Sodium-Ion Batteries: Technologies and Applications 9783527350612

Table of contents :
Cover
Half Title
Sodium-Ion Batteries: Technologies and Applications
Copyright
Contents
Preface
1. Introduction
1.1 Overview
1.2 The Birth and Development of Sodium-ion Batteries
References
2. Characteristics of Sodium-ion Batteries
2.1 Basic Features
2.2 Working Principle
2.3 Concepts and Equations
2.3.1 Cell Voltage
2.3.1.1 Electromotive Potential
2.3.1.2 Theoretical Voltage EΘ
2.3.1.3 Open Circuit Voltage Eocv
2.3.1.4 Operating Voltage Ecc
2.3.1.5 Cutoff Voltage
2.3.2 Cell Capacity and Specific Capacity
2.3.2.1 Theoretical Capacity (Co)
2.3.2.2 Actual Capacity (C)
2.3.2.3 Rated Capacity (Cr)
2.3.2.4 Specific Capacity (Cm or CV)
2.3.3 Cell Energy and Specific Energy
2.3.3.1 Theoretical Energy (Wo)
2.3.3.2 Actual Capacity (W)
2.3.3.3 Specific Capacity (Wm or Wv)
2.3.4 Cell Power and Specific Power
2.3.5 Charge and Discharge Rate
2.3.6 Constant Current Charge and Discharge
2.3.7 Constant Voltage Charge
2.3.8 Coulombic Efficiency
2.3.9 Energy Conversion Efficiency
2.3.10 Cell Internal Resistance
2.3.11 Cell Life
2.3.12 State of Charge (SOC)
2.3.13 Depth of Discharge (DOD)
2.4 Structural Composition
2.4.1 Cathode Materials
2.4.2 Anode Materials
2.4.3 Electrolytes
2.4.4 Separators, Binders, Conductive Agents, and Current Collectors
References
3. Cathode Materials of SIBs
3.1 Polyanion Cathode
3.1.1 Phosphates
3.1.1.1 Olivine‐type Phosphates (NaMPO4, MFe, Mn, etc.)
3.1.1.2 NASICON‐type Phosphates (Na3M2(PO4)3, MTi, V, Ni, Fe, Mn, etc.)
3.1.1.3 Pyrophosphate Na2MP2O7
3.1.2 Sulfates/Borates/Silicates
3.1.2.1 Sulfates
3.1.2.2 Borates
3.1.2.3 Silicate
3.1.3 Mixed Polyanions
3.1.3.1 Fluorophosphates
3.1.3.2 Mixed Phosphates
3.2 Oxide Cathode
3.2.1 Layered Transition Metal Oxides
3.2.1.1 Structural Classification
3.2.1.2 Key Issues of Layered Oxides
3.2.1.3 P2‐type Layered Oxides
3.2.1.4 O3‐type Layered Oxides
3.2.1.5 P3‐type Layered Oxides
3.2.1.6 Mixed‐phase Layered Oxides
3.2.2 Tunnel‐type Oxides
3.2.2.1 NaxMnO2
3.2.2.2 Nax[MnM]O2 (M=Ti, Fe, Co, etc.)
3.2.2.3 Tunnel Oxides for Aqueous SIBs
3.3 Prussian Blue and their Analogues
3.3.1 Prussian Blue in Non‐Aqueous SIBs
3.3.1.1 Iron Hexacyanoferrate (FeHCF)
3.3.1.2 Manganese Hexacyanoferrate (MnHCF)
3.3.1.3 Cobalt Hexacyanoferrate (CoHCF)
3.3.1.4 Nickle Hexacyanoferrate (NiHCF)
3.3.1.5 Other Hexacyanoferrates
3.3.1.6 Other Metal Hexacyanometallic Compounds
3.3.2 Prussian Blue in Aqueous SIBs
3.3.2.1 Single‐Redox‐Center PBAs
3.3.2.2 Two‐Redox‐Center PBAs
3.3.2.3 All‐PBA Aqueous Batteries
3.4 Perovskite Transition Metal Fluorides
3.4.1 Metal Fluorides
3.4.2 Sodium Metal Fluorides
3.5 Organic Cathode
3.5.1 Working Mechanism
3.5.2 Carbonyl Small Molecules
3.5.3 Conductive Polymers
References
4. Anode Materials of Sodium-ion Batteries
4.1 Carbon‐based Anode
4.1.1 Graphite Anode
4.1.2 Soft Carbon
4.1.3 Hard Carbon
4.1.3.1 The Doping of Heteroatoms
4.1.3.2 Structure and Morphology Designing
4.2 Titanium‐based Anode
4.2.1 The Exploring of TiO2 Samples
4.2.2 The Exploring of TiS2 and TiSe2 Samples
4.2.3 The Exploring of Other Ti‐based Samples
4.3 Conversion Anode
4.3.1 Co‐based Samples
4.3.1.1 The Exploring of Co‐based Oxides
4.3.1.2 The Exploring of Co‐based Sulfides and Selenides
4.3.1.3 The Exploring of Co‐based Phosphide
4.3.2 Ni‐based Samples
4.3.2.1 The Exploring of Ni‐based Oxides/Sulfides
4.3.2.2 The Exploring of Ni‐based Selenium, Phosphide, and Other Samples
4.3.3 Fe‐based Samples
4.3.3.1 The Exploring of Fe‐based Oxides
4.3.3.2 The Exploring of Fe‐based Sulfides and Selenides
4.3.3.3 The Exploring of Fe‐based Phosphides
4.3.3.4 The Exploring of Other Fe‐based Composites
4.3.4 Mo‐based Samples
4.3.4.1 The Exploring of Mo‐based Oxides
4.3.4.2 The Exploring of Mo‐based Sulfide and Selenides
4.3.4.3 The Exploring of Other Mo‐based Composites
4.3.5 Other Metal‐based Samples
4.3.5.1 The Exploring of Zn‐based Samples
4.3.5.2 The Exploring of Cu‐based Samples
4.3.5.3 The Exploring of Mn‐based Samples
4.3.5.4 The Exploring of Cr‐based Composites
4.3.5.5 The Exploring of W‐based Composites
4.3.5.6 The Exploring of V‐based Composites
4.3.5.7 The Exploring of Nb‐based Composites
4.3.5.8 The Exploring of In‐based Samples
4.4 Metal/Alloy Anode
4.4.1 Sb‐based Samples
4.4.1.1 The Exploring of Sb and Sb‐based Alloy Samples
4.4.1.2 The Exploring of Sb‐based Oxide, Sulfides, Selenium
4.4.2 Sn‐based Samples
4.4.2.1 The Exploring of Sn‐based Alloys and Sn@Carbon Materials
4.4.2.2 The Exploring of Sn‐based Oxides
4.4.2.3 The Exploring of Sn‐based Sulfides
4.4.2.4 The Exploring of Sn‐based Selenide, Phosphide
4.4.3 Bi‐based Samples
4.4.4 Ge‐based Samples
4.4.4.1 The Exploring of Ge and the Relative Alloying Materials
4.4.4.2 The Exploring of Ge‐based Oxides Samples
4.4.4.3 The Exploring of Other Ge‐based Samples (GeX, X=Se, S, OH, P)
References
5. Electrolyte, Separator, Binder and Other Devices of Sodium Ion Batteries
5.1 Introduction
5.2 Organic Liquid Electrolytes
5.2.1 Physical and Chemical Properties
5.2.2 Organic Solvents
5.2.2.1 Ester‐based Solvents
5.2.2.2 Ether‐based Solvents
5.2.3 Electrolyte Salt
5.2.4 Electrolyte Additives
5.2.4.1 Film Formation Additives
5.2.4.2 Flame Retardant Additives
5.2.4.3 Overcharge Protection Additives
5.2.4.4 Additives with Other Functions
5.2.5 New Electrolyte Systems
5.3 Solid State Electrolytes
5.3.1 Physical and Chemical Properties
5.3.2 Inorganic Solid Electrolyte
5.3.2.1 β‐alumina
5.3.2.2 NASICON
5.3.2.3 Sulfides
5.3.3 Polymer Electrolyte
5.3.3.1 Solid Polymer Electrolytes (SPEs)
5.3.3.2 Gel Polymer Electrolytes (GPEs)
5.3.4 Composite Solid Electrolyte
5.3.4.1 CSEs with Passive Fillers
5.3.4.2 CSEs with Active Fillers
5.3.5 Phase Interface Between Electrode and Electrolyte
5.3.5.1 Solid Electrolyte Interphase (SEI)
5.3.5.2 Cathode Electrolyte Interphase (CEI)
5.4 Separator
5.4.1 Glass Fiber
5.4.2 Polyolefin Separator
5.4.3 Nonwoven Separator
5.5 Binder
5.5.1 Poly(vinylidene fluoride) (PVDF)
5.5.2 Polyacrylic Acid (PAA)
5.5.3 Sodium Alginate (SA)
5.5.4 Sodium Carboxymethyl Cellulose (CMC)
5.5.5 Crosslinked Binders
5.5.6 Conductive Binders
5.5.7 Self‐healing Binders
5.6 Conductive Agent
5.6.1 Carbon Black
5.6.1.1 Acetylene Black (AB)
5.6.1.2 Super‐P (SP)
5.6.1.3 Ketjen Black (KB)
5.6.2 Graphene
5.6.3 Carbon Nanofibers (CNFs)
5.6.4 Carbon Nanotubes (CNTs)
5.7 Current Collector
5.7.1 Metal‐based Current Collector
5.7.2 Carbon‐based Current Collector
5.8 Conclusion and Perspectives
References
6. Advanced Characterization Techniques and Theoretical Calculation
6.1 Imaging and Microscopy
6.1.1 Fundamentals of Imaging and Microscopy
6.1.2 Electron Microscopy Studies of SIBs
6.1.3 Synchrotron X‐Ray Imaging Studies of SIBs
6.1.4 Neutron Imaging Studies of SIBs
6.1.5 Scanning Probe Microscopy Studies of SIBs
6.1.6 Optical Microscopy Studies of SIBs
6.2 Synchrotron Radiation X‐Ray Diffraction Technique
6.2.1 Principles of XRD
6.2.2 Characteristics of XRD
6.2.3 XRD studies of SIBs
6.2.4 Challenges and Opportunities
6.3 Synchrotron Radiation X‐ray Absorption Spectroscopy Technique
6.3.1 Principles of XAS
6.3.2 Characteristics of XAS
6.3.3 XAS Studies of SIBs
6.3.4 Challenges and Opportunities
6.4 Solid‐state Nuclear Magnetic Resonance Spectroscopy
6.4.1 Principles of ssNMR
6.4.2 NMR Interactions and Shift Ranges for Battery Materials
6.4.2.1 Shift Interactions (Nuclear Spin−Electron Spin)
6.4.2.2 Dipolar Coupling (Nuclear Spin−Nuclear Spin)
6.4.2.3 Quadrupolar Coupling
6.4.3 ssNMR Studies of SIBs
6.4.4 The Challenge of NMR Detection
6.5 Electrochemical Test Techniques
6.5.1 Cyclic Voltammetry
6.5.2 Galvanostatic Charge–Discharge
6.5.3 Electrochemical Impedance Spectroscopy
6.5.4 Other Electrochemical Testing Techniques
6.5.5 Electrochemical Analysis of SIBs
6.6 Other Characterization Techniques
6.6.1 Neutron Diffraction Technique
6.6.2 Fourier Transform Infrared Spectrometry
6.6.3 Raman
6.7 Theoretical Calculation
6.7.1 Classical Molecular Dynamics
6.7.2 Ab Initio Molecular Dynamics
6.7.3 Machine‐learning Molecular Dynamics
6.7.4 Applications of Theoretical Calculations
References
7. Practical Application of SIBs
7.1 Introduction
7.2 Commercial Sodium Battery
7.2.1 High‐Temperature Na–S Battery
7.2.2 Sodium–Nickel Chloride Battery
7.3 Design and Manufacture Process of SIBs
7.3.1 Laboratory Button Battery Assembly
7.3.1.1 Metal Na Anode Materials
7.3.1.2 Button Cell Assembly Order
7.3.1.3 The Matching of Positive and Negative Electrodes
7.3.2 Type of Cell for SIBs
7.3.2.1 Cylindrical Battery
7.3.2.2 Soft‐pack Battery
7.3.2.3 Prismatic Battery
7.3.3 Design Requirements for Cell
7.3.3.1 Basic Design Principles
7.3.3.2 Safety Design
7.3.4 Manufacturing Process of SIBs
7.3.4.1 Front‐end Electrode Fabrication Process
7.3.4.2 Back‐end Assembly Process
7.3.4.3 Formation and Sorting Process
7.3.4.4 Design of SIBs Pack
7.3.4.5 Battery Management System
7.4 Presodiation Techniques
7.4.1 EC/Chemical Methods
7.4.1.1 EC
7.4.1.2 Chemical Methods
7.4.2 Self‐sacrificial Additive
7.4.3 Other Novel Methods of Presodiation
7.4.4 Factors Need to be Improved
7.5 Performance Tests and Failure Analysis
7.5.1 Electrochemical Performances Test
7.5.2 Safety Performances Test
7.5.3 Failure Phenomenon
7.5.4 Failure Analysis Method
7.5.5 Cost Estimation
7.6 Commercial Application and Future Perspectives
7.6.1 Current State of Commercialization of SIBs
7.6.2 Application Prospect
7.6.2.1 Low‐Speed Electric Vehicle Market
7.6.2.2 Large‐scale ESSs
References
Index

Citation preview

Sodium-Ion Batteries

Sodium-Ion Batteries Technologies and Applications

Edited by Xiaobo Ji, Hongshuai Hou, and Guoqiang Zou

Editors Prof. Xiaobo Ji

Central South University College of Chemistry and Chemical Engineering No. 932, Lushan South Road YueLu District 410083 Changsha China

All books published by WILEY-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate. Library of Congress Card No.: applied for

Prof. Hongshuai Hou

British Library Cataloguing-in-Publication Data

Central South University College of Chemistry and Chemical Engineering No. 932, Lushan South Road YueLu District 410083 Changsha China

A catalogue record for this book is available from the British Library.

Prof. Guoqiang Zou

Central South University College of Chemistry and Chemical Engineering No. 932, Lushan South Road YueLu District 410083 Changsha China Cover Image: © Ohoishi/Shutterstock

Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at . © 2024 WILEY-VCH GmbH, Boschstraße 12, 69469 Weinheim, Germany All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law. Print ISBN: 978-3-527-35061-2 ePDF ISBN: 978-3-527-84166-0 ePub ISBN: 978-3-527-84167-7 oBook ISBN: 978-3-527-84168-4 Typesetting:

Straive, Chennai, India

v

Contents Preface xiii 1 1.1 1.2

2 2.1 2.2 2.3 2.3.1 2.3.1.1 2.3.1.2 2.3.1.3 2.3.1.4 2.3.1.5 2.3.2 2.3.2.1 2.3.2.2 2.3.2.3 2.3.2.4 2.3.3 2.3.3.1 2.3.3.2 2.3.3.3 2.3.4 2.3.5 2.3.6 2.3.7 2.3.8

Introduction 1 Jinqiang Gao, Wentao Deng, Guoqiang Zou, Hongshuai Hou, and Xiaobo Ji Overview 1 The Birth and Development of Sodium-ion Batteries 4 References 8 Characteristics of Sodium-ion Batteries 11 Haoji Wang, Wentao Deng, Hongshuai Hou, Guoqiang Zou, and Xiaobo Ji Basic Features 11 Working Principle 14 Concepts and Equations 15 Cell Voltage 16 Electromotive Potential 16 Theoretical Voltage EΘ 16 Open Circuit Voltage Eocv 16 Operating Voltage Ecc 16 Cutoff Voltage 16 Cell Capacity and Specific Capacity 17 Theoretical Capacity (Co ) 17 Actual Capacity (C) 17 Rated Capacity (Cr ) 18 Specific Capacity (Cm or CV ) 18 Cell Energy and Specific Energy 18 Theoretical Energy (W o ) 18 Actual Capacity (W) 18 Specific Capacity (W m or W v ) 18 Cell Power and Specific Power 19 Charge and Discharge Rate 19 Constant Current Charge and Discharge 19 Constant Voltage Charge 19 Coulombic Efficiency 19

vi

Contents

2.3.9 2.3.10 2.3.11 2.3.12 2.3.13 2.4 2.4.1 2.4.2 2.4.3 2.4.4

Energy Conversion Efficiency 20 Cell Internal Resistance 20 Cell Life 20 State of Charge (SOC) 20 Depth of Discharge (DOD) 20 Structural Composition 20 Cathode Materials 21 Anode Materials 23 Electrolytes 24 Separators, Binders, Conductive Agents, and Current Collectors 25 References 26

3

Cathode Materials of SIBs 29 Xu Gao, Wentao Deng, Guoqiang Zou, Hongshuai Hou, and Xiaobo Ji Polyanion Cathode 30 Phosphates 31 Olivine-type Phosphates (NaMPO4 , M=Fe, Mn, etc.) 31 NASICON-type Phosphates (Na3 M2 (PO4 )3 , M=Ti, V, Ni, Fe, Mn, etc.) 33 Pyrophosphate Na2 MP2 O7 35 Sulfates/Borates/Silicates 36 Sulfates 36 Borates 37 Silicate 37 Mixed Polyanions 38 Fluorophosphates 38 Mixed Phosphates 42 Oxide Cathode 43 Layered Transition Metal Oxides 43 Structural Classification 43 Key Issues of Layered Oxides 46 P2-type Layered Oxides 56 O3-type Layered Oxides 60 P3-type Layered Oxides 64 Mixed-phase Layered Oxides 64 Tunnel-type Oxides 67 Nax MnO2 67 Nax [MnM]O2 (M=Ti, Fe, Co, etc.) 69 Tunnel Oxides for Aqueous SIBs 70 Prussian Blue and their Analogues 70 Prussian Blue in Non-Aqueous SIBs 72 Iron Hexacyanoferrate (FeHCF) 72 Manganese Hexacyanoferrate (MnHCF) 73 Cobalt Hexacyanoferrate (CoHCF) 75 Nickle Hexacyanoferrate (NiHCF) 75

3.1 3.1.1 3.1.1.1 3.1.1.2 3.1.1.3 3.1.2 3.1.2.1 3.1.2.2 3.1.2.3 3.1.3 3.1.3.1 3.1.3.2 3.2 3.2.1 3.2.1.1 3.2.1.2 3.2.1.3 3.2.1.4 3.2.1.5 3.2.1.6 3.2.2 3.2.2.1 3.2.2.2 3.2.2.3 3.3 3.3.1 3.3.1.1 3.3.1.2 3.3.1.3 3.3.1.4

Contents

3.3.1.5 3.3.1.6 3.3.2 3.3.2.1 3.3.2.2 3.3.2.3 3.4 3.4.1 3.4.2 3.5 3.5.1 3.5.2 3.5.3

Other Hexacyanoferrates 77 Other Metal Hexacyanometallic Compounds 77 Prussian Blue in Aqueous SIBs 79 Single-Redox-Center PBAs 79 Two-Redox-Center PBAs 80 All-PBA Aqueous Batteries 81 Perovskite Transition Metal Fluorides 82 Metal Fluorides 82 Sodium Metal Fluorides 84 Organic Cathode 85 Working Mechanism 86 Carbonyl Small Molecules 88 Conductive Polymers 89 References 90

4

Anode Materials of Sodium-ion Batteries 109 Peng Ge, Shaohui Yuan, Guoqiang Zou, Hongshuai Hou, Yue Yang, and Xiaobo Ji Carbon-based Anode 109 Graphite Anode 110 Soft Carbon 111 Hard Carbon 112 The Doping of Heteroatoms 112 Structure and Morphology Designing 114 Titanium-based Anode 116 The Exploring of TiO2 Samples 116 The Exploring of TiS2 and TiSe2 Samples 117 The Exploring of Other Ti-based Samples 118 Conversion Anode 118 Co-based Samples 118 The Exploring of Co-based Oxides 118 The Exploring of Co-based Sulfides and Selenides 119 The Exploring of Co-based Phosphide 120 Ni-based Samples 121 The Exploring of Ni-based Oxides/Sulfides 122 The Exploring of Ni-based Selenium, Phosphide, and Other Samples 122 Fe-based Samples 123 The Exploring of Fe-based Oxides 124 The Exploring of Fe-based Sulfides and Selenides 124 The Exploring of Fe-based Phosphides 126 The Exploring of Other Fe-based Composites 127 Mo-based Samples 128 The Exploring of Mo-based Oxides 128 The Exploring of Mo-based Sulfide and Selenides 129

4.1 4.1.1 4.1.2 4.1.3 4.1.3.1 4.1.3.2 4.2 4.2.1 4.2.2 4.2.3 4.3 4.3.1 4.3.1.1 4.3.1.2 4.3.1.3 4.3.2 4.3.2.1 4.3.2.2 4.3.3 4.3.3.1 4.3.3.2 4.3.3.3 4.3.3.4 4.3.4 4.3.4.1 4.3.4.2

vii

viii

Contents

4.3.4.3 4.3.5 4.3.5.1 4.3.5.2 4.3.5.3 4.3.5.4 4.3.5.5 4.3.5.6 4.3.5.7 4.3.5.8 4.4 4.4.1 4.4.1.1 4.4.1.2 4.4.2 4.4.2.1 4.4.2.2 4.4.2.3 4.4.2.4 4.4.3 4.4.4 4.4.4.1 4.4.4.2 4.4.4.3

The Exploring of Other Mo-based Composites 130 Other Metal-based Samples 130 The Exploring of Zn-based Samples 130 The Exploring of Cu-based Samples 131 The Exploring of Mn-based Samples 132 The Exploring of Cr-based Composites 133 The Exploring of W-based Composites 133 The Exploring of V-based Composites 133 The Exploring of Nb-based Composites 134 The Exploring of In-based Samples 135 Metal/Alloy Anode 135 Sb-based Samples 135 The Exploring of Sb and Sb-based Alloy Samples 135 The Exploring of Sb-based Oxide, Sulfides, Selenium 137 Sn-based Samples 138 The Exploring of Sn-based Alloys and Sn@Carbon Materials 139 The Exploring of Sn-based Oxides 141 The Exploring of Sn-based Sulfides 142 The Exploring of Sn-based Selenide, Phosphide 142 Bi-based Samples 143 Ge-based Samples 145 The Exploring of Ge and the Relative Alloying Materials 145 The Exploring of Ge-based Oxides Samples 145 The Exploring of Other Ge-based Samples (GeX, X=Se, S, OH, P) 146 References 146

5

Electrolyte, Separator, Binder and Other Devices of Sodium Ion Batteries 171 Mingguang Yi, Mingjun Jing, Wentao Deng, Guoqiang Zou, Hongshuai Hou, Tianjing Wu, and Xiaobo Ji Introduction 171 Organic Liquid Electrolytes 173 Physical and Chemical Properties 173 Organic Solvents 175 Ester-based Solvents 175 Ether-based Solvents 177 Electrolyte Salt 180 Electrolyte Additives 183 Film Formation Additives 185 Flame Retardant Additives 186 Overcharge Protection Additives 187 Additives with Other Functions 187 New Electrolyte Systems 188 Solid State Electrolytes 191 Physical and Chemical Properties 191

5.1 5.2 5.2.1 5.2.2 5.2.2.1 5.2.2.2 5.2.3 5.2.4 5.2.4.1 5.2.4.2 5.2.4.3 5.2.4.4 5.2.5 5.3 5.3.1

Contents

5.3.2 5.3.2.1 5.3.2.2 5.3.2.3 5.3.3 5.3.3.1 5.3.3.2 5.3.4 5.3.4.1 5.3.4.2 5.3.5 5.3.5.1 5.3.5.2 5.4 5.4.1 5.4.2 5.4.3 5.5 5.5.1 5.5.2 5.5.3 5.5.4 5.5.5 5.5.6 5.5.7 5.6 5.6.1 5.6.1.1 5.6.1.2 5.6.1.3 5.6.2 5.6.3 5.6.4 5.7 5.7.1 5.7.2 5.8

Inorganic Solid Electrolyte 192 β-alumina 192 NASICON 193 Sulfides 194 Polymer Electrolyte 197 Solid Polymer Electrolytes (SPEs) 197 Gel Polymer Electrolytes (GPEs) 200 Composite Solid Electrolyte 203 CSEs with Passive Fillers 204 CSEs with Active Fillers 208 Phase Interface Between Electrode and Electrolyte Solid Electrolyte Interphase (SEI) 211 Cathode Electrolyte Interphase (CEI) 214 Separator 217 Glass Fiber 218 Polyolefin Separator 218 Nonwoven Separator 219 Binder 220 Poly(vinylidene fluoride) (PVDF) 220 Polyacrylic Acid (PAA) 221 Sodium Alginate (SA) 222 Sodium Carboxymethyl Cellulose (CMC) 222 Crosslinked Binders 223 Conductive Binders 224 Self-healing Binders 225 Conductive Agent 225 Carbon Black 225 Acetylene Black (AB) 226 Super-P (SP) 226 Ketjen Black (KB) 227 Graphene 228 Carbon Nanofibers (CNFs) 230 Carbon Nanotubes (CNTs) 231 Current Collector 232 Metal-based Current Collector 232 Carbon-based Current Collector 234 Conclusion and Perspectives 236 References 238

6

Advanced Characterization Techniques and Theoretical Calculation 247 Cheng Yang, Libao Chen, Hongshuai Hou, Guoqiang Zou, Xiaobo Ji, and Zhibin Wu Imaging and Microscopy 248 Fundamentals of Imaging and Microscopy 248

6.1 6.1.1

210

ix

x

Contents

6.1.2 6.1.3 6.1.4 6.1.5 6.1.6 6.2 6.2.1 6.2.2 6.2.3 6.2.4 6.3 6.3.1 6.3.2 6.3.3 6.3.4 6.4 6.4.1 6.4.2 6.4.2.1 6.4.2.2 6.4.2.3 6.4.3 6.4.4 6.5 6.5.1 6.5.2 6.5.3 6.5.4 6.5.5 6.6 6.6.1 6.6.2 6.6.3 6.7 6.7.1 6.7.2 6.7.3 6.7.4

Electron Microscopy Studies of SIBs 249 Synchrotron X-Ray Imaging Studies of SIBs 250 Neutron Imaging Studies of SIBs 251 Scanning Probe Microscopy Studies of SIBs 254 Optical Microscopy Studies of SIBs 255 Synchrotron Radiation X-Ray Diffraction Technique 256 Principles of XRD 256 Characteristics of XRD 257 XRD studies of SIBs 259 Challenges and Opportunities 262 Synchrotron Radiation X-ray Absorption Spectroscopy Technique 263 Principles of XAS 264 Characteristics of XAS 266 XAS Studies of SIBs 266 Challenges and Opportunities 268 Solid-state Nuclear Magnetic Resonance Spectroscopy 270 Principles of ssNMR 271 NMR Interactions and Shift Ranges for Battery Materials 272 Shift Interactions (Nuclear Spin−Electron Spin) 272 Dipolar Coupling (Nuclear Spin−Nuclear Spin) 273 Quadrupolar Coupling 273 ssNMR Studies of SIBs 273 The Challenge of NMR Detection 278 Electrochemical Test Techniques 279 Cyclic Voltammetry 279 Galvanostatic Charge–Discharge 281 Electrochemical Impedance Spectroscopy 282 Other Electrochemical Testing Techniques 285 Electrochemical Analysis of SIBs 286 Other Characterization Techniques 287 Neutron Diffraction Technique 287 Fourier Transform Infrared Spectrometry 290 Raman 292 Theoretical Calculation 293 Classical Molecular Dynamics 296 Ab Initio Molecular Dynamics 297 Machine-learning Molecular Dynamics 298 Applications of Theoretical Calculations 300 References 304

7

Practical Application of SIBs 311 Huanqing Liu, Wentao Deng, Hongshuai Hou, Guoqiang Zou, and Xiaobo Ji Introduction 311 Commercial Sodium Battery 311 High-Temperature Na–S Battery 312

7.1 7.2 7.2.1

Contents

7.2.2 7.3 7.3.1 7.3.1.1 7.3.1.2 7.3.1.3 7.3.2 7.3.2.1 7.3.2.2 7.3.2.3 7.3.3 7.3.3.1 7.3.3.2 7.3.4 7.3.4.1 7.3.4.2 7.3.4.3 7.3.4.4 7.3.4.5 7.4 7.4.1 7.4.1.1 7.4.1.2 7.4.2 7.4.3 7.4.4 7.5 7.5.1 7.5.2 7.5.3 7.5.4 7.5.5 7.6 7.6.1 7.6.2 7.6.2.1 7.6.2.2

Sodium–Nickel Chloride Battery 313 Design and Manufacture Process of SIBs 314 Laboratory Button Battery Assembly 314 Metal Na Anode Materials 314 Button Cell Assembly Order 315 The Matching of Positive and Negative Electrodes 315 Type of Cell for SIBs 316 Cylindrical Battery 316 Soft-pack Battery 317 Prismatic Battery 317 Design Requirements for Cell 317 Basic Design Principles 318 Safety Design 318 Manufacturing Process of SIBs 319 Front-end Electrode Fabrication Process 319 Back-end Assembly Process 320 Formation and Sorting Process 321 Design of SIBs Pack 321 Battery Management System 321 Presodiation Techniques 322 EC/Chemical Methods 323 EC 323 Chemical Methods 323 Self-sacrificial Additive 324 Other Novel Methods of Presodiation 324 Factors Need to be Improved 325 Performance Tests and Failure Analysis 325 Electrochemical Performances Test 326 Safety Performances Test 326 Failure Phenomenon 327 Failure Analysis Method 328 Cost Estimation 330 Commercial Application and Future Perspectives 332 Current State of Commercialization of SIBs 332 Application Prospect 333 Low-Speed Electric Vehicle Market 333 Large-scale ESSs 334 References 334 Index 337

xi

xiii

Preface Although originated from almost the same era around 1970s and held similar working principles, sodium-ion batteries (SIBs) had grasped too less attention than lithium-ion batteries (LIBs) before 2010s due to the inferiority of energy density, cycle life, and essentially the lack of suitable electrode materials. However, since the booming success of commercial LIBs brought with it the anxiety over lithium and cobalt sources and prices, SIBs have gained renewed and ever-increasing focus in the past decade owing to the unlimited sodium sources and the diverse choices of transition metals, giving birth to numerous research papers and some start-ups endeavoring to push for commercialization of SIBs. It can be foreseen that the predominance of LIBs in the market may be difficult to shake, at least in the next decade. Nonetheless, SIBs can be a good complement, or even a strong competitor, in specific application scenarios, including but not limited to large-scale grid energy storage, distributed energy storage, and low-speed electric vehicles, by virtue of their considerable advantages in cost-effectiveness compared with LIBs, lead-acid batteries, and vanadium redox flow batteries. The rapid scaling up of energy storage systems, which is critical to address the hour-to-hour variability of wind and solar electricity generation on the grid, is of great urgency in the global demand to develop low-carbon energy, offering golden opportunities for the development of SIBs. One of the vital preconditions that allow us to consider the real-world utilization of SIBs is the great breakthrough in battery materials we have achieved in the past few years. Regarding cathode materials, plenty of candidates, such as transition metal oxides/fluorides, Prussian blue analogs, and polyanionic compounds with improved specific capacity and prolonged cycle life, have been successfully discovered. On the anode side, the advent of hard/soft carbon has significantly promoted the practical application of SIBs, while the extensive explorations of non-ferrous-based conversion-type materials have furnished abundant choices of anode materials to equip SIBs for different applications. Other materials have also achieved tremendous progress toward the requirements of application-wise sceneries, while diverse characterization techniques have been successfully utilized to deepen the understanding of the fundamental working mechanisms of all these battery materials. Today, materials engineering for SIBs is still ongoing while research is updating superfast. Therefore, it is of great importance to look back from time to time at what we have done, where we are, and how we should

xiv

Preface

proceed for the advancement of SIBs. In this book, we would like to present a comprehensive review of the research history and the state-of-the-art progresses of SIBs, accompanied with in-depth discussions on key issues of materials and some perspectives for the future development of SIBs based on the best of our knowledge. This book is prepared with the support of many peer researchers who made important contributions to the final version of this book. Chapter 4 is authored by Associate Professors Peng Ge and Yue Yang, Chapter 5 is authored by Associate Professors Mingjun Jing and Tianjing Wu, and Chapter 6 is authored by Dr. Zhibin Wu. The other chapters are mainly authored by Professor Xiaobo Ji, Hongshuai Hou, and Guoqiang Zou, and they also completed the compilation and revision of this book. Finally, the authors would like to acknowledge the financial support of National Natural Science Foundation of China (5221101846, U21A20284). 17 August 2023 Changsha, China

Professor Xiaobo Ji Professor Hongshuai Hou Professor Guoqiang Zou Central South University, Changsha, China

1

1 Introduction Jinqiang Gao, Wentao Deng, Guoqiang Zou, Hongshuai Hou, and Xiaobo Ji Central South University, School of Chemistry and Chemical Engineering, Changsha, China

1.1 Overview Since humans have obtained energy by drilling wood for fire, every energy revolution has been accompanied by great progress in human civilization. However, the consumption of fossil energy has caused irreversible pollution and damage to the human environment, so it is urgent to replace fossil energy with renewable energy to get mankind out of the upcoming energy crisis and environmental disaster. In recent years, the technology of converting wind, solar, hydraulic, tidal, and other renewable energies into electric energy has rapidly developed [1]. However, the power generation is limited by natural conditions, owing to randomness, intermittent and fluctuating characteristics, leading to a great impact on the state grid if the generated electric energy were to be fed directly into the grid. The new energy power generation industries are still facing serious energy wastage problems, such as wind and light wastage. Therefore, in order to greatly improve the utilization of renewable energy and establish a green, low-carbon, efficient, and sustainable development society, it is necessary to develop an efficient and convenient large-scale energy storage technology and form an “energy internet” of renewable energy–energy storage system–smart grid–users. At present, the storage of electrical energy mainly includes physical energy storage, chemical energy storage, electrochemical energy storage, and other technologies [2]. Physical energy storage includes pumped hydro storage, compressed air energy storage, flywheel energy storage, and superconducting energy storage. Chemical energy storage includes various types of fossil fuels and hydrogen energy. Electrochemical energy storage includes secondary batteries and supercapacitors. Electrochemical energy storage, such as secondary batteries, has a wide range of application prospects in the energy field, owing to the advantages of high energy density, high energy conversion efficiency, and fast response speed. At present, there are four types of secondary batteries that have realized commercial applications: lead-acid batteries, high-temperature sodium batteries, vanadium flow batteries, and lithium-ion batteries. However, these batteries are limited by their Sodium-Ion Batteries: Technologies and Applications, First Edition. Edited by Xiaobo Ji, Hongshuai Hou and Guoqiang Zou. © 2024 WILEY-VCH GmbH. Published 2024 by WILEY-VCH GmbH.

