapter discusses the effects of impurities on poisoning and the methods used to mitigate them. The primary approach to increasing poisoning tolerance involves reducing impurity adsorption on the Pt surface and converting impurities into removable forms. Research strategies for anode cell reversal catalysts focus on the catalyst support, as well as the addition of OER catalysts and HOR selective catalysts. It is worth noting that the exploration of HOR selective catalysts represents a new research direction, as it has the potential to fundamentally solve corrosion issues. Lastly, the low platinum anode catalysts are briefly introduced as the most direct way to reduce cost, and the research directions are pointed out. The main idea is to increase Pt utilization by modifying exposed forms or creating new ACL structures, while maintaining output performance, resulting in reduced Pt loading and lower costs. Despite advancements in anode catalysts, challenges persist in meeting long-term targets for practical fuel cell vehicles. Most validations are limited to single-cell experiments, with a lack of stack-level evaluations. Additionally, a convincing testing standard or protocol for poisoning and cell reversal functions should be established to further assess and compare the functional performance of PEMFCs. Furthermore, while studies focus on developing individual functions, a more integrated and robust anode is desired since practical PEMFC operation often involves multiple factors. For instance, when CO-containing hydrogen is utilized in the anode, cell reversal and CO poisoning may occur simultaneously. An enhanced CO and cell reversal tolerance dual-functional anode that preserves output performance is preferred to minimize the repercussions. Therefore, a comprehensive and combined functional MEA is more attractive for future applications as it closely mimics real-world conditions. Anode catalysts have historically been undervalued, despite the potential for significant performance deterioration in various situations. This oversight has hindered the development of PEMFCs and limited their capability to achieve high performance with a low Pt mass loading. Therefore, an increase in awareness and attention towards anode catalyst research is necessary, and innovative ideas must be inspired to further advance the development of PEMFCs.

References 1. Li, Z.; Wang, Y.; Mu, Y.; Wu, B.; Jiang, Y.; Zeng, L.; Zhao, T. Recent Advances in the Anode Catalyst Layer for Proton Exchange Membrane Fuel Cells. Renew. Sustain. Energy Rev. 2023, 176, 113182. https://doi.org/10.1016/j.rser.2023.113182. 2. Mota, A.; Lopes, P. P.; Ticianelli, E. A.; Gonzalez, E. R.; Varela, H. Complex Oscillatory Response of a PEM Fuel Cell Fed With H2/CO and Oxygen. J. Electrochem. Soc. 2010, 157 (9), B1301. https://doi.org/10.1149/1.3463725. 3. Lu, H.; Rihko-Struckmann, L.; Sundmacher, K. Spontaneous Oscillations of Cell Voltage, Power Density, and Anode Exit CO Concentration in a PEM Fuel Cell. Phys. Chem. Chem. Phys. 2011, 13 (40), 18179. https://doi.org/10.1039/c1cp21984g. 4. Hassan, A.; Paganin, V. A.; Ticianelli, E. A. Investigation of Carbon Supported PtW Catalysts as CO Tolerant Anodes at High Temperature in Proton Exchange Membrane Fuel Cell. J. Power Sources 2016, 325, 375–382. https://doi.org/10.1016/j.jpowsour.2016.06.043.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Anodes

87

5. Zhang, D.; Liu, W.; Ye, K.; Li, X. High CO and Sulfur Tolerant Proton Exchange Membrane Fuel Cell Anodes Enabled by “Work along Both Lines” Mechanism of 2,6-Dihydroxymethyl Pyridine Molecule Blocking Layer. J. Colloid Interface Sci. 2024, 653, 413–422. https://doi.org/10.1016/j.jcis.2023.09.076. 6. Sun, M.; Lv, Y.; Song, Y.; Wu, H.; Wang, G.; Zhang, H.; Chen, M.; Fu, Q.; Bao, X. CO-Tolerant PtRu@h-BN/C Core-Shell Electrocatalysts for Proton Exchange Membrane Fuel Cells. Appl. Surf. Sci. 2018, 450, 244–250. https://doi.org/10.1016/j.apsusc.2018.04.170. 7. Wang, L.; Zhou, Y.; Yang, Y.; Subramanian, A.; Kisslinger, K.; Zuo, X.; Chuang, Y.-C.; Yin, Y.; Nam, C.-Y.; Rafailovich, M. H. Suppression of Carbon Monoxide Poisoning in Proton Exchange Membrane Fuel Cells via Gold Nanoparticle/Titania Ultrathin Film Heterogeneous Catalysts. ACS Appl. Energy Mater. 2019, 2 (5), 3479–3487. https://doi.org/10.1021/ acsaem.9b00264. 8. Yang, Z.; Chen, C.; Zhao, Y.; Wang, Q.; Zhao, J.; Waterhouse, G. I. N.; Qin, Y.; Shang, L.; Zhang, T. Pt Single Atoms on CrN Nanoparticles Deliver Outstanding Activity and CO Tolerance in the Hydrogen Oxidation Reaction. Adv. Mater. 2023, 35 (1), 2208799. https://doi.org/10.1002/adma.202208799. 9. Li, H.; Wang, X.; Gong, X.; Liu, C.; Ge, J.; Song, P.; Xu, W. “One Stone Three Birds” of a Synergetic Effect between Pt Single Atoms and Clusters Makes an Ideal Anode Catalyst for Fuel Cells. J. Mater. Chem. A 2023. https://doi.org/10.1039/D3TA01313H. 10. Huang, Z.; Lu, R.; Zhang, Y.; Chen, W.; Chen, G.; Ma, C.; Wang, Z.; Han, Y.; Huang, W. A Highly Efficient pH-Universal HOR Catalyst With Engineered Electronic Structures of Single Pt Sites by Isolated Co Atoms. Adv. Funct. Mater. 2023, 2306333. https://doi.org/10.1002/adfm.202306333. 11. Yang, X.; Wang, Y.; Wang, X.; Mei, B.; Luo, E.; Li, Y.; Meng, Q.; Jin, Z.; Jiang, Z.; Liu, C.; Ge, J.; Xing, W. CO-Tolerant PEMFC Anodes Enabled by Synergistic Catalysis between Iridium Single-Atom Sites and Nanoparticles. Angew. Chem. Int. Ed. 2021, 60 (50), 26177–26183. https://doi.org/10.1002/anie.202110900. 12. Zhai, J.; Hou, M.; Zhang, H.; Zhou, Z.; Fu, J.; Shao, Z.; Yi, B. Study of Sulfur Dioxide Crossover in Proton Exchange Membrane Fuel Cells. J. Power Sources 2011, 196 (6), 3172–3177. https://doi.org/10.1016/j.jpowsour.2010.11.103. 13. Liu, Y.; Du, L.; Kong, F.; Han, G.; Gao, Y.; Du, C.; Zuo, P.; Yin, G. Sulfur Dioxide-Tolerant Bimetallic PtRu Catalyst toward Oxygen Electroreduction. ACS Sustain. Chem. Eng. 2020, 8 (2), 1295–1301. https://doi.org/10.1021/acssuschemeng.9b06785. 14. Ioroi, T.; Yasuda, K. Highly Reversal-Tolerant Anodes Using Ti4O7-Supported Platinum With a Very Small Amount of Water-Splitting Catalyst. J. Power Sources 2020, 450, 227656. https://doi.org/10.1016/j.jpowsour.2019.227656. 15. You, S.-H.; Jung, S.-M.; Kim, K.-S.; Lee, J.; Park, J.; Jang, H. Y.; Shin, S.; Lee, H.; Back, S.; Lee, J.; Kim, Y.-T. Enhanced Durability of Automotive Fuel Cells via Selectivity Implementation by Hydrogen Spillover on the Electrocatalyst Surface. ACS Energy Lett. 2023, 2201–2213. https://doi.org/10.1021/acsenergylett.2c02656. 16. Jung, S.-M.; Yun, S.-W.; Kim, J.-H.; You, S.-H.; Park, J.; Lee, S.; Chang, S. H.; Chae, S. C.; Joo, S. H.; Jung, Y.; Lee, J.; Son, J.; Snyder, J.; Stamenkovic, V.; Markovic, N. M.; Kim, Y.-T. Selective Electrocatalysis Imparted by Metal-Insulator Transition for Durability Enhancement of Automotive Fuel Cells. Nat. Catal. 2020, 3 (8), 639–648. https://doi.org/ 10.1038/s41929-020-0475-4. 17. Hsueh, Y.-C.; Wang, C.-C.; Kei, C.-C.; Lin, Y.-H.; Liu, C.; Perng, T.-P. Fabrication of Catalyst by Atomic Layer Deposition for High Specific Power Density Proton Exchange Membrane Fuel Cells. J. Catal. 2012, 294, 63–68. https://doi.org/10.1016/j.jcat.2012.07.006. 18. Dang, D.; Zou, H.; Xiong, Z.; Hou, S.; Shu, T.; Nan, H.; Zeng, X.; Zeng, J.; Liao, S. High-Performance, Ultralow Platinum Membrane Electrode Assembly Fabricated by In Situ Deposition of a Pt Shell Layer on Carbon-Supported Pd Nanoparticles in the Catalyst Layer Using a Facile Pulse Electrodeposition Approach. ACS Catal. 2015, 5 (7), 4318–4324. https://doi.org/10.1021/acscatal.5b00030. 19. Tian, Z. Q.; Lim, S. H.; Poh, C. K.; Tang, Z.; Xia, Z.; Luo, Z.; Shen, P. K.; Chua, D.; Feng, Y. P.; Shen, Z.; Lin, J. A Highly Order-Structured Membrane Electrode Assembly with Vertically Aligned Carbon Nanotubes for Ultra-Low Pt Loading PEM Fuel Cells. Adv. Energy Mater. 2011, 1 (6), 1205–1214. https://doi.org/10.1002/aenm.201100371.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Cathodes Thomas Merzdorf, Elisabeth Hornberger, Sebastian Ott, and Peter Strasser, Department of Chemistry, Chemical Engineering Division, Technical University of Berlin, Berlin, Germany © 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. This is a update of S.H. Bergens, M.E.P. Markiewicz, FUEL CELLS – PROTON-EXCHANGE MEMBRANE FUEL CELLS | Cathodes, Editor(s): Jürgen Garche, Encyclopedia of Electrochemical Power Sources, Elsevier, 2009, Pages 616–625, ISBN 9780444527455, https://doi.org/10.1016/B978-044452745-5.00224-0.