2

1 Introduction

disadvantages, such as lead-acid batteries with low energy density (30–50 W h kg−1 ), high-temperature sodium batteries that need to operate at higher temperatures (300–350 ∘ C), and the low energy conversion efficiency of vanadium flow batteries (75–82%). The secondary battery represented by the lithium-ion battery has many advantages, such as high energy density, high energy storage efficiency and nonmemory effect, small self-discharge, long cycle life, and wide application range. Currently, lithium-ion batteries have been successfully used in small electronics, electric vehicles, and aerospace. At the same time, the research direction of lithium-ion batteries is gradually toward ultrahigh energy density and ultralong cycling life. However, lithium resources are relatively concentrated in a few countries, the overall reserves are limited, and the mining conditions are relatively harsh. So, it is difficult to support the development of electric vehicles and large-scale energy storage. In recent years, sodium-ion batteries (SIBs) with the same working principle and similar battery components as lithium-ion batteries have received widespread attention, owing to the advantages of abundant sodium resources, cost-effectiveness, and outstanding comprehensive performance. SIBs can meet the requirements of low cost, long cycling, and high-safety performance, alleviating the limited development of energy storage batteries caused by the shortage of lithium resources, which is a promising supplement to lithium-ion batteries and can gradually replace lead-acid batteries. Therefore, SIBs are expected to play an important role in renewable energy storage [3]. In the early 1970s, the research on SIBs was almost simultaneously carried out with that on lithium-ion batteries, and lithium-ion batteries were successfully commercialized in 1991, while SIBs have not yet been commercialized. The research on SIBs can learn from the research experience of lithium-ion batteries because the working principle, materials, and battery components of the two batteries are similar. It is worth noting that it cannot be fully copied due to differences in charge carriers (Li+ vs. Na+ ). Therefore, finding suitable materials for SIBs and building suitable SIB systems are the key to its practical application. In recent years, a series of advances have been made at home and abroad on the core material systems (cathode material, negative electrode, electrolyte, and separator), main auxiliary materials (binders, conductive agents, and current collectors), key battery technologies (nonaqueous, aqueous, and solid-state batteries), and analytical characterization, material prediction, and failure mechanism, which have laid a solid foundation for the commercialization of SIBs [4, 5]. With the deeper insight into this field, more and more potential advantages of SIBs have been found, which will give SIBs more characteristics and a favorable position in the future energy storage market [6]. Some advantages of SIBs are summarized (Figure 1.1): (1) Sodium resources are abundant, widely distributed, economical, and there are no bottlenecks for the development of SIBs. (2) The working principles of SIBs and lithium-ion batteries are similar, and they are compatible with the existing production equipment of lithium-ion batteries.

1.1 Overview

(3) Alloying reactions can be avoided between sodium and aluminum, and the current collectors of the positive and negative electrodes for SIBs can use cheap aluminum foil, which can further reduce costs with no overdischarge problems. (4) Bipolar SIBs can be constructed; that is, the positive and negative electrode materials can be coated on both sides of the same aluminum foil. The electrodes are periodically stacked under the isolation of solid electrolytes, which can achieve higher voltage, save inactive materials, and improve the energy density. (5) The Gibbs free energy of solvation of sodium ions is lower than that of lithium ions, which is beneficial for interface desolation. (6) The Stokes diameter of the sodium ion is lesser than that of the lithium ion, and a high ionic conductivity can be achieved with a low concentration of sodium salt electrolyte, making the low salt concentration electrolyte suitable for use in SIBs. (7) SIBs have excellent rate performance as well as outstanding cycling performance at high and low temperatures. (8) The SIB does not catch fire or explode in the safety test, and the safety performance is good.

Figure 1.1

Characteristics of sodium-ion batteries.

3

4

1 Introduction

1.2 The Birth and Development of Sodium-ion Batteries Since the concept of sodium batteries was proposed in the science fiction novel “Twenty Thousand Leagues Under the Sea,” the real emergence of sodium batteries has taken nearly 100 years. In 1967, Yao and Kummer [7] found the conduction of Na+ in Na-β′′ -Al2 O3 . In 1968, the Ford Company invented high-temperature sodium–sulfur battery (Na-Na-β′′ -Al2 O3 |S) (300–350 ∘ C) with sodium and sulfur as the negative and positive electrodes, respectively, and Na-β′′ -Al2 O3 as the solid electrolyte. In 1986, Coetzer [8] replaced sulfur with NiCl2 and invented the ZEBRA battery (Na|Na-β′′ -Al2 O3 |NiCl2 ). In 2003, NGK company realized the commercialization of high-temperature sodium–sulfur batteries. However, both sodium–sulfur batteries and ZEBRA batteries are sodium batteries that work at high temperatures. In order to reduce the working temperature of sodium batteries to improve their safety, a lot of research work began to develop sodium batteries that work at room temperature. Taking this into consideration, the development of room-temperature SIBs has undergone a long process (Figure 1.2). In 1976, Whittingham et al. [9] conducted a study of the behavior of Li+ intercalating TiS2 , followed by the electrochemical reversible deintercalation of Na+ in TiS2 at room temperature [10]. France Armand [11] proposed the concept of “rocking chair batteries” at the NATO Conference on Materials for Advanced Batteries held in 1979, which opened up the research on lithium-ion and SIBs. In 1981, French Delmas et al. [12] firstly reported the electrochemical properties of Nax CoO2 -layered oxide cathode materials and proposed a classification trend for layered oxide structures, according to the coordination environment of alkali metal ions. Layered oxides are divided into O type or P type (O refers to octahedron and P refers to triangular prisms), and numbers (such as 2 and 3.) represent the number of stacking layers of the least repeated oxygen units. During that period, a variety of sodium-containing transition metal-layered oxides, Nax MO2 (M = Ni, Ti, Mn, Cr, Nb), were reported. When studying the behavior of Na+ in the NaTi2 (PO4 )3 electrode material, it was found that NASICON-structured solid electrolyte Na3 M2 (PO4 )3 (M = Ti, V, Cr, Fe, etc.) [4] could also be used as electrode material. However, in the late 1980s, research reports on sodium-ion intercalating materials were very limited, and only a few papers and patents were published, mainly because [13]: (i) The research on lithium-ion intercalating materials was just beginning in this period, and a large number of researchers focused their research on lithium-ion batteries. (ii) Limited by the research conditions (such as the low purity of the electrolyte, the poor tightness of the glove box, and the low purity of argon.), it is difficult to use the active metal sodium as an electrode to accurately evaluate the performance of the electrode material in the half-batteries. (iii) The graphite successfully applied in lithium-ion batteries has almost no sodium storage capacity in carbonate electrolytes, resulting in the lack of suitable anode materials for the study of SIBs. In fact, before the successful commercialization of lithium-ion batteries, some companies in the United States and Japan carried out research on

1.2 The Birth and Development of Sodium-ion Batteries

Figure 1.2

The development of room-temperature sodium-ion batteries.

sodium-ion full batteries, such as P2-Nax CoO2 , which was used as positive electrode and Na–Pb alloy as negative electrode. Although the SIB can reach 300 cycles, its average discharge voltage is below 3 V, which has no advantage over C||LiCoO2 battery (3.7 V) and thus failed to attract the attention of researchers. In 2000, the SIB got its first opportunity. Stevens and Dahn [14] prepared a hard carbon anode material for SIB via pyrolysis of glucose for the first time and demonstrated a specific capacity of 300 mAh g−1 . It is worth noting that, up to now, hard carbon materials are still the most promising anode materials for SIBs. The second important finding was the reversible variability of the Fe4+ /Fe3+ pair in NaFeO2 reported by Okada et al. [13], which has no electrochemical activity in LiFeO2 . Except for the layered oxides, Na2 FePO4 F polyanionic material reported in 2007 by Nazar and coworkers [15] exhibits only 3.7% volumetric change

5

Figure 1.3

Research articles about sodium-ion batteries between 2000 and 2021.

1.2 The Birth and Development of Sodium-ion Batteries

during the deintercalating/intercalating of sodium ions, which is lower than that of olivine-type NaFePO4 (15% volumetric change). So far, the papers published between 2000 and 2009 on SIB materials have shown a slow growth trend and are mainly concentrated in a few laboratories. Since 2010, the research of SIBs has entered a period of revival, and the number of related articles has increased rapidly (Figure 1.3), mainly due to the following reasons [16]: (i) The research on lithium-ion battery materials at this time mainly focuses on the application improvement and the in-depth analysis of electrochemical processes, and the difficulty of developing new materials has significantly increased. So, many researchers turned to the exploration of SIB material systems. (ii) Concerns about lithium resources and the demand for new large-scale energy storage applications also make researchers to develop new battery systems. On this background, SIBs developed rapidly with the research experience of the lithium-ion battery. So far, researchers have reported a variety of SIB cathode materials, anode materials, and electrolyte systems [17]. Among them, cathode materials mainly include layered and tunneled transition metal oxides, polyanionic compounds, Prussian blue analogs, and organic materials. Anode materials mainly include carbon materials, alloys, phosphorus compounds, and organic carboxylates. Except for new material systems, the research and development of SIBs is also working in the direction of low cost and practicality. In 2011, Komaba et al. [18] firstly reported the electrochemical performances of hard carbon||NaNi0.5 Mn0.5 O2 full-cell. In the same year, the world’s first SIB company, FARADION, was established in the United Kingdom. In 2013, Goodenough and coworkers [19] proposed a Prussian white cathode material with high voltage and excellent magnification performance. In 2017, China’s first SIB company (HiNa Battery Technology Co., Ltd.) was founded, which built the first low-speed electric vehicle powered by SIB and the first 100 kW h SIB energy storage power station in 2018 and 2019, respectively [20]. As of 2020, more than 20 companies around the world are committed to the research and development of SIBs, indicating that SIBs are moving toward practical application. At the same time, in order to develop more secure SIBs for large-scale energy storage, the research and development of aqueous SIBs and solid-state SIBs that replace organic electrolytes with aqueous electrolytes and solid electrolytes, respectively, are also being carried out simultaneously. Nowadays, the development of SIBs has attracted the attention of many countries around the world, and China is one of the strongest competitors in the research and development of SIB technology. HiNa Battery, Natrum Energy, and CATL have accelerated the commercialization of SIBs. In the near future, SIBs are expected to be applied in commercialization in China first, providing a strong guarantee for national energy security.

7

8

1 Introduction

References 1 Zhao, C., Wang, Q., and Hu, Y.S. (2020). Rational design of layered oxide materials for sodium-ion batteries. Science 370: 708–711. 2 Dunn, B., Kamath, H., and Tarascon, J.M. (2011). Electrical energy storage for the grid: a battery of choices. Science 334: 928–935. 3 Lu, Y.X., Zhao, C.L., Rong, X.H. et al. (2022). Compositionally complex doping for zero-strain zero-cobalt layered cathodes. Nature 610: 67–73. 4 Delmas, C. (2018). Sodium and sodium-ion batteries: 50 years of research. Advanced Energy Materials 8: 1703137. 5 Pan, H.L., Hu, Y.S., and Chen, L.Q. (2013). Room-temperature stationary sodium-ion batteries for large-scale electric energy storage. Energy & Environmental Science 6: 2338–2360. 6 Winter, M., Bamett, B., and Xu, K. (2018). Before Li ion batteries. Chemical Reviews 118: 11433–11456. 7 Yao, Y.F.Y. and Kummer, J.T. (1967). Ion exchange properties of and rates of ionic diffusion in beta-alumina. Journal of Inorganic and Nuclear Chemistry 29: 2453–2475. 8 Coetzer, J. (1986). A new high-energy density battery system. Journal of Power Sources 18: 377–380. 9 Whittingham, M.S. (1976). Electrical energy storage and intercalation chemistry. Science 192: 1126–1127. 10 Newman, G.H. and Klemann, L.P. (1980). Ambient temperature cycling of an Na-TiS2 cell. Journal of the Electrochemical Society 127: 2097–2099. 11 Armand, M.B. (1980). Intercalation Electrodes//Murphy DW. Materials for Advanced Batteries, 145–161. New York: Springer. 12 Delmas, C., Braconnier, J.J., Fouassier, C. et al. (1981). Electrochemical intercalation of sodium in Nax CoO2 bronzes. Solid State Ionics 3: 165–169. 13 Okada S, Takahashi Y, Kiyabu T, et al. 210th ECS Meeting Abstracts, 2006, MA 2006-02, 201. 14 Stevens, D. and Dahn, J.R. (2000). High capacity anode materials for rechargeable sodium-ion batteries. Journal of the Electrochemical Society 147: 1271–1273. 15 Ellis, B.L., Makahnouk, W.R.M., Makimura, Y. et al. (2007). A multifunctional 3.5V iron-based phosphate cathode for rechargeable batteries. Nature Materials 6: 749–753. 16 Kubota, K. and Komaba, S. (2015). Review—practical issues and future perspective for Na-ion batteries. Journal of the Electrochemical Society 162: A2538–A2550. 17 Palomares, V., Casas-Cabanas, M., Castillo-Martinez, E. et al. Update on Na-based battery materials.(2013). A growing research path. Energy & Environmental Science 6: 2312–2337.

References

18 Komaba, S., Murata, W., Ishikawwa, T. et al. (2011). Electrochemical Na insertion and solid electrolyte interphase for hard-carbon electrodes and application to Na-ion batteries. Advanced FunctionalMaterials 21: 3859–3867. 19 Wang, L., Lu, Y., Liu, J. et al. (2013). A superior low-cost cathode for a Na-ion battery. Angewandte Chemie International Edition 52: 1964–1967. 20 Lu, Y.X., Rong, X.H., Hu, Y.S. et al. (2019). Research and development of advanced battery materials in China. Energy Storage Materials 23: 144–153.

9

11

2 Characteristics of Sodium-ion Batteries Haoji Wang, Wentao Deng, Hongshuai Hou, Guoqiang Zou, and Xiaobo Ji Central South University, College of Chemistry and Chemical Engineering, Changsha, China

2.1 Basic Features Sodium-ion batteries (SIBs) have attracted attention in large-scale energy storage applications due to the abundance and global distribution of sodium resources in the earth’s crust. The crustal abundance of sodium resources reaches 2.74%; however, it is only 0.0065% for lithium resources. And lithium resources are unevenly distributed, 70% of which are concentrated in South America and a few countries and regions. In addition, the reserves of elements commonly used in SIBs, such as iron and manganese, are relatively high in the crust, while the nickel and cobalt commonly used in lithium-ion batteries (LIBs) are relatively poor, as shown in Figure 2.1 [1]. In theory, sodium ions do not react with aluminum, so cheap aluminum foil can be used as the anode collector, substituting copper foil in SIBs. In general, compared with LIBs, the material cost of SIBs can be reduced by about 30–40%, with the largest difference in the cathode materials, as shown in Figure 2.2. At the same time, large-scale production of SIBs is not limited by geographical factors, which is conducive to the sustainable development of large-scale energy storage. In the periodic table, the sodium element is located in the third period of the first main group. Its atomic mass and atomic radius are second only to lithium. Sodium has two different electron clouds (spherical and dumbbell shaped) outside the atomic nucleus, and the outermost electrons occupy 3s orbits. SIBs and LIBs have similar working principles, but the differences in physical and chemical properties between Na and Li will inevitably lead to the differences in electrochemical performances as shown in Table 2.1. The standard hydrogen potential of Na+ is −2.71 V (Na+ /Na, standard hydrogen electrode, SHE), which is slightly higher than that of Li+ (−3.04 V, Li+ /Li, standard hydrogen electrode, SHE). In addition, the weight and volume energy density of SIBs are not comparable to that of LIBs due to the heavier mass of Na (23 g mol−1 ). However, the large ion radius of Na+ makes the detrimental migration resistance and sluggish kinetic reaction in host lattice. Meanwhile, it also

Sodium-Ion Batteries: Technologies and Applications, First Edition. Edited by Xiaobo Ji, Hongshuai Hou and Guoqiang Zou. © 2024 WILEY-VCH GmbH. Published 2024 by WILEY-VCH GmbH.

2 Characteristics of Sodium-ion Batteries 70000

63 000

crust (ppm) Abundance in

60000 50000

40000 30000

23 000

20000

15 000 6 600

79 Zn

68 Cu

Co

90 Ni

30

Fe

Mn

V

Ti

K

190 140 1100

Na

0

17

Cr

10000 Li

12

Figure 2.1 Reserve content of elements in the earth’s crust. Source: Reproduced from Li et al. [1]/with permission of Elsevier.

Lithium-ion battery Anode Cathode

Electrolyte

16% 26%

Sodium-ion battery

26%

Cathode

Anode

Current collector

11%

Electrolyte

u ed tr

4% Separator

Others

s Co

10%

18%

43%

n

io

ct

15% 0% ~4

30 5%

13% Separator

Others 13%

Current collector

Figure 2.2

Cost components of lithium-ion batteries (LIBs) and sodium-ion batteries (SIBs).

leads to the severe lattice stresses, complex structural evolution, and drastic interface reactions in the process of sodium de-/intercalation. For example, NaCoO2 and LiCoO2 have the same layered crystal structure because the ion radius of Co (0.54 Å) is smaller than that of Na (1.02 Å) and Li (0.76 Å). The theoretical specific capacity of NaCoO2 cathode material in SIBs is 235 mAh g−1 , which is less than LiCoO2 in LIBs (274 mAh g−1 ). And the working voltage of NaCoO2 is about 1.0 V lower than that of LiCoO2 . After 0.5 Li+ /Na+ is removed, the potential difference between NaCoO2 and LiCoO2 drops to about 0.4 V. During the charging and discharging process, there are multiple platforms in the electrochemical curve of NaCoO2 that correspond to different phase transitions. In particular, the ordered and disordered distributions of Na+ and vacancies have a great impact on the performance of electrode materials, as shown in Figure 2.3a [2].

2.1 Basic Features

Table 2.1

Comparison of the properties of Li and Na.

Parameters

Li

Na

Relative atomic mass

6.94

23.00

EΘ (A+ /Aaq )(V vs. SHE)

−3.04

−2.71

EΘ (A+ /Apc )(V vs. Li+ /Lipc )

0

0.23

Shannon’s Radii/Å

0.76

1.02

Theoretical mass specific capacity of ACoO2 (mAh g−1 )a)

274

235

Stokes radii (H2 O)/Å

2.38

1.84

Stokes radii (PC)/Å

4.8

4.6

Desolvation free energy (PC)/(KJ mol−1 ) Melting point (∘ C)

215.8

158.2

180.5

97.8

Precursor price/($/t)

5 000

150

Geographical distribution

70% in South America

Global distribution

−1

a) The theoretical mass specific capacity of ACoO2 (mAh g ) based on LiCoO2 and NaCoO2 .

5 3

Li/Li1–xCoO2

A B C A

3 CoO2 Layer Li or Na ions

2

1 (a)

IV III

II 2

Na/Na1–xCoO2

0

50

100

Capacity (mAh

I

1

B C

c b

E (V)

Voltage (V)

4

0

A B

a

150 g–1)

200

0 (b)

100

200 Q (mAh

300

400

g–1)

Figure 2.3 (a) Comparison of charge/discharge curves of Li||LiCoO2 and Na||NaCoO2 half-cells. Source: Reproduced from Yabuuchi et al. [2]/with permission of American Chemical Society. (b) Charge–discharge curves of graphite electrodes in (I) Li (blue), (II) Na (green), and (III, IV) K cells (black and red), respectively. Source: Reproduced from Komaba et al. [3]/with permission of Elsevier.

As for anode materials, the Na- and Li-storage behavior of graphite materials is different. As shown in Figure 2.3b, although graphite has excellent lithium storage capacity in LIBs, it has almost no sodium storage capacity [3]. The reason is that Li+ can embed in graphite electrodes and form a stable first-order intercalation compound, LiC6 , with a theoretical capacity of 372 mAh g−1 , while Na+ cannot form a first-order stable intercalation compound (NaC70 ) with graphite due to the large radius of Na+ and thermodynamic factors. Consequently, the sodium storage activity

13

14

2 Characteristics of Sodium-ion Batteries

of graphite is low, and the theoretical capacity is only 31 mAh g−1 . All these indicate that it is necessary to explore new systems different from LIBs in order to give play to the advantages of SIBs in the development of electrode materials. Every coin has two sides. The effect of large Na+ radius is not all negative. During the process of development, some differences exist between LIBs and SIBs: (i) It is easier to separate sodium ions from transition metals to form a layered structure at high temperatures due to the large radius difference between sodium and transition metal ions. In the layered structure, lithium-containing oxides are mainly O-type structures, while sodium-containing oxides have an extensive category of O-type and P-type materials, providing more options for developing advanced cathode materials for SIBs. For example, P2/O3 biphasic layered oxides Na0.7 Ni0.2 Cu0.1 Fe0.2 Mn0.5 O2−δ show excellent electrochemical performances [4]. (ii) Many transition metal elements that are not electrochemically active in lithium-containing layered oxides are active in sodium-containing layered oxides within the normal charge/discharge voltage range. Only three elements, including Ni, Co, and Mn, are able to reversibly gain or lose electrons in lithium-containing oxides, while Ni, Co, Mn, Cr, V, Fe, Cu, and Ti elements are electrochemically active with a high degree of reversibility in sodium-containing oxides. (iii) The diffusion rate of large-radius Na+ in electrode materials is not necessarily lower than small radius Li+ , because it is related to the crystal structure of electrode materials. For example, the diffusion rate of Na+ in layered Na2/3 [Ni1/3 Mn2/3 ]O2 is higher than the diffusion rate of Li+ in spinel Li4 Ti5 O12 [5]. (iv) The large Na+ insertion/extraction during the charge and discharge process does not necessarily result in a dramatic volume change. For example, the volume change of layered P2-Na0.66 [Li0.22 Ti0.78 ]O2 after Na+ extraction is only 0.77%, which is much smaller than that of LiMn2 O4 (V: ∼5.6%), LiCoO2 (c: ∼2.6%), and LiFePO4 (V: ∼6.8%) [6]. (v) The large Na+ radius, on the one hand, gives it a strong ability to desolvate in polar solvents, leading to a high conductivity in the electrolyte, and on the other hand, allows to use low salt concentration electrolytes to achieve the same conductivity, which further reduces the cost.

2.2 Working Principle SIBs and LIBs have a similar principle that uses alkali metal ions to migrate back and forth between the cathode and anode electrodes, which was named as “rocking chair battery” proposed by M. Armand, as shown in Figure 2.4. It is essentially a concentration cell where the cathode and anode materials are made up of compounds with different sodium-ion contents. During charging, Na+ is removed from the sodium-rich cathode and inserts into the sodium-poor anode through the electrolyte and separator, along with a valence increase of transition metal ions in the cathode (loss of electrons), and the electrons migrate from the cathode to anode via an external circuit. This process generally presents an overall voltage increase. During discharging, the whole process is reversed. Na+ is extracted from the sodium-rich anode, passes through the electrolyte and separator, and comes

2.3 Concepts and Equations

Charge e–

e–

e–

e–

Separator

e– Al

Cathode

SEI

Electrolyte

e– SEI

e–

Anode

Al

e– Discharge

Figure 2.4 The “rocking chair” working principle of rechargeable sodium-ion batteries. Source: Reproduced from Li et al. [1]/with permission of Elsevier.

back to the sodium-poor cathode electrode with the simultaneous movement of electrons and decrease of transition metal valence in the cathode electrode (gaining electrons). Therefore, Na+ migration inside the cell accompanied by electron transfer in the external circuit maintains charge balance of the whole system. Similar to the LiCoO2 //graphite battery, the SIBs can be expressed as an Nax TMO2 //hard carbon battery, with Nax TMO2 as the cathode electrode and hard carbon as the anode electrode. The reaction equations can be expressed as: Cathode: Nax TMO2 → Nax−y TMO2 + yNa+ + ye−

(2.1)

Anode: nC + yNa+ + ye− → Nay Cn

(2.2)

Battery: Nax TMO2 + C ↔ Nax−y TMO2 + Nay C

(2.3)

Na+

extraction/insertion between the cathode and anode materials Ideally, the does not destroy the crystal structure; thus, SIBs are viable rechargeable secondary batteries that can be used in a variety of applications such as large-scale energy storage and low-speed electric vehicles. Due to the similar structural components between SIBs and LIBs, it can highly borrow from the LIBs equipment, technology, and methods to accelerate the industrialization process of SIBs.

2.3 Concepts and Equations When describing the electrochemical performance of SIBs, some terms are usually involved.

15

16

2 Characteristics of Sodium-ion Batteries

2.3.1

Cell Voltage

2.3.1.1 Electromotive Potential

The electrochemical charging and discharging process of cell is actually achieved through chemical reactions. The relationship between the Gibbs free energy and cell potential is as follows: ΔGΘ = −nFEΘ (at standard condition)

(2.4)

where n is the amount of transferred electrons in the electrode reaction; F is Faraday’s constant, F = 96 485 C mol−1 (or F = 26.8 A h mol−1 ); and EΘ is standard electrode potential. The output voltage is equal to the cell potential EΘ when the discharge current tends to zero. Equation (2.4) shows the maximum limit of the conversion of chemical energy into electrical energy, which provides a theoretical basis for improving the cell performance. At nonstandard condition: ΔG = −nFE (E < EΘ )

(2.5)

2.3.1.2 Theoretical Voltage E 𝚯

ΔGΘ = −nFEΘ

(2.6)

Cathode (reduction potential) + Anode (oxidation potential) = Standard cell electromotive potential. The voltage difference between the two electrodes is called the cell voltage. The theoretical voltage is the maximum limit of the cell voltage, and the theoretical cell voltage of different materials is also different. Beyond these, cell voltages also include the following. 2.3.1.3 Open Circuit Voltage E ocv

The open circuit voltage is the voltage difference between the cathode and anode of the cell without loading. The open circuit voltage is always less than the cell electromotive potential. 2.3.1.4 Operating Voltage E cc

The operating voltage is the voltage difference between the cathode and anode of cell with loading. It is the actual output voltage when the cell is operating, and it changes with the current and the discharge depth. The operating voltage is always lower than the open circuit voltage since the existence of the polarization resistance and ohmic resistance (Ecc = Eocv − IRi ) and the cell operating voltage is influenced by the discharge regime and the ambient temperature. 2.3.1.5 Cutoff Voltage

The cutoff voltage refers to the maximum charge voltage or minimum discharge voltage specified when the cell is charged or discharged.

2.3 Concepts and Equations

2.3.2

Cell Capacity and Specific Capacity

The cell capacity is the amount of electricity obtained from the cell under certain discharge conditions, expressed as ampere per hours (A h). The cell capacity also contains theoretical capacity, actual capacity, and rated capacity. 2.3.2.1 Theoretical Capacity (C o )

The theoretical capacity is the amount of capacity provided by all active materials fully participating in the electrochemical reaction. The actual capacity is only a fraction of the theoretical capacity, corresponding to the real capacity that cells can provide. Faraday’s law states that the amount of material participating in the electrochemical reaction at the electrode is directly proportional to the amount of released electricity. For example, 1 mol of active material participating in the electrochemical reaction will release the electricity of F = 96 485 C mol−1 (or F = 26.8 A h mol−1 ). Therefore, the theoretical capacity is calculated as follows: Co = 96 485 ×

1 m m × ne ÷ 3600 = 26.8ne = m (A h) M M q

(2.7)

where m is the mass of active materials in complete reaction; M is the molar mass of the active materials; ne is the number of electrons gained or lost during the electrode reaction; and q is the electrochemical equivalent of the active materials. For NaFeO2 , the theoretical capacity is 26.8 A h mol−1 . 2.3.2.2 Actual Capacity (C)

The actual capacity is the amount of electricity actually released by the cell under certain discharge conditions (e.g. 0.2 C). The actual capacity of a cell in an unspecified discharge regime is usually expressed as the nominal capacity. The nominal capacity is only an approximate representation of the actual capacity. The discharge current intensity, temperature, and cutoff voltage of the cell are called the discharge regime of the cell. Different discharge regimes result in different capacities, calculated as follows: Constant current discharge: (2.8)

C = It Constant resistance discharge: t

C=

∫0

t

Idt =

1 Vdt R ∫0

(2.9)

Approximate calculation formula: Vave t (2.10) R where I is the discharge current; R is the discharge resistance; t is the time discharging to the cutoff voltage, and V ave is the average discharge voltage of the cell. C=

17

18

2 Characteristics of Sodium-ion Batteries

2.3.2.3 Rated Capacity (C r )

Rated capacity is the minimum amount of electricity discharging under certain discharge conditions as the cell is designed and manufactured. 2.3.2.4 Specific Capacity (C m or C V )

In order to compare the electrochemical performance differences of different cells, the specific capacity is introduced. The specific capacity is the capacity per unit mass or volume of cell (or active material), referred to as mass specific capacity (A h kg−1 ) or volume specific capacity (A h l−1 ), respectively. Cm = C∕m or CV = C∕V

(2.11)

where m is the cell mass (kg); and V is the cell volume (L).

2.3.3

Cell Energy and Specific Energy

The cell energy refers to the electrical energy output from the external work of the cell under certain discharge conditions, usually expressed as watt per hour (W h). 2.3.3.1 Theoretical Energy (W o )

The cell output energy is the theoretical energy W o when the discharge capacity is equal to the theoretical capacity. That is to say: Wo = Co EΘ

(2.12)

This is also the maximum external work of a reversible cell at constant temperature and pressure: Wo = −ΔGΘ = nFEΘ

(2.13)

2.3.3.2 Actual Capacity (W )

The actual energy is the actual output energy as the cell discharge, which is numerically equal to the product of the actual capacity and average operating voltage of the cell. The cell operating voltage is always less than the electromotive potential, and the actual energy is always less than the theoretical energy because the active material cannot be fully utilized. W = CVave

(2.14)

2.3.3.3 Specific Capacity (W m or W v )

Specific energy, also known as energy density, is the amount of energy per unit mass or volume of a cell, referred to as mass specific energy (W m ) and volume specific energy (W v ), respectively, and is often expressed as W h kg−1 or W h L−1 . The specific energy also contains theoretical specific energy and actual specific energy. Wm = CVave ∕m or Wm = CVave ∕V

(2.15)

2.3 Concepts and Equations

2.3.4

Cell Power and Specific Power

Cell power is the output energy per unit time, expressed as watts (W) or kilowatts (kW) under a certain discharge regime, and the output power per unit mass or unit volume of the cell is the specific power, expressed as W kg−1 or W l−1 . The theoretical power can be expressed as Po = Wo ∕t = Co EΘ ∕t = ItEΘ ∕t = IEΘ

(2.16)

where t is the discharge time; Co is the theoretical capacity of the cell; I is the constant current; and EΘ is the electromotive potential. And the cell actual power can be expressed as P = IV = I(EΘ − IRi ) = IEΘ − I 2 Ri

(2.17)

where I 2 Ri is the consumed power by the internal resistance of the cell, which is useless to the load.

2.3.5

Charge and Discharge Rate

The charge/discharge rate is generally expressed as time rate or rate. The time rate is the hours that it takes for a cell to discharge its rated capacity at a given current. And the rate is the current that is required to discharge its rated capacity within a given time. The rate is usually denoted by the letter C, and 0.2 rate is also called 0.2C. Time rate and rate are reciprocals of each other, C = 1/h.

2.3.6

Constant Current Charge and Discharge

Constant current charge and discharge is the process of charging or discharging cell at a constant current. The charge and discharge process generally completes as reaching the set cutoff voltage.

2.3.7

Constant Voltage Charge

Constant voltage charge is the process of charging cell at a constant voltage. The cutoff current is set, and the charging process ends as the current is less than this value.

2.3.8

Coulombic Efficiency

Under certain charge and discharge conditions, the percentage of the charge released from the discharge (discharge capacity) to the charge charged during charging (charge capacity) is called the coulombic efficiency, also called the charge and discharge efficiency. The coulombic efficiency is affected by many factors, such as electrolyte decomposition, passivation of electrode interface, and changes in the structure, morphology, and conductivity of electrode materials.

19

20

2 Characteristics of Sodium-ion Batteries

2.3.9

Energy Conversion Efficiency

Energy conversion efficiency refers to the ratio of cell discharge to charge energy. It is a dimensionless number ranging from 0 to 1 and is sometimes expressed as a percentage.

2.3.10 Cell Internal Resistance The cell internal resistance consists of ohmic resistance (RΩ ) and polarization resistance (R). Ohmic resistance contains the resistance of electrode material, electrolyte, separator, current collector, and contact resistance between components. Polarization resistance refers to the resistance caused by polarization during an electrochemical reaction. Polarization resistance includes the resistance induced by electrochemical polarization and concentration polarization. In order to compare the internal resistance of different cells, the specific resistance ′ R is introduced. That is, the internal resistance of cells per unit capacity: R′ = Ri ∕C

(2.18)

where C is cell capacity, A h.

2.3.11 Cell Life Cell life includes cycle life and rest life. Cycle life refers to the number of cycles that a battery can charge and discharge to a given value (e.g. 80% of the initial value) under certain conditions (e.g. a certain voltage range, charge/discharge rate, and ambient temperature). Rest life refers to the aging time for the cell without load to reach a given index in a given environment.

2.3.12 State of Charge (SOC) State of charge (SOC) is the ratio of the remaining capacity to the initial charge state used for a period of time or aging for a period of time, usually expressed as a percentage. SOC = 100% means the fully charged state of cells.

2.3.13 Depth of Discharge (DOD) Depth of discharge (DOD) is the degree of discharge. It reflects the percentage of the cell discharge capacity and the rated discharge capacity.

2.4 Structural Composition SIBs mainly include the cathode material, anode material, electrolyte, separator, binder, conductive agent, and current collector.