1 2 3 4 5 6 7 8 9 References

Introduction Cathode potential during operation How oxygen is converted into water at the cathode Cathode materials, structure and fabrication Platinum-based electrocatalyst Ionomer Carbon support materials Fabrication methods Conclusion

89 90 91 92 93 93 94 95 95 95

Abstract An introduction to cathodes in proton-exchange membrane fuel cells (PEMFCs) is presented. The topics covered include what cathodes in PEMFCs are, how they are made and how they work. After introducing the fundamentals, the mechanism on the atomic scale is explained. Furthermore, the different parts of the cathode catalyst layer are described as well as their interactions with each other. Lastly, the fabrication of the cathode is elucidated.

Glossary Active surface area Surface atoms electrochemically available for the reaction. Density functional theory Computational modelling method to study the electronic structure and properties of atoms, molecules and materials and is capable to predict reaction energies, transition states and reaction pathways. Ionomer poisoning Thin ionomer layer or larger agglomeration of ionomer on and around electrochemical active nanoparticles hindering the oxygen transport. Ostwald ripening Material transfer from smaller to larger particles, leading to particle growth. Oxygen transport resistance Measure for the impediment of the flow of oxygen toward the active sites. Polymer electrolyte membrane Selective barrier that only allows proton transport through in proton exchange membrane fuel cells (PEMFCs). Proton-coupled electron transfer Chemical reaction that involves the transfer of electrons and protons often in a concerted process. Triple phase boundary Location of contact between three different phases, gas, liquid and solid.

Key points

• • • • • • • •

88

Introduction of proton-exchange membrane fuel cells in terms of structure and functionality Explanation of the cathode potential during fuel cell operation Oxygen reduction mechanism with the impact of the catalyst, particle shape and particle composition Cathode structure in the PEMFC and reactant and product transport Platinum-based electrocatalyst features and degradation processes Ionomer interactions with the catalyst layer Carbon support material structure and improvements of the ionomer support and catalyst support interactions Presentation of different fabrication methods used for the preparation of MEAs

Encyclopedia of Electrochemical Power Sources, 2nd Edition

https://doi.org/10.1016/B978-0-323-96022-9.00158-4

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Cathodes

89

Abbreviations AFC CCM CCS DFT Eanode Ecathode Ecell GDL HOR MEA NASA ORR PEMFC RDE

1

Alkaline fuel cell Catalyst coated membrane Catalyst coated substrate Density functional theory Anode potential Cathode potential Cell potential Gas diffusion layer Hydrogen oxidation reaction Membrane electrode assembly National Aeronautics and Space Administration Oxygen reduction reaction Proton-exchange membrane fuel cell Rotating disk electrode

Introduction

This article provides an introduction to cathodes in proton-exchange membrane fuel cells (PEMFCs) and is intended for the reader with a background or working knowledge in science, and the technical expert in the field. As shown in Fig. 1A, a PEMFC that operates on hydrogen as fuel consists primarily of three components: the anode, where the hydrogen is split into protons (H+) and electrons (e−); a polymer electrolyte membrane that transports the protons through the cell from the anode to the cathode; and the cathode, where the protons, electrons, and oxygen (from air) combine to form the only

Electric Current

(A)

e–

Hydrogen Input

O2

H2

H2

Anode: 2 H2 4 H+ + 4e– 0 E = 0.00 V

(C)

Silicon gaskets

H2 e–

Excess Hydrogen Output

(B)

e–

Air Input

MEA Eight M6 bolts

e– H+

O2

Fuel Cell React ion: 2 H2 + O2 2 H2O Ecell = 1.23 V

O2 H2O

ExcessAir, Water and Heat Output

Bipolar plates Current collector plates

Electrolyte Cathode: O2 + 4 H+ + 4e– 2 H2O E0 = 1.23 V

(D)

Fig. 1 (A) Simplified fuel cell operational scheme. (B) Compartments of a fuel cell1 (C) Picture of a single cell fuel cell hardware (10 cm²) (outside). (D) Picture of a single cell fuel cell hardware (inside).

90

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Cathodes

chemical product of the fuel cell, water. The electrons travel through a circuit outside the PEMFC where their energy is used to do work, such as powering a light bulb or an electric vehicle. Compared to other types of fuel cells, hydrogen PEMFCs operate at relatively low temperatures, 1.0 M miles, respectively. Therefore, the fuel cell durability under real-life operating conditions is essential for the hydrogen-powered PEMFC systems to be commercially viable. One of the predominant material degradation mechanisms in PEMFC applications is the deterioration of carbon-supported platinum-based electrocatalysts. Carbon-supported (e.g., Vulcan or Ketjen black (KB)) platinum or platinum-alloy (e.g., PtCo) catalysts are used in PEMFCs to catalyze the hydrogen oxidation reaction (HOR) at the anode and the oxygen reduction reaction (ORR) at the cathode. Although some non-platinum catalyst systems have demonstrated impressive ORR activity, for mobile applications, mass transport and durability are still substantially limited compared to platinum-based catalysts. Fortunately, the total platinum loading in membrane-electrode assemblies (MEAs) of PEMFCs has been successfully reduced by more than an order of magnitude (to
0.17mgPt cm−2 at the cathode (1mgcm−2 ¼ 0.01kgm−2)) compared with the PEMFC technology in the early 1990s. Losses in power density due to suppressed oxygen transport were observed for cells with further reduced platinum loadings (to 25,000 h and the given numbers increase accordingly (the same applies to the numbers given below). ii) Open-circuit voltage: Time when no current is drawn from the stack and the cathode potential increases to 0.95 V versus RHE, which can occur just before fuel cell system shutdown or immediately after startup. The expected cumulative time spent at open-circuit voltage (OCV) condition will be over 100 h over a vehicle’s lifetime. iii) Load cycling: Automotive PEMFC systems operate with frequent power changes (e.g., between idling and full power), which translate into cell voltage changes. For fully dynamic, nonhybridized (i.e., no propulsion battery) automotive conditions, most frequently the cathode electrode voltage will cycle between 0.55 and 0.9 V versus RHE, and it is estimated to be about 300,000 times during a vehicle’s lifetime. iv) Startup/shutdown: Automotive fuel cell systems will experience frequent startup and shutdown events. If no countermeasures are taken, during vehicle shutdown, air in the environment can penetrate the anode compartment of the fuel cell stack leading to an air front displacing hydrogen; conversely, during startup after long shutdown periods, a hydrogen front will displace air in the anode compartment. The H2/air front developed in the anode, opposite an air-filled cathode, results in local cathode interfacial potential (solid phase to electrolyte phase potential difference) reaching as high as 1.5 V versus RHE. This has been shown to severely limit system lifetime. It is estimated that there are approximately 38,500 start/stop cycles during the life of automotive fuel cell systems. In the last years, operating technologies such as voltage cut-off and lean air operation have been demonstrated to effectively mitigate the drastic performance loss due to start/stop cycles. As the anode catalyst is exposed to a hydrogen environment in PEMFCs, the anode potential remains close to 0 V (vs RHE) in most of the transient operations. Therefore, anode catalyst degradation is not typically observed in PEMFC durability tests. The only factor that affects the anode is the cell reversal caused by global anode fuel starvation, and the damage done by it to the anode catalyst is very similar to that caused by the local anode fuel starvation to the cathode. Thus, in this article, we only discuss the various degradation mechanisms of cathode platinum (or platinum-alloy) catalysts and the carbon support in the context of automotive fuel cell applications. The performance decay induced by the degradation of catalyst and carbon support will be highlighted. It should be noted that the degradation mechanisms discussed in the following also apply to stationary PEMFC systems.

2 2.1

Activity loss of the kinetically active catalyst components Loss of active platinum surface area

During long-term fuel cell operation, a significant loss of active platinum surface area is generally observed, particularly for the cathode catalyst, which can be attributed to two simultaneously occurring processes caused by the small but finite platinum solubility (represented schematically in Fig. 1): (1) a nanometer-scale Ostwald-ripening process in which smaller platinum particles dissolve into the ionomer phase and redeposit on larger platinum particles on the carbon support that are separated by a few nanometers for a well-dispersed catalyst; and (2) a micrometer-scale diffusion process in which dissolved platinum ions diffuse in the ionomer phase and subsequently precipitate in the ionomer phase of the electrode or the membrane via reaction with hydrogen permeating through the membrane.1 In addition to the above two mechanisms, platinum active surface area loss may also be caused by particle aggregation, coalescence, and ripening as well as the corrosion of the carbon support. In the latter case, the interaction between platinum nanoparticles and the carbon support, which maintains the high dispersions of platinum particles, disappears because of carbon corrosion.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Catalysts: Life Limitations

H2 anode

Pt Pt Pt carbon Carbon Pt support Pt support Pt Pt Pt

Pt Pt Pt carbon Carbon Pt Pt support support Pt Pt Pt

Membrane

Cathode

Time

Pt Pt carbon Carbon support support

Time

Pt Pt carbon Carbon support Pt support Pt Pt

Diffusion medium (DM)

Pt 2+

H2 + Pt2+ p Pt + 2H+

99

Membrane/cathode |10 Pm Cathode/DM interface interface

Fig. 1 Schematic representations of the two different Pt area loss mechanisms occurring in the cathode electrode of a proton-exchange membrane fuel cell (PEMFC): (1) Pt dissolution (left-hand side) and Ostwald ripening on the carbon support (right-hand side) and (2) diffusion of dissolved Pt species into the ionomer phase and precipitation via reduction with crossover H2. Ferreira, P. J.; La, O0 . G. J.; Yang, S. H. et al. Instability of Pt/C Electrocatalysts in Proton-Exchange Membrane Fuel Cells – A Mechanistic Investigation. J. Electrochem. Soc. 2005, 152, A2256–A2271. Reproduced by permission of The Electrochemical Society.

12 Pm

Anode

Pt line

14.5 Pm

18 Pm

7 Pm

Cathode

Fig. 2 Cross-sectional scanning electron microscopy (SEM) photomicrograph of a membrane-electrode assembly (MEA) after operation for 2000 h at open-circuit voltage (OCV) conditions with H2/air at 80  C, 100% relative humidity (RH), and 150kPaabs. Gore series 5510 MEA with 0.4/0.4 mgPt cm−2 loading for anode/ cathode. Zhang, J.; Litteer, B. A.; Gu, W.; Liu, H.; Gasteiger, H. A. Effect of Hydrogen and Oxygen Partial Pressure on Pt Precipitation within the Membrane of PEMFCs. J. Electrochem. Soc. 2007, 154, B1006–B1011. Reproduced by permission of The Electrochemical Society.

The platinum catalyst is stable at low electrode potentials such as those of the anode electrode at normal operating conditions (0–0.05 V vs. RHE), whereas platinum dissolution poses significant concerns at high potentials occurring in the cathode electrode (
0.7–0.95 V vs RHE), particularly at idle or OCV conditions. Fig. 2 demonstrates a cross-sectional scanning electron microscopy (SEM) photomicrograph of an MEA after operation for 2000 h at OCV conditions with a cathode potential of
0.95 V versus RHE.2 It was found that a platinum band (metallic platinum deposits) is formed within the membrane, close to the membrane/cathode interface (
3–4 mm away from the cathode electrode; see the bright white line in Fig. 2). These platinum particles are electrically disconnected from the cathode electrode and, therefore, inactive in the electrochemical reaction. The mechanism of platinum dissolution and precipitation within the membrane is depicted schematically in Fig. 1 and is analogous to the precipitation of platinum inside the electrolyte matrix of phosphoric acid fuel cells (PAFCs) observed previously. The location of the platinum band is determined by the relative local flux of H2 and O2 within the membrane. It has been shown by both model calculations and experimental measurements that the platinum band location is the point where the ratio of the local flux of hydrogen versus oxygen is 2:1, which is the stoichiometry to form water. As mentioned in the Introduction, platinum dissolution is a transient process and mainly occurs if the electrode potential is continuously cycled between oxidizing and reducing potentials as in the case of load cycling. With the help of electrochemical

100

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Catalysts: Life Limitations

scanning flow cell (SFC) half-cell measurements directly coupled to an inductively coupled plasma mass spectrometer (ICP-MS), it was demonstrated that the amount of anodically (during a positive going potential sweep) dissolved platinum is strongly related to the number of low-coordinated surface sites, whereas cathodic platinum dissolution depends on the amount of oxide formed and the timescale. Typically, the amount of platinum dissolution in the cathodic sweep is significantly larger than in the anodic sweep. Thus, a mechanism for platinum dissolution was suggested based on a place exchange of oxygen atoms from surface to sub-surface positions (see Fig. 3).3 In an H2/air PEMFC, dynamic variation of cell current results in dynamic voltage cycling of the cathode electrode potential owing to the slow ORR kinetics. Besides the platinum loss into the ionomer/membrane phase, the platinum particle size can grow. Consequently, platinum catalysts lose active surface area also because of a growth in particle size. The active platinum surface area (APt, in units of m2 g−1 Pt ) is usually quantified from the H-adsorption/desorption region via cyclic voltammetry and decreases over time owing to platinum particle growth and platinum precipitation into the ionomer phase. In the last years, increasingly advanced analytical characterization techniques have been developed and employed to study the structural changes of the catalyst to unravel the mechanisms that lead to catalyst degradation on a fundamental level. In electrochemical half-cells, identical location–transmission electron microscopy (IL-TEM) and the above-mentioned SFC ICP-MS technique have been introduced. IL-TEM has been successfully applied to study PEMFC catalysts before and after accelerated stress tests (AST). This technique, however, is restricted to ex-situ studies. SFC ICP-MS measurements allow the online monitoring of metal dissolution. High-energy X-rays, by comparison, are a powerful tool to investigate the structure of the catalyst operando, i.e., under operating conditions while monitoring the performance, and can be applied both to electrochemical half-cells as well as MEA configurations. Due to the high penetration depth of high-energy X-rays, structural changes can be investigated at different length scales relevant to PEMFCs. To distinguish the different growth mechanisms, such as Ostwald ripening, particle aggregation, and sintering, a combination of small angle X-ray scattering (SAXS) and X-ray diffraction (XRD), (also called wide-angle X-ray scattering (WAXS)) can be used. SAXS is an X-ray scattering technique that probes material structures with relatively larger sizes (typically 1–100 nm) by detecting the X-rays scattered at small angles. In this range, SAXS allows to quantitatively follow changes in nanostructures in realistic reaction environments. Thereby, changes in particle size of PEMFC catalysts can be quantified under operation.4,5 XRD probes the atomic structure of crystalline materials. Diffraction is based on the interference of X-rays scattered by the atoms arranged periodically in a crystal lattice. With XRD, changes in lattice parameters and microstrain of PEMFC catalysts can be followed. An example of combining the two techniques is shown in Fig. 4, where it was demonstrated that the dominant degradation mode changes over time during an AST from aggregation to coalescence to ripening.6 The loss of active platinum surface area at the cathode electrode leads to cell voltage losses owing to reduced kinetics of the ORR, −2 in mA cm−2 ¼ 10−2Am−2)), defined as which is commonly quantified by monitoring the platinum-specific activity (i(0.9V) s Pt (1mA cm the current at a cathode potential of 0.9 V at given temperature and hydrogen and oxygen partial pressures, normalized by the