2.4 Structural Composition

2.4.1

Cathode Materials

The cathode material is an important part of the secondary SIBs, which not only participates in the electrochemical reaction as an electrode but also provides sodium sources. But it is not appropriate to directly replace lithium with sodium in the electrode material of LIBs. Specific energy, cyclic performance, safety, cost, and environmental impact should be considered comprehensively when designing and selecting cathode materials for SIBs. The ideal cathode material for SIBs should meet the following conditions: (a) Large specific capacity, which requires the cathode material with a low relative molecular mass, and a large amount of Na can be inserted into the host structure; (b) High operating voltage, which requires the largely negative Gibbs free energy of the discharge reaction; (c) Good rate charge and discharge performance, which requires a high Na+ diffusion rate in the electrode material and surface; (d) Long cycle life, which requires the small structural change during the process of Na+ extraction/insertion; (e) Good safety, which requires the electrode materials to have high chemical stability and thermal stability; (f) Easy manufacture, environmental friendliness, and cheapness. The cathode materials mainly include oxides, polyanions, Prussian blue analogs, and organics. The oxides mainly include layered structure and tunnel structure oxides, and polyanions include phosphate, fluorinated phosphate, pyrophosphate, and sulfate. The operating voltages of common cathode materials are summarized in Figure 2.5 [7]. Similar to LIBs, almost all cathode materials of SIBs contain variable valence transition metals, which can maintain the electrical neutrality of the host materials during the Na+ extraction/insertion. Among them, the layered oxides with periodic layered structures have attracted much attention due to the simple preparation method, high specific capacity, and voltage. In addition, the lattice oxygen redox can be used to further improve the energy density of such materials. However, most of the layered materials are sensitive to water and moist air, which affects the stability and electrochemical performance of the electrode materials. Type tunnel oxides with a unique “S”-shaped channel are resistant to water and moist air, have excellent ability, and have the advantage of high electrochemical stability in both organic and aqueous environments. But their practical application is still restricted by the low first charging capacity. Polyanions with a solid three-dimensional skeleton have the characteristics of high operating voltage and high-rate charge/discharge. In general, the stronger the electronegativity of polyanions, the higher the voltage the material possesses. The combination of different polyanions can produce new hybrid polyanion units with different electronegativities, thus enhancing the electrochemical properties of the material, which is important for the design of high-voltage and energy-density materials. However, the electrical conductivity of these compounds is generally poor. In order to improve their electronic and ionic conductivity, carbon coating and doping

21

5

200 300 400 500600 (Wh kg–1) Layered

1

2

Hexacyanometalate Phosphate or fluorophosphate

3

Voltage (V vs. Na)

4

5

4 8

3

i 12

a 10 ii 9 b

13

6

7 iii

11 c d e

14 g h mn

i o 15

2

50

100

150

iv f j

v pqr

200

Specific capacity (mAh g–1)

k s vi

250

1. Na4Co0.24Mn0.3Ni0.3(PO4)2P2O7 2. Na3Co3(PO4)2P2O7 3. Na2CoPO4F 4. Na7V4(P2O7)4(PO4) 5. Na3(VO)2(PO4)2F 6. Na3V2(PO4)2F3 7. Na3(VO0.8)2(PO4)2F1.4 8. Na2MnP2O7 9. NaVPO4F 10. Na4Fe3(PO4)2P2O7 11. Na3V2(PO4)3 12. Na2CoP2O7 13. Na2FeP2O7 14. Na2FePO4F 15. NaFePO4 a. Na0.85Li0.17Ni0.21Mn0.64O2 b. NaFeO2 c. Na(Ni1.5Sb0.3)O2 d. Na(Ni0.5Ti0.5)2 e. Na(Ni0.25Fe0.5Mn0.25)O2

f. Na0.66Li0.18Mn0.71Ni0.21Co0.08O2 g. NaCrO2 h. NaNiO2 i. Na(Ni0.5Mn0.5)O2 j. Na(Ni0.60Co0.05Mn0.35)O2 k. Na0.78Li0.18Ni0.25Mn0.583Ow I. Na0.61Ti0.48Mn0.52O2 m. Na0.74CoO2 n. Na(Ni0.33Mn0.33Co0.33)O2 o. Na0.44MnO2 p. Na0.95Li0.15(Mn0.55CO0.10Ni0.15)O2 q. Na0.66(Fe0.5Mn0.5)O2 r. NaMnO2 s. Na2MnMn(CN)6 i. NaNiFe(CN)6 ii. Na1.32Mn[Fe(CN)6]0.835H2O iii. NaxCo(Fe(CN)6]0.909H2O iv. Fe(Fe(CN)6)0.79 v. Na0.61Fe(Fe(CN)6)0.94 vi. Na2MnMn(CN)6

Figure 2.5 Voltage, specific capacity, and energy density of classified cathode materials for SIBs. Source: Reproduced from Tian et al. [7]/with permission of American Chemical Society.

2.4 Structural Composition

methods are often employed, which will lead to the reduction in their volume energy density. Prussian blue compounds with a more open skeleton have a moderate discharge capacity and operating voltage. The interstitial sites in the crystal structure provide space for Na+ transport during the charge and discharge process, but the presence of interstitial water will also occupy the sodium storage sites, thus reducing the sodium storage capacity. It is important to investigate the mechanism of interstitial water in Prussian blue compounds to improve their electrochemical performances. Organic cathode materials generally have the characteristics of multielectron reaction, resulting in high specific capacity, but they generally suffer from the problems of easily dissolving in organic electrolytes and poor electronic conductivity.

2.4.2

Anode Materials

The anode material is also the main component of SIBs. The ideal negative electrode material should meet the following conditions: (a) Low redox potential for the Na extraction/insertion reaction to cater to the high output voltage of SIBs; (b) Small potential change of the electrode during Na+ extraction/insertion to ensure low-voltage fluctuations in the charge and discharge process; (c) Good structural and chemical stability during Na+ extraction/insertion to enable a high cycle life and safety of cell; (d) High reversible specific capacity; (e) Good ion and electron conductivity to obtain high-rate charge/discharge and excellent low-temperature charge/discharge performances; (f) Ability to form a dense and stable solid electrolyte film (SEI) with the electrolyte to prevent continuous reduction of the electrolyte on the anodic surface and irreversible consumption of Na from the cathode electrode; (g) Simple manufacture and low cost; (h) Abundant resources and environmental friendliness. The anode materials mainly include carbon-based, organic, alloy, and other anode materials. As the sodium storage capacity of graphite is too low to be used as anode materials for SIBs, carbon-based materials mainly refer to amorphous carbon and nanocarbon materials. Amorphous carbon-based anode materials with large disorder have low sodium storage potential, moderate storage capacity (higher than titanium-based materials and lower than alloy-based materials), small volume deformation after Na+ extraction, and excellent cycling performance, as shown in Figure 2.6 [8]. Nanocarbon materials mainly include graphene and carbon nanotubes (CNTs), relying on the surface adsorption to realize Na+ storage achieving rapid charge and discharge. However, low initial coulombic efficiency and poor cycling performances make it difficult to obtain practical applications. Titanium-based materials have good stability in air and low strain after Na+

23

2 Characteristics of Sodium-ion Batteries

3.0 Carbon-based materials Alloy-based materials

2.5

Metal oxides and sulfides

2.0

Ti-based materials

Na0.66[Li0.22Ti0.78]O2

Organic materials

1.5 1.0 0.5 0.0

Na2Ti3O16

Voltage (V vs. Na+/Na)

24

NiCo2O4

FeS2

Sb2O4

NiO

MoS2 Na2C16H10O4 Ni S Sb2O3 2 3 SnS2 TiO2 Li4Ti5O5 Sb CuO γ-Fe2O3 Mo3Sb7 Co3O4 Fe3O4

Na4Ti5O12

Carbon

Na2Ti3O7

Ge

Na2C8H4O4SnO

SnSb Sn4P3

SnS

NiP3

Sn

P

SnO2

0 100 200 300 400 500 600 700 800 9001000 1600 2000 Sepcific capacity (mAh g–1)

Figure 2.6 Average voltage (discharge) versus capacity plot of anode materials for SIBs. Source: Reprinted with permission from Kang et al. [8]. Copyright 2015 Royal Society of Chemistry.

extraction but poor capacity. Organic compounds have chemical richness, low cost, environmental friendliness, multielectron reaction, and tunable electrochemical window but suffer from poor electronic conductivity and are easily soluble in electrolytes. Na-M (M = Sn, Pb, P, Sb, and Bi) alloys have a high theoretical capacity, low sodium storage potential, and good electrical conductivity as anode materials for SIBs. It also avoids the dendritic problem with good safety, but the sodium alloy has a large volume expansion, electrode pulverization, and capacity rapid decays during repeated cycles. Other anodic materials include metal oxides such as Fe2 O3 , CuO, CoO, MoO3 , and NiCo2 O4 and metal sulfides such as MoS2 and SnS. These materials have high theoretical capacity but poor conductivity, easy agglomeration, irreversible conversion reactions, and poor cycle and rate performances.

2.4.3

Electrolytes

Electrolytes are an important part of the cell, undertaking the bridge role of transporting ions to form a good ion channel. The electrolytes include liquid and solid electrolytes. The liquid electrolyte is also customarily referred to as the electrolyte, which generally consists of nonaqueous organic solvent and sodium salt. The nonaqueous solution electrolyte should meet the following conditions as used in the system of SIBs: (a) High ionic conductivity; (b) Good thermal stability without decomposition over a wide temperature range; (c) Wide electrochemical window, which is stable in the range of 0–4.5 V (relative to Na/Na+ ); (d) High chemical stability, which does not react with the cathode, anode, current collector, separator, binders, etc.;

2.4 Structural Composition

(e) (f) (g) (h)

Good-solvating properties of ions; No toxicity, low vapor pressure, and safety; Good reversible reaction; Simple manufacture and low cost.

Among the above factors, chemical stability, safety, and reaction rate are pivotal. The solvents currently used in SIBs are mainly ester and ether solvents. Ester solvents, especially cyclic and chain carbonates, are commonly used. Electrolytes based on carbonate solvents often have high ionic conductivity and good oxidation resistance. Cyclic carbonates with relatively high viscosity have a significantly higher dielectric constant than other types of solvents and are able to dissolve sodium salts better. Ether solvents have a lower dielectric constant than cyclic carbonates but higher than chain carbonates. And ether solvents with relatively low viscosity are limited in practical application due to their poor resistance to oxidation and easy decomposition under high-voltage state. For sodium salts, the features of large-radius anions and weak anion–cation linkage are conducive to salts well dissolving in solvents and providing sufficient ionic conductivity to obtain good ion transport. Sodium salts include inorganic and organic sodium salts. Inorganic sodium salts are commonly used but have some problems, such as high oxidation and easy decomposition. Organic sodium salts are more thermally stable but have the drawbacks of corrosive collectors and relatively high cost. A small amount of additive can make up for some of the abovementioned defects of solvents or sodium salts to form a protective film on the surface of the electrode material, reduce the flammability of the organic electrolyte, or prevent overcharge, which makes the study of additives more and more important.

2.4.4

Separators, Binders, Conductive Agents, and Current Collectors

At present, the separators, binders, conductive agents, and current collectors used in SIBs are generally borrowed from the relatively mature system of LIBs. The separators are a very critical component of the liquid SIBs. They are able to not only separate the cathode and anode electrodes to prevent short circuits but also ensure the penetration of electrolyte solvent and transport of solvated Na+ . Requirements for separators in SIBs: (a) Good chemical stability and certain mechanical strength (including tensile strength and puncture resistance) in the electrolyte; (b) Resistant to the oxidation and reduction of electrode materials and corrosion of the electrolyte; (c) Small resistance to the ions transport, accordingly reducing the internal resistance and energy loss in the high-current discharge of cells; (d) Good insulator of electrons and blocking the growth of dendrites and particles detaching from the electrode; (e) Good thermal stability, physical properties (including wettability, chemical stability, permeability, and safety performance), and thickness uniformity; (f) Abundant and inexpensive material sources.

25

26

2 Characteristics of Sodium-ion Batteries

Currently, the commonly used separators in SIBs are polyolefin polymer materials (such as polyethylene PE, polypropylene PP, and PP-PE-PP composite film.) and glass fiber (the main components are inorganic oxides such as silica and alumina). The binders bind the active material to the conductive agent and current collector in the process of the electrode preparation. The main requirements of SIB binders are easy processing, low toxicity, low cost, and environmental friendliness; moderate swelling capacity in the electrolyte; sufficient thermal stability during the drying process; stability to sodium salts, solvents, and decomposition products in the electrolyte; certain electron and ionic conductivity; good electrochemical stability over a wide voltage range; good safety performance; controlled pH to prevent collector corrosion; and flexibility to mitigate volume changes in the electrode during the charge and discharge process. The binders commonly used in SIBs include PVDF, sodium carboxymethyl cellulose (CMC), and polyacrylic acid (PAA). The conductive agent mainly functions as an electricity conductor in the electrode. The requirements for SIB conductive agents are high electronic conductivity; good chemical/electrochemical stability without side reactions; a certain ability to retain electrolyte (fully wetted by the electrolyte); uniformly disperse into the slurry; and easy production, low cost, and environmental friendliness. At present, the conductive agents commonly used in SIBs are some carbon materials, including acetylene black, Super P, conductive graphite KS, conductive graphite SFG, Cochin black, CNTs, carbon nanofibers, and graphene. The current collector is to gather the generated current by the electrode to realize external current conduction. The contact between the collector and active material will directly affect the electrochemical performance of the cell. The requirements for SIB conductive agent are good electrical conductivity and low internal resistance; good chemical and electrochemical stability over a wide voltage range; resistance to corrosion of the electrolyte; good flexibility and easy processing; and stable mechanical properties. In SIBs, the cathode and anode collectors can use cheaper aluminum foil because Na+ does not alloy with Al.

References 1 Li, Y., Lu, Y., Zhao, C. et al. (2017). Recent advances of electrode materials for low-cost sodium-ion batteries towards practical application for grid energy storage. Energy Storage Materials 7: 130–151. 2 Yabuuchi, N., Kubota, K., Dahbi, M., and Komaba, S. (2014). Research development on sodium-ion batteries. Chemical Reviews 114: 11636–11682. 3 Komaba, S., Hasegawa, T., Dahbi, M., and Kubota, K. (2015). Potassium intercalation into graphite to realize high-voltage/high-power potassium-ion batteries and potassium-ion capacitors. Electrochemistry Communications 60: 172–175. 4 Gao, X., Liu, H., Chen, H. et al. (2022). Cationic-potential tuned biphasic layered cathodes for stable desodiation/sodiation. Science Bulletin 67: 1589–1602.

References

5 Lee, D.H., Xu, J., and Meng, Y.S. (2013). An advanced cathode for Na-ion batteries with high rate and excellent structural stability. Physical Chemistry Chemical Physics 15: 3304–3312. 6 Sun, Y., Zhao, L., Pan, H. et al. (2013). Direct atomic-scale confirmation of three-phase storage mechanism in Li4 Ti5 O12 anodes for room-temperature sodium-ion batteries. Nature Communications 4: 1–10. 7 Tian, Y., Zeng, G., Rutt, A. et al. (2020). Promises and challenges of next-generation “beyond Li-ion” batteries for electric vehicles and grid decarbonization. Chemical Reviews 121: 1623–1669. 8 Kang, H., Liu, Y., Cao, K. et al. (2015). Update on anode materials for Na-ion batteries. Journal of Materials Chemistry A 3: 17899–17913.

27

29

3 Cathode Materials of SIBs Xu Gao, Wentao Deng, Guoqiang Zou, Hongshuai Hou, and Xiaobo Ji Central South University, College of Chemistry and Chemical Engineering, Changsha, China

In line with lithium-ion batteries (LIBs), cathode materials for sodium-ion batteries (SIBs) are generally comprised of a stable two-dimensional (2D) or three-dimensional (3D) framework structure, transition metal ions with variable valences, and movable alkali metal (sodium) ions as charge carriers, while the design of structure and composition significantly determine the working voltage, energy density, power density, cycle life, and manufacturing cost of a SIB. However, despite similar working mechanisms, directly inheriting those cathode materials for LIBs is not realistic for SIBs due to the gap between the standard electrode potential and the ionic size of sodium and lithium metal/ion. In this case, discovering suitable cathode materials has become one of the critical missions since the renaissance of SIBs. Theoretically, an ideal cathode material for a SIB is expected to possess: (i) a high working potential with low voltage hysteresis which is important to achieve a high energy density and energy efficiency for a SIB; (ii) a large gravimetric/volumetric specific capacity which is also crucial to the energy density; (iii) a good chemical stability in either non-aqueous or aqueous electrolytes that ensures the stable operation; (iv) a stable host structure allowing for reversible intercalation and deintercalation of Na ions which is related to the cycling stability; (v) a high electronic and ionic conductivity that is favorable for fast electrochemical reactions and thereby the power density; (vi) non-toxic compositions, easy synthesis methods, and low material costs corresponding to promising real-world applications [1–3]. So far, the as-reported cathode materials for sodium-ion batteries mainly include transition metal oxides, polyanionic compounds, Prussian blue and corresponding derivatives, and organic compounds (Figure 3.1). Distinct properties in crystal structure, ionic conductance, electronic conductivity, and charge compensation mechanism endow these candidates with different electrochemical behaviors and also offer a big room for the exploration of advanced cathode materials for SIBs [4].

Sodium-Ion Batteries: Technologies and Applications, First Edition. Edited by Xiaobo Ji, Hongshuai Hou and Guoqiang Zou. © 2024 WILEY-VCH GmbH. Published 2024 by WILEY-VCH GmbH.

3 Cathode Materials of SIBs

Cathodes materials for SIBs

phosphates Sulfates borates silicates mix-polyanion

layered oxides tunnel oxide multiphasicoxides

Polyanion

Oxide

Prussian blue Prussian blueanalogues

Prussian Blue

Fluorides Organic materials

Other-type

(a)

Cathode Working potential/V (vs. Na+/Na)

30

5

4

3

2 (b)

0

50

100

150

200

250

–1)

Specific capacity (mAh g

Figure 3.1 (a) Classification and (b) comparison of cathode materials for SIBs. (b) Source: Reproduced from Hwang et al. [1]/with permission of Royal Society of Chemistry/CC BY 3.0.

3.1 Polyanion Cathode The general structural formula of the sodium polyanion compounds can be expressed as Nax Ma (Xb Oc )d Ye , wherein M is mainly one or several transition metals with variable valence (such as Fe, Ni, Mn, Co, Ti, V, and Cr); X is mainly P, S, Si, As, Mo, W, etc.; Y is F or OH, etc. [5]. Among them, the common polyanionic cathode materials for sodium-ion batteries are phosphates, pyrophosphates, sulfates, and

3.1 Polyanion Cathode

silicates. Most of these materials have a three-dimensional framework structure constructed by (Xb Oc )d n− polyanions and MO6 polyhedrons, endowing the materials with excellent structural stability, thermal stability, a fast ion transfer rate, and outstanding rate and cycle performances. Benefiting from the strong inductive effect of polyanionic groups, which weakens the covalent feature of M—O bonding, transition metal redoxes in polyanion-type materials usually exhibit higher working voltages compared to those in layered oxides. In addition, by introducing two or more anionic groups, mixed polyanionic compounds such as fluorophosphate, phosphate, and pyrophosphate, can be obtained, further enhancing the inductive effects and elevating the redox potentials. However, polyanion-type materials are commonly plagued by low electronic conductivity, requiring essential modifications like bulk doping or surficial carbon coating to improve their electrochemical performances.

3.1.1

Phosphates

3.1.1.1 Olivine-type Phosphates (NaMPO4 , M=Fe, Mn, etc.)

As an analogue of LiFePO4 , olivine-type NaFePO4 has attracted wide attention in view of the remarkable success of the former in commercial LIBs. In a typical olivine-type structure (Figure 3.2a), corner-sharing FeO6 octahedra construct zigzag chains parallel to the c-axis, which are bridged by PO4 tetrahedra via sharing either edge or corner, building 1D tunnels along the b-axis that allow for fast and stable migration of sodium or lithium ions [6]. As expected, olivine-type NaFePO4 is able to deliver a reversible specific capacity of over 140 mAh g−1 , accompanied by a long discharge plateau at around 2.75 V, while good cycling stability is also evidenced [7]. Different from LiFePO4 , olivine-type NaFePO4 exhibits two charge plateaus (Figure 3.2b), which are likely rooted in a distinct two-stage phase evolution process separated by an intermedia phase Na2/3 FePO4 [7, 9]. Within the desodiation region of 2/3 ≤ x ≤ 1, a high solid solubility of Na+ /vacancy allows for a solid-solution process, accounting for the first charge plateau (slope); When further desodiated to x < 2/3, a two-phase reaction from Na2/3 FePO4 to FePO4 occurs, corresponding to the second charge plateau. Kinetics discrepancy is suggested as the dominant factor separating the voltage of these two plateaus, while the single plateau appearing in discharge is attributed to the overlap of the kinetic-restrained two-phase plateau and the kinetic-fast solid-solution plateau. Worthy to mention, two discharge plateaus can be observed at high current rates, due to the different polarization of two sodiation processes. Although olivine-type NaFePO4 demonstrates good potential for sodium storage, unfortunately, it is not a thermal stable phase and cannot be obtained by conventional synthetic methods. To date, ion exchange, whether in a chemical or electrochemical way, is the only route to prepare olivine-type NaFePO4, while such a complicated synthesis process raises the concern of production cost due to the use of expensive and/or toxic reagents and therefore limits industrial applications. Instead, maricite has been proven to be thermodynamically favored phase of NaFePO4 , which can be obtained by high-temperature or hydrothermal reactions [10, 11].

31

3 Cathode Materials of SIBs

1st cycle 2nd cycle 15th cycle 50th cycle 100th cycle

0.1V

32

(a)

(b) Olivine-NaFePO4

a) 1st 2nd 3rd 10th 50th 100th 200th 300th

(c)

(d) Maricite-NaFePO4

Figure 3.2 Crystal structure and voltage profiles for (a, b) olivine- and (c, d) maricite-NaFePO4 . Source: Reproduced from Yabuuchi et al. [6]/with permission of American Chemical Society, (b) Reproduced from Tang et al. [7]/with permission of The Royal Society of Chemistry. (d) Reproduced from Liu et al. [8]/with permission of John Wiley & Sons.

In maricite NaFePO4 (Figure 3.2c), the site occupations of Fe2+ and Na+ ions are completely reversed in contrast to olivine NaFePO4 , while the position of PO4 tetrahedra remains unchanged, leading to the absence of 1D channels for the migration of sodium ions [6]. In this case, maricite NaFePO4 was deemed to be electrochemically inactive. However, nanosized maricite NaFePO4 surprisingly exhibits good electrochemical activity. A maximum reversible capacity of ∼142 mAh g−1 has been obtained through slope-like charge–discharge profiles, followed by outstanding cycling stability (Figure 3.2d) [8, 10]. On the basis of density function theory (DFT) calculations, such an unexpected activity was ascribed to the lowered energy barrier for the migration of sodium ions in amorphous FePO4 , which simultaneously forms during the desodiation of nanosized NaFePO4 Maricite [10]. Notably, the dedicated design of nanostructured FePO4 /C or NaFePO4 /C composites could further boost the specific capacity, rate capability, and cycling stability of maricite-type cathodes [8, 11]. Similar to NaFePO4 , NaMnPO4 also has two analogue phases, olivine and maricite, while the latter is thermodynamically favored but electrochemically inert. Olivine-type NaMnPO4 or NaMn1−x Fex PO4 can be obtained by ion exchange or other soft chemistry methods like the phosphate-format precursor method

3.1 Polyanion Cathode

and the topotactic molten salt reaction, however, demonstrating unsatisfactory performances [12, 13]. Nanosizing the particles and optimizing carbon coating are believed to be the most promising routes to improve the electrochemical performances of Mn-based Phosphates, but more investigations are still needed. 3.1.1.2 NASICON-type Phosphates (Na3 M2 (PO4 )3 , M=Ti, V, Ni, Fe, Mn, etc.)

Sodium super ionic conductor (NASICON) type materials derive from a family of materials possessing similar structure and superior ionic conductivity. Originating from Na1+x Zr2 P3−x Six O12 , which was first reported by Goodenough in the 1960s [14], NASICON-type materials were initially utilized as solid-state electrolytes due to the fast sodium-ion transport but afterwards developed as one of the important cathode materials by introducing transition metal ions with electrochemical activities. For NASICON-type phosphate, the general formula can be described as Na3 M2 (PO4 )3 , where M refers to Ti, V, Ni, Fe, Mn, etc. [15]. In a typical structure (Figure 3.3a) [16], two MO6 octahedra and three PO4 tetrahedra are connected by sharing the corners, forming a so-called “lantern unit” ([M2 (PO4 ]3 ). This unit cell periodically extends along the c axis while some of the sodium ions occupy space between adjacent MO6 octahedra, denoted as Na1 sites, and other sodium ions occupy space on the sides of the units, denoted as Na2 sites. The lantern units well construct a firm 3D framework with large interstices that allows for fast migration of sodium ions, endowing NASICON-type cathode materials with excellent rate capability and cycling stability. Moreover, both the cation (Mn+ ) and anion (O2− ) in the NASICON-type structure are highly replaceable, offering great feasibility for material design and engineering. Na3 V2 (PO4 )3 is one of the typical NASICON-type phosphates. Inside the 3D framework built by VO6 octahedra and PO4 tetrahedra, one Na ion occupies Na1 sites with six-fold coordination (6b) are occupied by one Na ion while two Na ions lie in Na2 sites with eightfold coordination (18e), as illustrated in Figure 3.3a. However, during electrochemical desodiation, only the Na ions in Na2 sites can be reversibly extracted from the host structure, accompanied by the one-electron reaction of the V4+ /V3+ redox couple at around 3.4 V (vs. Na+ /Na). This corresponds to a theoretical capacity of 117 mAh g−1 , which can be fully utilized with the support of conductive carbon networks (Figure 3.3b) [17]. Considering that Na ions in Na1 sites remain almost unchanged during (de)sodiation, it was believed that the migration of Na+ in Na3 V2 (PO4 )3 mainly occurs through the Na2–Na2 pattern [20]. By means of DFT calculations, X. Ji et al. [18] further proposed a three-dimensional transport that comprises two pathways along the x and y directions and one possible curved route for Na migration (Figure 3.3c). Chen et al. [19] further suggested that the Na ions at both Na1 and Na2 sites are evolved in the Na ion transportation via a concerted ion-exchange route (Figure 3.3d). It is worthy to mention that the immobility of the Na ion in the Na1 site is likely rooted in the difficulty of oxidizing V4+ to V5+ . In this case, these results may indicate a lingering research topic on the ion migration mechanism in Na3 V2 (PO4 )3 . Moreover, Na3 V2 (PO4 )3 can accommodate one more Na+ if lowering down the discharge cut-off voltage, resulting in a plateau at around 1.6 V and an end member of Na4 V2 (PO4 )3 [21]. In this case, Na3 V2 (PO4 )3

33

3 Cathode Materials of SIBs 4.8 0.1 C 0.2 C 0.5 C 1C

4.4

Voltage (V vs. Na+/Na)

34

4.0 3.6 3.2

3.44 3.42

2.8

3.38

3.32 3.30

(c)

(d)

Discharge plateau

3.34

1.6

(b)

ΔV = 0.04 V

3.36

2.0

(a)

Charge plateau

3.40

2.4

0

Current density: 0.1 C 0

10

20

20

30

40

40

50

60

60

80

100

120

Specific capacity (mAh g–1)

Figure 3.3 (a) Typical crystal structure of NASCON-type Nax M2 (AO4 )3 , M refers to transition metals, A refers to P for phosphate. Source: Reproduced from Ouyang et al. [16]/with permission of Springer Nature/CC BY 4.0. (b) Voltage profiles for Nax M2 (AO4 )3 /C composite. Source: Reproduced from Song et al. [17]/with permission of Royal Society of Chemistry. (c) Possible Na ion migration paths in Na3 V2 (PO4 )3 along x, y, and z directions. Source: Reproduced from Song et al. [18]/with permission of Royal Society of Chemistry/CC BY 3.0. (d) Sketch map of the direct diffusion route (path A) and stepwise ion exchange route (path B) for Na migration, top view on the left and side view on the right. Source: Reproduced from Wang et al. [19]/with permission of American Chemical Society.

can also be utilized as an anode material despite its low capacity and poor cycling stability. During the (de)sodiation process, Na3 V2 (PO4 )3 undergoes a reversible two-phase reaction between Na3 V2 (PO4 )3 and Na1 V2 (PO4 )3 , along with an overall volume change of ∼8.26%, which is highly favorable for stable cycling [22]. However, the real rate and cycling performance of Na3 V2 (PO4 )3 are significantly plagued by low electronic conductivity. By adopting proper carbon-coating modification, a high-rate capacity up to 500 C and a long cycle life over 20 000 cycles at 30 C can be achieved [23]. Besides, substituting V with other transition metals like Mn [24] and Fe [25] is another popular strategy to modify Na3 V2 (PO4 )3 , either from the consideration of electrochemistry or economy issue as V is expensive and toxic. For example, by replacing 50% of V by Mn2+ , Goodenough et al. [26] successfully obtained Na4 MnV(PO4 )3 , which delivers an initial reversible capacity of 101 mAh g−1 at 1 C, followed by a capacity retention of 89% after 1000 cycles. The introduction of Mn2+ leads to a new voltage plateau at 3.6 V, which corresponds to reversible Mn3+ /Mn2+ redox, elevating the working voltage of Na3 V2 (PO4 )3 . Chen et al. [27] further reported a Na4 MnCr(PO4 )3 , which records a high specific capacity of ∼160 mAh g−1 , contributed by Mn2+ /3+ , Mn3+ /4+ , and Cr3+ /Cr4+ redox. Benefiting from the easy replacement of transition metal ions, a large amount of

3.1 Polyanion Cathode

NASICON type Na3 M2 (PO4 )3 with various transition metal compositions has been reported. 3.1.1.3 Pyrophosphate Na2 MP2 O7

Phosphates are not stable at high temperatures above 500 ∘ C, where oxygen loss would be triggered for PO4 3− , leading to the formation of pyrophosphate (P2 O7 4− ), which has higher thermostability. Sodium-transition-metal pyrophosphates (Na2 MP2 O7 , M=Fe, Co, Mn, et al.) have also been developed as cathode materials for SIBs since 2012, when the electrochemistry behavior of Na2 FeP2 O7 was discovered [28]. Na2 MP2 O7 has a variety of crystal structures that depend on the species of transition metal and synthesis conditions [5]. For Na2 FeP2 O7 , it generally forms in a triclinic P-1 structure built by Fe2 O11 dimers (corner-shared FeO6 octahedra) and the P2 O7 group (corner-shared PO4 tetrahedra) via face-sharing or corner-sharing, while Na ions occupy in four crystallography sites [29]. The as-obtained 3D framework gives birth to Zig-Zag type channels, allowing for fast transportation of Na ions. Theoretically, one Na ion can be reversibly extracted from Na2 FeP2 O7 along with the redox of Fe3+ /Fe2+ , corresponding to a capacity of 97 mAh g−1 . In practice, a reversible capacity of 92 mAh g−1 as well as good cycling stability and rate capability can be achieved with composited carbon [30]. During the charge/discharge process, two voltage plateaus appear at around 3.0 V while another one appears at around 2.5 V, which were attributed to three biphasic reactions through the intermedia phase emerging at Na1.75 , Na1.5 , and Na1.25 [29]. Moreover, both pristine Na2 FeP2 O7 and desodiated Na2 FeP2 O7 are thermally stable up to 500 ∘ C, which is superior to LiFePO4 (∼400 ∘ C) and comparable to FePO4 (∼500 ∘ C). Notably, the rest of the Na in NaFeP2 O7 is not accessible due to the depletion of Fe3+ /Fe2+ redox and the difficulty of further oxidizing Fe3+ . One of the possible strategies to extend the available capacity is to tune the Na/Fe ratio in the formula of Na4−a Fe2+a/2 (P2 O7 )2 (2/3 ≤ a ≤7/8) [31]. Supported by a carbon coating, Na3.32 Fe2.34 (P2 O7 )2 delivers a reversible capacity of ∼100 mAh g−1 at a current rate of 12 mA g−1 , which is slightly higher than the theoretical capacity of Na2 FeP2 O7 [32]. Among other sodium-based pyrophosphates, Na2 CoP2 O7 attracted much attention as it has three different polymorphs, namely triclinic (P-1), tetragonal (P42/mnm), and orthorhombic (Pna21 ), strongly dependent on the constituent 3d cation size. Thermodynamically speaking, the triclinic phase is most unstable, while the orthorhombic form is most stable for Na2 CoP2 O7 . When employed as cathode materials for SIBs, orthorhombic Na2 CoP2 O7 demonstrates a reversible capacity of ∼80 mAh g−1 in a voltage window of 1.5−4.7 V and features in the slop-type voltage profiles [33]. Kim et al. stabilize the rose form of Na2 CoP2 O7 through the introduction of Na deficiencies [34]. This triclinic (P-1) Na2 CoP2 O7 exhibits a reversible capacity of ∼80 mAh g−1 mainly contributed by the long plateau above 4.0 V, which brings an average voltage of 4.3 V vs. Na/Na+ and therefore an improved energy density compared to its blue polymorph. In addition, Na2 MnP2 O7 in the space group P-1, isostructural with Na2 FeP2 O7 and rose Na2 CoP2 O7 , shows a reversible capacity of 90 mAh g−1 contributed by Mn3+ /Mn2+ redox in the voltage range of 4.5−1.5 V (vs. Na/Na+ ) [35]. Despite of the superior thermal stability and rich

35

36

3 Cathode Materials of SIBs

structural diversity arousing much research interest, the real-world applications of pyrophosphates may be hindered by their drawbacks, including low capacity and relatively sluggish ion-transportation properties.

3.1.2

Sulfates/Borates/Silicates

3.1.2.1 Sulfates

In contrast to other polyanion compounds, sodium-transition-metal sulfates exhibit a higher working potential owing to the stronger electronegativity of SO4 2− and the corresponding greater inductive effects. As a typical example, alluaudite Na2 Fe2 (SO4 )3 demonstrates a working potential of 3.8 V (vs. Na/Na+ ), which is the highest redox potential of a Fe3+ /Fe2+ couple ever discovered [36]. Notably, alluaudite Na2 Fe2 (SO4 )3 has a special AA′ BM2 (XO4 )3 -type structure under the P21 /C space group, where A, A′ , and B refer to Na ions in Na2, Na3, and Na1 sites, respectively, while M=Fe and X=S (Figure 3.4a). In this 3D framework, two edge-sharing FeO6 octahedra constitute a Fe2 O10 dimer, while all these isolated Fe2 O10 dimers are connected by SO4 tetrahedra through sharing the corners, forming large 1D channels along the c axis. Na ions are assigned to three crystallographic sites, namely, fully occupied Na1 sites and partially occupied Na2/Na3 sites. During the charging process, Na ions are mainly extracted from Na3 sites and only a small part from Na2 and Na1 sites (Figure 3.4b,c) [37]. Once Na ions in Na1 sites are extracted after charging above 4.06 V, Fe3+ ions in Fe1 sites (8f) are prone to migrate into Na1 vacancies. In the following discharge process, Na ions will be intercalated back into Na2 sites at first, and then into Na3 sites. The sodium extraction from Na1 sites and the accompanying migration of Fe3+ ions are found to be irreversible. However, it is also evidenced that a small portion of Na ions can be reversibly (de)intercalated from/into the as-formed Fe1 vacancies, starting from the

(b)

(a)

(c)

Figure 3.4 (a) Crystal structure, (b) voltage profiles (inset: dQ/dV curves), and (c) Rate performances (inset: voltage profiles) for Na2 Fe2 (SO4 )3 . Source: Reproduced from Barpanda et al. [36]/with permission of Springer Nature/CC BY 4.0.

3.1 Polyanion Cathode

initial discharge. Therefore, the host structure is stabilized, allowing for a reversible capacity of 102 mAh g−1 and good rate capabilities [37]. Notably, impurities like FeSO4 are commonly detected in the as-obtained Na2 Fe2 (SO4 )3 . The use of excess Na2 SO4 can help to decrease the FePO4 impurities, leading to the formation of off-stoichiometric Na2+2x Fe2−x (SO4 )3 , whose electrochemical performances have grasped much research interest as well [38]. It is anticipated that Mn-based sulfates will exhibit better working potential in SIBs over 4 V than Fe-based sulftes, which is favorable for approaching high energy density [39]. Nonetheless, experimental evidence for the sodium-storage electrochemistry of Mn-based sulfates has not been reported to date. By introducing Mn into Na2 Fe2 (SO4 )3 , Yamada [40] obtained a series of solid solutions in the formula of Na2.5 (Fe1−y Mny )1.75 (SO4 )3 . The voltage of Fe3+ /Fe2+ redox is elevated along with the increase of Mn, whereas the capacity is decreased due to the inactivity of Mn3+ /Mn2+ redox. It is worthy to mention that one of the drawbacks of sulfates is their relatively lower thermostability, as decomposition would occur at around 400 ∘ C, leading to SO2 gas evolution. However, on the other hand, sulfates can be synthesized at lower temperatures via solid-state reactions, ball milling, or solvothermal methods, favoring the low-energy consumption and low-cost production. Similar to other polyanion compounds, coating or compositing with carbon is effective in improving the rate capability of sulfates [41]. 3.1.2.2

Borates

In contrast to other polyanion compounds, borates demonstrate a relatively higher theoretical capacity thanks to the low molar mass of Boron (9.012 g mol−1 ). The diversity of [Bx Oy ]n− groups (e.g. [BO3 ]3− , [B3 O6 ]3− [B2 O4 ]4− , [BO4 ]5− , [B3 O7 ]5− et al. gives birth to various structures of borates, which enables versatile design of materials but complicates the synthesis as well [42, 43]. To date, some borates have been explored as either cathode or anode materials for LIBs however, much less attention has been paid to the borate-type cathodes for SIBs. Notably, pentaborate Na3 FeB5 O10 showing electrochemical activity in SIBs was first reported by the Tarascon group [44]. This material has an orthorhombic structure under the Pbca space group built by FeO4 tetrahedra and [B5 O10 ]5− groups through corner-sharing manner while Na ions are occupied in the interspace of FeO4 –B5 O10 layers. Tested in the voltage range of 1.5–4.5 V, a reversible capacity of ∼30 mAh g−1 , corresponding to ∼0.3 mol Na ions removal and insertion, can be reached while the participation of Fe3+ /Fe2+ redox is captured. However, this material displays huge voltage hysteresis, possibly rooted in its poor electronic and ionic conductivity. 3.1.2.3 Silicate

The formula of sodium orthosilicates can be described as Na2 MSiO4 , in which M refers to transition metals like Fe, Mn, et al. Different from phosphates and sulfates, the framework of Na2 MSiO4 is constructed by corner-sharing MO4 and SiO4 tetrahedra, while Na+ ions are distributed in the voids of the skeleton along the c axis, forming a monoclinic structure under the space group of Pn (Figure 3.5). Benefiting from the super stability of the tetrahedral units, silicates

37

38

3 Cathode Materials of SIBs

a

a

c

b

(a)

(b)

Figure 3.5 Crystal structure of monoclinic Na2 MSiO4 , projected from (a) [010] and (b) [001] direction. Source: Reproduced from Barpanda et al. [42]/with permission of John Wiley & Sons.

exhibit outstanding thermostability up to 1000 ∘ C and low-strain structural evolution during (de)sodiation. Theoretically, a high specific capacity of 278 mAh g−1 can be reached if two Na ions are removed from the material, accompanied by a two-electron transfer reaction of the transition metals. Na2 FeSiO4 is able to give an initial discharge capacity of 119 mAh g−1 at 1.5–4.5 V, supported by carbon coating modification. By means of morphology engineering, a higher capacity of 179 mAh g−1 was also achieved for mesoporous Na2 FeSiO4 /carbon nanotube composites. However, the working potential is around 2 V (vs. Na/Na+ ), which obviously limits their utilization as cathode materials. More notably, Na2 MnSiO4 exhibits a reversible capacity of ∼206 mAh g−1 in 1.5–4.5 V based on the two-electron transfer reaction of Mn2+ /Mn3+ /Mn4+ , featuring in the slop-like voltage profiles. Crystalline Na2 MnSiO4 is transformed to an amorphous phase during charging to 4.5 V and well recovered after fully discharged to 1.5 V, demonstrating reversible structural evolution. Like other Mn-based electrode materials, Mn dissolution would cause serious capacity decay for Na2 MnSiO4 , which can be further improved by surficial engineering by adding electrolyte additives. In contrast, Na2 CoSiO4 shows two discharge plateaus at 3.4 and 3.1 V, yielding ∼125 mAh g–1 within 2.0–3.75 V, which is contributed by Co2+ /Co3+ redox. Co3+ /Co4+ redox might be triggered when charged above 4.0 V, however, followed by structural collapse and fast capacity decay. In addition, Na2 NiSiO4 is theoretically predicted to be a promising cathode material for SIBs with high operating potential over 4 V, but experimental evidence has not yet been found due to the limited voltage window of common organic electrolytes.

3.1.3

Mixed Polyanions

3.1.3.1 Fluorophosphates

Fluorophosphate is one of the most promising polyanion-type cathodes for SIBs due to the high working potential originating from the stronger inductive effects of fluorinions. The research on fluorophosphate-type cathodes for SIBs can be dated back to 2003, when NaVPO4 F was first reported by Barker et al. [45]. Tested in a Li/NaVPO4 F cell, it gave a reversible capacity of 82 mAh g−1 with an average discharge voltage of ∼3.7 V vs. Na/Na+ , followed by fast capacity fading [45].

3.1 Polyanion Cathode (VO)2P2O7 (113)

Na3V2(PO4)3 (113)

Na3V2(PO4)3 (226)

Na3V2(PO4)3 Na4P2O7 (116) (005)

high

Intensity (a.u.)

900 800 700 600 500

low 400 300 25

12

16

20

24

32

36

48

52

Angle (2θ) T (°C)

(a)

(002) Na3V2(PO4)2F3

(112) VPO4

(111) VPO4

(002) NaF 4.5

O

[PO4]

F

z x

(b)

y

[V2O8F3]

Voltage (V)

4.0

Na

3.5

2.5

V

2.0

(c)

1C

3.0

P

NVPF NVPF@C NVPF@C@CNTs

0

20

40

60

80

100

120

140

Capacity (mAh g–1)

Figure 3.6 (a) Structure and compositional evolution of NaF-VPO4 mixture (1 : 1) upon heating. (b) Structural illustration of Na3 V2 (PO4 )2 F3 . (c) The voltage profiles of Na3 V2 (PO4 )2 F3 , Na3 V2 (PO4 )2 F3 @C, and Na3 V2 (PO4 )2 F3 @C@CNTs electrodes. Source: (a) Reproduced from Li et al. [46]/with permission of John Wiley & Sons. (b) Gao et al. [47]/John Wiley & Sons/Public Domain (CC BY 4.0).

Notably, the latest report has demonstrated that the assignments of single-phase NaVPO4 F obtained from solid-state synthesis might be doubtful. As evidenced by in situ XRD (Figure 3.6a) collected during the solid-state synthesis of a NaF-VPO4 mixture (1 : 1), neither tetragonal nor monoclinic NaVPO4 F was obtained at any temperature. Instead, Na3 V2 (PO4 )2 F3 with a tetragonal P42 /mnm structure was found to be the main fluorophosphate-type product, which merely forms at a low synthesis temperature around 400 ∘ C, while higher temperatures lead to the loss of F via VF3 gas release [46]. In addition, tavorite-type NaVPO4 F can be obtained by electrochemical ion exchange from LiVPO4 F. However, it shows quite limited reversible capacity as the cathode material for SIBs [48]. In fact, Na3 V2 (PO4 )2 F3 is one of the widely studied fluorophosphate-type cathodes for SIBs. The typical crystal structure of Na3 V2 (PO4 )2 F3 , as shown in Figure 3.6b, is built by the V2 O8 F3 dioctahedron and PO4 groups via corner-sharing manner while Na ions are located in the channels along [110] and [1–10] directions. Generally, the reversible removal of two Na ions from Na3 V2 (PO4 )2 F3 can be expected, corresponding to a calculated capacity of 128 mAh g−1 supported by V4+ /V3+ redox. Like other polyanion materials, the practical capacity of Na3 V2 (PO4 )2 F3 is constrained by its low conductivity and is highly dependent on the modification of particle morphology and electron transfer properties. Ji et al. [49] reported a Na3 V2 (PO4 )2 F3 /C composite synthesized by a solution-based carbon-thermal reduction method, which delivers an initial discharge capacity of 111.6 mAh g−1

39

40

3 Cathode Materials of SIBs

at 0.091 C in the voltage window of 1.6–4.6 V vs. Na/Na+ , followed by 97.6% capacity retention after 50 cycles, manifesting good cycling performances. Recently, they further proposed an ingenious strategy to cross-link Na3 V2 (PO4 )2 F3 with multi-dimensional nanocarbon frameworks composed of amorphous carbon and carbon nanotubes. Profiting from the favorable particle size that the shortens Na-ion transmission pathway as well as an improved electronic transfer network, a high specific capacity (126.9 mAh g−1 at 1 C, 1 C = 128 mA g−1 ) and remarkably improved long-term cycling stability with 93.9% capacity retention after 1000 cycles at 20 C are recorded (Figure 3.6c) [47]. Supported by in situ synchrotron-based XRD, Croguennec et al. [50] revealed a complicated phase transition process involving at least four intermedia phases instead of the previously recognized single-phase reaction mechanism evidenced by ex situ XRD. Specifically, multiple two-phase transitions mainly occur in the composition between Na3 V2 (PO4 )2 F3 and Na2 V2 (PO4 )2 F3 , where two intermedia phases appear at the composition of Na2.4 and Na2.2 . Further desodiation undergoes a solid-solution process between Na1.8 and Na1.3 . As the end member of charge, NaV2 (PO4 )2 F3 possesses a Cmc21 structure, while the co-existence of V3+ and V5+ ions is detected by nuclear magnetic resonance (NMR) and X-ray absorption spectroscopy (XAS), which is ascribed to the disproportionation reaction of V4+ . Meanwhile, ex situ 23 Na NMR spectra demonstrate that two Na sites are both involved in the desodiation from Na3 V2 (PO4 )2 F3 to Na2 V2 (PO4 )2 F. Tarascon et al. [51] further revealed that the third Na ion in Na3 V2 (PO4 )2 F3 can also be removed when charged up to 4.8 V accompanied with the oxidation of V ions beyond V4+ , leading to a disordered phase having tetragonal symmetry (space group: I4/mmm), as illustrated in Figure 3.7a,b. In the following discharge process, three Na ions can be inserted back into the material when discharged to a low cut-off voltage of 1 V since the insertion of the third sodium occurs below 1.6 V. This plateau appears after a sudden potential drop, displaying large hysteresis in contrast to subsequent charge profiles, which were attributed to different thermodynamics or reaction pathways on charge and discharge. During the following cycling between 4.4 and 1.0 V, approximately 3.13 mol Na ions can be reversibly inserted and removed into/from this disordered phase, generating an initial capacity of 200 mAh g−1 with good cycling stability (Figure 3.7c) and a 14% increase of the energy density. Fluorine atoms in Na3 V2 (PO4 )2 F3 can also be replaced by oxygen atoms, leading to Na3 (VO1−x PO4 )2 F1+2x (0 ≤ x ≤ 1). As a typical example, Na3 (VOPO4 )2 F has a similar framework to Na3 V2 (PO4 )2 F3 but belongs to a different space group (I4/mmm). Due to the relatively lower atomic mass of O than F, Na3 (VOPO4 )2 F has a slightly higher theoretical capacity of 130 mAh g−1 in line with the removal of two Na ions. Nevertheless, the (de)sodiation potential of Na3 (VOPO4 )2 F is slightly reduced at the same time owing to the weaker inducive effects of O. Composited with graphene, Na3 (VOPO4 )2 F exhibits a reversible capacity of 120 mAh g−1 within 2.5–4.5 V [52]. Zhao et al. [53] successfully developed a bubble-assisted soft template method that achieves large-scale production of hollow core-shell Na3 (VOPO4 )2 F spheres with outstanding rate performance (81 mAh g−1 at 15 C) and cycling stability (70% capacity retention after 3000 cycles).

3.1 Polyanion Cathode

(a)

(b)

(c)

Figure 3.7 (a) Initial charge/discharge profiles of Na3 V2 (PO4 )2 F3 with opened charging cut-off voltages. (b) An illustration of the structural evolution of Na3 V2 (PO4 )2 F3 during initial charge/discharge cycle in 1.0−4.8 V. (c) Cycling performances of Na3 V2 (PO4 )2 F3 tested within different (de)sodiation depth. Source: Reproduced from Yan et al. [51]/with permission of Springer Nature/CC BY 4.0.

41

42

3 Cathode Materials of SIBs

Another type of fluorophosphate can be described as Na2 MPO4 F, where M represents bivalent transition metals such as Fe2+ , Co2+ , and Mn2+ . Na2 FePO4 F has a 2D layered structure that can be assigned to an orthorhombic system under the space group (Pbcn). Each [FePO4 F] layer is composed of dioctahedral Fe2 O7 F2 units bridged by PO4 tetrahedral ions, while two Na ions are occupied in the interlayer space with 2D migration pathways [54]. Allowing for one sodium removal, this material is expected to have a theoretical capacity of ∼124 mAh g−1 based on Fe3+ /Fe2+ redox. Supported by carbon coating, a reversible capacity of 110 mAh g−1 was recorded by Komaba et al. [55]. Two pairs of voltage plateaus appear at around 3 V during the charge and discharge process, indicative of two stages of biphasic transition between which a monoclinic intermedia phase is detected at Na1.5 [56]. Isostructural to Na2 FePO4 F, Na2 CoPO4 F is able to give an initial reversible capacity of ∼100 mAh g−1 , mainly due to a high voltage plateau above 4.3 V. However, it suffers from fast capacity decay in the following cycles, likely due to the side reaction triggered by such a high operating voltage [57]. Different from Fe- or Co-based fluorophosphates, Na2 MnPO4 F has a distinct 3D tunnel-type structure belonging to the P21 /n space group. Carbon coated Na2 MnPO4 F delivers a reversible capacity of 120 mAh g−1 , equaling 96% of the theoretical capacity with one Na removal (∼125 mAh g−1 ). Unfortunately, this material is also plagued by fast capacity deterioration and sluggish kinetics [58]. 3.1.3.2 Mixed Phosphates

By importing P2 O7 groups into a PO4 -based framework, or in the converse way, one can obtain a mixed phosphate which exhibits improved structural reversibility and thermal stability compared with the former. One of the typical formulas of mixed phosphates can be described as Na4 M3 (PO4 )2 P2 O7 , where M refers to bivalent transition metals including Fe2+ , Mn2+ , Co2+ , and Ni2+ . Taking Na4 Fe3 (PO4 )2 P2 O7 as an example, the framework is built by [M3 P2 O13 ] layers parallel to the b-c plane and the interconnected P2 O7 groups, while the [M3 P2 O13 ] unit is composed of FeO6 octahedra and PO4 tetrahedra [59]. Notably, this framework enables sodium migration along 3D channels along a, b, and c directions. Supported by carbon coating, the theoretical capacity of 129 mAh g−1 can be fully accessed, accompanied by good cycling stability which stemmed from the solid-solution reaction mechanism with only slight local distortion and a small volume change [60]. Na4 Co3 (PO4 )2 P2 O7 has a capacity of 95 mAh g−1 , which is entirely contributed by voltage plateaus between 4.1 and 4.7 V, making it a choice of high voltage cathode for SIBs [61]. Despite a complex structural evolution including four biphasic transitions and one solid-solution region, this material presents excellent rate performance and cycling stability, benefiting from the steady transport properties. Compensated by V4+ /V3+ redox, Na4 V4 (P2 O7 )4 PO4 delivers a capacity of 92 mAh g−1 together with good rate capabilities up to 10 C [62]. It was reported that partial substitution of V with Al could release the activity of the V4+ /V5+ redox, however, deteriorating the electrochemical reversibility [63]. In addition, the experiences in the research of various mixed phosphates have inspired great interest in exploring other mixed-polyanion type cathode materials for SIBs. Theoretical analysis has

3.2 Oxide Cathode

demonstrated the possibility to mix various polyanions such as PO4 3− , P2O7 4− , CO3 2− , SO4 2− , and SiO4 4− to form stable compounds, indicating a great deal of choice to build novel mix-polyanion materials. As a proof, some trials on the synthesis of Na4 MnPO4 CO3 and Na3 FePO4 CO3 have met with success, showing high capacity over 120 mAh g−1 and considerable cycling stabilities [64].

3.2 Oxide Cathode Na-containing transition metal oxides are mainly divided into layer-type oxides and tunnel-type oxides. Among them, the general formula of sodium layered oxide can be described as Nax TMO2 (TM: transition metal, 0.5 ≤ x ≤ 1), which is constructed by alternately stacking NaO2 layers and TMO2 layers [65]. Sodium layered oxides are one of the most promising cathode materials for SIBs owing to their high theoretical specific capacities up to 300 mAh g−1 , a wide range of operating voltages, and a large diversity of structures and compositions favoring material design and modification. Meanwhile, lots of transition metals, including V, Ti, Ni, Fe, Co, Cr, Mn, and Cu, are proved to be electrochemically active in sodium layered oxides, offering broader choices for designing high-performance and low-cost cathode materials [66, 67]. Moreover, similar to lithium layered oxides, sodium layered oxides can be synthesized by conventional solid-state reactions supported by various methods, such as ball milling, co-precipitation, sol–gel, and spray drying, for the preparation of precursors, which is facile for industrial production. However, the real-world applications of sodium layered oxides are still plagued by diverse issues involving limited practical capacity, sluggish kinetics, severe structural deterioration accompanied by fast capacity fading, and moisture sensibility. In contrast, tunnel oxides have better rate performance, cycling performance, and storage stability due to their stable “S”-shaped framework (channel). Nevertheless, the disadvantage is that tunnel oxides generally have low sodium content (such as Na0.44 MnO2 ), leading to very limited initial charge capacity and therefore constrained practical utilization.

3.2.1

Layered Transition Metal Oxides

3.2.1.1 Structural Classification

Due to the large difference in the ionic radii of sodium ions and transition metal ions, sodium layered oxides are more likely to form cationic ordered rock-salt superstructures. As shown in Figure 3.8, the common feature of the structures of sodium layered oxides is that the oxygen atoms are stacked in a hexagonal close-packed arrangement, in which TMO6 octahedra are connected by sharing edges to form a TMO2 layer, while sodium ions are distributed between adjacent TMO2 layers, forming a NaO2 layer. In 1980, Delmas [68] first proposed to divide the layered structure into P-type (NaO6 is a triangular prism) and O-type (NaO6 is an octahedron) according to the surrounding environment of sodium ions. Meanwhile, a corresponding number like 2 or 3 is used to describe the number of non-repeatable TMO2 sheets within each unit cell, thereby classifying common layered structures as P2, P3, O2,

43

44

3 Cathode Materials of SIBs

(a)

(b)

(c)

(d)

(e)

Figure 3.8 Illustration of (a) structural components, (b) P2-type structure, (c) O2-type structure, (d) O3-type structure, and (e) P3-type structure for sodium layered oxides. Source: Reproduced from Gao et al. [65]/with permission of John Wiley & Sons.

and O3 phases. This classification method has been widely adopted by subsequent researchers. In addition, some distorted phases derived from P2, P3, and O3 phases with the change of space groups, such as P’2 and O’3, etc., are often marked with an apostrophe (’) for distinguishment. In general, O3 and P2 type structures can be simply differentiated according to their sodium content. The sodium content of O3-type layered oxides is generally in the range of 0.8 < x ≤ 1. In this structure, three TMO6 layers (AB, CA, and BC) are repeatedly stacked, while Na ions occupy octahedral (O) sites between two adjacent TMO6 layers. The sodium content of P2-type layered oxides is generally in the range of 0.6 ≤ x ≤ 0.8, and is formed by stacking two TMO2 layers (AB and BA), and the sodium ions are located in the triangular prism position between adjacent TMO2 layers. Specifically, the prismatic sites that can be occupied by sodium ions are classified into two different types: one is the prismatic site face-sharing with two adjacent TMO6 octahedra, labeled as Naf sites; the other is the prismatic site face-sharing with the two adjacent tetrahedra spaces formed by three edge-sharing TMO6 octahedra, marked as Nae sites. The transition metal ions above and below the Naf site has a greater repulsion effect on the sodium ions between them, leading to the unfavorable occupation of Na ions in the Naf site. In general, Na ions at the Naf site and the Nae site are roughly distributed in a ratio of 1 : 2. The P3 phase is usually an intermediate phase produced during the desodiation of O3-type materials, but it can also be directly synthesized by controlling the composition and reaction temperature. In this structure, TMO2 layers are stacked in the order of “AB-BC-CA” while sodium ions occupy prismatic positions between two adjacent TMO2 layers. Each NaO6 prism is face-sharing with a TMO6 octahedron on one side and a tetrahedron space formed by three edge-sharing TMO6 octahedra on the opposite side. The O2 phase is a metastable phase generated by the interlayer slip after deep desodiation (normally occurs when x ≤ 0.35) of the P2 phase. The TMO2 layers are stacked in the order “AB-AC,” while sodium ions occupy the octahedral positions between the layers. Unlike the P3 phase, the O2 phase cannot be directly synthesized by solid-state reactions. Instead, soft chemical methods like molten-salt ion exchange

3.2 Oxide Cathode

are a feasible route to obtain O2-type materials from P2 precursors. Other derived phases, such as O’3, P’2, OP4, and OP2, which are generated by interlayer slip or structural distortion of P2, O3, and P3 phases, will be introduced in subsequent sections. It is worth noting that roughly distinguishing or predicting P2 and O3 structures by sodium content is possible for most materials. However, with the emergence of low sodium content O3 type materials (x ≤ 0.7) [70] and high sodium content P2 phases (x > 0.8) [71], the limitation of this criterion is apparently exposed, indicating the requirement for deeper understanding of the original difference between P2 and O3 phases. By comparing the structures of a series of P2 and O3 layered oxides, Hu et al. [70] revealed that the two types of stacking are strongly relevant to the thickness of TMO2 slab (dO-M-O ) and NaO2 slab (dO-Na-O ), and a ratio of ∼1.62 for dO-Na-O /dO-M-O can be used as an indicator to predict P2 and O3 structure (Figure 3.9a,b). This finding agrees well with the fact that the stacking manner of layered oxides is mainly determined by the electrostatic interactions between TMO2 slabs together with the electrostatic shielding effects of Na ions. Further, they proposed a concept of “cationic potential” to evaluate the electron cloud density and polarization state of cations in the layered oxides, successfully describing the correlations between the stacking structures and the chemical compositions of layered oxides [69]. According to the definition (Eq. (3.1)), a cationic potential is a normalization of the composition-weighted ionic potential of cations including TM (Eq. (3.2), wi , ni , and Ri refer to the content, charge number, and radius of TM i , respectively) and Na (Eq. (3.3), x is the content of Na) to the ionic potential of anions (ΦO ), in which the ionic potential is the ratio of the charge number with the ion radios indicating the charge density at the surface of an ion. Consequently, P2 and O3 structures can be predicted by comparing cationic potential and mean Na ionic potential since the former implies the strength of TM electron cloud extension and interlayer electrostatic repulsion and the latter indicates the degree of the electrostatic shielding effects of Na ions. A phase diagram of P2 and O3 oxides can be obtained by plotting the cationic potential (ΦTM ) vs. the weighted average Na ionic potential (ΦNa ), providing a more accurate prediction of P2 and O3 structure (Figure 3.9c). It is worth noting that, cationic potential only predicts a preferred thermodynamic-stable phase for a given composition, while the structure of the product also depends on the synthesis process, especially the solid-state reaction dynamics. Besides, other structures such as entropy-dominated phases, metastable phases, and disordered phases are not predictable to date through cationic potential theory, implying the complexity of layered oxides and a large space for further investigation. Φcation =

ΦTM ΦNa

Φ ∑ wi nOi ΦTM = Ri x ΦNa = RNa

(3.1) (3.2) (3.3)

45

3 Cathode Materials of SIBs

P2 Phase O3 Phase

2.2 2.0 Ratio

46

1.8 1.6 1.4

(a)

(b)

1.2

2

4

6 8 Number

10

(c)

Figure 3.9 (a) Illustration the thickness of TMO2 slab (dO-M-O ) and NaO2 slab (dO-Na-O ). Source: Adapted from Zhao et al. [69]. (b) the ratio between the interlayer distances of dO-M-O and dO-Na-O for the typical P2- and O3- type compounds. Source: Reproduced from Zhao et al. [70]/with permission of John Wiley & Sons. (c) Cationic potential of representative P2and O3-type Na-ion layered oxides. Source: Adapted from Zhao et al. [69].

3.2.1.2 Key Issues of Layered Oxides Structural Evolution

Similar to lithium layered oxides, sodium layered oxides are intercalation-type cathode materials, in which Na ions can be reversibly extracted and inserted from/into sodium layers during charge and discharge processes. Due to the large radius of the Na ion (1.02 Å), sodium layered oxides often undergo drastic structural and volumetric changes during sodium intercalation and de-intercalation, significantly affecting the structural reversibility and ion-diffusion properties and thereby the cycling and rate performances. Therefore, revealing the structural evolution mechanism of sodium layered oxides is essential for understanding the electrochemical behaviors and further designing better materials. In P2 configuration, oxygen ions from the two adjacent TMO2 layers are coaxially distributed along the c axis, stabilized by Na ions located in the interlayer prismatic sites. Along with the extraction of Na ions, oxygen ions in the adjacent layers will lose the shielding effect of Na ions and become “face-to-face,” leading to an increase in interlayer electrostatic repulsion. This will first induce the expansion of interlayer distance without phase transition, which is also called the P2-type solid-solution

3.2 Oxide Cathode

(a)

(b)

Figure 3.10 (a) Illustration of the P2-“Z”-O2 the phase transition process. (b) In situ XRD contour map (002 peak) and corresponding charge–discharge curves for (b) Na2/3 [Ni1/3 Mn2/3 ]O2 and (c) Na2/3 [Ni1/6 Mn1/2 Fe1/3 ]O2 . Source: Reproduced from Somerville et al. [72]/with permission of Royal Society of Chemistry/CC BY 3.0.

region. Once it goes to a deep desodiation state, e.g. x ≤ 0.35 in P2-Nax TMO2 , the strong electrostatic repulsion will drive the slip of TMO2 layers along (2/3, 1/3, 0) or (1/3, 2/3, 0) directions to accommodate the close-packed arrangement of oxygen ions. At this time, Na vacancies are transformed into an octahedral type, indicative of the formation of O-type stacks. When all the P-type stacks are transformed into O-type stacks, a uniform O2 phase with a low concentration of sodium and a small interlayer distance can be obtained. The appearance of the OPx (or denoted as “Z” phase) phase can be deemed an incomplete P2–O2 phase transition, which produces a periodical distribution of P-type stacks and O-type stacks (Figure 3.10a) [72]. One of the typical materials experiencing a P2–O2 phase transition is P2-Na2/3 Ni1/3 Mn2/3 O2 , in which the phase transition occurs after the removal of 1/3 Na+ , accompanied by a long charging plateau around 4.2 V (Figure 3.10b). In contrast, other P2-type materials, for instance Na2/3 [Ni1/6 Mn1/2 Fe1/3 ]O2 , commonly experience a P2-OPx (or P2-“Z”) phase transition even after deep desodiation (Figure 3.10c). Even if the sodium removal goes to x < 0.2, the OPx phase remains without further transformation to the O2 phase, and the internal reason is still

47

48

3 Cathode Materials of SIBs

Figure 3.11 Summary of phase evolution vs. Na content upon first charge for typical P2-type materials with and without oxygen-redox contribution. Source: Reproduced from Bai et al. [75]/with permission of Elsevier.

unclear. Somerville et al. [72] suggested that the P2-O2 phase transition is likely related to the ordered structure of cations, but theoretical and experimental evidence are required. Numerous studies have shown that both P2-O2 and P2-OPx phase transitions are reversible at least within initial cycles, but the P2-OPx phase transition is relatively preferred due to the smaller volume change than the P2-O2 phase transition and the resultant better cycling stability [73, 74]. However, as for long-term cycling, the P2-OPx transition is also detrimental as increased defects and even particle cracks will be induced after repeated volume changes. Therefore, much effort has been put into pursuing a complete solid-solution reaction. In most cases, this can only be achieved by limiting the desodiation depth. Notably, some reports have claimed that the participation of oxygen redox may delay or suppress P–O transition, well maintaining P2 structure during deep desodiation (Figure 3.11), but more proofs and corresponding explanations are still needed [75, 76]. For Mn-based P2-type materials, the deep sodiation process is usually accompanied by a P2–P’2 phase transition [77]. It is generally believed that the strong Jahn-Teller effect of Mn3+ ions is the driving force for this phase transition. That is, the twisting of the MnO6 octahedron leads to the elongation of the unit cell along the b axis, leading to the transformation into a twisted orthorhombic structure (b/a > 1.732) (Figure 3.12a). This phase transition is also a reversible process. However, due to the small interlayer spacing of the P’2 phase with high sodium content, the kinetics in this region are relatively slow, resulting in an unfavorable rate and cycling performance [79]. It is worth noting that P’2 phase with similar sodium contents to P2 phase can be obtained through a high-temperature quench process when having a high concentration of Mn3+ , exhibiting even better electrochemical performances, which is likely related to the effects of TM vacancies [80, 81]. In addition, multi-step phase transitions with stair-like voltage

3.2 Oxide Cathode

profiles can be induced by Na+ /vacancy ordering or charge ordering. Intrinsically, Na+ /vacancy indicate the formation of superlattice at special sodium contents like 1/3, 1/2, 5/8, and 2/3, which can be found in some layered oxides irrelevant to the original stacking structure (P2, P3, and O3). For example, P2-Na2/3 Ni1/3 Mn2/3 O2 undergoes three Na+ /vacancy ordering processes at the sodium content of 1/3, 1/2, and 2/3, accompanied by a sudden voltage drop or increase (Figure 3.12b–d) [78]. It is proposed that such Na+ /vacancy ordering is detrimental to sodium diffusion and structural reversibility, which can be efficiently improved by cationic substitution. Compared with P2-type materials, the structural evolution of O3-type materials during (de)sodiation is even more complex, which is significantly affected by the composition of transition metals. The most general evolution during desodiation is the O3–P3 phase transition, followed by the P–O transition. Taking NaFe1/2 Co1/2 O2 as an example (Figure 3.13a) [82], the O3 structure is maintained before the removal of 0.13 Na ions. Along with further extraction of sodium ions, the coordination of sodium ions changes from octahedra to triangular prisms through the slip of TMO2 layers, which is more conducive to stabilizing the crystal structure, leading to a complete O3–P3 phase transition after the removal of 0.36 Na ions. When sodium extraction goes deeper, P3 phase first experiences a solid-solution process followed by an abrupt weakening and broadening in diffraction peaks, corresponding to the transformation into an O3-like phase with lots of stacking faults and a greatly shrunk interlayer distance, which is generally denoted as the O3’ or O’3 phase [87, 88]. Based on structural simulation, Yabuuchi et al. proposed that O3-NaFe1/2 Mn1/2 O2 transforms from a P3 phase to an OP2 phase with alternating O and P layers after deep desodiation [89]. Recently, Kim et al. [83] successfully observed the crystal structure of deep desodiated Na0.22 Ti0.25 Fe0.25 Co0.25 Ni0.25 O2 using high-resolution scanning transmission electron microscopy (STEM), confirming the existence of the OP2 phase (Figure 3.13b). However, the phase evolution of O3-type materials during deep desodiation is related to chemical composition. Specially, after extracting 0.75 Na+ from O3-NaNi0.5 Mn0.4 Ti0.1 O2 , a new (00l) diffraction peak appears around 20∘ , which demonstrates larger cell contraction than the P3-OPx transition and is quite similar to the P2-O2 transition. Wang et al. assigned this to a new O3 phase according to the similar symmetry (Figure 3.13c) [84]. However, such a small interlayer spacing more likely indicates the formation of an O1 phase with “ABAB” stacking configuration, which is commonly found in some cation-ordered O3 phase materials, such as honeycomb-type Na3 Ni2 SbO6 [90] and Na2 RuO3 (Figure 3.13d,e) [86]. Chen et al. [91] and Sato et al. [92] also found this phase transition process in the deep charging of manganese-based O3-type materials (O’3-NaMnO2 ). In addition, there are also some O3-type materials that do not directly experience the O3-P3 phase transition in the initial stage of sodium removal, instead leading to a twisted monoclinic phase (space group C2/m), such as the O′ 3 phase and the O′′ 3 phase, which are often accompanied by complex multiphase transformation processes. Such phase transitions are likely rooted in Na+ /vacancy ordering or charge ordering, the migration of transition metal ions, or the structural distortion caused by the strong Jahn-Teller

49

3 Cathode Materials of SIBs

bH aH cO

P2

bO Shrink

Stretch eg t2g

bO

Mn3+

aO P’2

(a)

P’2

δ in NaδNi1/3Mn2/3O2 0.6

0.5

0.4 P2-Na1/3Ni1/3Mn2/3O2

4.0

Voltage (V)

50

3.5

P2-Na1/2Ni1/3Mn2/3O2

3.0 P2-Na2/3Ni1/3Mn2/3O2

2.5 0

20

(b)

(c)

40

60

80

Capacity (mAh g–1)

δ= 2 3

δ= 1 2

δ= 1 3

Figure 3.12 (a) Illustration of P’2 phase and a comparison with P2 phase. (b) Typical charge/discharge profiles of P2-Naδ Ni1/3 Mn2/3 O2 and (c) schematic of in-plane Na+ /vacancy orderings in the triangular lattice at 𝛿 = 2/3, 𝛿 = 1/2, and 𝛿 = 1/3, respectively (empty red circles, Na+ on Nae sites; solid blue circles, Na+ on Naf sites; thick green lines, unit cell). Source: Reproduced from Wang et al. [78]/American Association for the Advancement of Science/CC BY 4.0.