Fig. 3 Scheme of Pt dissolution due to the alternation of the platinum surface state during: (a) anodic polarization, above ca. 1.1 VRHE dissolution and passivation of the surface compete; and (b) cathodic polarization, during surface reduction below ca. 1.0 VRHE (re-)deposition and dissolution compete. Topalov, A. A.; Cherevko, S.; Zeradjanin, A. R.; Meier, J. C.; Katsounaros, I.; Mayrhofer, K. J. J. Towards a Comprehensive Understanding of Platinum Dissolution in Acidic Media. Chem. Sci. 2014, 5, 631–638. Reproduced by permission of the Royal Society of Chemistry.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Catalysts: Life Limitations

101

Fig. 4 (a) Ratio of particle size determined by SAXS to the crystallite size determined by XRD in a PEMFC cathode over the course of an accelerated stress test (AST). (c) Normalized quantity of microstrain lattice distortion calculated by XRD. (b) TEM images of the Pt catalyst from the fresh MEA and (d) after 5000 cycles of stepping between 0.6 and 1.10 V at 80  C. Martens, I.; Chattot, R.; Drnec, J. Decoupling Catalyst Aggregation, Ripening, and Coalescence Processes Inside Operating Fuel Cells. J. Power Sources 2022, 521, 230851. Reproduced by permission of Elsevier.

cathode active platinum surface area, APt. Although the platinum-specific activity is a fundamentally important catalyst property, analogous to the TOF in gas-phase catalysis, the more important catalyst property for engineering applications is the platinum mass 2 −1 ¼ 103 m 102 kg 10−1)). The latter is defined as the current at a cell voltage of 0.9 V at a given activity (im(0.9V) in Amg−1 Pt (1 m g temperature and hydrogen and oxygen partial pressures, normalized by cathode platinum loading. These two quantities are related by the active platinum surface area, APt, according to


(1) im ð0:9 VÞ A mgPt −1 ¼ is ð0:9 VÞ mA cmPt −2  A Pt m2 gPt −1  10−5 To quantify i(0.9V) and im(0.9V), it is generally necessary to correct the cell voltage for ohmic losses using, for example, high-frequency s resistance (HFR) measurements and to correct the current obtained at 0.9 V for the parasitic hydrogen crossover current, which derives from the hydrogen permeation through the membrane from the anode to the cathode side. The latter can be measured in a H2/N2-fed cell (anode/cathode) by applying approximately 0.5 V across the cell. Thus, as a catalyst degrades during PEMFC durability tests, the voltage loss associated with the ORR can be caused by a loss in electrochemical surface area, APt (i.e., fewer , owing to compositional or morphological changes of the catalyst (i.e., available sites for reaction), a loss in specific activity, i(0.9V) s reduced activity per surface atom, equivalent to a reduced TOF), or a combination of these. As is obvious from Eq. (1), the mass activity, im(0.9V), will be affected by both degradation phenomena.

3

Activity loss of platinum-alloy catalysts

In PEMFC testing, carbon-supported platinum-alloy catalysts (e.g., PtCo/C and PtNi/C) have been shown to have higher mass activity compared to pure Pt/C catalysts. This observation is analogous to the previously observed higher platinum-specific activity of polycrystalline platinum alloys Pt3M(M ¼ Co, Ni, etc) over polycrystalline platinum in aqueous electrolyte. It is also consistent with the observed enhanced activity of PtM/C over Pt/C in both liquid electrolyte and MEA fuel cell tests. In addition, PtM/C catalysts also showed enhanced stability in maintaining high platinum active surface area compared with pure Pt/C, which was initially believed to be related to the inherent stability of the alloy, but was later demonstrated to be primarily due to the heat treatment effect during PtM/C synthesis in addition to the inherently higher stability of the generally larger catalyst particles of PtM/C compared with Pt/C. The platinum-alloy catalysts are typically prepared by co-depositing or sequentially depositing platinum and the second metal from salt solutions onto the carbon-support surface, followed by annealing at high temperature (
900  C) to obtain the alloy form, leading to larger particle sizes.

102

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Catalysts: Life Limitations

With the demonstrated initial mass activity advantage and enhanced dissolution/sintering stability of PtM/C, these cathode catalysts are very attractive for PEMFC applications. Hence, current state-of-the-art PEMFCs for passenger fuel-cell vehicles use PtCo/C as the cathode catalyst. However, as can be seen in Fig. 5(a), PtCo/C catalysts are found to lose specific activity with time even during steady-state operation, amounting to a factor of
1.8 over 1000 h.7 Nevertheless, the mass activity advantage of PtCo/C over Pt/C is maintained in this durability experiment (see Fig. 5(b)). Specific activity degradation in platinum-alloy catalysts is associated to the slow loss of the second metal over time by leaching from the catalyst surface, leaving a Pt-enriched surface or skin. Cobalt was indeed found in the membrane at the end of the test of PtCo/C catalysts using electron probe microanalysis (EPMA). To confirm the specific and mass activity loss owing to the loss of cobalt from the catalyst surface, an ex-situ leaching experiment was conducted in 0.5 mol L−1 sulfuric acid at 80  C for 24 h. It was found in a later work that the amount of cobalt leached out from PtCo/C catalysts is linearly dependent upon the platinum surface area, indicating that cobalt is primarily removed from the surface layer of the particles. This is further supported by both X-ray diffraction (XRD) and extended X-ray absorption fine structure (EXAFS) analyses that reveal lattice parameter changes and the presence/absence of Co-O bonds. As the leaching of cobalt from PtCo alloys became apparent, experiments were conducted to compare the mass activity of leached PtCo/C with that of unleached PtCo/C and also with that of pure platinum sample with similar particle size (Fig. 6). It was found that the majority of the high activity associated with unleached PtCo/C alloys disappears and the activity of leached PtCo/C samples approaches that of pure Pt/C.8 The loss in specific activity observed in the steady-state experiments shown in Fig. 5 was also observed in voltage cycling experiments between 0.7 and 0.9 V versus RHE, which are thought to mimic the voltage cycles that occur during automotive operating conditions. Although the PtCo/C catalyst demonstrated the capability of maintaining the platinum active surface area, there is a significant mass and specific activity loss after 83,000 cycles. Despite these challenges, PtCo/C catalysts have been successfully implemented into commercial fuel cells and several reports have been published describing the development of the catalyst, its support as well as the whole stack, see for example the work of Borup et al.,9 Cullen et al.,10 or Yoshizumi et al.,11

4

Factors affecting platinum dissolution

As platinum dissolution is accelerated when the electrode is cycled between the oxide formation/oxide desorption potential range, operating conditions that influence the adsorption of platinum oxides would be expected to affect the platinum area loss. Reduced platinum surface area loss for Pt/C cathode electrodes was indeed observed at reduced relative humidity (RH), which may be is (PA cmpt2)

1000 h

200 h

0h

1000 h

700 h

200 h

700 h

PtCo/C

Pt/C

0h

500 450 400 350 300 250 200 150 100 50 0

(a)

im (A mgpt1)

0.30

PtCo/C

Pt/C

0.25 0.20 0.15

1000 h

700 h

200 h

0h

1000 h

700 h

0.00

0h

0.05

200 h

0.10

(b)

Fig. 5 Changes in (a) specific activity and (b) mass activity of
50 wt% Pt/C and
50 wt% PtCo/C cathode catalysts during a steady-state durability test in a full active area short stack (20cells) operated at 0.2 A cm−2 at 80  C, 150kPaabs, and 100% relative humidity (RH). Here, Pt-specific activity and mass activity are defined at 0.9 V, 80  C, and 100 kPaabs H2 and O2. Note: 1 A cm−2 ¼ 104 Am−2; 1 A mg−1 ¼ 106 A kg−1. Mathias, M. F.; Makharia, R.; Gasteiger, H. A. et al. Two Fuel Cell Cars in Every Garage. Interface 2005, 14(3), 24–35. Reproduced by permission of The Electrochemical Society.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Catalysts: Life Limitations