3.2 Oxide Cathode

(a)

(b)

(c)

(d)

(e)

Figure 3.13 (a) In situ XRD for NaFe0.5 Co0.5 O2 . (b) HR-STEM image of Nax Ti0.25 Fe0.25 Co0.25 Ni0.25 O2 charged to x = 0.22 (4.2 V). (c) In situ XRD for NaNi0.5 Mn0. 4Ti0.1 O2 . (d) Structure of O1-type Na1 RuO3 and (e) In situ XRD for Na2 RuO3 . Source: (a) Reproduced from Kubota et al. [82]/with permission of John Wiley & Sons. (b) Reproduced with permission from Kim et al. [83], © 2020/JOHN WILEY & SONS, INC. (c) Reproduced from Wang et al. [84]/with permission of John Wiley & Sons. (d) Reproduced from Mortemard de Boisse et al. [85]/with permission of Springer Nature/CC BY 4.0 and (e) Reproduced from Mortemard de Boisse et al. [86]/with permission of Springer Nature/CC BY 4.0.

effects, usually accompanied by the generation of defects and poor reversibility. Notably, these complex phase transitions are commonly found in the many unary O3-type materials such as NaFeO2 [93], NaNiO2 [94], and NaCoO2 [95], and some binary materials like NaNi1/2 Mn1/2 O2 [96], resulting in fast capacity decay. Doping or substituting active or inactive metal elements to construct multi-component systems is an effective route to inhibit irreversible phase transitions and alleviate local structural distortions [84, 97, 98]. It is hard and unprofitable to avoid the O3-P3 phase transition as it appears at the initial stage of the desodiation and the P3 phase generally supports faster Na-ion diffusion than the O3 phase. However, it is highly suggested to suppress further P3-OPx or P3-O1 transitions considering the detrimental volume change.

51

52

3 Cathode Materials of SIBs

Charge Compensation

Similar to lithium layered oxides, both cationic redox and anionic redox can be accessed in sodium layered oxides to compensate the charge change during (de)sodiation, which is essentially rooted in the unique band structures (Figure 3.14a) [99]. Generally, the redox of layered oxides is mainly dependent on the overlap of metal (M) nd orbitals and O 2p orbitals. As known, there are five M nd orbitals consisting of two eg orbitals and three t2g orbitals, among which the two eg orbitals are prone to strongly overlap with O 2p (σ) orbitals, while the three t2g orbitals prefer to weakly interact with O 2p (π) orbitals according to adapted symmetries and energy. Consequently, the former gives birth to new eg and eg * orbitals with high splitting energy, and the latter results in t2g and t2g * orbitals with less splitting. At the same time, the electrons from the original M nd orbitals and O 2p orbitals will fill in these new orbitals in line with the energy levels, forming antibonding (M–O)* bands with high-lying energy close to Fermi level and bonding (M–O) bands with low energy. Specifically, antibonding (M–O)* bands are mainly filled by M nd electrons, thus showing metal character, while bonding (M–O) bands are principally occupied by O 2p electrons, displaying O character. Once Li+ or Na+ ions are extracted from the layered oxides, equal electrons will be preferentially removed from the high-lying (M–O)* bands near Fermi level, accompanied by the oxidation of M, which is the well-recognized cationic redox process. However, the band structure can be changed in some cases, leading to different electron-removing processes. For example, in Li-rich systems like Li2 MnO3 , Li ions occupy both in Li layers and TM layers, giving rise to Li–O–Li configurations (in contrast to conventional Li–O–M configurations), in which the particular O 2p orbital pointing toward Li–O–Li direction cannot overlap with Li 2s orbitals due to the unmatched energy level (Figure 3.14b,c). As a consequence, non-bonding (NB) O 2p states are obtained, which have an energy lying between (M–O) and (M–O)* states and thereby can offer additionally removable electrons once the (M–O)* states are depleted, termed as anionic redox process [100]. In fact, the reported sodium layered-oxide cathodes mainly rely on cationic redox for charge compensation. One of the merits of sodium layered oxides is that most of the 3d transition metal redox couples, including Fe3+ /Fe4+ and Cu2+ /Cu3+ couples, are electrochemically active due to the suitable redox potential and the low covalency between sodium and oxygen [89], providing wide choices for material design. However, when solely relying on cationic redox, the materials energy density will be highly dependent on the type and amount of active TM ions, as they determine the working potential and accessible capacity, respectively. Moreover, the valence evolution of TM ions commonly induces volume change as well as local structural distortions when involving Jahn-Teller active ions, which are all detrimental to the electrochemical performances. Reversible anionic redox is an important alternative for the charge compensation of layered-oxide cathodes, which greatly extends the accessible capacity. To date, various strategies, such as synthesizing alkalis-rich materials (Figure 3.15a) [85, 103], creating TM vacancies (Figure 3.15b) [101], and substituting with d0 or d10 metals (Figure 3.15c) [102, 104], have been explored to trigger reversible oxygen redox. Some of the reported materials, like P2-type

3.2 Oxide Cathode

(a)

(b)

(c)

Figure 3.14 (a) The schematic band structure of transition-metal oxides and a comparison of (d) LiMO2 and (e) Li-rich Li2 MO3 . Source: Reproduced from Sathiya et al. [99]/with permission of Springer Nature.

Na0.72 [Li0.24 Mn0.76 ]O2 , exhibit ultra-high discharge capacities up to 271 mAh g−1 [105]. Another potential advantage of anionic redox is the related smaller structure change than that relates to cationic redox, while the intrinsic reason is still yet to be clarified. Nevertheless, the issues of anionic redox are also obvious. One is the poor reversibility of pure anionic redox, leading to fast capacity decay. Indeed, it is difficult to completely avoid oxygen release or the corresponding surface or bulk deterioration. Moreover, anionic redox usually exhibits large voltage hysteresis, sluggish kinetics, and fast voltage decay, resulting in low energy efficiency and poor rate capability. Lots of work has been focused on mitigating these issues by means of compositional modulation, surficial modification, and electrolyte optimization, but more efforts are still required. TM Migration and Dissolution

During the desodiation of sodium layered oxides, special ions like Fe3+ and Cr3+ are prone to migrate from TMO6 octahedra to the face-sharing tetrahedral spaces in the adjacent NaO2 layer to stabilize the structure (Figure 3.16a) [108, 109]. In unary O3-NaFeO2 and O3-NaCrO2 , such migration leads to irreversible phase transformation and constrained sodium insertion/diffusion, accompanied by large polarization and fast capacity loss (Figure 3.16b). Li et al. [110] studied the Fe migration in various Fe-based materials supported by DFT calculations. The gathering of Fe ions was demonstrated as the main detriment since the Jahn-Teller

53

54

3 Cathode Materials of SIBs

(a)

(b)

(c)

Figure 3.15 Structural illustration and voltage profiles for (a) Ordered Na1 [Na1/3 Ru2/3 ]O2 (or denoted as Na2 RuO3 ), (b) Na4/7 [◽1/7 Mn6/7 ]O2 where refers to Mn vacancy (or denoted as Na2 Mn3 O7 ), (c) Na2/3 [Mg0.28 Mn0.72 ]O2 . Source: (a) Reproduced from Mortemard de Boisse et al. [85]/with permission of Springer Nature/CC BY 4.0. (b) Reproduced from Mortemard de Boisse et al. [101]/with permission of John Wiley & Sons. (c) Reproduced from Yabuuchi et al. [102]/with permission of Royal Society of Chemistry.

distortion of the Fe4+ O6 octahedral might facilitate the migration of neighboring Fe3+ . This conclusion gave birth to the suggestion that Fe migration would be largely avoided when lowering down the proportion of iron to 33% (in the TM layer), as Fe ions would be significantly isolated in such hexagonal configurations. Therefore, the modifications of Fe- and Cr-based sodium layered oxides were generally performed by substituting Fe or Cr with other cations so as to achieve better electrochemical performances. Noteworthy, some reports showed that Fe migration may also occur in those systems with Fe proportions less than 33%,

3.2 Oxide Cathode

A B C A Fe3+ is stable at octahedral sites. (a)

A face-shared vacant tetrahedral site is formed by Na extraction.

Fe3+ migrates into the vacant tetrahedral site.

Voltage (V)

x in Na1–xFeO2

(b)

Capacity (mAh g–1)

Figure 3.16 (a) The schematic illustration of Fe migration during Na extraction in Nax FeO2 . Source: Reprinted with permission from Yabuuchi and Komaba [106] under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 license. (https:// creativecommons.org/licenses/by-nc-sa/3.0/); © 2014, the National Institute for Materials Science, (b) Initial charge/discharge curves of Na/NaFeO2 cells with different cut-off voltage at a rate of 12 mA g−1 . Source: Yabuuchi et al. [107]/The Electrochemical Society of Japan/Public domain.

e.g. P2-Na0.7 [Cu0.15 Fe0.3 Mn0.55 ]O2 [111] and P2-Na0.67 [Mn0.66 Fe0.2 Ni0.15 ]O2 [112] as supported by X-ray absorption spectroscopy (XAS), pair distribution function (PDF), or Mössbauer spectroscopies, implying a requirement to further understand the migration in sodium layered oxides. The dissolution of TM ions into organic electrolytes is another issue for sodium layered oxides. Proton-induced disproportionation of Jahn-Teller active TM ions like Mn3+ and Ni3+ is generally believed to be the origin of corresponding TM dissolution as the produced low-valence TM ions (Mn2+ and Ni2+ ) are highly soluble [113]. Notably, in organic electrolytes, protons mainly come from the residual-water-induced hydrolysis of PF6 − or the oxidation of electrolytes activated by highly oxidized TM ions. Moreover, it is also reported that such dissolution may also stem from the phase transition to soluble metal oxides at deep charged states. Such TM dissolution will lead to not only the loss of redox centers and unstable solid-electrolyte interfaces but also the deposition of transition metals on the anode, resulting in the deterioration of both cathode and anode and thereby severe capacity fading [114]. The promising routes to suppress TM dissolution mainly include

55

56

3 Cathode Materials of SIBs

improving the stability of the electrolyte and modifying the surficial structure or chemical composition of cathode materials. Air Stability

Sodium layered oxides normally suffer from poor stability in air, especially in a moist environment, arising the issues for material storage and transport. Most of the sodium layered oxides exhibit obvious structural deterioration accompanied by the loss of Na, which was attributed to spontaneous Na+ /H+ exchange followed by surficial reactions with CO2 [115], and the insertion of CO3 2− species [116]. Recently, Yang and coworkers [117] revealed that the air-stabilities of P2-type layered oxides are correlated to the redox potential in the first cycles, which corresponds to the requested energy for chemical Na+ extraction (Figure 3.17). For those layered materials relying on low-voltage redox couples such as V3+/4+ , Fe3+/4+ , and Mn3+/4+ , the loss of Na+ from the bulk to the surface spontaneously occurs during air exposure. Meanwhile, Na+ /H+ exchange simultaneously happens for charge compensations, forming a NaCO3 /NaHCO3 layer at the surface of particles and leading to structural changes in the bulk. By contrast, materials containing high-voltage redox couples, such as Ni2+/4+ , Cu2+/3+ , and O2−/n− , generally show much better air-stability, thanks to the higher energy barriers for chemical Na+ extraction. Although it is found that the structural deterioration in moist air can be recovered by thermal treatments, the as-resulted increased cost of material storage is surely unfavorable. In this case, optimizing the TM component or coating with stable surface layers is suggested as an effective strategy to enhance the air-stability of sodium layered materials. 3.2.1.3 P2-type Layered Oxides

P2-type materials are sodium-deficient materials, so only transition metal elements with multiple stable valences (such as Co, Mn, and V) can form unary P2-type materials. P2-Na0.7 CoO1.96 [118] is one of the earliest reported cathode materials

Figure 3.17 The comparison of the main redox couples in layered sodium transition metal oxides. Source: Reproduced from Zuo et al. [117]/with permission of Springer Nature/CC BY 4.0.

3.2 Oxide Cathode

for Na-ion batteries, which allows for reversible (de)intercalation of 0.37 mol Na ions in the voltage range of 3.5–2.0 V accompanied by reversible Co3+ /Co4+ redox. The charge–discharge voltage profile of the material has obvious “stair-like” characteristics, corresponding to a multi-step phase transition related to Na+ /vacancy ordering [119]. P2-Nax MnO2 has a reversible specific capacity close to 200 mAh g−1 within 1.5–4.4 V; however, it suffers from fast capacity decay during cycling due to complex phase transition, Jahn-Teller effects, and Mn-dissolution issues [80]. P2-Na0.71 VO2 is also electrochemically active based on V3+ /V4+ redox, presenting a reversible capacity of ∼110 mAh g−1 in the range of 1.4−2.5 V. However, the low working voltage limits its application as cathode materials [120]. Binary P2-phase materials usually consist of two (transition) metal elements in different valence states. As mentioned above, Na2/3 Ni1/3 Mn2/3 O2 is a classic binary P2-type material, in which each Ni2+ is surrounded by six Mn4+ in the TM layer, forming a honeycomb ordered structure. The formation of this superlattice structure usually requires two conditions: one is a large charge difference between the two cations in the TM layer (|V1–V1| ≥ 2), and the other is a certain mole ratio for the two ions, such as 1 : 2, 2 : 1, and 1 : 3 [121]. Based on this principle, a large family of TM-ordering materials such as Na3/4 Li1/4 Mn3/4 O2 [103], and Na2/3 Ni2/3 Te1/3 O2 [122], has been prepared. Such TM-ordering structures can generally be identified by the superlattice diffraction peaks from XRD patterns. However, in some cases, like P2-Na2/3 Ni1/3 Mn2/3 O2 , the superlattice diffraction peaks are not detectable from laboratory XRD data, while synchrotron-based XRD or neutron diffraction are further required [72]. P2-Na2/3 Ni1/3 Mn2/3 O2 can provide an initial charge/discharge capacity of 160/215 mAh g−1 in the voltage range of 1.5–4.5 V contributed by both Ni2+ /Ni3+ /Ni4+ redox and Mn3+ /Mn4+ redox, however, followed by fast capacity fading due to the P2–O2 phase transition and the induced drastic volume change (∼23%). By limiting the charge–discharge window, the cycling stability of the material can be significantly improved at the expense of capacity. Besides, cationic doping or substitution is widely preformed in order to improve the cycle stability of P2-Na2/3 Ni1/3 Mn2/3 O2 . Plenty of studies have proved that partially replacing Ni or Mn with Li [123], Cu [124], Mg [73], Fe [125], Ti [78], and other cations can significantly inhibit the P2–O2 phase transition and result in a P2–OP4 phase transition with a smaller volume change, thereby improving the cycling stability. As mentioned in an earlier part, the charge–discharge curves of P2-Na2/3 Ni1/3 Mn2/3 O2 also manifest multi-plateau features, implying Na+ /vacancy ordering during the (de)intercalation process. This super-superstructure will hinder the migration of sodium ions, affecting the rate capability. Replacing part of Mn4+ with Ti4+ can significantly suppress the Na+ /vacancy ordering, smoothing the charge–discharge curves and greatly enhancing the rate performance [78]. Na2/3 [Fe1/2 Mn1/2 ]O2 is another eye-catching P2-type layered oxide reported by Yabuuchi et al. in 2012, which exhibits a reversible specific capacity as high as 190 mAh g−1 . An average discharge voltage of 2.74 V is higher than that of P2-Nax MnO2 (about 2.5 V) due to the contribution of Fe3+ /Fe4+ redox, giving an energy density of 520 Wh kg−1 , which is close to that of the commercial LiFePO4 (∼530 Wh kg−1 ). In addition, since Fe and Mn are both abundant and inexpensive

57

58

3 Cathode Materials of SIBs

elements, Fe/Mn-based layered oxides have been considered as one of the most promising cathode materials for SIBs. However, their practical applications still face big challenges from fast capacity decay, poor kinetics, large voltage hysteresis, and moisture sensibility due to Fe migration and unfavorable phase transitions. Cationic substitution is widely utilized to improve the cycling stability of Fe/Mn based layered oxides, and the key is to decrease the concentration of Fe, for instance, to less than 1/3. Recently, X. Ji et al. [126] revealed that, Fe migration is merely triggered after the oxidation of ∼0.3 mol Fe3+ in O3-Nax Fe1/2 Mn1/2 O2 , while Fe4+ -activated and Mn-dissolution boosted surficial passivation is the major origin of the fast decay in the Fe-migration-free region, updating the cognitions of Fe/Mn based materials and offering clues for further material modification. In 2014, Hu et al. [127] reported a new type of P2-type binary material, Na2/3 Cu1/3 Mn2/3 O2 and demonstrated for the first time that the Cu2+ /Cu3+ redox couple is electrochemically active and reversible in sodium layered oxides, arousing wide research interests in Cu-based oxide cathode materials. Recently, Ji et al. [79] studied the role of Cu in Cu/Mn based P2-type layered oxides, revealing that the expansion/contraction of the MnO6 octahedron are dramatically reduced along with increasing Cu content, which results in the suppressed P2-P′ 2 phase transition at the end of discharge, accompanied by significantly boosted cycling stability. It was also reported that the incorporation of Cu into P2-type Nax MnO2 induces reversible redox at high voltage, region while the mechanism requires further clarification. When it comes to anionic redox behaviors, it is necessary to introduce P2-type Li/Mn based layered oxides, which are a family of anionic-redox-based cathodes featuring high capacities. As a notable example, P2-type Na0.72 [Li0.24 Mn0.76 ]O2 exhibits a reversible capacity as high as 271 mAh g−1 , which is quite close to its theoretical capacity (291 mAh g−1 ) based on the (de)intercalation of 1 Na+ [105]. Both anionic redox and cationic redox contribute to charge compensation, as Mn3+ /Mn4+ redox merely support a maximum (de)intercalation of 0.76 Na+ . Impressively, the host structure undergoes a near solid-solution reaction while the P2–O2 phase transition is significantly inhibited. However, such a high initial capacity fails to be stabilized in the following cycles due to the poor reversibility of oxygen redox. Similar anionic-redox-based P2-type cathodes, including P2-Na5/6 Li1/4 Mn3/4 O2 (∼200 mAh g−1 ) [103], P2-Na2/3 Mg0.28 Mn0.72 O2 (∼160 mAh g−1 ) [102], and P2-Na2/3 Zn1/4 Mn3/4 O2 (∼200 mAh g−1 ) [128], are also reported, while the common feature is the existence of O 2p NB states as mentioned in the earlier part. The design of multicomponent materials is the main strategy to mitigate the issues of unary and binary layered oxides, as it may couple the advantages of each transition metal element and weaken its adverse effects, thus achieving better electrochemical performances. For instance, by importing Ni and Fe into P2-type Nax MnO2 , Sun et al. [129] obtained a P2-type Na0.55 [Ni0.1 Fe0.1 Mn0.8 ]O2 cathode, which delivers a high discharge capacity of ∼221 mAh g−1 and a high average potential of ≈2.9 V (vs. Na/Na+ ) for SIBs, accompanied by good rate performance at a high current density of 2.4 A g−1 and long-term cycling stability with ≈80% capacity retention after 500 cycles at 600 mA g−1 . Cerder et al. [130]

3.2 Oxide Cathode

reported a P2-Type Na2/3 (Mn1/2 Fe1/4 Co1/4 )O2 , which demonstrated outstanding rate capabilities up to 30 C thanks to the incorporation of Co into P2-Nax Fe1/2 Mn1/2 O2 . Moreover, by importing Fe into P2-Na2/3 Cu1/3 Mn2/3 O2 materials, Hu et al. [131] successfully invented a Ni/Co free P2-type Na7/9 Cu2/9 Fe1/9 Mn2/3 O2 , which delivers a reversible capacity of 89 mAh g−1 with a high discharge average voltage of 3.6 V contributed by Fe3+ /Fe4+ and Cu2+ /Cu3+ redox. During the removal of ∼0.36 Na+ , the host structure experiences a highly reversible solid-solution evolution, in consistent with the smooth voltage profile and good cycling stability. However, the solid-solubility of Cu in the Cu–Fe–Mn system is constrained to ∼22%, which can be even decreased by increasing Mn4+ due to the large difference in ionic diameters, likely limiting the material design [132]. In application wise, X. Ji et al. [133] proposed a short-process route to directly obtain Cu–Fe–Mn based layered oxides from natural chalcopyrite. The as-prepared P2-type produced with “self-doped” impurity elements demonstrates comparable electrochemical performances and even better structural stability against air and water. It is worth highlighting that this work presents a prototype to shorten the path from natural ores to high-performance and low-cost cathode materials. In addition to the compositional optimization, the research hotspots on P2-type materials also include the design of high-sodium content P2-type materials, sodium-site engineering, and the search for suitable additional sodium suppliers in full cells. The capacity fading of P2-type materials is largely due to the violent phase transition in the high voltage range. One of the potential strategies to improve the reversible capacity and cycle stability of P2-type materials is to increase the sodium content, which can broaden the concentration range for sodium removal. Zhao et al. [71] reported a P2-type Na0.85 Li0.08 Ni0.30 Mn0.62 O2 in which the sodium content is increased to 0.85 by adjusting TM composition. When charged to 4.6 V, about 0.58 mol of Na+ can be removed, corresponding to an initial charge capacity of over 150 mAh g−1 , which benefits full-cell applications. Although only about 0.25 mol of Na ions remain in the structure after being fully charged, the material does not undergo the transformation of P2 to O2 or OP4 phase, indicating that high sodium content helps to broaden the solid-solution reaction phase region and delay the P–O phase transition. It is also suggested that high sodium content in pristine material also modifies the electronic structure and enabling the redox of Ni2+ /Ni4+ couples within a lower voltage window, favoring the material stability against electrolytes. At the same time, Jin et al. [134] reported a high-sodium-content P2 material with similar composition (Na0.85 Li0.12 Ni0.22 Mn0.66 O2 ), confirming a stable capacity of ∼120 mAh g−1 with smooth voltage profiles and complete solid-solution reactions (Figure 3.24b,c). It is worth noting that the highest sodium content of P2 phase materials reported so far is 0.85. According to the prediction of cation potential [69], this may be the maximum sodium content in the P2 structure as it requires the incorporation of a cation with higher ion potentials than Mn4+ ions (only Si4+ is matched), which is experimentally infeasible. Na-site doping is another effective way to improve the structural stability of layered oxides. Wang et al. [136] incorporated K+ into the sodium ion sites of P2-Na0.7 MnO2 , which inhibited the P–O phase transition in the high-voltage region through the “pillar” effect. A

59

60

3 Cathode Materials of SIBs

reversible specific capacity as high as 238 mAh g−1 was recorded, accompanied by a capacity retention of 98% after 100 cycles. Similar studies have shown that both Mg2+ [137, 138] and Fe3+ [139] can be incorporated into the sodium layer to act as “pillars,” suppressing the P–O phase transition and improving the cycling stability of P2-type materials.

3.2.1.4 O3-type Layered Oxides

O3-type layered oxides are generally more conducive to full battery applications considering their higher initial sodium content. However, higher sodium content also comes with some drawbacks. On the one hand, the high occupation of sodium ions induces smaller interlayer spacings, curved sodium ion diffusion pathways, and complex phase transitions in O3-type structures, thereby resulting in poor rate and cycling performances. On the other hand, higher sodium ion content generally brings with it a low average valence of TM ions and sodium-rich surfaces, leading to serious instability in air, especially moist air. Therefore, the research focuses of O3 phase materials mainly include improving sodium ion transport kinetics, inhibiting unfavorable phase transitions, and enhancing air stability. Unary O3-type materials are commonly composed of trivalent TM ions such as Ti3+ , V3+ , Cr3+ , Mn3+ , Fe3+ , Co3+ , Ni3+ , and Cu3+ . Among them, NaTiO2 [140] and NaVO2 [141] have low working voltages and therefore limited application values as cathode materials. NaCuO2 [142] can be obtained with an O3-like structure (Cu and the nearest neighbor and next neighbor O are regarded as a violently distorted CuO6 octahedron). As Cu is already in the highest +3 state, a few studies have paid attention to its application potential as anode materials. NaCoO2 is one of the earliest reported O3-type oxide cathode materials. Like P2-Nax CoO2 , the charge–discharge curve of O3-NaCoO2 exhibits a distinct step-like shape, corresponding to Na+ /vacancy ordering and the O3-O’3-P’3 phase transition [95]. It gives a specific capacity of about 140 mAh g−1 in the voltage range of 4.0–2.0 V, followed by fast capacity fading (89% retention after 30 cycles). Due to the strong Jahn-Teller effects of Mn3+ and Ni3+ , NaMnO2 [92] and NaNiO2 [94] generally form distorted monoclinic structures termed as O′ 3 phase (space group C2/m). During (de)sodiation, O′ 3 undergoes more complicated structural evolution than that of the O3 phase, leading to poor cycling stability. NaFeO2 has attracted extensive attention due to the discovery of reversible Fe3+ /Fe4+ redox [107]. However, as introduced above, this material suffers from irreversible Fe migration, which leads to a limited reversible capacity of ∼80 mAh g−1 . A similar TM migration issue was also found in NaCrO2 [143]. Nonetheless, NaCrO2 exhibits a higher available capacity of around 120 mAh g−1 with good structural reversibility (O3-O’3-P’3 phase transition). Another shortcoming of NaCrO2 is its instability in the air. Yu et al. [144] improved the stability and electronic conductivity of NaCrO2 by carbon coating, significantly improving the cycling and rate performances. However, the environmental hazards of the Cr element limit the development and application of Cr-based materials. The voltage profiles of some unary O3-type materials are shown in Figure 3.25a.

3.2 Oxide Cathode

Profiting from the solid-solution nature, various cations can be incorporated into the layered structure, giving birth to binary, ternary, and even multi-component layered oxides with combined features and improved electrochemical performances [145]. Typical binary materials such as NaFe0.5 Co0.5 O2 [82], NaFe0.3 Ni0.7 O2 [146], NaNi0.5 Mn0.5 O2 [96], and NaNi0.5 Ti0.5 O2 [147], present much improvement in reversible capacity and/or cycle stability in comparison with unary materials, but some problems remain. For example, the reversible specific capacity of O3-NaNi0.5 Mn0.5 O2 reaches 185 mAh g−1 (4.5–2.0 V) [96], much higher than that of O3-NaNiO2 (147 mAh g−1 , 4.5–2.0 V) [94] and O′ 3-NaMnO2 (about 175 mAh g−1 , 4.5–1.2 V) [91]. However, issues like Na+ /vacancy ordering and complex phase transitions still exist, as reflected by the characteristic stair-like voltage profiles, leading to fast capacity fading. Numerous studies have shown that further replacing part of Ni2+ or (and) Mn4+ in NaNi0.5 Mn0.5 O2 with Co3+ [148], Fe3+ [149] Ti4+ [97] Sn4+ [98], etc. can significantly suppress the Na+ /vacancy ordering and mitigate phase transition, thus improving the electrochemical performance. The ternary material NaNi1/3 Fe1/3 Mn1/3 O2 obtained by introducing Fe into NaNi0.5 Mn0.5 O2 is one of the classic O3-type materials with industrial-application prospects. This material provides a specific capacity of 132 mAh g−1 in the voltage range of 2.0–4.0 V, accompanied by a simple O3-P3 phase transition and good cycling stability [88, 150]. Hu et al. [151] further adjusted the element ratio to obtain the O3-type “nickel-rich” NaNi0.60 Fe0.25 Mn0.15 O2 , increasing the stable specific capacity to ∼150 mAh g−1 (2.0–4.0 V). By incorporating Li into TM layers, You et al. obtained a quaternary Na0.85 Li0.1 Ni0.18 Mn0.54 Fe0.18 O2 , which shows a high capacity of 159 mAh g−1 within 2.0–4.5 V, followed by good cycling performances, demonstrating outstanding high-voltage stability [152]. Deng et al. [154] reported a pentanary O3-type NaLi0.05 Mn0.50 Ni0.30 Cu0.10 Mg0.05 O2 , recording a reversible specific capacity of 172 mAh g−1 and outstanding cycling stability. Notably, such multi-component materials are to some extent benefiting from the good compatibility of O-type structures for TM-ion species [155]. Different TM ions can easily cause local distortion of the crystal structure due to the difference in radius or electronegativity. However, the distortion of the octahedral position for Na+ will not change the overall O-type configuration, resulting in the capability to accommodate more TM species. This unique property is further confirmed by the successful synthesis of a high-entropy O3-type NaNi0.12 Cu0.12 Mg0.12 Fe0.15 Co0.15 Mn0.1 Ti0.1 Sn0.1 Sb0.04 O2 , which consists of nine TM species (Figure 3.18a–c) [153]. The material manifests a reversible specific capacity of about 110 mAh g−1 in the voltage range of 3.9–2.0 V with outstanding rate capability and cycling stability, which is attributed to the so-called high-entropy-stabilization effects. That is, the high-entropy composition will cause the random assignment of the redox elements, which is beneficial for accommodating the local changes of the interaction between TMO2 slabs and NaO2 slabs, thereby delaying the possible phase transition (Figure 3.18d). A recent report further claimed that the high-entropy configuration can significantly strengthen the host structure by mitigating Jahn-Teller distortion, Na+ /vacancy ordering, and lattice parameter changes, bringing with it improved Na+ transport kinetics and

61

3 Cathode Materials of SIBs

thermal stability [156]. Application wise, Cu–Fe–Mn based O3-type layered oxides stand out for their relatively lower material cost and good structural stability. As a porotype material, O3-Na0.9 Cu0.22 Fe0.3 Mn0.48 O2 presents good stability in air and water due to the incorporation of a Cu2+ /Cu3+ couple with high redox potential [87]. A reversible specific capacity of ∼100 mAh g−1 is obtained at 4.05–2.5 V, accompanied by good cycle stability. Apart from the optimization of average compositions, improvements to the surficial and interfacial structure or particle morphology are also important approaches to better electrochemical performances for O3-type materials. Sun et al. [157] successfully prepared O3-type Ni–Co–Mn-based materials with radially aligned hierarchical columnar structure (RAHC) via a well-controlled co-precipitation method. The inner core of the secondary particles consists of a nickel-rich component (mean composition Na[Ni0.75 Co0.02 Mn0.23 ]O2 ), which ensures a high capacity contribution, while the outer surface layer is composed of less nickel and more

Intensity (Counts)

Observed Calculated Difference Bragg peaks

+

60 2θ (°)

80

Na

Mg

Ni

Cu

Mn

Fe

Co

Ti

Sn

Sb

O

100

4.0 3.5 3.0 1st cycle 2nd cycle

2.5

∼89%

2.0 0

(b)

40

20

(a) Voltage (V vs. Na /Na)

62

20

40

60

80

100

0.5 μm

120

Capacity (mAh g–1)

(c) HEO-O3 Nax[TM1TM2...TMn]O2

Typical-O3 Nax[TM1TM2TM3]O2 Redox

Redox Pristine phase

Pristine phase TMO2 slab NaO2 slab

Redox

Redox Desodiated phase

Desodiated phase TMO2 slab NaO2 slab

TM (types)

TMn+ (oxidation states)

Interaction between TMO2 and NaO2 slabs

(d)

Figure 3.18 (a) XRD and SEM, (b) initial charge/discharge profiles, and (c) EDS elemental maps for NaNi0.12 Cu0.12 Mg0.12 Fe0.15 Co0.15 Mn0.1 Ti0.1 Sn0.1 Sb0.04 O2 , (d) Possible mechanism of high-entropy composition in facilitating layered O3-type structure. Source: Reproduced with permission from Zhao et al. [153], © 2020/JOHN WILEY & SONS, INC.