103

0.3

Mass activity A mgPt1 at 900 mV

Pt PtCo

0.25

Leached PtCo Pt trendline

0.2

PtCo trendline

0.15 0.1 0.05 0

0

10

20

30

40

50

60

70

80

90

100

MEA CO eca m2 gPt1

Fig. 6 Relationship of O2 mass activity at 900 mV of unleached PtCo/C, leached PtCo/C, and Pt/C catalysts as a function of membrane-electrode assembly (MEA) electrochemical area. Data are collected at 80  C with H2/O2 (2:10 stoichiometric flows) at 150kPaabs and 100% relative humidity (RH). ‘MEA CO eca m2 g−1 Pt ’ is the Pt electrochemical surface area measured by cyclic voltammetry using CO adsorption. Pt/C catalysts were annealed similarly to the PtCo/C catalysts to vary the Pt particle size. Ball, S. C.; Hudson, S. L.; Theobald, B. R. C.; and Thompsett, D. PtCo, a Durable Catalyst for Automotive Proton Electrolyte Membrane Fuel Cells? ECS Trans. 2007, 11(1), 1267–1278. Reproduced by permission of The Electrochemical Society.

attributed to a decrease in surface oxide formation at lower RH. Thus, PEMFC operation at low RH conditions could be a viable means of mitigating platinum dissolution, provided that sufficient proton conduction can be obtained at low RH conditions. Another factor to consider is the effect of voltage cycling ranges and dynamics on platinum dissolution rate. The platinum dissolution mechanism during voltage cycling has been investigated using SFC ICP-MS half-cell measurements. The amount of platinum dissolved is dependent on the anodic and cathodic sweep rate (i.e., the potential cycling pattern). The same was observed in the voltage cycling experiments with PEMFC MEAs. As a consequence, the question arises of how to design ASTs that substantially accelerate the observed catalyst degradation but which results at the same time can be extrapolated to “real life usage” of fuel cells. Typically one separates AST protocols to stress the active phase of the catalyst and AST protocols to induce carbon corrosion. Different protocols have been proposed by the US Department of Energy (DOE), and the Fuel Cell Commercialization Conference of Japan (FCCJ). A typical protocol to stress the active phase is to apply a square-wave potential sequence jumping the potential between 0.6 and 0.95 V or 1.0 V vs RHE. In the work of Stahira et al. it has been shown that by applying such a square-wave AST, the degradation could be accelerated by a factor of 5 in time over triangle-wave AST protocols.12 This observation is in agreement with half-cell measurements and can be explained by different degrees of oxidation of the catalyst during an AST cycle. It is important to note that not only the upper potential limit influences the observed degradation, but the lower potential limit as well. For example, it has been shown by Ronowski et al. that the stability of Pt/C and PtNiX/C catalysts was substantially higher if a square-wave AST between 0.7 and 0.95 V instead of the typical 0.6–0.95 V vs RHE was applied.13 With the help of operando XRD, it could be shown that this difference in degradation is related to an incomplete reduction of the catalysts in the case of 0.7 V vs RHE as the lower potential limit. Furthermore, applying cyclic voltammetry during or as part of an AST protocol (e.g., to determine the ECSA), leads to substantially increased degradation as well. Thus the question arises of how well the different AST protocols are reflected by a real fuel cell drive cycle. In addition, it has to be pointed out that the comparisons between the tests of different groups are difficult to compare as a proper iR compensation is essential to actually reach the desired upper and lower potential limits during the AST cycles. When applying thousands of AST cycles, even minor deviations in the reached potential limits can have a pronounced effect on the obtained results. Last but not least, also the catalyst conditioning or activation can have a substantial influence on the stability of PtM/C catalysts. Yet, there is no consensus on the applied protocols in the literature. For example, Chattot et al. showed using operando XRD in both liquid-electrolyte half cells and PEMFCs that for PtNi/C catalysts, the loss in ORR performance can be substantially mitigated by adjusting the initial chemistry and structure of the catalyst.14

5 5.1

Catalyst carbon-support corrosion Effect of carbon-support corrosion on MEA performance

Besides the dissolution/sintering of platinum (or platinum-alloy) catalysts discussed above, the corrosion of catalyst carbon supports is a critical issue in PEMFC system operations. In PAFCs which operate at
200  C, corrosion rates of the commonly

104

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Catalysts: Life Limitations

used high-surface-area carbon supports (e.g., Vulcan XC72 and KB) are unacceptably high, necessitating the implementation of more corrosion-resistant graphitized carbon supports. Because of the significantly lower operating temperature of PEMFCs (currently
80  C), it has largely been assumed that carbon-support corrosion would be negligible. This assumption is correct for cathode potentials in the normal PEMFC operating range. However, if automotive applications involve cathode potential ranges that far exceed the stability region of these carbon supports this can lead to pronounced degradation. As discussed in the Introduction, without mitigation, automotive fuel cell systems undergo an estimated 38,500 startup/shutdown cycles over the life of a vehicle, which can lead to short-term potential excursions of the cathode electrode to 1.2–1.5 V owing to H2/air fronts in the anode compartment. The carbon corrosion kinetics of 50 wt% Pt/C catalysts on various carbon supports has been determined at 95  C and a constant potential of 1.2 V, and is shown in the inset of Fig. 7. Under these accelerated carbon corrosion conditions, it takes 50 mV in performance within a narrow window of carbon weight loss for both the non-graphitized and graphitized carbon supports, at approximately 11  2%, even though the graphitized Vulcan support corrodes nearly 2 orders of magnitude slower than the KB support (inset of Fig. 7). It should be noted, though, that the percentage of carbon loss beyond which significant performance decay occurs also depends on the operating conditions at which the performance is evaluated. At any given carbon weight loss (in %) owing to corrosion, the performance loss increases with increasing RH and current density and with decreasing oxygen partial pressure. Therefore, this threshold carbon loss value should be preferably cited against given operating conditions and current densities. Similar to the fact that potential cycling accelerates the dissolution of platinum, it is found that dynamic voltage cycling also accelerates carbon-support corrosion. Experiments were conducted for voltage cycles between 0.6 and 0.9 V (square waves at 5 cycles h−1) and the results are shown in Fig. 8. It is observed that the carbon dioxide evolution rate (i.e., carbon corrosion rate) is much higher (
3–4 times) during voltage cycling than at constant potential hold at either 0.6 or 0.9 V. The carbon dioxide peaks correspond well with the potential step change and appear at both the upward and downward potential transients. This is similar to the carbon dioxide evolution that was also observed when the cathode feed streams were switched from nitrogen to air, in which case the cathode potential cycles between
0.1 and 0.95 V. The measured carbon corrosion rate and the cumulative carbon weight loss during voltage cycling and constant potential hold are demonstrated in Fig. 9. It is observed that the carbon corrosion rate during voltage cycling decreases dramatically with time, following the same power law dependency as observed under steady-state conditions. Clearly, voltage cycling alone can lead to a considerable amount of carbon weight loss for a cathode electrode with conventional carbon support, whereby the degradation rate is indicated to be
3-fold larger for voltage cycles between 0.6 and 0.9 V when compared with corrosion at a constant potential of 0.9 V (see Fig. 9(a)). Therefore, voltage cycling-induced carbon corrosion needs to be considered when evaluating the overall carbon loss in automotive fuel cell applications. The fundamental mechanism responsible for the accelerated carbon corrosion with potential cycling is not yet fully understood. It is generally attributed to the repeated reduction and oxidation of platinum and/or carbon surface groups during voltage cycles.