3.2 Oxide Cathode

manganese (mean composition Na[Ni0.58 Co0.06 Mn0.36 ]O2 ), which reduces surface side reactions and improves the stability of the electrode/electrolyte interface. Most importantly, it presents a gradient distribution of chemical compositions from the surface to the inside, instead of the regular separate configurations of coating or surficial doping, remarkably alleviating the difference in stress/strain in the local area during (de)sodiation process and thereby suppressing particle cracking. Consequently, this material exhibits a reversible capacity of 157 mAh g−1 (based on cathode materials) in combination with a hard carbon anode, together with remarkably enhanced cycling stability and rate performances compared with conventional bulk materials. Guo and coworkers [158] successfully synthesized nanoflake-like O3-NaLi0.05 Mn0.5 Ni0.3 Cu0.1 Mg0.05 O2 through a thermal polymerization method followed by a solid-state reaction. The exposed dominant crystal plane and smaller particle size shorten the Na-ion diffusion pathway, thus improving the rate capability. TM ordering is also commonly found in O3-type structures. Yang and coworkers reported a honeycomb-type layered oxide with a formula of Na3 Ni2 SbO6 , which can be assigned to a monoclinic lattice under the space group of C2/m. Based on the classification rules mentioned earlier, this material can also be labeled as an O′ 3 phase with the formula NaNi2/3 Sb1/3 O2 . In the TMO2 sheets, each SbO6 octahedron is surrounded by six NiO6 edge-sharing octahedrons, forming the superstructure lattice. It gives a reversible capacity of 117 mAh g−1 contributed by Ni2+ /Ni3+ redox, accompanied by good cycling stability and excellent rate capability due to the super sodium ion conductivity [90]. During the (de)sodiation, it undergoes a reversible O3-P3-O1 phase transition, different from those O3-type materials without TM ordering. Sb can be replaced by other pentavalent cations like Bi5+ , without changing the structure and performance [159]. However, among bivalent TM ions, only Ni2+ presents good electrochemical activity and reversibility in this structure. NaLi1/3 Mn2/3 O2 is another typical honeycomb-type material in which Li+ and Mn4+ ions are located at TM sites while each LiO6 octahedron is surrounded by six MnO6 edge-sharing octahedrons. Although NaLi1/3 Mn2/3 O2 was predicted to be a thermodynamically stable phase, the practical synthesis by the conventional solid-state method failed and gave birth to P2-type products [103, 160]. Recently, Tarascon et al. [161] successfully obtained pure O3-type NaLi1/3 Mn2/3 O2 (here denoted as the O′ 3 phase in accordance with the above classification) by carefully controlling the reaction atmosphere. This material exhibits a reversible capacity of 190 mAh g−1 , contributed by both anionic redox and Mn3+ /Mn4+ redox, surprisingly accompanied by good cycling stability and negligible voltage decay since TM migration occurs in a favorable “intralayer” manner instead of the detrimental “interlayer” way. Sodium rich layered oxides Na2 TMO3 , isostructural to Li2 MnO3 , were also investigated inspired by the unique electrochemical behaviors of Li2 MnO3 . Due to the larger ionic diameter of Na+ than Li+ , correspondingly larger tetravalent cations like 4d or 5d TM ions are required to form the honeycomb ordering in TMO2 layers. As a typical example, O′ 3-type (space group C2/m) Na2 RuO3 with honeycomb ordering of NaO6 octahedra and RuO6 octahedra can be synthesized by thermal decomposition from Na2 RuO4 . It delivers a reversible

63

64

3 Cathode Materials of SIBs

capacity of 180 mAh g−1 based on both oxygen redox and Ru4+ /Ru5+ redox, followed by good cycling stability (Figure 3.15a) [85]. Noteworthy, O3-type (space group R-3m) Na2 RuO3 with disordered distribution of Na and Ru in TMO2 slabs can also be prepared by conventional solid-state reactions. However, it merely gives a capacity of 135 mAh g−1 accompanied by a solely cationic redox of the Ru4+ /Ru5+ couple, implying the crucial role of TM ordering in triggering oxygen redox. 3.2.1.5 P3-type Layered Oxides

Despite the lack of clear guidance for materials design and synthesis, P3-type materials can be empirically obtained by appropriately reducing the sodium content and synthesis temperature of an O3 phase material with the targeted TM composition [95]. The structural evolution of P3-type materials during (de)sodiation can also be regarded as a part of the structural evolution of the corresponding O3 phase materials, demonstrating quite similar electrochemical behaviors when working in the same (de)intercalation depth [162]. Specially, the typical P2-type composition of Na2/3 Ni1/3 Mn2/3 O2 can also be obtained in a P3-type structure by adjusting the precursor and reaction temperature, however, presenting quite similar electrochemical behaviors to its P2-type analogue [163]. Another interesting P3-type material is Na0.6 Li0.2 Mn0.8 O2 [164, 165], which has a specific capacity of about 80 mAh g−1 in the voltage range of 4.5–3.5 V solely contributed by anionic redox. During the (de)sodiation process, the P3 structure is well maintained except for the formation of some stacking faults. The migration of Li/Mn ions can also be significantly avoided, as prismatic sodium vacancies are too large for them. Noticing these merits, P3-Na0.6 Li0.2 Mn0.8 O2 has been regarded as a good prototype to pointedly delve into the behaviors and mechanisms of anionic redox. Supported by ex situ neutron powder diffraction (NPD) technology, Rong et al. [165] found that the crystal structure factors play an important role in stabilizing the oxidized species, inhibiting the irreversible transformation of the oxidized species to O2 gas. However, despite the absence of oxygen release, the oxygen redox suffers from poor reversibility, displaying fast capacity fading. 3.2.1.6 Mixed-phase Layered Oxides

As mentioned above, sodium transition metal oxides can form in various structures, including P-type, O-type, tunnel-type, and some derived phases, showing different electrochemical behaviors. In brief, P-type layered and tunnel-type materials generally have better sodium ion diffusion kinetics and a wider solid-solution reaction range with better cycling stability, however, they suffer from limited initial charge capacities. In contrast, O-type materials exhibit almost the opposite characteristics. For the sake of coupling the different merits of different structures, the development of mixed-phase oxides could be a good choice. So far, various biphasic and multiphasic materials involving P/O, P/P, and layered/tunnel composites have been reported, demonstrating improved electrochemical performances as expected. Among various mixed-phase configurations, the P/O hybrid structure has attracted the most research interest. Guo et al. [166] reported a P2/O3-Na0.66 Li0.18

3.2 Oxide Cathode

Mn0.71 Ni0.21 O2+𝛿 and recorded a reversible capacity of ∼200 mAh g−1 . Moreover, the P2/O3 intergrown structure along the [001] direction was clearly captured by HADDF-STEM (Figure 3.19a). Later, Keller et al. [168] reported a P3/P2/O3 composite material, Na0.76 Mn0.5 Ni0.3 Fe0.1 Mg0.1 O2 , exhibiting a specific reversible capacity of 155 mAh g−1 in 2.0–4.3 V accompanied by a capacity retention of 90.2% after 601 cycles, which is more superior than the controlled samples of P2-Na2/3 Mn0.7 Ni0.1 Fe0.1 Mg0.1 O2 and O3-NaMn0.5 Ni0.3 Fe0.1 Mg0.1 O2 . Nonetheless, it should be noted that the different TM composition between P3/P2/O3 mixed-phase materials and a single P2 or O3 sample was not taken into consideration. Wang et al. [169] further obtained P2-, O3-, and P2/O3 mixed-phases with the same TM composition but different sodium concentrations, well confirming the improved rate capability of P2/O3 mixed-phase compared to O3 phase and elevated capacity compared to P2 phase. Although plenty of P2/O3 biphasic materials have been reported to date, some intrinsic questions like how the mixed-phase forms and how it improves the electrochemical performances are yet to be focused. Note that P2/O3 biphasic materials were empirically obtained with sodium content between 0.7 and 0.8. However, this is not a reliable rule since many layered oxides with such a high sodium content were obtained in either P2 or O3 single phases. Recently, X. Ji et al. [167] reported a comprehensive study on the formation mechanism of P2/O3 biphasic materials. In combination with cationic potential theory and various techniques, it is demonstrated that the P2/O3 biphasic structure is essentially stemmed from the internal heterogeneity of cationic potential induced by locally compositional non-uniformity (Figure 3.19b). Such compositional non-uniformity is rationalized by the thermodynamic driving force, which differ the reaction barriers between sodium and transition metal sources and largely determines the ion diffusion kinetics during solid-state reactions (Figure 3.19c). Notably, it had been recognized that synthesis temperature will greatly affect the formation of P2 and O3 phases while this work further answered how it works. A guidance for the rational design and efficient preparation of P2/O3 biphasic materials was also proposed, which consists of the customization of a multi-element composition (more than one metal ion in TM sites) with a critical overall cationic potential between the characteristic region of P2 and O3 phases and an inhomogeneous local composition created by reaction temperature or other synthetic factors. Moreover, as-obtained P2/O3 biphasic Na0.7 Ni0.2 Cu0.1 Fe0.2 Mn0.5 O2–𝛿 (denoted as NCFM) with well-designed quaternary composition shows outstanding rate capabilities (62 mAh g−1 at 2.4 A g−1 ), cycling stabilities (84% capacity retention after 500 cycles), and full cell performances (190 Wh kg−1 coupled with soft carbon anode). Regarding how the P2/O3 biphasic structure improves electrochemical performances, Keller [112] attributed it to the “synergistic effect” of the P2 and O3 structures without further explanations. Based on the in-situ or ex-situ XRD results, Sun [107], Chen [113], and Wang [114], et al. proposed that the P2/O3 intergrown structure can alleviate the phase transition in the high voltage region and reduce the volume change to a certain extent, thereby improving the structural reversibility.

65

66

3 Cathode Materials of SIBs

(a)

(b)

(c)

Figure 3.19 (a) HADDF-STEM image of P2/O3 intergrown structure. (b) Illustration of the correlations between cationic potential, synthesis temperature, and structure. (c) XRD of NCFM obtained at different temperature. Source: (a) Reproduced with permission from Guo et al. [166], © 2015/JOHN WILEY & SONS, INC. (b–f) Reproduced from Gao et al. [167]/with permission of Elsevier.

However, it failed to explain the improvement of the initial specific capacity (especially the initial charge capacity). Ji and coworkers further described the “synergistic effect” as the competitive reactions between P2, O3, and the intermedia phases based on the finding that physically mixed P2–O3 powders also present better performances than single P2 and O3 phase. Namely, the sodium extraction and insertion would preferentially occur in those phases with lower barriers associated with better structural stability or integrity, for instance, O3′ phase in the high-voltage region and P2/P3 phases in the mild voltage region. Then, the overall performances would largely depend on the testing voltage window due to the

3.2 Oxide Cathode

distinct phase evolutions of the single P2 and O3 phases. It should be highlighted that, for some components, constructing a P2/O3 biphasic structure cannot ensure the improvement of battery performances [115], implying that the rational design of TM composition is of vital importance. In addition, P2/P3 mixed-phase materials were also reported, showing considerable enhancement in kinetic properties. In view of the formation characteristics of P3-type materials, P2/P3 mixed-phase materials are usually obtained at low sodium content or low reaction temperatures. The P2/P3-Na0.67 Mn0.64 Co0.30 Al0.06 O2 composite phase material reported by Jiang et al. [117] displays a reversible specific capacity of ∼160 mAh g−1 in 1.5–4.0 V, accompanied by good rate and cycling performances. Zhu et al. [118] reported a multiphasic composite Na0.5 Ni0.1 Co0.15 Mn0.65 Mg0.1 O2 , in which a small amount of spinel phase was also integrated into the P2/P3 configuration, demonstrating good cycle stability.

3.2.2

Tunnel-type Oxides

Tunnel-type sodium oxides are generally obtained with a low Na/TM ratio (e.g. 800 mAh g−1 [123]. Complexing with CNTs, the ultrafine FeS2 @CNT composites could show a capacity of ∼750 mAh g−1 at 0.1 A g−1 after 100 cycles in Figure 4.5A [124]. Meanwhile, the reduced graphene oxide-wrapped FeS2 shows a capacity of 600 mAh g−1 after 100 cycles in Figure 4.5B [125]. Moreover, FeS2 @N-doped graphene microspheres displayed fast sodium-ions storage abilities (251.7 mAh g−1 over 10 000 cycles at

125

126

4 Anode Materials of Sodium-ion Batteries

5 A g−1 ) [127]. It could be found that the introduced carbon would facilitate the increase of conductivity and alleviate volume expansion, bringing about structural stability and reduced side reactions, thus resulting in long-term cycling properties and rate abilities. But, owing to the rich-sulfur state of FeS2 samples, the serious polysulfides could result in the fast fading of capacity. From the experiences of Li–S systems, it could be noted that other metal-oxides/sulfides would be beneficial for the capturing polysulfides through the utilization of strong polar bonds. From the previous reports, some metal-based materials have been explored, such as TiO2 , MoS2 , ZnS, and MnS samples [128–130]. For example, the capacity of FeS2 @TiO2 nanorods shows long-term cycling stability (637.8 mAh g−1 at 0.2 A g−1 after 300 cycles and 374.9 mAh g−1 at 5.0 A g−1 after 600 cycles) [129]. For Fe-based selenide, considering the relatively low capacity of Se, FeSe2 was regarded as the promising candidate. Interestingly, oxygen-doped FeSe2 nanosheets were prepared, and the addition O-atoms induced the enhancement of solid electrolyte interfaces (SEI). Moreover, the introduction of O-atoms brought the broadening of energy distribution, finally resulting in remarkable pseudo-capacitive behaviors [131]. Li’s groups reported the preparation of ZnSe–FeSe2 /RGO nanocomposites through the use of MOF-precursors. Benefiting from the synergistic effect of ZnSe and carbon, the materials displayed a high reversible capacity of 439 mAh g−1 at 0.1 A g−1 [132]. Moreover, Co0.85 S and MoSe2 were also explored for their high electrochemical properties in Figure 4.5C [126]. More interestingly, rod-like FeSe2 @N-doped carbon were successfully prepared through the pyrolyzation of Prussian blue. Meanwhile, through the control of heating rate (1, 5, and 10 ∘ C min−1 ), the size of rods was further tailored. Supported by the ether-based electrolytes, the as-designed samples displayed excellent cycling stability (308 mAh g−1 over 10 000 cycles) [133]. 4.3.3.3 The Exploring of Fe-based Phosphides

Compared to the materials above, Fe-based phosphides have been devoted numerous attention recently due to the relatively high capacity of P samples (2596 mAh g−1 ). Nowadays, considering the weak electronegativity of P-atoms, FeP, FeP2 and FeP4 samples have been successfully explored, especially the relatively stable FeP sample. In fact, through the facile ball-milling of manners, FeP/graphite could be successfully prepared, displaying a initial capacity of about ∼400 mAh g−1 [134]. Considering that the simple mechanism manners hardly induce the enhancements of active sites, the chemical manners were always explored in detail. For example, through chemical solution manner, FeP quantum dots were confined in P-doped carbon matrix. As a result, the sample shows a capacity of 674 mAh g−1 at 0.1 A g−1 and the record high-rate performances (262 mAh g−1 at 20 A g−1 ) in Figure 4.6A [135]. Meanwhile, CoP and NiP were also introduced to improve their capacity [92, 138]. Also, through the doping of heteroatoms, their capacity could be further improved. Yang’s groups reported Se-doped Fe2 P samples that displayed a capacity of 395.1 mAh g−1 at 1.0 A g−1 after 1000 cycles in Figure 4.6B [136]. Moreover, series of FePS3 and FePSe3 were also carried out [139–141]. FePS3 Nanosheets@MXene show a capacity of 676.1 mAh g−1 at the current of 100 mA g−1 after 90 cycles in Figure 4.6C [137].

4.3 Conversion Anode (A) (a1)

Carbonization

Hydrothermal

+

Phosphorization

Cl Fe MIL-101@CNTs

Cl

FeP@OCF-charge FeP@OCF-discharge FeP-charge FeP-discharge

1000 800 600 400

80 100 mA g–1 60

400 40 200

0

20

0

0 10 20 30 40 50 60 70 80 90 100 Cycle number

60 30

0

0

10

20 30 Cycle number (n)

800

(c2)

0.6

100 80

100

600

200

500

400

100 60

1000 2000

40 200

20 Unit: mA g–1

0 0

10

40

0 50

P

20 30 40 Cycle number

50

0 60

Se

(c3) 1 mV s–1

0.4 Current (mA)

100

90

200

Coulombic efficiency (%)

Specific capacity (mAh g–1)

120

1200

Fe

(C) 800(c1) 600

FeP@OCF/CNTs

(a3)

(b1)

Coulombic efficiency (%) Specific capacity (mAh g–1)

(B)

Specific capacity (mAh g–1)

(a2)

Coulombic efficiency (%)

Cl

0.2

58%

0.0 –0.2 –0.4 –0.6 0.0

0.5 1.0 1.5 2.0 2.5 Potential (V vs. Na+/Na)

3.0

Figure 4.6 For FeP quantum dots@P-doped carbon: the preparing mechanism, TEM images, and cycling stability (A1–A3). Source: Shi et al. [135]/with permission of John Wiley & Sons, Inc; For Se-doped Fe2 P samples: Mapping images (B1). Source: Gao et al. [136]/with permission of Elsevier; For FePS3 Nanosheets@MXene: cycling stability and rate properties (C1–C3). Source: Ding et al. [137]/Springer Nature.

4.3.3.4 The Exploring of Other Fe-based Composites

Despite the discussed materials above, other Fe-based samples were also explored, such as metal nitrides and carbides, resulting in high conductivity and ion-discussion coefficients. For example, with the assistance of melamine, Fe3 N was fabricated. Incorporating with carbon materials, Fe3 N@3D N-doped carbon was successfully prepared, displaying a capacity of 374 mAh g−1 , mainly ascribed to the construction of Fe—N bonds. Moreover, through the analysis of ex-situ XRD and TEM images, the high reversible crystalline-phase transformation could be noted [142]. And, Fe3 C@N-doped graphitic shell core-shell nanoparticles could deliver a capacity of 637 and 525 mAh g−1 at current density of 0.1 and 3.0 A g−1 . Obviously, with the increased rate abilities, the reducing capacity was smaller than that of other samples, further confirming the merits of this type materials [143].

127

128

4 Anode Materials of Sodium-ion Batteries

4.3.4

Mo-based Samples

Molybdenum (Mo), as an important metals resource, has attracted plenty of attention in energy-storage fields. And, the relative Mo-based materials displayed well conductivity, significant conducive crystal structure, and great capacity. But, like the merits of other metal-based samples, bare Mo-based samples still suffer from volume expansion and insufficient cycling lifespan, and so on. Generally, the constructing of unique architecture and complexing with carbon have been deemed as effective manners to promote their capacity. Differing from other metal-based samples, Mo-based materials exhibited series of significant traits, perhaps resulting in considerable electrochemical properties, such as the large space structure of MoO3 , the interlayer structure of Mo-based sulfide and selenide, and so on. Especially for Mo-based sulfides, MoS2 samples possess three kinds of crystalline phases: one-layer-stacked trigonal phase (named as 1T phase), two-layer-stacked hexagonal polymorph phase (regarded as 2H phase), and three-layer-stacked rhombohedral phase (deemed as 3R phase). More significantly, the band gaps of Mo-based samples (containing Mo-phosphide, nitride, 1T-MoS2 , and 1T MoSe2 , etc.) were more narrow than those of other metal-based samples (like Co/Fe/Ni-based samples), that is to say, the former has better electronic conductivity and reaction kinetics, which is also related to the designing of architectures. The detailed discussion about Mo-based samples was carried out as follows. 4.3.4.1 The Exploring of Mo-based Oxides

Up to now, MoO2 and MoO3 samples have captured much attention, due to their superior theoretical capacities, low toxicity, and excellent chemical stability. In particles, the theoretical specific capacity of MoO3 particles could reach up to 1117 mAh g−1 , larger than that of other metal-based samples. Through the analysis of discharge platforms, it could be noted that the region from 1.5 to 3.0 V was associated with the reactions (MoO3 + xNa+ + xe− → Nax MoO3 ), while that below 1.5 V was associated with (Nax MoO3 + (6–x)Na+ + (6–x)e− → Mo + 3Na2 O). As known, compared to metal sulfides and selenides, metal oxides still suffer from sluggish reaction kinetics and inferior reversibility. Thus, nanoscale engineering and complexing carbon were always carried out. For the controlling of structure, nanoparticles, nanosheets, and interesting morphology (such as hollow spheres and flower ) have been explored in detail. Rao’s groups reported that a few-layer MoO3 nanosheets incorporating with 3D graphene displayed a capacity of ∼800 mAh g−1 at 0.05 A g−1 [144]. Owing to their rich oxygen state, rich-defects MoO3−x samples were prepared and coated by the ultrathin Al2 O3 layers with great cycling stabilities [145]. Moreover, the heteroatoms were also introduced to improve their capacity. One-dimensional bunched Ni−MoO2 @Co−CoO−N-doped carbon was successfully prepared through the pyrolyzation of MOF-based precursors with strong pseudo-capacitive behaviors [70]. And, through the tailoring of interlayer engineering between carbon and MoO3 samples, the as-obtained samples displayed considerable cycling stability (larger than 1000 cycles) [146].

4.3 Conversion Anode

4.3.4.2 The Exploring of Mo-based Sulfide and Selenides

Differing from metal-oxides, metal-sulfides and selenides displayed remarkable conductivity and electrochemical activities. Moreover, note that the large interlayer distance (0.63 nm) was beneficial for the insertion of ions, leading to the extra capacity contribution. Among them, MoS2 molecules were composed of layer structure, and stacked through van der Waals’ interaction. Thus, for the evolution of MoS2 and MoSe2 , increasing interlayer distance, structural designing, and complexing were always used. For instance, through the hydrothermal reactions, MoS2 @C hollow nanorods were successfully prepared, showing the expanded interlayer distance (1.02 nm) and bringing about the initial capacity of ∼1000 mAh g−1 [147]. Of course, when the interlayer distance increases up to the limited value, multi-layer samples transform into thin-layer samples, or even single-layer structures. Yu’s groups reported single-layered ultrasmall nanoplates of MoS2 @carbon through electrospinning after heat treatments, where unique significant architecture was beneficial for the exposure of active sites, further bringing about the considerable electrochemical properties. Even at 10.0 A g−1 , their capacity could be still kept at about 300 mAh g−1 after 100 cycles [148]. Moreover, for the structural designing, yarn ball-like MoS2 nanospheres were coated with N-doped carbon and displayed remarkable rate abilities (352.1 mAh g−1 at 10.0 A g−1 ) [149]. Also, 3D hierarchical framework [150], graphene-like structure [151], etc. were further explored [152–154]. Obviously, the controlling of morphology would be conductive to promote the increase of active sites and alleviating volume expansion, boosting the improvement of electrochemical properties. And, some metal-based materials were further introduced for their increased electrochemical properties, such as SnS2 [155], Cu2 S [156], MoOx [157], FeS2 [128], and so on. For instance, MoS2 /Ti3 C2 hybrid films could deliver an exceptional volumetric specific capacity of 1510 mAh cm−3 at 0.28 mA cm−3 [158]. Meanwhile, porous hierarchical TiO2 @MoS2 @RGO nanoflowers displayed a long-term cycling stability (460 mAh g−1 ) with a capacity fluctuation of 0.03% per cycle within 350 cycles at 1.0 A g−1 [159]. Compared to 2H and 3R phases, 1T-MoS2 after the design process could display the greater electrochemical properties. 1T/2H MoS2 @SnO2 heterostructures were successfully designed, showing a capacity of 262 mAh g−1 at 2.0 A g−1 for 500 cycles [160]. For Mo-based selenides, they displayed further improvements in conductivity of MoSe2 . Similar to the designing manners of MoS2 @composities, MoSe2 was also modified in the same manner. For example, Ji’s groups reported the fabrication of MoSe2 nanospheres, displaying remarkable rate abilities [161]. Meanwhile, profiting from the introduction of interfacial chemical bonds, their capacity could increase up to 491 at 1.0 A g−1 . 1D-MoSe2 @TiC@C branch-core arrays were further explored, displaying the capacity of ∼184 mAh g−1 at 10.0 A g−1 [162]. Through the utilization of MOF-based template, the ZnSe and few-layer MoSe2 were homogeneously confined in the carbon matrix. At 4.0 A g−1 , the capacity could stabilize at 381 mAh g−1 after 250 cycles [163]. Moreover, considering the advantages of MoS2 , designing MoSx Se2−x structure was expected to achieve excellent physical–chemical properties. Zhou’s group successfully

129

130

4 Anode Materials of Sodium-ion Batteries

obtained MoS1.5 Se0.5 nanosheets, which were then wrapped by graphene to form MoS1.5 Se0.5 /carbon 3D nanostructure. The as-obtained samples displayed outstanding rate abilities, and the capacity of 375 mAh g−1 at 10 C (1 C = 680 mAh g−1 ) [164]. Luo’s report, nanocages-rich MoSe2 were tightly anchored on the surface of 3D B, N-doped rGO. Firstly, the nanocages mainly come from the defect of MoSe2 and dislocation of layers, which could be separated into three types. Note that, they show the attractive volume above 3.5 nm, rendering >10 times of Na-ions storage, which parallels the construction of inter-planner spacing [165]. 4.3.4.3 The Exploring of Other Mo-based Composites

Owing to the unstable chemical properties of other Mo-based samples, they were less explored. Among them, Mo–P and Mo–N samples were investigated. For example, hollow MoP@3D carbon was successfully prepared, displaying a capacity of 183.4 mAh g−1 . But, even after 3000 cycles could be still kept about stabilities at 10.0 A g−1 [166]. Through the use of metal-organic frameworks as precursors, porous yolk–shell MoP/Cu3 P@carbon micro cages were firstly prepared, showing a capacity of 307.8 mAh g−1 at 1.0 A g−1 . Even at 5.0 A g−1 , the capacity could be still kept at about 132.1 mAh g−1 over 6000 loops [167]. From their points, the excellent electrochemical properties are mainly derived from the high conductivities of Cu-particles (5.7 × 107 S m−1 ), bringing about considerable electrochemical properties.

4.3.5

Other Metal-based Samples

Of course, Zn-based, Cu-based, W-based, Mn-based, Cr-based, V-based, and Nb-based samples were explored for oxides, sulfides, and selenides. As known, all of them are regarded as conversion materials, which suffer from volume expansion and sluggish reaction kinetics. The modifying manners could be divided into the controlling of morphology, introduction of heteroatoms, and complexing of other metal samples. 4.3.5.1 The Exploring of Zn-based Samples

ZnO, ZnS, and ZnSe have been explored. For ZnO samples, series of manners were carried out for the improvement of electrochemical properties. Assisted by in situ TEM, the detailed sodiation behaviors revealed that the sodiated ZnO nanowires showed profuse dislocation plasticity. Meanwhile, the observed high-density dislocations provide more ductility to the sodiated ZnO anode, thus bringing about improved energy-storage ability [168]. Shao’s group reported ultrasmall ZnO nanocrystals incorporating honeycombed N-doped carbon, delivering a capacity of 166 mAh g−1 at 0.3 A g−1 . Even at 0.7 A g−1 , their cycling stability could be kept about 1000 cycles [169]. Moreover, considering the interfacial gaps of composites, Zn–O–C bonds were successfully constructed in ZnO@rGO samples due to the rich functional groups of rGO, boosting the quickening of electron transport and enhancing structural stability [170]. Moreover, the designed rod-like structure serves as the transferring channel (like highway), facilitating the shuttling of

4.3 Conversion Anode

ions. Profiting from series of traits above, their capacity could be kept above 100 mAh g−1 at 0.5 A g−1 . Obviously, the sluggish kinetics would inhibit the release of energy-storage capacity, thus inducing the exposure of active sites. Through hydrothermal reactions and CVD manners, ZnO porous nanosheets@amorphous carbon are successfully fabricated, displaying strong volume strains and large electrode/electrolyte contact interfaces [171]. Moreover, Zhang et al. prepared flower-like NiO/ZnO@n-doped carbon materials, showing the porous structure resulting in the quickening ion transport capability. At 2.5 A g−1 , the capacity could be retained at about 154 mAh g−1 after 2500 cycles [172]. In addition, Co3 O4 and Ni-films, respectively, were investigated,. Based on the discussions and reports above, it could be summarized, ZnO electrodes hardly displayed the considerable energy-storage abilities, mainly ascribed to unsuitable capacity. Thus, the exploration of ZnS and ZnSe was further explored. For example, Zhang’s groups reported urchinlike ZnS microspheres@N-doped carbon, delivering a capacity of 690 mAh g−1 after 100 cycles at 0.1 A g−1 [173], while the as-prepared ZnS nanoparticles@rGO displayed a capacity of 481 mAh g−1 after 500 loops [174]. Certainly, through the tailoring of carbon, [email protected] co-doped carbon was prepared with the formation of Zn—O—S/Zn—S—C bonds, exhibiting a capacity of 347 mAh g−1 at 0.8 A g−1 [175]. Further reducing the particle size of active martials ZnS, the capacity of ZnS quantum dots (sub-10-nm-scale) @GO could reach up to 759 mAh g−1 at 0.1 A g−1 after 100 cycles [176]. More interestingly, utilizing ether-based electrolytes, ZnS nanospheres could show long-term cycling stability [177]. After assembling full cell (Na3 V2 (PO4 )3 vs. hollow sphere ZnS-Sb/C), their capacity could be kept at about ∼130 mAh g−1 . But, metal-sulfide still suffers from inferior rate abilities, mainly associated with the poor conductivity. Thus, further introducing Se elements, ZnSe@carbon nanofibers were successfully obtained through electrospinning manners. At large current density of 2.0 A g−1 , the sodium-ions storage capacity could be up to 369 mAh g−1 after 200 cycles [178]. Meanwhile, willow-leaf-like ZnSe@N-doped carbon nanoarchitecture could show a capacity of 242.2 mAh g−1 at 8.0 A g−1 even after 3200 cycles for SIBs anodes [179]. Even if the operating temperature was reduced to −20 ∘ C, the capacity of ultrafine ZnSe@3D porous N-doped carbon could be still kept at about ∼300 mAh g−1 [180]. And, using ZIF-8/67 as precursors, the capacity of ZnSe/CoSe@N-doped porous carbon could be kept about at 236.4 mAh g−1 at 10.0 A g−1 [180]. Of course, Zn—O—C bonds were further used to induce the improvements of ZnSe@carbon capacity, delivering a capacity of 281 mAh g−1 after 5.0 A g−1 after 1000 cycles [181]. 4.3.5.2 The Exploring of Cu-based Samples

Copper (Cu), as a highly conductive metal, displayed promising potential. But, owing to their high cost, the practical applications of anodes were still limited for SIB anodes. Whereas, considering the traits of Cu-based composites, they were still explored. For example, CuO array@N-doped carbon and CuO/C nanospheres were fabricated [182, 183]. Among them, the latter showed a reversible capacity of 402 mAh g−1 after 600 cycles at a current density of 0.2 A g−1 , and 304 mAh g−1 at 2.0 A g−1 . The rare exploration was due to their sluggish kinetic behaviors, and thus

131

132

4 Anode Materials of Sodium-ion Batteries

CuS and CuSe samples were further explored. Through the use of template methods, the as-obtained CuS@CoS2 double-shelled nanoboxes displayed a capacity of ∼400 mAh g−1 at 0.5 A g−1 after 500 cycles [184]. And, the large-scale free-standing CuS nanotube arrays@graphdiyne deliver a capacity of ∼200 mAh g−1 at 1.0 A g−1 after 1000 loops [185]. Meanwhile, N-doped carbon-coated CuS nanowires were further prepared through the refluxing methods, displaying a capacity of 294.4 mAh g−1 even at 20 A g−1 [186]. Remarkably, a competitive long-life cyclic stability could be noted (216.7 mAh g−1 at 20 A g−1 , retaining 81.7% of its capacity retention over 10 000 cycles). Supported by the detailed kinetic analysis, their pseudo-capacitive contributions were found to be larger than ∼90%. Erythrocyte-like CuS microspheres@PANI were fabricated, where PANI serves as elastic buffering layer to protect the volume expansion. Benefiting from the design of structure, the as-optimized samples displayed ultra-large rate properties (500 mAh g−1 at 0.1 A g−1 and 215 mAh g−1 at 40.0 A g−1 ). Meanwhile, excellent cycling stability could be found over 7500 cycles at 20 A g−1 . From their points, it could be noted that SEI films could be stabilized and that the growth of Cu-nanoparticles was inhibited by PANI swollen [187]. For CuSe samples, the nanosheets-assembled CuSe crystal pillar and ultrathin CuSe nanosheets have been explored as SIBs anodes [188, 189]. The former could display a capacity of 295 mAh g−1 at current density of 10 A g−1 and the latter deliver a capacity of 276 mAh g−1 at 20.0 A g−1 . And, with the use of in situ XRD and ex-situ TEM, their structural transition and phase evolution were further revealed [188]. 4.3.5.3 The Exploring of Mn-based Samples

Owing to their relatively high theoretical specific capacity and low cost, manganese (Mn)-based materials have greater potential to be explored in comparison to other transition-metal oxide anodes. Owing to multi-valence state of manganese (Mn), series of Mn-based composites were investigated, such as MnO/MnO2 / Mn2 O3 /Mn3 O4 , MnS, MnSe, and so on. For Mn-based oxide composites, plenty of MnO@C samples were investigated, like ultrasmall MnO nanoparticles@N-doped carbon, MnO@C nanorods, Ni-doped carbon@MnO@graphene ribbons and so on. For example, Xia’s group prepared double-shell hollow MnO nanospheres with a capacity of 97 mAh g−1 at 10.0 A g−1 [190]. Moreover, using a rapid and simple hydrothermal route, feather-like MnO2 nanostructure@carbon cloth was prepared, displaying a capacity of 300 mAh g−1 at 0.1 A g−1 after 400 cycles [191]. Free-standing MnO2 nanosheets@carbon nanofibers displayed a capacity of 135 mAh g−1 at 1.0 A g−1 [192]. It could be found that Mn-based composites hardly provided high enough capacity. Thus, MnS and MnSe were further investigated. Among them, nano-MnS@N-doped carbon showed a capacity of 461.2 mAh g−1 after 200 cycles at 0.1 A g−1 [193], while rod-like MnS@C from the thermal chemical reaction of MOF exhibited a capacity of 148.3 mAh g−1 at 1.0 A g−1 after 5000 loops [194]. In addition, using Prussian blue analog as the precursor, cubic MnS-FeS2 composites exhibited a capacity of 134 mAh g−1 at 4.0 A g−1 after 14 500 cycles [195]. Benefiting from C—S—Mn bonds, the as-obtained single MnS nanocubes@N,S co-doped carbon could display a capacity of 329.1 mAh g−1 at 1.0 A g−1 after

4.3 Conversion Anode

3000 cycles [196]. Moreover, thorn-ball-like alpha-MnSe/C nanospheres, coaxial alpha-MnSe@N-doped carbon, and rock-salt MnS0.5 Se0.5 nanocubes have been explored, respectively. Their energy-storage abilities were mainly about 405 mAh g−1 at 0.5 A g−1 , 405 mAh g−1 at 14.0 A g−1 , and 457 mAh g−1 at 0.1 A g−1 . In summary, considering the non-active characteristic traits of Mn-elements, the relative research about Mn-based samples was hard to be explored. 4.3.5.4 The Exploring of Cr-based Composites

The relative Cr-based samples were relatively less. For example, MCr2 S4 (M = Cr, Ti, Fe) was first prepared, displaying a capacity of 470, 375, and 524 mAh g−1 at 0.5 A g−1 after 200 cycles [197]. Using the ball mill method, Na2 CrO4 /C displayed relatively high working voltage (similar to 1 V), while showing a capacity of 228 mAh g−1 [198]. 4.3.5.5 The Exploring of W-based Composites

Similar to Mo-based samples, W-based samples also displayed unique electrochemical traits, such as significant interlayer distance, large crystalline space, and so on. And their energy-storage properties have been widely explored. For W-based oxides, nano-faceted WO3–x nanorods were in-situ complexed with carbon nanosheets, displaying a capacity of 184.6 mAh g−1 at 30.0 A g−1 [199]. Meanwhile, using NVP as cathodes, the capacity of full cell could be kept at about 285.5, 182.6, 112.5 mAh g−1 at 0.5/1.0/5.0 A g−1 . From their points, the excellent may come from the advantages as follows: (i) the larger interplanar spacing induced the shortening of ordered ion-diffusion paths; (ii) rich oxygen vacancies and unsaturated active sites could improve reaction efficiency and the strengthen structural stability. But, owing to the sluggish reaction behaviors, more research focuses on the WS2 and WSe2 samples. For example, with the assistance of facile solvothermal method and heat treatments, the capacity of the as-obtained WS2 nanowires could reach up to 605.3 mAh g−1 at 0.1 A g−1 [200]. Meanwhile, Zhang’s groups reported WS2 @hollow carbon sphere (575 mAh g−1 at 0.1 A g−1 ) [201]. After the thinning of WS2 layers, layered WS2 in hollow beaded carbon nanofibers could deliver a capacity of 351 mAh g−1 at 2.0 A g−1 [202]. In order to expose the activity of active materials WS2 , they were further reduced to form WS2 nanocrystals@N, P co-doped carbon, displaying a capacity of 311 mAh g−1 at 3.0 A g−1 [203]. For their large conductivity, WSe2 -based composites were further explored. For instance, 3D free-standing WSe2 /C hybrid nanofibers were prepared, and the great long-term rate properties could be noted at about 257.0 mAh g−1 at 25 A g−1 after 10 000 cycles [204]. Moreover, ultrathin few-layered WSe2 @N,P dual-doped carbon exhibited a capacity of 210 mAh g−1 at 0.5 A g−1 for 120 cycles [205]. 4.3.5.6 The Exploring of V-based Composites

For pursuing the high energy/power density, vanadium (V)-based samples with multi-electron reactions (V2+ –V5+ ) have been regarded as the promising candidates for next-generation energy-storage systems, mainly ascribed to their consolidated frameworks and high theoretical capacities. And their energy-storage properties have been widely explored. Of course, bare V-based composites hardly meet the high

133

134

4 Anode Materials of Sodium-ion Batteries

demand of energy-storage systems, so structure-function and state-of-the-art were further designed for the improvements of electrochemical properties. Moreover, as known, V (No.23 in periodic table) comes from VB groups, meanwhile the valence electron layer was about 3d3 4s2 , resulting in the relatively high valence, which has a relationship with energy-storage capacity. Firstly, for V-based oxides, VO2 has been explored, respectively. About VO2 -based composites, VO2 /rGO nanorods were successfully prepared through microwave-assisted solvothermal method, which could retain a capacity of 200 mAh g−1 after 200 cycles [206]. Moreover, with the assistance of simple solution-derived method, graphene-wrapped porous VO2 microspheres were fabricated. Surprisingly, the as-optimized samples showed a high reversible capacity of 373 mAh g−1 , and 138.8 mAh g−1 at 24.0 A g−1 [207]. Moreover, attracted by the synergistic effect of the VO2 and MXene, 3D flower-like VO2 /MXene was firstly prepared, displaying a capacity of 206 mAh g−1 at 1.6 A g−1 [208]. Moreover, amorphous VO2 were reported by Fan’s group, exhibiting the excellent pseudo-capacitive behaviors for high-rate symmetric batteries [209]. From their points, amorphous materials could provide open architecture with the possibility of multiple percolation directions, boosting the shuttling of ions and alleviating volume expansion. Even at high current density (20.0 A g−1 ), the capacity could remain about 120–125 mAh g−1 ). For V-based sulfides, the exploring of VS2 has been carried out. Based on DFT calculations, metallic VS2 monolayer polytypes displayed the promising potential (232.91 and 116.45 mAh g−1 for 1H and 1T phases, respectively) and open-circuit voltage (1.30 and 1.42 V for 1H and 1T phases, respectively) [210]. And, VS2 nanoarchitectures were assembled by single-crystal nanosheets, showing a high capacity of 193 and 172 mAh g−1 at a high rate of 0.5 and 1.0 A g−1 , and 403 mAh g−1 after the following cycling test of 200 cycles at 0.2 A g−1 [211]. In order to alleviate the volume expansion, the battery, their pristine particle size was further reduced through the formation of spherical nanoflowers VS2 [212]. The as-obtained composites showed a capacity of 329 mAh g−1 at 0.2 A g−1 with the ICE of 88.52%. Owing to their layer traits, the complexing of other layer MoS2 samples was explored. Such as MoS2 /VS2 samples, even at 10.0 A g−1 , a capacity of 644 mAh g−1 could be noted [213]. Moreover, metallic VSe2 and conductive polypyrrole were complexed, delivering a superior rate capability with a capacity of 260.0 mAh g−1 at 10 A g−1 in SIBs. After 2800 loops at 4.0 A g−1 , the long-term cycling stability could be found at about 325 mAh g−1 [214]. In summary, owing to the poor solution of vandic salts, the exploring of V-based composites was hardly carried out. 4.3.5.7 The Exploring of Nb-based Composites

Similar to the traits of V-based samples, those of Nb-based composites were further explored as SIBs anodes, especially their rich redox chemistry (Nb5+ /Nb4+ , Nb4+ /Nb3+ ). For V-based oxides, the ordered-mesoporous Nb2 O5 /carbon composite showed mesoporous structure, which facilitated the enhancement of volume stress and the infiltration of electrolyte/electrodes [215]. And the as-obtained sample displayed a capacity of 150 mAh g−1 at 0.05 A g−1 . Moreover, assisted by electrospinning manners, T-Nb2 O5 /C nanofibers exhibited a relatively stable capacity of

4.4 Metal/Alloy Anode

150 mAh g−1 at 1 A g−1 after 5000 loops. Even at an ultrahigh charge-discharge rate of 8 A g−1 , a high reversible capacity of 97 mAh g−1 could be still achieved [216]. Moreover, after surface engineering, black Nb2 O5–x @graphene nanosheets were prepared, showing great rate properties of 123 mAh g−1 at 3.0 A g−1 [217]. 4.3.5.8 The Exploring of In-based Samples

The flexible In2 S3 nanoplates@multi-walled carbon nanotubes were firstly prepared through simple vacuum-assisted assembly. The capacity of the as-fabricated In2 S3 @ MWCNTs could reach up to 410 mAh g−1 at 0.05 A g−1 . Meanwhile, through the complexing of In2 S3 and g-C3 N4 , the relative composites were successfully prepared, where g-C3 N4 matrix could improve the transferring of ions, alleviating the volume alteration happening during cycling [218, 219].

4.4 Metal/Alloy Anode 4.4.1

Sb-based Samples

Antimony, as the primary member of alloying materials, has captured numerous attentions due to its high theoretical capacity (660 mAh g−1 ) [220]. Note that low puckered-layer structure of Sb always shows the operating platforms (08–0.9 V), accompanied by the release of structural strains, resulting in promising potential. But, considering its serious volume expansion (293%) from hexagonal Na3 Sb as the final stable phase, the fast capacity fading would be detected. Moreover, to pursue the demand for higher capacity, series of SbX samples were constructed (X = O, S, Se, etc.), bringing about considerable energy-storage abilities. Frustratingly, all of them were still limited by volume expansion, the shuttling effect of polysulfides, and other side reactions. Like other alloying electrodes, the controlling of architecture and other heterostructures was introduced to promote their capacity, as displayed in Figure 4.7 [221, 222]. 4.4.1.1 The Exploring of Sb and Sb-based Alloy Samples

For Sb metals, series of tailoring manners were carried out, including different architectures, various alloys, and incorporating other samples. Among them, the designing of structure was widely carried out, which would be divided into unique morphology, reduced particle sizes, and the preparing amorphous. For the control of morphology, nanoparticles, leaf-like, rod-like, and so on have been explored, with considerable electrochemical performances. It could be deduced that the advantages of unique architecture could induce the increase of active sites and the reduction of structural stress. Certainly, their morphology was closely associated with their preparation manners. Nowadays, they could be summarized as mainly being about the substitution method (using the higher active metals), reduction manners (using NaBH4 or other gases), and hot carbon reduction. For example, Ji’s groups reported Sb hollow spheres and leaves from the reaction of SbCl3 with Zn or Mg particles [223, 224]. Benefiting from their increased contacting area and rich active sites,

135

136

4 Anode Materials of Sodium-ion Batteries

Figure 4.7 The simple sodium-storage mechanism of different Sb-type about materials. Source: Liu et al. [221]/Royal Society of Chemistry.

their capacity could be obviously improved, displaying a capacity of >600 mAh g−1 with excellent cycling stability. Moreover, through the Cu2 Sb-mediated growth strategy, the highly uniform Sb nanotubes were prepared. Owing to the enhanced structural strain and accelerating ion transport, the optimized samples deliver remarkable rate abilities (286 mAh g−1 ) and cycling stability (342 mAh g−1 after 6000 cycles at 1.0 A g−1 ) [225]. Utilizing the high conductivity of other metals (like Ni, Cu, Co, Al, Fe, or Mo), they only serve as the stable framework to alleviate the volume expansion. Supported by the solution reactions, Sb–Ni–C samples were successfully prepared, delivering a capacity of 468 mAh g−1 at 1.0 A g−1 in Figure 4.8A [226]. Moreover, carbon materials, as the typical buffering matrix, were always introduced into the active samples. Using MOF as the precursor, Sb@porous carbon octahedrons were prepared. After the optimized voltage (0.5–0.8 V), the capacity of the as-obtained sample could reach up to 634.6, 474.5, and 451.9 mAh g−1 at 0.1, 0.2, and 0.5 A g−1 after 200, 500, and 250 cycles, respectively in Figure 4.8B [227]. Moreover, despite the introduction of SC matrix, other hard matrices were also introduced, such as TiO2 and MoS2 , alleviating the volume expansions [228, 229].

4.4 Metal/Alloy Anode (A) (a1)

(a3) NaBH4

Sb–Ni Alloy

Sb-Ni cyanogel

Sb–Ni framework

In situ immobilization GO

rGO

NaBH4

GO@Sb–Ni cyanogel

(B) (b1)

rGO@Sb–Ni framework

HAADF

1000 Capacity (mAh g–1)

(a2)

(a4) rGO@Sb–Ni (40GO) rGO@Sb–Ni (20GO) rGO@Sb–Ni (80GO) Sb–Ni

800 100 mA g–1

600 400 200 0

C

0

20

40 60 Cycle number

80

100

(b2) 200 nm

5 μm C Sb

1500 100 1200

Sb@C

80 60

600

40 300 0

Microsized Sb 0

40

80 120 160 Cycle number

20 0 200

Coulombic efficiency (%)

Sb

Specific capacity (mAh g–1)

(b3)

Figure 4.8 For Sb–Ni–C samples: the preparing mechanism, TEM images and cycling stabilities (A1–A4). Source: Wu et al. [226]/with permission of American Chemical Society; For Sb@porous carbon octahedron: Mapping images, SEM images, and cycling stabilities (B1–B3). Source: Li et al. [227]/with permission of American Chemical Society.

For example, double-walled Sb@TiO2−x nanotubes were successfully designed, displaying a capacity of 424 mAh g−1 after 1000 cycles at 6.6 A g−1 [229]. 4.4.1.2 The Exploring of Sb-based Oxide, Sulfides, Selenium

The elements (O, S, Se, etc.) of VI Groups serve important role in the energy-storage ability due to their relatively high reaction activities and suitable binding capabilities. Among them, it could be noted that, with the increasing of radii, the conductivity of Sb-based samples improved from Sb2 O3 to Sb2 S3 , further to Sb2 Se3 samples. Moreover, owing to their growth anisotropy, they always showed octahedral structure for Sb2 O3 and rod-like structure for Sb2 S3 and Sb2 Se3 . For obtaining other unique morphology, extreme fabrication processes were investigated, such as hydrothermal methods, microwave manners, and so on. Certainly, Sb-based samples still suffer from more serious volume swelling and inferior reversibility, thus various protecting matrices were utilized, effectively inhibiting their side

137

138

4 Anode Materials of Sodium-ion Batteries

reactions and particle pulverizations. Among them, different kinds of carbon were used, such as 1D structures (CNTs, carbon fibers), 2D structures (graphene, nanosheets), and 3D structures (hollow sphere, carbon matrix). For Sb2 O3 samples, the designing of architecture was about coralline-like, spherical tremella, and so on. Mesoporous Sb2 O3 @Sb nanocomposite was successfully fabricated through the one-step dealloying of two-phase Mg-Sb precursor, with ultrafine capacity (659 mAh g−1 ) and excellent rate abilities (200 mAh g−1 at 29.7 A g−1 ) [230]. With the assistance of spraying method and one-step partial reduction process, Sb2 O3 @Sb nanoparticles with N-doped carbon microcages were prepared, delivering considerable cycling stability (245.2 mAh g−1 after 10 000 cycles under 10.0 A g−1 ) [231]. And, most of the reports about Sb2 O3 were about the incorporation between active materials and carbon, including Sb2 O3 nanosheet@graphene, Sb2 O3 @CNTs@rGO, free-standing Sb2 O3 @carbon cloth, and so on. Moreover, other Sb-based oxides (Sb4 O5 Cl2 , Sb8 O11 Cl2 , and SnO-GeO2 -Sb2 O3 glass) were explored, but with inferior cycling stabilities [232–234]. For Sb2 S3 or Sb2 Se3 samples, rod-like structures were always found. For example, 1D Sb2 S3 @C and Sb2 Se3 @C were prepared by Ji’s group [235, 236]. Of course, other composites were noted about Sb-based sulfides@carbon, such as amorphous Sb2 S3 @graphite [237], Sb2 S3 @PPy [238], Sb2 S3 nanorods@rGO [239], Sb2S3 @P-doped carbon [240], Sb2 S3 @S-doped carbon [241], Sb2 S3 @CNTs@N-doped carbon [242], and so on. Those about Sb-based selenide@carbon mainly contain Sb2 Se3 @C [243], Sb2 S3 @N-doped graphene [244], 2D nanosheets Sb2 Se3 @carbon [245], rod-like Sb2 Se3 @carbon [246], and cube-like Sb2 Se3 @carbon [247]. For example, Qiao’ groups reported the preparation of multi-shell hollow structured Sb2 S3 through the sulfuration of ZIF-8 [248]. Moreover, the relative hybridized composites were carried out due to their synergistic effects, such as improved surface-controlling behaviors, enhanced conductivity, and so on. Nowadays, series of heterostructures were explored, including In2 S3 @Sb2 S3 [218], Sb2 S3 @Ti3 C2 Tx nanowire [249], Sb2 S3 @Bi2 S3 [250], Sb2 S3 @MoS2 [251], Sb2 S3 @SnS2 @C [252], and Sb@Sb2 Se3 @TiC@C [253]. For instance, the as-reported In2 S3 @Sb2 S3 @MCNTs displayed remarkable pseudo-capacitive behaviors. Even at 3.2 A g−1 , the capacity could remain about 355 mAh g−1 [218]. It could be deduced from about three points: (i) the porous morphology, (ii) the synergistic effect of insertion/conversion/alloying reactions, and (iii) the lowered migration barrier of Na+ ions. Moreover, owing to the obvious interfacial gaps, the Sb—C or S—C bonds were established [254]. But it should be acknowledged that the chemical preparation of Sb2 S3 was limited by serious pollution and high energy consumption. Thus, the derivation of stibnite’s was taken from places, such as natural Sb2 S3 @graphite, Sb@Sb2 S3 @double carbon, and Sb2 S3 @C with Sb-C bonds.

4.4.2

Sn-based Samples

Attracted by the high theoretical capacity of tin (Sn) samples (847 mAh g−1 for Na15 Sn4 in SIBs), plenty of exploring activities were triggered. But, it should be acknowledged that Sn still suffers from serious volume expansion (larger than ∼420%), perhaps resulting in particle pulverization, capacity fading, and inferior

4.4 Metal/Alloy Anode

cycling stabilities. Up to now, series of Sn-based samples have been explored, such as Sn composites, SnOx , and SnS. Moreover, plenty of modified manners were carried out to improve their electrochemical properties, including carbon in complexing, structure controlling, and so on. Firstly, the reaction mechanism was revealed, and it could be found that, one Sn atom could host 3.75 sodium atoms, that is to say, forming Na15 Sn4 samples with a delivering capacity of 847 mAh g−1 . Based on the previous reports, the phase transformations (β-Sn→NaSn2 →α-Na1.2 Sn→Na5–x Sn2 →Na15+x Sn4 ) were confirmed through Ab initio random structure searching (AIRSS) and species-swap methods [255]. Of course, with the developments of advanced analyzing technologies, in situ TEM images and ex-situ EDS were further taken place to investigate the visual phase transformation [256–258]. 4.4.2.1 The Exploring of Sn-based Alloys and Sn@Carbon Materials

For solving the relative issues during cycling, series of improved manners were carried out. Nowadays, it could be summarized as follows: (i) the introduction of other metals, (ii) the carbon complexing. For the former, it could be divided into two aspects: one is about the introduction of no-active metal, where other metals hardly offer extra capacity, but only serve as buffering matrix; another is about the complexing of active metals (like Sb, P, etc.), where they play roles as active sites to provide higher capacity, accompanying with the synergistic effect. For instance, copper (Cu), as the high conductive material, has been used to form Cu–Sn alloys, delivering a capacity of 660 mAh g−1 at 2.5 A g−1 after 400 loops, mainly ascribed to the formation of relatively stable 3D structure Cu6 Sn5 samples [259]. Meanwhile, through the control of preparing manners, Co–Sn, Ni–Sn, Fe–Sn, and Cu–Sn were further obtained with interesting designing architecture (like microcages and 3D porous). Moreover, in order to further improve their capacity, Sb was introduced to form Sb–Sn samples. Utilizing ethylamine, choline chloride, and ethylene glycol as new-type electrolytes, its capacity could be kept stable with the residual capacity (95% with the first capacity as standard) after 300 cycles. Even after 800 loops, the capacity ratio could be about 80% [260, 261]. For the latter, by using carbon as buffering matrix, their energy-storage properties could be effectively improved, and it is ascribed that the low melting point of Sn (231.89 ∘ C) could induce the agglomeration of Sn materials in the preparing process. Moreover, carbon matrices also displayed high conductivity and well ion-passing abilities. Of course, the loading manners serve important role in controlling of electrochemical properties. And, the synthesis methods could be concluded to be: aerosol spray pyrolysis, self-assembly methods, microwave-assisted reduction, and electrospinning technology. Firstly, aerosol spray pyrolysis could be illustrated with the pyrolysis of metalorganics, where Sn elements were distributed as single atoms. With the increase in temperature, plenty of tiny Sn nanoparticles could be uniformly anchored in the carbon matrix. Chen’s groups firstly reported the preparation of ultrasmall Sn particles@carbon through the pyrolysis of resorcinol and Sn-salts, where the size of Sn was only 8 nm, boosting the enhancements of

139

4 Anode Materials of Sodium-ion Batteries (A)

(B)

(a2)

(a1)

(a3)

GQDs with “octopus” structure

(b1)

Graphene oxide

(b2)

Complexation Sn2+

O atom C atom N atom

Self assembly

NGQD/Sn-NG

700 °C

Chemical confinement

NH3

Spatial confinement

100

600

80 400

60 NGQD/Sn-NG NGQD/Sn Sn-NG Sn

200

40 20

Coulombic efficiency (%)

(b3) Capacity (mAh g–1)

140

0 0

50

100

150 Cycle number

200

250

0 300

Figure 4.9 The ultrasmall Sn@carbon from aerosol spray pyrolysis: SEM and TEM images (A1–A3). Source: Liu et al. [262]/with permission of John Wiley & Sons, Inc; The Sn@carbon samples from self-assembly methods: the fabrication mechanism, TEM images, and long-term cycling stability (B1–B3). Source: Liu et al. [263]/with permission of John Wiley & Sons, Inc.

volume stress. At 1.0 A g−1 , the capacity could be kept at about 415 mAh g−1 after 500 loops in Figure 4.9A [262]. Meanwhile, self-assembly methods could illustrate that the tailoring of preparing conditions could induce the reduction of Sn2+ elements. Assisted by the NH3 and graphene quantum dots, Sn nanodots (2–5 nm) were anchored in carbon matrix, accompanied by the construction of interfacial bonds (Sn–O–C and Sn–N–C bonds). Benefiting from the ultratiny structure, more active sites were exposed as expected, thus bringing about the high capacity (555 mAh g−1 at 0.1 A g−1 ) in Figure 4.9B [263]. Moreover, microwave-assisted reduction could be interpreted so that, the ultrafast of heating and cooling could inhibit the growth of

4.4 Metal/Alloy Anode

Sn nanocrystals. Zhang et al. reported monolithic Sn/C composites with long-term cycling stability (1000 cycles at 1.0 A g−1 ). From their points, the aggregation of Sn particles was effectively inhibited by the N–C layers [264]. With electrospinning technology, it could be understood that, through the complexing of organic solution and Sn salts, the rod-like structure could be formed. After heating and pyrolyzation, Sn2+ ions could be reduced in carbon matrix to form Sn@C CNF samples. After 1000 cycles, the capacity could still be about 80% of the 1st cycling capacity as standard [265]. 4.4.2.2 The Exploring of Sn-based Oxides

In order to further improve their capacity, anions with extra storage ability were introduced, inducing O, S, etc. Among them, considering the relative mass of O-atoms, the Sn-based oxides were always explored, but they still suffer from serious volume expansion. In order to solve the problem above, the reduction of particle size and designing of architectures were also undertaken, which would facilitate the structural stress. Based on their kinds of dimensions, they could be divided into dost-like, rod-like, sheet-like, and 3D frameworks [266, 267]. Nanodots SnO2 @carbon were firstly prepared and used as free-standing electrodes [268]. Assisted by hydrothermal reaction, SnO2 submicrorods were successfully confined on the surface of Ni foam, displaying a capacity of 424 mAh g−1 at 0.1 A g−1 after 200 loops [269]. Moreover, Guo’s groups reported the preparation of Sn/SnO nanosheets, incorporating carbon matrix [270]. Despite the capacity improvements from the increasing active sites, their cycling stabilities were not effectively enhanced. Thus, complexing with carbon matrix has been regarded as the effective modification manners. And the kinds of carbon could be divided into 1D CNTs, 2D CNSs, and 3D CFWs [271]. For instance, with the support of liquid-phase deposition and in situ chemical-polymerization methods, the CNT@SnO2 @PPy were successfully prepared, where CNTs and PPY serve roles as electrons fast-moving paths and volume-swelling matrices [272]. Compared to CNTs, graphene, as the popular material, has been widely used to improve their electrochemical properties, and series of SnO2 @rGO or SnO2 @carbon cloth was prepared. Chen et al. prepared N-doped silk wadding-derived carbon/SnOx@reduced graphene oxide, which displayed 572.2 mAh g−1 at 0.1 A g−1 . Even after 1000 cycles, the capacity could be still kept at about 245.7 mAh g−1 at 1.0 A g−1 . From their points, the excellent energy-storage ability could be deduced from three points: (i) the reduced particle size (< 100 nm), (2) the designed double carbon layers, and (ii) the highly effective conductive networks in Figure 4.9B [273]. Compared to 1D and 2D structures, 3D CFWs were always prepared, mainly ascribed to the facile preparing process. For the prepared SnO2 @C@TiO2 hollow spheres, their capacity could be kept at about 380 mAh g−1 at 0.2 A g−1 . Herein, carbon and TiO2 were regarded as the SC matrix and robust protecting shells, collectively inhibiting the volume expansion [274]. Moreover, some glass anodes were also investigated, such as Sn@SnO–P2 O5 and SnO–V2 O5 –SiO2 samples with considerable energy-storage properties [275, 276]. In summary, through the control of architecture and carbon layers, their electrochemical efficiency could be effectively improved [274].

141

142

4 Anode Materials of Sodium-ion Batteries

4.4.2.3 The Exploring of Sn-based Sulfides

Owing to the directional assembly characteristic of SnS2 , they were always going to form nanosheets. Through the solvothermal methods, hierarchical ultrathin SnS2 nanosheets were formed [277]. For their crystalline structure, the building of defects and heteroatom doping were introduced. With the exposure of active sites, the pseudo-capacitive behaviors of SnS2 could be effectively enhanced, boosting its remarkable rate abilities. Even at 12.8 A g−1 , the as-obtained samples displayed a capacity of 12.8 mAh g−1 . Obviously, the increase in active sites could induce the enlargement of energy distribution [278]. Meanwhile, with the introduction of Co-ions, the conductivity of the as-prepared Co-doped SnS2 would be improved, accompanied by the enhanced activity of redox reactions. At 0.2 A g−1 , the capacity could reach up to 1288 mAh g−1 after 100 cycles [196]. In addition, complexing other phases displayed the interesting synergistic effect, and series of composites were explored, including SnS2 @Co3 S4 , SnS2 @NiS2 , SnS2 @Sb2 S3 , and so on. By solvothermal method and an oil bath method, the SnS2 @Co3 S4 samples were prepared, showing a capacity of 637 mAh g−1 after ∼760 loops at 2.0 A g−1 [279]. More interestingly, their electrochemical performances could be affected by the selection of binder. After the exploring of CMC-Na, PAA-Na, CMC-Na-PAA-Na (1/1), PVDF, and PTFE, PAA-CMC binder electrodes exhibited considerable rate performances [280]. Meanwhile, series of SnS2 @C were further designed. Based on their distributions, they could be divided into three kinds: (i) growing on the surface; (ii) coated in the internal; (iii) forming the composites. Meanwhile, in order to further improve their abilities, the tailoring of carbon was further taken place, especially heteroatom doping (N, S, and P elements) [281, 282]. Utilizing electrospinning technique, 2D SnS2 nanosheets were successfully grown on the surface of 1D S, N-doped carbon fibers, displaying a capacity of 380.1 mAh g−1 at 0.5 A g−1 after 200 cycles [283, 284]. Meanwhile, ultrasmall SnS2 @ N-doped 2D graphene was prepared with a capacity of 400 mAh g−1 at 0.2 A g−1 after 100 cycles [285]. And, through solution reactions, SnS2 @3D carbon was fabricated successfully, showing a reversible of 753.8 mAh g−1 at 0.1 A g−1 [286]. But, it should be acknowledged that the interfacial gaps serve important roles in the improvement of energy-storage abilities. Thus, the interfacial C—Sn and C—S bonds were further constructed for the high energy abilities [201]. For instance, Sun’s group reported SnS2 @rGO samples with the formation of C—S bonds and obviously lowered Warburg impedance coefficient from 89.7 to 56.7 Ω rad1/2 s−1/2 [201]. 4.4.2.4 The Exploring of Sn-based Selenide, Phosphide

Differing from S-anions, the introduction of Se-ions would further induce the increase of conductivity, while introducing P-anions would bring about the remarkable extra capacity. Up to now, Cu-doped SnSe, amorphous SnSe have been explored with high capacity, but displaying inferior cycling stability [202, 287]. Of course, various kinds of carbon have been introduced for alleviating the volume expansion and inhibiting the shuttling effect of polyselenides, accompanied by the control of architecture. SnSe with unique architecture (nanoplates, nanoparticles, microrods,

4.4 Metal/Alloy Anode

CQDs, and so on) were designed in succession through the controlling of atoms growth [288–290]. Moreover, some heterostructures were further explored for their considerable electrochemical properties, including SnSe@Mo@C, SnSe@TiO2 @C, and SnSe@SnO@C [291–293]. Huo’s group reports that SnSe nanoplates were vertically grown on nitrogen-doped carbon (SnSe/NC). From the shifting of Raman peaks (A3 g), the strong Sn—C bonding was confirmed, resulting in high-performance anode in SIBs. Benefiting from unique architectures, the interlayer Na+ diffusion barriers were reduced to 0.1 eV, accompanied by the low energy barrier (0.14 e/uc). Moreover, Sn—C bonds brought about enhancements of conductivity, and the rate capacity could reach up to 88 mAh g−1 at 20.0 A g−1 [294]. For obtaining the higher capacity, P-anions were introduced with Sn cations to form Sn3 P4 samples (theoretical specific capacity, 1132 mAh g−1 ), but they still suffer from serious volume swelling. Supported by ex-situ TEM images, the aggregation of Sn particles was found to fade from 6 to 50 nm during cycling, bringing about the capacity fading [295, 296]. Meanwhile, the detailed reaction mechanism was carried out through first-principles molecular dynamics study, where Sn3 P4 was firstly used to form Na3 P subphase and Na15 Sn4 subphase, then bringing about the construction of core (Na15 Sn4 )/matrix (Na3 P) morphologies [297]. Thus, the designing of architecture was carried out, such as nanoparticles, nanospheres, and so on, accompanying with the complexing of carbon [298–302]. Through coating and reductions, the yolk–shell structures Sn3 P4 @C were prepared with excellent rate properties. Even at ∼4.0 A g−1 , the capacity could remain about 421 mAh g−1 [303]. Through the control of electrolytes (ether and carbonate electrolytes), the typical Sn3 P4 samples showed a capacity of 550 mAh g−1 at 0.05 A g−1 [304]. Of course, considering the synergistic effect of heterojunctions, TiC has been demonstrated to be an effective additive to enhance the cycle stability of Sn4 P3 by suppressing Sn agglomeration during cycling [305].

4.4.3

Bi-based Samples

As the typical members of V groups, B metal shows the high volumetric capacity of 3756 mAh g cm−1 . Similar to the traits of Sb, they were also limited by the volume expansion (∼280%), resulting in the contact loss between active materials and current collectors and finally resulting in the fading of capacity. Up to now, Bi nanoparticles, spheres, and dots were systematically explored. Through facile annealing process, Bi dots (3–18 nm) were in situ embedded in carbon matrix. Interestingly, benefiting from the introduction of KOH, the hosting carbon matrix displayed strong absorption abilities, inducing the suppressing of Bi dost agglomerations. Moreover, the as-obtained samples showed the high ICE (88.13%) and ultra-stable cycle life (10 000 cycles at 2.5 A g−1 ). Using Na3 V2 (PO4 )3 as cathodes, the as-obtained full cell displayed excellent energy density (up to 195 Wh kg−1 ), accompanied by the long-term cycling stabilities (600 loops) [306]. Supported by solution reactions and thermal-reeducation manners, hierarchical Bi nano/submicron spheres were assembled. Incorporating with carbon samples,

143

144

4 Anode Materials of Sodium-ion Batteries

ICE of the optimized samples could be up to 288 mAh g−1 at 0.5 A g−1 . Using the capacity of first cycles as standard, its ratio could be kept at about 95% after 17 300 cycles at 5.0 A g−1 [307]. As the typical modification manners, other heterostructure materials were also introduced for the promotion of their electrochemical properties, such as Bi@1D carbon nanofibers [308], Bi@rGO [309], Bi@hollow carbon [310], Bi@Fe2 O3 [311] Bi@amorphous TiO2 samples [312] and so on. Moreover, from the reports of Bi-based samples in other fields, the relative Bi-metal acid materials were designed, including hollow BiFeO3 @graphene [313], BiPO4 @rGO [314], Bi2 WO6 microflowers [315], and black BiVO4 with reduced particles size and rich oxygen vacancies [316]. Among them, the latter displayed a capacity of >400 mAh g−1 at 0.5 A g−1 after 600 cycles, which mainly ascribed to the broadened energy distribution that induces the enhancement of surface/near-surface redox reactions. But it should be acknowledged that their capacity should be further improved through the introduction of other anions with energy capability. Thus, Bi-based oxides/sulfides/selenides were further explored. For example, the Bi2 O3 nanoflake@carbon film was investigated as free-standing electrodes, delivering a high energy density of 18.94 Wh kg−1 [317]. Meanwhile, Bi2 O3 @CNFs were successfully prepared, and the capacity could be retained at about 504 mAh g−1 at 0.025 A g−1 after 100 cycles. Although series of Bi-based oxides@other samples (BiSb@ Bi2 O3 /SbOx , yolk–shell Bi2 O3 @TiO2 submicrospheres, Bi2 O3 @graphene) were carried out, and the capacity hardly matched the demand of high capacity. As known, when metal-oxides were used as SIBs electrodes, the compact Na2 O film would be formed on the surface of electrodes, inhibiting the insertion of sodium ions [318]. Significantly, about the Na2 S and Na2 Se films, sodium ions could be free to travel in the Na2 S or Na2 Se products. For Bi-based sulfides, considering their serious volume expansion and inferior electrochemical reversibility, the matrix with limited effect was always used as hosting materials, like rod-like Bi2 S3 @C [319], Bi2 S3 @CNT [320], Bi2 S3 -PPy yolk–shell [321], coupled flower-like Bi2 S3 @graphene aerogels [322], nanostructured Bi2 S3 @3D [323], and so on. For example, Dandelion-like Bi2 S3 /rGO hierarchical microspheres were successfully prepared through hydrothermal reactions [324]. Benefiting from unique laminar structure with relatively large interlayer distance, the capacity could be kept at about ∼207 mAh g−1 at 0.1 A g−1 after 1200 cycles. Even at high current density of 10.0 A g−1 , its capacity could remain about 120 mAh g−1 . Of course, in order to improve their inferior reversibility, some metal-based samples were introduced, accompanied with the functions of catalytic effect and improved conductivity, such as Bi2 S3 /Mxene [325], minimal Bi2 S3 /TiO2 [326], heterogeneous structured Bi2 S3 /MoS2 @NC nanoclusters [327], heterostructure VS4 /Bi2 S3 @C nanorods [328], and so on. Through the utilization of ZIF-67 samples as coating layer, the Bi2 S3 @Co9 S8 /NC samples were successfully prepared with long-term cycling stabilities (458 mAh g−1 after 1000 cycles at 1.0 A g−1 ) [329]. From their points, the excellent energy-storage abilities mainly come from the confinement effect of carbon and synergistic effect of Co9 S8 . For Bi-based selenides, series of selenides were carried out, including Bi2 Se3 /C nanocomposite [330], Bi2 Se3 /Mo3 Se4 composite [331], Bi2 Se3 /Bi2 O3 heterostructure [332], and so on. Integrated with carbon, the

4.4 Metal/Alloy Anode

as-prepared Bi2 Se3 /C composite exhibits great electrochemical cycling stabilities and rate properties, delivering an initial capacity of 527 mAh g−1 at 0.1 A g−1 and retaining ∼90% of this capacity after 100 loops [330].

4.4.4

Ge-based Samples

As one of the primary members of IV groups, germanium (Ge) elements have been regarded as promising candidates for high-capacity electrode materials. Similar to the reaction mechanism of Si electrode materials, the multi-electrons reactions were noted for Ge-materials. But, considering its relatively large molecular mass, the obvious reduction in capacity could be found at about 369 mAh g−1 (Ge + Na+ − 2e− → NaGe), but with a large volumetric capacity of 1974 mAh cm−3 . But, compared to Si-materials, Ge-materials displayed higher conductivity, revealing their unique advantages. 4.4.4.1 The Exploring of Ge and the Relative Alloying Materials

Although Ge metal has been widely explored, its relative sodium-storage abilities were carried out in 2016. Firstly, compared to other highly conductive materials (such as Sb and Sn), Ge metal displayed relatively weak conductivity. Thus, through the controlling of surface traits, Sb materials were electroless deposited on the surface of Ge, forming the designed Gex Sb1−x samples, bringing about considerable diffusion behaviors and stable interfacial structure. Assisted by the selection of ionic liquid electrolyte NaFSI-[Py1,4 ]FSI. At 0.83 A g−1 , the first capacity could be kept at about 290 mAh g−1 , and it could remain at about 225 mAh g−1 after 50 loops [333]. And, from the previous reports of Kulova’s groups, Ge0.91 In0.09 samples were synthesized from the aqueous solution by electrolysis, delivering a capacity of 590 mAh g−1 [334]. Further introducing N-doped carbon materials, the Cu3 Ge@carbon nanorods were prepared with the enhanced cycling stability (500 cycles) [335]. As known, owing to the serious volume expansion, series of manners were always used to improve capacity, including the tailoring of particle size and the complexing of other materials. Herein, mesoporous Ge, Ge@Fe2 O3 , and Ge@carbon materials were successfully designed with improved energy-storage materials. In particular, the mesoporous Ge samples showed a high reversible capacity of 803 mAh g−1 at ∼300 mAh g−1 after 100 cycles [336]. More interestingly, with the addition of vinylene carbonate, their cycling stability could be further promoted [337]. 4.4.4.2 The Exploring of Ge-based Oxides Samples

In order to further improve the capacity of Ge-based samples, other anions were introduced. Qian’s groups reported the obtaining of GeOx , which delivered a capacity of 270 mAh g−1 after 1000 cycles at 1.0 A g−1 [338]. Moreover, SnO–GeO2 –Sb2 O3 glass anode was further prepared. Through the control of Ba2+ fractions, their energy-storage abilities were successfully improved [232]. Considering the interesting structure and well-defined electrochemical activities, Mx GeOy (M = Zn, Cu, Cd, Co, Ge, and ln) were successfully prepared. For instance, Zn2 GeO4 micron-rods coated carbon were successfully prepared, displaying a capacity of 150 mAh g−1 at 2.0 A g−1 [339].