KB Vulcan Gr-KB Gr-Vulcan

20 40 100

60

Carbon weight loss (%)

Voltage loss at 0.8 A cm2 (mV)

0

80 100

1

120 140

10

1

10

100

1000

Time at 1.2 V, 95 qC

0

2

4

6

8

10

12

14

16

Carbon weight loss (%)

Fig. 7 Voltage loss vs. carbon weight loss of the cathode electrode measured at 0.8 A cm−2 with H2/air (2:2 stoichiometric flows) at 80  C, 175 kPaabs, and 50% inlet relative humidity (RH). Cathodes with Pt loadings of 0.4 mgPt cm−2 were based on 50 wt% Pt/C catalysts using conventional Vulcan XC72 (Vulcan) and Ketjen black (KB) supports as well as their fully graphitized analogs (Gr-Vulcan and Gr-KB). The inset shows the carbon weight loss vs. time under accelerated carbon corrosion conditions of 1.2 V and 95  C. Yu, P. T.; Gu, W.; Makharia, R.; Wagner, F. T.; Gasteiger, H. A. The Impact of Carbon Stability on PEM Fuel Cell Startup and Shutdown Voltage Degradation. ECS Trans. 2006, 3(1), 797–809. Reproduced by permission of The Electrochemical Society.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Catalysts: Life Limitations 0.5

1 0.9 V

0.9 0.8

0.4 0.35

0.7

0.6 V

0.3

0.6

0.25

0.5

0.2

0.4

0.15

0.3

0.1

0.2

0.05

0.1

0 1020

1030

1040

1050

1060

1070

Cell voltage (V)

CO2 concentration (ppm)

0.45

105

0 1080

Time (min) CO2 (0.6 V)

CO2 (0.9 V)

CO2 (0.6 V 0.9 V cycling)

Potential profile (0.6 V  0.9 V cycling)

Fig. 8 Carbon dioxide concentration during constant potential hold at either 0.6 or 0.9 V vs. reversible hydrogen electrode (RHE) as well as during voltage cycling between 0.6 and 0.9 V (5 cycles h−1). Experiments were conducted on a 50 cm2 active area single cell using a
50 wt% Pt/C-based membrane-electrode assembly (MEA) with 0.4/0.4 mgPt cm−2 loadings (anode/cathode). Test conditions: H2/N2 (1slpm flows) at 120 kPa, 80  C, and 80% RHinlet. The CO2 concentration was measured using a Midac Fourier transform infrared (FTIR) analyzer: the cathode outlet gas stream of the cell is vaporized at 120  C before admission into the FTIR chamber. Note: There is a time delay (
1–2 min) between the CO2 signal and the cell voltage profile owing to sampling delay in the FTIR.

6

Carbon-support corrosion during startup/shutdown cycles

Unmitigated startup/shutdown in vehicle operation has been identified as the most severe process leading to fuel cell degradation. Fig. 10 shows a schematic of a PEMFC during the transient of either a startup or a shutdown process. During vehicle shutdown, air from the environment diffuses into the anode compartment of the fuel cell stack, leading to an air front displacing hydrogen; and during a vehicle startup after a long shutdown period, a hydrogen front displaces air in the anode compartment. Because the in-plane proton transport resistance over hundreds of millimeters is significantly higher than that of the through-plane direction over FeOH2++H+ Fe3++H2O>Fe(OH)+2 +2H+ + 2Fe3++2HO2>Fe2(OH)4+ 2 +2H + Fe3++H2O2>Fe(HO)2+ +H 2 FeOH2++H2O2>Fe(OH)(HO2)++H+ 2+ 2+ Fe(HO2) !Fe +HO•2 • Fe(OH)(HO2)+!Fe2++ HO2 +OH− Fe(III)( )+HO•2 ! Fe2++O2+H+ Fe(III)+O•2 −! Fe2++O2 HO•2 +HO•2 !H2O2+O2 • • HO2 +O2 −+H+!H2O2+O2 • OH+HO•2 !H2O+O2 • OH+O•2 !OH• +O2 • OH+• OH!H2O2 − H++ SO2− 4 >HSO4 HSO−4 +• OH!H2O+SO•4 − SO•4 −+ H2O•2 !HSO−4 +HO•2 • • SO4 −+ HO2 !HSO−4 +O2 •− 2+ 3+ SO4 +Fe !SO2− 4 +Fe

63 3108 3.3107 1.2106 1.58105 11010 11017 K¼2.910−3 K¼7.6210−6 K¼810−4 K¼3.110−3 K¼2.010−4 2.710−3 2.710−3 100
C) in order to remove the residual traces of water and eventually the surfactant.

Hydrolysis Process The precursor polymer powder, after drying, is pelletized and melt-extruded into films of different thicknesses in a screw extruder at a temperature at least 30
C above the complete melting of the polymer as determined by DSC (depending on EW, different complete melting temperatures are observed). The film is then hydrolyzed with a process comprising a first step in hot, strong bases (usually potassium hydrooxide or sodium hydroxide) and a second step in strong acid. The first step is a nucleophilic substitution of second order (SN2) followed by a salification that gives the group –SO3Me from the –SO2F one. The SN2 reaction is relatively slow because of the high activation energy required for the formation of the transition stage –SO2FL OH; its rate is proportional to the concentration of the incoming nucleophile, the operating temperature, and the nucleophilic strength of the incoming group (in this case OH−).

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Membranes: Fluorinated

129

Figures 5 and 6, showing the section of a 50% hydrolyzed membrane, evidence that the reaction has a planar front and the very center of the membrane is the last hydrolyzed part; this means that the reaction is under kinetic control and the diffusion of reagents is probably not really determining. One can also notice in Figure 5 a change in morphology of the membrane once hydrolyzed; in fact, the precursor form is thermoplastic (and melt processable), the –SO3Me form is not. Figure 6 shows the profile of potassium contained in the membrane section of the same sample. It is important to notice that no potassium is present in the precursor form below the reaction front; it means that the reaction is a ‘surface’ reaction. It is well known in organic chemistry literature that the presence of polar aprotic solvents enhances the velocity of the SN2 reactions, probably because of the stabilization effect of the solvents on the counterions of the incoming nucleophile or because of the reduction of the intermediate formation activation energy; examples often reported of such solvents are dimethylacetamide (DMAc), dimethylformamide (DMF), and dimethyl sulfoxide. In Figure 7 the effect of these solvents is shown and the difference of almost 2 orders of magnitude in the reaction velocity is measured. Dimethyl sulfoxide is frequently used for this application, while DMAc and DMF show the limitation of decomposition in the presence of strong basis, for example: CH3CON(CH3)2+KOH!CH3COOK+(CH3)2NH. The control of complete conversion of the –SO2F groups is usually done by infrared analysis of the film.

−SO3K

173 μm

−SO2F

−SO3K

100 μm

Electron image 1

Figure 5 Scanning electron microscopy (SEM) section of partially hydrolyzed extruded membrane.

90 80 70 60 50 40 30 20 10 0 0

50

100 Potassium Ka1

Figure 6 Profile of potassium signal in section of partially hydrolyzed extruded membrane.

150

μm

130

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Membranes: Fluorinated Effect of polar aprotic solvents (80 °C) 120 KOH 14% − DMAc 30% KOH 14% − DMSO 30% KOH 14% − DMF 30% KOH 15%

Residual − SO2F (%)

100 80 60 40 20 0 0

2

4

6 Time (h)

8

10

12

Figure 7 Effect of polar aprotic solvents on the kinetics of hydrolysis of an extruded membrane. DMAc, dimethylacetamide; DMSO, dimethyl sulfoxide; DMF, dimethylformamide.

Short-Side-Chain Ionomer Properties Differential Scanning Calorimetry Analysis Differential scanning calorimetry experiments were performed using a PerkinElmer DSC7 calorimeter equipped with a refrigeration unit for controlled cooling to –20
C. Scans were made on samples weighing around 10 mg of SO3H form polymer in the temperature range of 20–280
C at a heating rate of 10
C min−1. Before measurement, all samples were pretreated at 150
C for 60 min. The reported data represent therefore the second polymer fusion; the EW of the SSC sample is 870, the LSC one is 1100 g per equivalent (Figure 8). A single glass transition temperature cannot be determined precisely, but a softening range has to be considered; the extrapolation of the glass transition temperature shows anyway a large difference (40
C) between SSC and LSC, which is crucial for high-temperature operation (see Mechanical Analysis and Dynamic Mechanical Spectroscopy). Similar DSC analyses (with a pretreatment of 15 min at 350
C) were implemented on the precursor form of the polymer; they were used to determine processing parameters and crystallinity level of these semicrystalline polymers. A lot of different EWs were examined and it can be seen that while the temperature of complete melting is well identified, the initial melt temperature is not clearly defined; this is in line with findings by M. R. Tant and coworkers. These authors had investigated the melting behavior by variable temperature WAXS, and detected by this method a single initial melt temperature of 130
C for different EW ionomer precursors. This temperature could therefore be used as a reference initial integration temperature for the calculation of the heat of fusion from the DSC curves. Qualitatively, measurements carried out on SSC Aquivion ionomer precursors confirm the higher

0.90

Normalized heat flow (W g−1)

0.85 Tg (half Cp extrapolated) = 65.63 °C

0.80 0.75

LSC ionomer

0.70 0.65 0.60 0.55 0.50

SSC ionomer Tg (half Cp extrapolated) = 104.60 °C

0.45 0.40 20

40

60

80

100

120 140 160 180 Temperature (°C)

200

220

240

260

Figure 8 Differential scanning calorimetry (DSC) spectrum of long-side-chain (LSC) and short-side-chain (SSC) polymers in acid form.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Membranes: Fluorinated

131

Heat of fusion (cal g−1)

5 SSC ionomer LSC ionomer

4

3

2

1

0 600

800

1000

1200

EW (g

1400

1600

eq−1)

Figure 9 Heat of fusion of precursor forms of long-side-chain (LSC) and short-side-chain (SSC) polymer depending on the equivalent weight (EW). Reproduced with permission from Ghielmi A, Vaccarono P, Troglia C, and Arcella V (2005) Proton exchange membranes based on the short-side-chain perfluorinated ionomer. Journal of Power Sources 145(2): 108–115.

crystalline content of SSC ionomers compared to LSC ionomer, as shown by the heat of fusion data reported in Figure 9. Note that these data must be considered only qualitative due to the uncertainties in selecting the baselines for the DSC spectra integration. No exact extrapolation to zero crystallinity is attempted due to this uncertainty. However, it can be seen that zero crystallinity occurs approximately around 700 g per equivalent of EW for SSC and between 900 and 1000 g per equivalent for LSC. Therefore, there is a wide region of EWs between 700 and 900 where LSC is completely amorphous while SSC still retains some crystallinity. In this EW range, SSC ionomers can be obtained with high conductivity and good mechanical properties and resistance to solubilization (due to crystallinity). The combination of these properties would not be possible in this EW interval with an LSC ionomer. More exact numbers (compared to the heat of fusion) can be reported for the temperature of complete melting. These data are shown as a function of EW in Figure 10. It can be seen that there is a good correlation between the EW and the Tcm, with Tcm increasing with EW. Due to the linear behavior of this dependency, a line has empirically been used to fit the data in Figure 10. The Tcm data obtained by M. R. Tant and coworkers by WAXS on Dow polymers have also been reported for comparison and are in good agreement with the data found in this work. Such an agreement can in principle be considered unexpected due to the different SFVE feed strategies used during the polymerization in the two cases (batch for the Dow ionomers and semi-batch for the Aquivion of the present work).

300 SSC ionomer (AquivionTM) SSC ionomer (DOW)

Tcm (°C)

280

260

240

220

200 600

700

800

900

1000

1100

1200

1300

1400

EW (g eq−1)

Figure 10 Temperature of complete melting, Tcm, of short-side-chain (SSC) precursor polymer (Aquivion and literature data of Dow polymer) depending on the EW. Reproduced with permission from Ghielmi A, Vaccarono P, Troglia C, and Arcella V (2005) Proton exchange membranes based on the short-side-chain perfluorinated ionomer. Journal of Power Sources 145(2): 108–115.

132

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Membranes: Fluorinated

Mechanical Analysis and Dynamic Mechanical Spectroscopy It has already been reported that SSC membranes of EW around 850 have similar mechanical properties compared to 1100-EW LSC, both in the dry and hydrated state. This is notable considering the higher water uptake of an 850-EW membrane (see Small-Angle X-ray Scattering Spectroscopy). It is believed that this coincidence in mechanical properties is related to equivalent levels of crystallinity, as indicated by the heat of fusion in Figure 9. Figure 11 shows the stress–strain analysis of a 50-mm SSC extruded membrane in comparison with an N112 sample. Higher domain orientation can be observed in the SSC sample (cf. membrane behavior in machine direction); the linear part of the curve indicates a substantially equal elastic module. The biggest difference in the mechanical properties between the SSC and the LSC ionomer can be observed when the temperature increases to values close to the glass transition temperature of the polymer. There are different analyses that are usually considered to evaluate the glass transition temperature of a polymer; most typical is the DSC analysis shown in Figure 8 where the extrapolated Tg are 104
C for SSC and 65
C for LSC. One other typical analysis is the dynamic mechanical scan. Measurements were carried out according to ASTM D4065 by means of a controlled strain mechanical spectrometer (Rheometrics ARES) equipped with a very sensitive transducer (low torque resolution: 0.02 g cm) that allows to perform measurements on very thin samples. The specimen has a rectangular shape (length 30 mm; width 10 mm) and is subjected to mechanical oscillations of constant amplitude and constant frequency (1 Hz). The mode of deformation is torsional. The damping and shear module are obtained from measurements of the torque that the sample generates in response to the applied deformation. The spectrometer is operated with a continuous flow of liquid nitrogen which enables the samples to be scanned from –150 to 200
C at a heating rate of 2
C min−1. Measurements are taken at 2
C intervals. The measurements were started after drying the specimens inside the spectrometer under a flow of dry nitrogen at 150
C for 30 min. Preliminary tests showed that this was the procedure giving the most highly reproducible results. Figure 12 shows the dynamic mechanical spectrum of the SSC and LSC ionomers. The relaxation phenomena that can be observed are g peak (occurring at the same temperature of PTFE), b peak (probably due to residual water), and a peak, commonly

80 °C dry, TD 35 30

SSC ionomer 850 EW 50 μm

Stress (MPa)

25 20 15 10

LSC ionomer 1100 EW 50 μm

5 0 0

50

100

(a)

150 Strain (%)

200

250

300

250

300

80 °C dry, MD 35 SSC ionomer 850 EW 50 μm

30

Stress (MPa)

25 20 15

LSC ionomer 1100 EW 50 μm

10 5 0 0

(b)

50

100

150

200

Strain (%)

Figure 11 Stress–strain analysis of a 50 mm short-side-chain (SSC) extruded membrane in comparison with an N112 membrane. LSC, long-side-chain.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Membranes: Fluorinated

1010

133

102

109

101

108

100

107

10−1

106 −200

−150

−100

−50

(a)

0

50

100

150

200

Damping

G′ (Pa)

SSC ionomer

10−2 250

Temp (°C)

1010

102

109

101

108

100

107

10−1

106 −200

(b)

−150

−100

−50

0

50

100

150

200

Damping

G′(Pa)

LSC ionomer

10−2 250

Temp (°C)

Figure 12 Dynamic mechanical spectroscopy (DMS) measurement of the short-side-chain (SSC) and long-side-chain (LSC) ionomers. Reproduced with permission from Arcella V, Ghielmi A, Merlo L, and Gebert M (2006) Membrane electrode assemblies based on perfluorosulfonic ionomers for an evolving fuel cell technology. Desalination 199: 6–8.

considered as the glass transition of the ionic phase of the material and in relation with the decrease in mechanical performance of the material. When comparing these data to those previously obtained in the literature, the great difference between the literature value of the a-transition temperature Ta and the value found in this work is immediately evident. This is true for both ionomer types. More in detail, G. A. Eisman reports a Ta of 165
C for the SSC and of 110
C for the LSC ionomer, against 130
C for the SSC and 70
C for the LSC as detailed in Figure 12. This temperature difference is probably due to the different hydration states in the literature works. As a matter of fact, the measurements carried out by G. A. Eisman were performed on samples which were water-soaked and quickly cooled to the starting measurement temperature – that is, highly hydrated samples. On the contrary, in the present work, samples have been thoroughly dried in nitrogen before the measurement in order to optimize reproducibility. Lower Ta values for dry samples are in line with decreasing Ta values observed on undried samples upon successive thermal cycling in the spectrometer. In spite of the difference found on Ta between this work and the literature, it must be noted that a difference of about 40–60
C between SSC and LSC still remains, very near to the difference found by G. A. Eisman. This difference suggests SSC membranes to be more suitable than LSC membranes for high-temperature FC operation (>100
C), as already indicated in the literature for the Dow membranes.

134

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Membranes: Fluorinated

Small-Angle X-ray Scattering Spectroscopy The SAXS typical scan in Figure 13 confirms the different orientation of the domains in machine and transverse direction of the SSC extruded membranes already noticed in the stress–strain characterization. Small-angle X-ray scattering spectroscopy analysis can be used as a third method suitable for estimating the glass transition temperature of the polymer by analyzing the polymer at different temperatures (starting from the lower one). The SAXS spectrum typically shows two peaks: One can be related with the ionic domains, the other one to the crystalline domains (also called peak of the ‘matrix’) at lower q, difficult to be detected when intensity is plotted on Y axis. When increasing temperature, the usual behavior of ionomers is the progressive disappearing of the ionomeric domains (most probably because of the water lost) and a progressive increase in intensity of the matrix domain’s peak. If an arbitrary temperature is chosen (in case of the analysis in Figure 14, the reference temperature is 50
C), it is possible to plot [Imax(T )/Imax(Tref)]1/2 versus temperature, where Imax corresponds to the value

1200 Average intensity X direction intensity 1000

Y direction intensity

800

600

400

200

0.1

1 q (nm−1)

Figure 13 Small-angle x-ray scattering (SAXS) scan of short-side-chain (SSC) extruded membrane.

1.50 LSC ionomer SSC ionomer

1.45 1.40

(lmin(T )/lmin (50 °C))112

1.35 1.30 1.25 1.20 1.15 1.10 1.05 1.00 0.95 0.90 0.85 0.80 20

40

60

80

100

120

140

160

180

200

T (°C)

Figure 14 Estimation of glass transition temperature of short-side-chain (SSC) and long-side-chain (LSC) membranes via small-angle X-ray scattering (SAXS) analysis.

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Membranes: Fluorinated

135

of scattering intensity of the domain matrix. Figure 14 shows that there is a temperature value where a change in slope of the curves is observed: it is believed that this point corresponds to the glass transition temperature of the ionomeric material because the material is free to move its polymeric chains and recrystallize. Also in this case it is shown that the SSC Tg is around 140
C, while that of the LSC one is between 100 and 110
C, one other time the gap is maintained.

Water Uptake The water absorption behavior of acid-form membranes from liquid water and saturated water vapor at 100
C was analyzed. Different EW SSC ionomers were studied and compared to LSC. Water uptake in liquid water at 100
C as a function of EW is reported in Figure 15 and compared to the absorption of LSC membranes as reported in the literature. The water absorption is notably lower for the SSC membranes at a given EW. Therefore, the same water uptake is obtained with a lower-EW SSC ionomer. A water uptake of 35–37%, typical of an 1100-EW LSC membrane, is obtained with an 900-EW SSC one. Typical hydration at 100
C for an 850-EW SSC is 45%. From the curve shown for SSC ionomers in Figure 15, it can be noted that at EW moving downward of 850, the water uptake curve becomes progressively steeper. Therefore, proper compositional control in polymerization is necessary in this interval of EWs to obtain membranes which have a reproducible water uptake behavior. Water uptake from liquid water has been compared to the uptake from saturated water vapor for both SSC and LSC membranes in the acid form; it is usual that the water uptake observed from vapor is lower than the one observed from liquid water at the same temperature (Schroeder paradox). Measurements have been performed on different EW SSC ionomers, and in Table 4 the ratio between the uptake from the liquid (DWliq) and the saturated vapor (DWvap) is reported. It can be seen that the Schroeder paradox is increasingly important at decreasing EWs. For high-EW SSC membranes, the uptake from liquid and saturated vapor is virtually identical. In terms of FC operation, this would mean that high-EW membranes would behave similarly in the presence or absence of condensed water in the cell, while the behavior of low-EW ionomers would be significantly different in the two cases, this difference increasing at decreasing EWs.

Short-Side-Chain Aquivion Membrane Characterization Ionomer Dispersion Preparation Ionomer precursor polymers similar to that described earlier for membranes were synthetized, recovered from the latex, washed, and dried. The obtained polymer powder was then hydrolyzed converting it to the salt form by immersion in a potassium hydroxide/water 10/90 w/w solution at 80
C for a time long enough to detect complete disappearance of the SO2F groups and then acidified in an excess amount of nitric acid. The polymer powder was then washed and dried. The polymer dispersion was then obtained by dissolving the polymer powder in an autoclave by a high-temperature process similar to what is described by Grot in patent US 4,433,082 and used for electrode ink production as described below.

120 SSC ionomer LSC ionomer

Water uptake (%)

100 80 60 40 20 0 600

700

800

900

1000

1100

1200

1300

1400

EW (g eq−1) Figure 15 Water uptake of short-side-chain (SSC) and long-side-chain (LSC) vs equivalent weight (EW). Reproduced with permission from Ghielmi A, Vaccarono P, Troglia C, and Arcella V (2005) Proton exchange membranes based on the short-side-chain perfluorinated ionomer. Journal of Power Sources 145(2): 108–115.

136

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Membranes: Fluorinated Table 4

Ratio between liquid (DWliq) and the saturated vapor (DWvap) hydration for different EWs

Membrane

EW (geq−1)

DWliq/DWvap

SSC ionomer membrane

670 770 825 1170 1350 1100

2.5 1.8 1.8 1.0 1.1 1.8

LSC ionomer membrane

SSC, short-side-chain; LSC, long-side-chain. Reproduced with permission from Arcella V, Troglia C, and Ghielmi A (2005) Hyflon ion membranes for fuel cells. Industrial & Engineering Chemistry Research 44: 7646–7651.

Membrane–Electrode Assembly Preparation and Cell Assembly The testing apparatus consists of 25 cm2 single cells with triple serpentine pattern flow fields, mounted on 50 W test stations. The SSC membrane thickness was 50 mm and the EW was 870 g per equivalent. The electrodes were fabricated on a PTFE support by casting hydro-alcoholic inks produced from ionomer dispersion 830 EW and a commercial Pt/C 50% by weight on Vulcan XC-72. The casting blade height was adjusted in order to have a platinum load of 0.25 mg cm−2 of platinum on both anode and cathode electrodes. The two electrodes were transferred by high-temperature ‘decal’ onto the membrane; the cells were assembled using two commercial carbon felt gas diffusion layers 0.4 mm thick, with a thin ‘micro-diffusion layer’ on one side. The gas diffusion layers were assembled with the micro-diffusion layers facing the electrodes. The cells were closed with a torque of 5 N m on each of the eight tie rods. Rigid gaskets 0.26 mm thick were present on both anode and cathode side. The membrane-electrode assemblies (MEAs) were protected with a thin, rigid subgasket on both sides before assembly; the active area was reduced to 20 cm2 by the presence of the subgasket. Before starting the test, the MEAs were conditioned for 8 h in the following operating conditions: Fixed current: 1000 mA cm−2 Cell temperature: 75
C Air flow: 1300 sccm (2.5 bara, dew point 80
C) Hydrogen flow: 650 sccm (2.5 bara, dew point 80
C)

Chemical Stability Analysis It is quite usual, in the ionomers’ development, to test the chemical stability of the ionomers with the so-called Fenton test. This consists in immersion of an acid form membrane sample in a solution containing hydrogen peroxide and ferric ions. The following Fenton test conditions were adopted: Iron salt: Fe(NH4)2(SO4)2 Iron concentration: 36 ppm of Fe2+ in solution Hydrogen peroxide concentration: 15 or 30% by volume Bath temperature: 55
C Reaction time (without inserting fresh hydrogen peroxide): 6 h Membrane sample weight: 1 g The reagents are mixed at ambient temperature by correcting the pH of the Fe2+ solution with sulfuric acid in order to reach a pH level below 3 before the addition of hydrogen peroxide. The membrane sample is then added and the solution is heated to 55
C. Six hours of reaction are considered to start when the set temperature of 55
C is reached. The fluoride level is measured with an ion-selective electrode after the removal of the nonreacted hydrogen peroxide and the addition of a buffer. As shown from the error bars in Figure 16, the Fenton tests gave quite spread results (the SSC values are the average of 30 tests). The fluoride release value of SSC extruded membrane (Aquivion E87-05) is aligned to the value obtained on the LSC (Nafion N112) membrane, the cast LSC membrane (NR112) showing a slightly lower value. It might be questioned whether the lower value of the cast membrane reflects an actual lower tendency to degradation or could be affected by the apparently different surface behavior of the cast membrane (lower hydrophilic character) that could slow down the Fenton reaction. A substantially lower fluoride release is observed by testing a stabilized SSC membrane, named Aquivion E87-05 S. In order to check the evolution in time of the degradation of the polymer when subject to peroxide radical attack, a second kind of test was carried out by extracting the membrane from the Fenton solution every 6 h, purifying it by acid treatment and washing, and reinserting the membrane in fresh Fenton reagents. The fluoride amount in the different solutions obtained in this way was measured. The result of this test, shown in Figure 17, is a linear increase of the cumulative fluoride emissions with time for both

Fuel Cells – Polymer-Electrolyte Membrane Fuel Cell | Membranes: Fluorinated

137

Fluoride emission in mg F g−1 of polymer

2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 SSC extruded

SSC