145

146

4 Anode Materials of Sodium-ion Batteries

4.4.4.3 The Exploring of Other Ge-based Samples (GeX, X=Se, S, OH, P)

Obviously, compared to Na2 O samples, the relatively high conductivity of other Na2 X could be found, perhaps bringing about considerable electrochemical properties. For instance, Jiang’s groups reported GeSe nanowires with uniform particle sizes. Meanwhile, through the introduction of rods-like structure, the ions would move in the fixed direction, like on expressway. Thus, the optimized samples could be kept at about ∼433.4 mAh g−1 at 200 mA g−1 after 50 cycles with ∼85.3% capacity retention. Meanwhile, through the analysis of samples after cycling, the stable structure was further demonstrated, resulting from unique architecture [340]. Meanwhile, 2D monolayer GeSe and Cu–Ge–Se were further explored with this interesting promotion [341, 342]. But, for Ge-based sulfides, the practical materials were not prepared, and most exploring activities focused on the DFT calculation about the structure traits [343–345]. More significantly, ultrafine Cu2 GeS3 @carbon was explored, but its capacity still needed to be improved [346]. Obviously, benefiting from unique formation traits of double-metal materials, cobalt germanium hydroxide was obtained through hydrothermal method [347]. Profiting from the rich active sites of the active materials, the capacity could be kept at about 416 mAh g−1 after 100 loops at 0.1 A g−1 . However, because the capacity of Ge–Se/S/OH samples hardly met the demand of advanced energy-storage samples, the Ge–P samples were further investigated. Ascribed to the rich valence of Ge elements, GeP, GeP3 , and GeP5 were further obtained. For instance, GeP5 /C composite showed a remarkable capacity (>1200 mAh g−1 ), larger than that of other samples [348]. But, owing to its serious volume expansion, its capacity was hardly kept stable after long-term cycles. Of course, in order to further improve the volumetric capacity, the binder-free Ge–Co–P anode materials were prepared, accompanied by their considerable capacity (425 mAh g−1 ) [349].

References 1 Asher, R. and Wilson, S. (1958). Lamellar compound of sodium with graphite. Nature 181: 409–410. 2 Doeff, M.M., Ma, Y., Visco, S.J., and De Jonghe, L.C. (1993). Electrochemical insertion of sodium into carbon. Journal of The Electrochemical Society 140: L169–L170. 3 Wen, Y., He, K., Zhu, Y. et al. (2014). Expanded graphite as superior anode for sodium-ion batteries. Nature Communications 5: 4033. 4 Liu, Y., Merinov, B.V., and Goddard, W.A. III, (2016). Origin of low sodium capacity in graphite and generally weak substrate binding of Na and Mg among alkali and alkaline earth metals. Proceedings of the National Academy of Sciences 113: 3735–3739. 5 Chen, M.S., Fu, W., Hu, Y. et al. (2020). Controllable growth of carbon nanosheets in the montmorillonite interlayers for high-rate and stable anode in lithium ion battery. Nanoscale 12: 16262–16269.

References

6 Sun, F., Wang, K., Wang, L. et al. (2019). Hierarchical porous carbon sheets with compressed framework and optimized pore configuration for high-rate and long-term sodium and lithium ions storage. Carbon 155: 166–175. 7 Shen, Y., Huang, C., Li, Y. et al. (2021). Enhanced sodium and potassium ions storage of soft carbon by a S/O co-doped strategy. Electrochimica Acta 367. 8 Zhu, Y.Y., Wang, Y.H., Wang, Y.T. et al. (2022). Research progress on carbon materials as negative electrodes in sodium- and potassium-ion batteries. Carbon Energy. 9 Stevens, D.A. and Dahn, J.R. (2000). High capacity anode materials for rechargeable sodium-ion batteries. Journal of the Electrochemical Society 147: 1271–1273. 10 Bommier, C., Surta, T.W., Dolgos, M., and Ji, X. (2015). New mechanistic insights on Na-ion storage in nongraphitizable carbon. Nano Letters 15: 5888–5892. 11 Qiu, S., Xiao, L., Sushko, M.L. et al. (2017). Manipulating adsorption-insertion mechanisms in nanostructured carbon materials for high-efficiency sodium ion storage. Advanced Energy Materials 7. 12 Lu, H., Ai, F., Jia, Y. et al. (2018). Exploring sodium-ion storage mechanism in hard carbons with different microstructure prepared by ball-milling method. Small 14. 13 Morikawa, Y., Nishimura, S.-I., Hashimoto, R.-I. et al. (2020). Mechanism of sodium storage in hard carbon: an X-ray scattering analysis. Advanced Energy Materials 10. 14 Wang, P., Qiao, B., Du, Y. et al. (2015). Fluorine-doped carbon particles derived from lotus petioles as high-performance anode materials for sodium-ion batteries. Journal of Physical Chemistry C 119: 21336–21344. 15 Zhu, Y., Huang, Y., Chen, C. et al. (2019). Phosphorus-doped porous biomass carbon with ultra-stable performance in sodium storage and lithium storage. Electrochimica Acta 321. 16 Fu, L., Tang, K., Song, K. et al. (2014). Nitrogen doped porous carbon fibres as anode materials for sodium ion batteries with excellent rate performance. Nanoscale 6: 1384–1389. 17 Wu, D., Sun, F., Qu, Z. et al. (2022). Multi-scale structure optimization of boron-doped hard carbon nanospheres boosting the plateau capacity for high performance sodium ion batteries. Journal of Materials Chemistry A 10: 17225–17236. 18 Qie, L., Chen, W., Xiong, X. et al. (2015). Sulfur-doped carbon with enlarged interlayer distance as a high-performance anode material for sodium-ion batteries. Advanced Science 2. 19 Wu, F., Dong, R., Bai, Y. et al. (2018). Phosphorus-doped hard carbon nanofibers prepared by electrospinning as an anode in sodium-ion batteries. Acs Applied Materials & Interfaces 10: 21335–21342. 20 Yang, H.-S., Kim, S.-W., Kim, K.-H. et al. (2021). S and P dual-doped carbon nanospheres as anode material for high rate performance sodium-ion batteries. Applied Sciences-Basel 11.

147

148

4 Anode Materials of Sodium-ion Batteries

21 Li, Y., Hu, Y.-S., Titirici, M.-M. et al. (2016). Hard carbon microtubes made from renewable cotton as high-performance anode material for sodium-ion batteries. Advanced Energy Materials 6. 22 Ge, P., Hou, H., Cao, X. et al. (2018). Multidimensional evolution of carbon structures underpinned by temperature-induced intermediate of chloride for sodium-ion batteries. Adv Sci (Weinh) 5: 1800080. 23 Hou, H., Shao, L., Zhang, Y. et al. (2017). Large-area carbon nanosheets doped with phosphorus: a high-performance anode material for sodium-ion batteries. Advanced Science 4. 24 Hou, H., Banks, C.E., Jing, M. et al. (2015). Carbon quantum dots and their derivative 3D porous carbon frameworks for sodium-ion batteries with ultralong cycle life. Advanced Materials 27: 7861–7866. 25 Abou-Rjeily, J., Laziz, N.A., Autret-Lambert, C. et al. (2019). Bituminous coal as low-cost anode materials for sodium-ion and lithium-ion batteries. Energy Technology 7. 26 Li, Y., Xu, S., Wu, X. et al. (2015). Amorphous monodispersed hard carbon micro-spherules derived from biomass as a high performance negative electrode material for sodium-ion batteries. Journal of Materials Chemistry A 3: 71–77. 27 Wu, L., Buchholz, D., Bresser, D. et al. (2014). Anatase TiO2 nanoparticles for high power sodium-ion anodes. Journal of Power Sources 251: 379–385. 28 Chadha, T.S., Dutta, P.K., Raliya, R. et al. (2018). Sustainable one step process for making carbon-free TiO2 anodes and sodium-ion battery electrochemistry. Sustainable Energy & Fuels 2: 1582–1587. 29 Fan, M., Lin, Z., Zhang, P. et al. (2021). Synergistic effect of nitrogen and sulfur dual-doping endows TiO2 with exceptional sodium storage performance. Advanced Energy Materials 11. 30 Gan, Q., He, H., Zhao, K. et al. (2018). Plasma-induced oxygen vacancies in urchin-like anatase titania coated by carbon for excellent sodium-ion battery anodes. Acs Applied Materials & Interfaces 10: 7031–7042. 31 Wang, S., Zhu, Y., Sun, X. et al. (2021). Microwave synthesis of Fe-doped anatase TiO2 /N-doped modified graphene composites with superior sodium storage properties. Diamond and Related Materials 116. 32 Yan, D., Yu, C., Bai, Y. et al. (2015). Sn-doped TiO2 nanotubes as superior anode materials for sodium ion batteries. Chemical Communications 51: 8261–8264. 33 Bayhan, Z., Huang, G., Yin, J. et al. (2021). Two-dimensional TiO2 /TiS2 hybrid nanosheet anodes for high-rate sodium-ion batteries. Acs Applied Energy Materials 4: 8721–8727. 34 Wang, L., Lin, C., Yang, G. et al. (2022). SnO2 nanosheets grown on in-situ formed N-doped branched TiO2 /C nanofibers as binder-free anodes for sodium-ion storage. Electrochimica Acta 411. 35 Zhao, H., Qi, Y., Liang, K. et al. (2021). Interface-driven pseudocapacitance endowing sandwiched CoSe2 /N-doped carbon/TiO2 microcubes with ultra-stable sodium storage and long-term cycling stability. Acs Applied Materials & Interfaces 13: 61555–61564.

References

36 He, H., Sun, D., Zhang, Q. et al. (2017). Iron-doped cauliflower-like rutile TiO2 with superior sodium storage properties. Acs Applied Materials & Interfaces 9: 6093–6103. 37 Huang, J.P., Yuan, D.D., Zhang, H.Z. et al. (2013). Electrochemical sodium storage of TiO2 (B) nanotubes for sodium ion batteries. Rsc Advances 3: 12593–12597. 38 Lee, J., Lee, J.K., Chung, K.Y. et al. (2016). Electrochemical investigations on TiO2 -B nanowires as a promising high capacity anode for sodium-ion batteries. Electrochimica Acta 200: 21–28. 39 Huo, J., Ren, Y., Zhang, G. et al. (2022). Boosting sodium storage of hierarchical nanofibers with porous carbon-supported anatase TiO2 /TiO2 (B) nanowires. Acs Applied Energy Materials 5: 3447–3453. 40 Liao, J.-Y., De Luna, B., and Manthiram, A. (2016). TiO2 -B nanowire arrays coated with layered MoS2 nanosheets for lithium and sodium storage. Journal of Materials Chemistry A 4: 801–806. 41 Yan, Z., Liu, L., Tan, J. et al. (2014). One-pot synthesis of bicrystalline titanium dioxide spheres with a core-shell structure as anode materials for lithium and sodium ion batteries. Journal of Power Sources 269: 37–45. 42 Chaturvedi, A., Edison, E., Arun, N. et al. (2018). Two dimensional TiS2 as a promising insertion anode for Na-ion battery. Chemistryselect 3: 524–528. 43 Huang, C., Liu, Y., Zheng, R. et al. (2022). Interlayer gap widened TiS2 for highly efficient sodium-ion storage. Journal of Materials Science & Technology 107: 64–69. 44 Tao, H., Zhou, M., Wang, R. et al. (2018). TiS2 as an advanced conversion electrode for sodium-ion batteries with ultra-high capacity and long-cycle life. Advanced Science 5. 45 Zhang, D., Zhao, G., Li, P. et al. (2018). Readily exfoliated TiSe2 nanosheets for high-performance sodium storage. Chemistry-a European Journal 24: 1193–1197. 46 Yang, J., Zhang, Y., Zhang, Y. et al. (2017). S-doped TiSe2 nanoplates/Fe3 O4 nanoparticles heterostructure. Small 13. 47 Xu, E., Zhang, J., Liu, Y. et al. (2021). Electron oriented injection TiSe2 -C laminated heterojunctions derived from terminal functionalized MXene for high-rate sodium ion storage. Journal of Materials Chemistry A 9: 27684–27691. 48 Pan, J., Wang, N., Li, L. et al. (2021). Improved Na storage and Coulombic efficiency in TiP2 O7 @C microflowers for sodium ion batteries. Nano Research 14: 139–147. 49 Kim, S.-O. and Manthiram, A. (2016). High-performance red P-based P-TiP2 -C nanocomposite anode for lithium-ion and sodium-ion storage. Chemistry of Materials 28: 5935–5942. 50 Chang, L., Wang, K., Huang, L.-a. et al. (2017). Hierarchical CoO microflower film with excellent electrochemical lithium/sodium storage performance. J. Mater. Chem. A 5: 20892–20902. 51 Chang, L., Wang, K., Huang, L. et al. (2017). Hierarchically porous CoO microsphere films with enhanced lithium/sodium storage properties. Journal of Alloys and Compounds 725: 824–834.

149

150

4 Anode Materials of Sodium-ion Batteries

52 Jiang, J., Ma, C., Ma, T. et al. (2019). A novel CoO hierarchical morphologies on carbon nanofiber for improved reversibility as binder-free anodes in lithium/sodium ion batteries. Journal of Alloys and Compounds 794: 385–395. 53 Wang, Z., Sun, M., Ni, J., and Li, L. (2020). Nature-inspired Cu2 O@CoO tree-like architecture for robust storage of sodium. Journal of Materials Science & Technology 53: 126–131. 54 Guo, C., Zhang, W., Liu, Y. et al. (2019). Constructing CoO/Co3 S4 heterostructures embedded in N-doped carbon frameworks for high-performance sodium-ion batteries. Advanced Functional Materials 29. 55 Li, J., Wang, D., Zhou, J. et al. (2019). Ti-doped ultra-small CoO nanoparticles embedded in an octahedral carbon matrix with enhanced lithium and sodium storage. ChemElectroChem 6: 917–927. 56 Wen, J.-W., Zhang, D.-W., Zang, Y. et al. (2014). Li and Na storage behavior of bowl-like hollow Co3 O4 microspheres as an anode material for lithium-ion and sodium-ion batteries. Electrochimica Acta 132: 193–199. 57 Klavetter, K.C., Garcia, S., Dahal, N. et al. (2014). Li- and Na-reduction products of meso-Co3 O4 form high-rate, stably cycling battery anode materials. J. Mater. Chem. A 2: 14209–14221. 58 Liu, Y., Wan, H., Zhang, H. et al. (2020). Engineering surface structure and defect chemistry of nanoscale cubic Co3 O4 crystallites for enhanced lithium and sodium storage. ACS Applied Nano Materials 3: 3892–3903. 59 Xin, D., Dai, J., Liu, J. et al. (2017). Mesocrystal hexagonal Co3 O4 nanosheets for high performance lithium and sodium-ion batteries. Materials Letters 209: 388–391. 60 Jian, Z., Liu, P., Li, F. et al. (2014). Monodispersed hierarchical Co3 O4 spheres intertwined with carbon nanotubes for use as anode materials in sodium-ion batteries. Journal of Materials Chemistry A 2: 13805. 61 Wang, G., Zhang, M., Deng, Z. et al. (2020). Poplar branch bio-template synthesis of mesoporous hollow Co3 O4 hierarchical architecture as an anode for long-life lithium ion batteries. Ceramics International 46: 29033–29040. 62 Liu, Y., Wang, H., Cheng, L. et al. (2016). TiS2 nanoplates: a high-rate and stable electrode material for sodium ion batteries. Nano Energy 20: 168–175. 63 Wu, X., Wu, W., Wang, K. et al. (2015). Synthesis and electrochemical performance of flower-like MnCo2 O4 as an anode material for sodium ion batteries. Materials Letters 147: 85–87. 64 Chen, J., Ru, Q., Mo, Y. et al. (2016). Design and synthesis of hollow NiCo2 O4 nanoboxes as anodes for lithium-ion and sodium-ion batteries. Physical Chemistry Chemical Physics 18: 18949–18957. 65 Zhang, L. and Zhu, S. (2019). Hollow polyhedral ZnCo2 O4 anode materials for lithium-ion batteries. Materials Letters 236: 337–341. 66 Islam, M., Ali, G., Jeong, M.G. et al. (2019). Unveiling the mechanism of sodium ion storage for needle-shaped Znx Co3-x O4 nanosticks as anode materials. Nanoscale 11: 1065–1073.

References

67 Kong, H., Wu, Y., Hong, W. et al. (2020). Structure-designed synthesis of Cu-doped Co3 O4 @N-doped carbon with interior void space for optimizing alkali-ion storage. Energy Storage Materials 24: 610–617. 68 Sultana, I., Rahman, M.M., Mateti, S. et al. (2017). K-ion and Na-ion storage performances of Co3 O4 -Fe2 O3 nanoparticle-decorated super P carbon black prepared by a ball milling process. Nanoscale 9: 3646–3654. 69 Fang, G., Zhou, J., Cai, Y. et al. (2017). Metal–organic framework-templated two-dimensional hybrid bimetallic metal oxides with enhanced lithium/sodium storage capability. Journal of Materials Chemistry A 5: 13983–13993. 70 Fei, B., Chen, C., Hu, C. et al. (2020). Engineering one-dimensional bunched Ni-MoO2 @Co-CoO-NC composite for enhanced lithium and sodium storage performance. Acs Applied Energy Materials 3: 9018–9027. 71 Zhang, W., Yue, Z., Miao, W. et al. (2018). Carbon-encapsulated tube-wire Co3 O4 /MnO2 heterostructure nanofibers as anode material for sodium-ion batteries. Particle & Particle Systems Characterization 35: 1800138. 72 Li, J.-X., Xie, Q., Zhao, P., and Li, C. (2020). EDTA-Co(II) sodium complex derived Co(OH)2 /Co3 O4 /Co nanoparticles embedded in nitrogen-enriched graphitic porous carbon as lithium-ion battery anode with superior cycling stability. Applied Surface Science 504: 144515. 73 Zhu, X., Jiang, X., Liu, X. et al. (2017). Amorphous CoS nanoparticle/reduced graphene oxide composite as high-performance anode material for sodium-ion batteries. Ceramics International 43: 9630–9635. 74 Jia, H., Dirican, M., Chen, C. et al. (2018). Carbon-coated CoS@rGO anode material with enhanced cyclic stability for sodium storage. Materials Letters 233: 158–161. 75 Zhang, Y., Lv, C., Wang, X. et al. (2018). Boosting sodium-ion storage by encapsulating NiS (CoS) hollow nanoparticles into carbonaceous fibers. ACS applied materials & interfaces 10: 40531–40539. 76 Yao, L., Gu, Q., and Yu, X. (2021). Three-dimensional MOFs@MXene aerogel composite derived MXene threaded hollow carbon confined CoS nanoparticles toward advanced alkali-ion batteries. ACS nano 15: 3228–3240. 77 Wang, Y., Wang, Y., Wang, Y.X. et al. (2019). In situ formation of Co9 S8 nanoclusters in sulfur-doped carbon foam as a sustainable and high-rate sodium-ion anode. ACS applied materials & interfaces 11: 19218–19226. 78 Zhang, Y., Wang, N., Xue, P. et al. (2018). Co9 S8 @carbon nanospheres as high-performance anodes for sodium ion battery. Chemical Engineering Journal 343: 512–519. 79 Lu, S.J., Wang, Z.T., Zhang, X.H. et al. (2020). In situ-formed hollow cobalt sulfide wrapped by reduced graphene oxide as an anode for high-performance lithium-ion batteries. ACS applied materials & interfaces 12: 2671–2678. 80 Guo, Q., Ma, Y., Chen, T. et al. (2017). Cobalt sulfide quantum dot embedded N/S-doped carbon nanosheets with superior reversibility and rate capability for sodium-ion batteries. ACS nano 11: 12658–12667.

151

152

4 Anode Materials of Sodium-ion Batteries

81 Pan, Y., Cheng, X., Gong, L. et al. (2018). Double-morphology CoS2 anchored on N-doped multichannel carbon nanofibers as high-performance anode materials for Na-ion batteries. ACS applied materials & interfaces 10: 31441–31451. 82 Lian, Y., Xin, W., Zhang, M. et al. (2019). Low-content Ni-doped CoS2 embedded within N,P-codoped biomass-derived carbon spheres for enhanced lithium/sodium storage. Journal of Materials Science 54: 8504–8514. 83 Li, X., Pan, D., Deng, J. et al. (2021). Phase-junction engineering boosts the performance of CoSe2 for efficient sodium/potassium storage. Journal of Materials Chemistry A 9: 25954–25963. 84 Liu, X., Xiang, Y., Li, Q. et al. (2021). SnS2 -CoS2 @C nanocubes as high initial coulombic efficiency and long-life anodes for sodium-ion batteries. Electrochimica Acta 387: 138525. 85 Liu, T., Li, Y., Hou, S. et al. (2020). Building hierarchical microcubes composed of one-dimensional CoSe2 @nitrogen-doped carbon for superior sodium ion batteries. Chemistry 26: 13716–13724. 86 Chen, B., Meng, Y., Xie, F. et al. (2018). 1D sub-nanotubes with anatase/bronze TiO2 nanocrystal wall for high-rate and long-life sodium-ion batteries. Advanced Materials 30: 1804116. 87 Liu, B., Cao, J., Li, J. et al. (2021). Highly conductive Co3 Se4 embedded in N-doped 3D interconnected carbonaceous network for enhanced lithium and sodium storage. Journal of colloid and interface science 586: 630–639. 88 Zhang, S., Sun, L., Yu, L. et al. (2021). Core-shell CoSe2 /WSe2 heterostructures@carbon in porous carbon nanosheets as advanced anode for sodium ion batteries. Small 17: e2103005. 89 Xiao, Q., Song, Q., Zheng, K. et al. (2022). CoSe2 nanodots confined in multidimensional porous nanoarchitecture towards efficient sodium ion storage. Nano Energy 98: 107326. 90 Park, S.K. and Kang, Y.C. (2018). MOF-templated N-doped carbon-coated CoSe2 nanorods supported on porous CNT microspheres with excellent sodium-ion storage and electrocatalytic properties. ACS applied materials & interfaces 10: 17203–17213. 91 Wang, B., Chen, K., Wang, G. et al. (2019). A multidimensional and hierarchical carbon-confined cobalt phosphide nanocomposite as an advanced anode for lithium and sodium storage. Nanoscale 11: 968–985. 92 Han, L., Zhang, M., Wang, H. et al. (2020). Electrospun hetero-CoP/FeP embedded in porous carbon nanofibers: enhanced Na(+) kinetics and specific capacity. Nanoscale 12: 24477–24487. 93 Liu, J.-J., Li, J.-L., Guo, S.-M. et al. (2021). Porous CoP@RGO with pseudocapacitance characteristics for lithium ion storage. Scripta Materialia 201: 113983. 94 Zhang, K., Park, M., Zhang, J. et al. (2017). Cobalt phosphide nanoparticles embedded in nitrogen-doped carbon nanosheets: promising anode material with high rate capability and long cycle life for sodium-ion batteries. Nano Research 10: 4337–4350.

References

95 Park, J.-S., Park, G.D., and Kang, Y.C. (2021). Exploration of cobalt selenitecarbon composite porous nanofibers as anode for sodium-ion batteries and unveiling their conversion reaction mechanism. Journal of Materials Science & Technology 89: 24–35. 96 Wang, M., Peng, A., Jiang, J. et al. (2022). Heterointerface synergistic Na+ storage fundamental mechanism for CoSeO3 playing as anode for sodium ion batteries/capacitors. Chemical Engineering Journal 433. 97 Nayak, P.K., Yang, L., Brehm, W., and Adelhelm, P. (2018). From lithium-ion to sodium-ion batteries: advantages, challenges, and surprises. Angew Chem Int Ed Engl 57: 102–120. 98 Li, L., Fu, L., Li, M. et al. (2022). B-doped and La4 NiLiO8 -coated Ni-rich cathode with enhanced structural and interfacial stability for lithium-ion batteries. Journal of Energy Chemistry 71: 588–594. 99 Li, L., Chen, J., Huang, H. et al. (2021). Role of residual Li and oxygen vacancies in Ni-rich cathode materials. ACS Appl Mater Interfaces 13: 42554–42563. 100 Bai, Z., Lv, X., Liu, D.H. et al. (2020). Two-dimensional NiO@C-N nanosheets composite as a superior low-temperature anode material for advanced lithium-/sodium-ion batteries. ChemElectroChem 7: 3616–3622. 101 Cai, Z., Peng, Z., Liu, X. et al. (2021). Improving Na+ transport kinetics and Na+ storage of hierarchical rhenium-nickel sulfide (ReS2 @NiS2 ) hollow architecture by assembling layered 2D-3D heterostructures. Chinese Chemical Letters 32: 3607–3612. 102 Zhao, C., Shen, Z., Tu, F., and Hu, Z. (2019). Template directed hydrothermal synthesis of flowerlike NiSex /C composites as lithium/sodium ion battery anodes. Journal of Materials Science 55: 3495–3506. 103 Zhou, H., Li, Z., Wang, K. et al. (2019). Phase boundary-enhanced Ni3 N–Co3 N@CNT composite materials for lithium-ion batteries. Journal of Materials Chemistry A 7: 1779–1784. 104 Li, Y., Wang, X., Wang, L. et al. (2021). Ni@Ni3 N embedded on three-dimensional carbon nanosheets for high-performance lithium/sodiumsulfur batteries. ACS Appl Mater Interfaces 13: 48536–48545. 105 Wang, X., Kim, H.-M., Xiao, Y., and Sun, Y.-K. (2016). Nanostructured metal phosphide-based materials for electrochemical energy storage. Journal of Materials Chemistry A 4: 14915–14931. 106 Yang, J., Yang, N., Xu, Q. et al. (2019). Bioinspired controlled synthesis of NiSe/Ni2 P nanoparticles decorated 3D porous carbon for Li/Na ion batteries. ACS Sustainable Chemistry & Engineering 7: 13217–13225. 107 Jian, Z., Zhao, B., Liu, P. et al. (2014). Fe2 O3 nanocrystals anchored onto graphene nanosheets as the anode material for low-cost sodium-ion batteries. Chem Commun (Camb) 50: 1215–1217. 108 Qin, G., Duan, J., Yang, Y., and Liu, F. (2018). Magnetic field facilitated resilient chain-like Fe3 O4 /C/Red P with superior sodium storage performance. ACS applied materials & interfaces 10: 6441–6452. 109 Liu, H., Jia, M., Zhu, Q. et al. (2016). 3D-0D graphene-Fe3 O4 quantum dot hybrids as high-performance anode materials for sodium-ion batteries. ACS applied materials & interfaces 8: 26878–26885.

153

154

4 Anode Materials of Sodium-ion Batteries

110 Xiong, P., Zhang, F., Zhang, X. et al. (2021). Atomic-scale regulation of anionic and cationic migration in alkali metal batteries. Nature Communications 12. 111 Wang, N., Liu, Q., Li, Y. et al. (2017). Self-crosslink assisted synthesis of 3D porous branch-like Fe3 O4 /C hybrids for high-performance lithium/sodium-ion batteries. RSC Adv. 7: 50307–50316. 112 Liu, D., Wang, C., Jeong, K.-J., and Lee, J. (2020). Inner-conductivity optimized core-shell Ag@Fe3 O4 nanospheres for high-performance lithium-/sodium-ion batteries. Journal of Alloys and Compounds 832: 152824. 113 Kong, D., Cheng, C., Wang, Y. et al. (2017). Fe3 O4 quantum dot decorated MoS2 nanosheet arrays on graphite paper as free-standing sodium-ion battery anodes. Journal of Materials Chemistry A 5: 9122–9131. 114 Hwang, I., Jung, S.-K., Cho, S.-P., and Kang, K. (2019). In operando formation of new iron-oxyfluoride host structure for Na-ion storage from NaF–FeO nanocomposite. Energy Storage Materials 23: 427–433. 115 Pan, Q., Zheng, F., Ou, X. et al. (2017). MoS2 decorated Fe3 O4 /Fe1–x S@C nanosheets as high-performance anode materials for lithium ion and sodium ion batteries. ACS Sustainable Chemistry & Engineering 5: 4739–4745. 116 Bu, F., Xiao, P., Chen, J. et al. (2018). Rational design of three-dimensional graphene encapsulated core-shell FeS@carbon nanocomposite as a flexible high-performance anode for sodium-ion batteries. Journal of Materials Chemistry A 6: 6414–6421. 117 Wang, B., Zhang, S., Wang, G. et al. (2019). The morphology and electrochemical properties of porous Fe2 O3 @C and FeS@C nanofibers as stable and high-capacity anodes for lithium and sodium storage. Journal of colloid and interface science 557: 216–226. 118 Wang, Q., Zhang, W., Guo, C. et al. (2017). In situ construction of 3D interconnected FeS@Fe3 C@graphitic carbon networks for high-performance sodium-ion batteries. Advanced Functional Materials 27: 1703390. 119 Xu, X., Liu, Z., Ji, S. et al. (2019). Rational synthesis of ternary FeS@TiO2 @C nanotubes as anode for superior Na-ion batteries. Chemical Engineering Journal 359: 765–774. 120 Wang, Y., Wen, Z., Wang, C.C. et al. (2021). MOF-derived Fe7 S8 nanoparticles/ N-doped carbon nanofibers as an ultra-stable anode for sodium-ion batteries. Small 17: e2102349. 121 Chen, L., Han, L., Liu, X. et al. (2021). General synthesis of sulfonate-based metal-organic framework derived composite of Mx Sy @N/S-doped carbon for high-performance lithium/sodium ion batteries. Chemistry 27: 2104–2111. 122 Shadike, Z., Zhou, Y.-N., Ding, F. et al. (2014). The new electrochemical reaction mechanism of Na/FeS2 cell at ambient temperature. Journal of Power Sources 260: 72–76. 123 Walter, M., Zund, T., and Kovalenko, M.V. (2015). Pyrite (FeS2) nanocrystals as inexpensive high-performance lithium-ion cathode and sodium-ion anode materials. Nanoscale 7: 9158–9163. 124 Yin, J., Yang, H., and Wen, Z. (2021). FeS2 @CNT composite with synergistic battery/capacitor builds outstanding sodium-ion storage. Materials Letters 284: 128926.

References

125 Wang, Q., Guo, C., Zhu, Y. et al. (2018). Reduced graphene oxide-wrapped FeS2 composite as anode for high-performance sodium-ion batteries. Nano-micro letters 10: 30. 126 Wu, H., Yuan, R., Li, M. et al. (2022). Co0.85 Se-Fe7 Se8 nanocuboids embedded in reduced graphene oxides as cycle-stable anodes for sodium-ion batteries. Carbon 198: 171–178. 127 Hou, T., Yue, S., Sun, X. et al. (2020). Nitrogen-doped graphene coated FeS2 microsphere composite as high-performance anode materials for sodium-ion batteries enhanced by the chemical and structural synergistic effect. Applied Surface Science 505: 144633. 128 Yuvaraj, S., Veerasubramani, G.K., Park, M.-S. et al. (2020). Facile synthesis of FeS2 /MoS2 composite intertwined on rGO nanosheets as a high-performance anode material for sodium-ion battery. Journal of Alloys and Compounds 821: 153222. 129 Lu, Z., Wang, W., Zhou, J., and Bai, Z. (2020). FeS2 @TiO2 nanorods as high-performance anode for sodium ion battery. Chinese Journal of Chemical Engineering 28: 2699–2706. 130 Cao, D., Kang, W., Wang, W. et al. (2020). Okra-like Fe7 S8 /C@ZnS/N-C@C with core-double-shelled structures as robust and high-rate sodium anode. Small 16: e1907641. 131 Wang, Z.-Q., Zeng, B., Zhou, D. et al. (2022). Rich-oxygen-doped FeSe2 nanosheets with high pseudocapacitance capacity as a highly stable anode for sodium ion battery. Chemical Engineering Journal 428: 132637. 132 Yuan, J., Liu, W., Zhang, X. et al. (2020). MOF derived ZnSe–FeSe2 /RGO nanocomposites with enhanced sodium/potassium storage. Journal of Power Sources 455: 227937. 133 Ge, P., Hou, H., Li, S. et al. (2018). Tailoring rod-like FeSe2 coated with nitrogen-doped carbon for high-performance sodium storage. Advanced Functional Materials 28: 1801765. 134 Yang, Q.-R., Li, W.-J., Chou, S.-L. et al. (2015). Ball-milled FeP/graphite as a low-cost anode material for the sodium-ion battery. RSC advances 5: 80536–80541. 135 Shi, S., Sun, C., Yin, X. et al. (2020). FeP quantum dots confined in carbon-nanotube-grafted P-doped carbon octahedra for high-rate sodium storage and full-cell applications. Advanced Functional Materials 30. 136 Gao, L., Liu, Y., Hao, Y. et al. (2022). Self supported selenium doped Fe2 P hierarchical microsphere arrays enabling high performance sodium ion batteries. Journal of Alloys and Compounds 918: 165647. 137 Ding, Y., Chen, Y., Xu, N. et al. (2020). Facile synthesis of FePS3 nanosheets@MXene composite as a high-performance anode material for sodium storage. Nano-micro letters 12: 54. 138 Chen, G., Gao, L., Zhang, L., and Yang, X. (2021). Heterogeneous Fe-Ni-P nanosheet arrays as a potential anode for sodium ion batteries. Journal of Electroanalytical Chemistry 895: 115420. 139 Huang, S., Ye, M., Zhang, Y. et al. (2022). Achieving ultrahigh-rate and low-temperature sodium storage of FePS3 via in situ construction of graphitized porous N-doped carbon. ACS applied materials & interfaces 14: 42048–42056.

155

156

4 Anode Materials of Sodium-ion Batteries

140 Ma, Y., Lian, X., Xu, N. et al. (2022). Rational design of few-layer FePS3 nanosheets@N-doped carbon composites as anodes for sodium-ion batteries. Chemical Engineering Journal 427: 130882. 141 Xing, S., Yang, J., Wang, C. et al. (2020). Fabrication of van der Waals heterostructured FePSe3 /carbon hybrid nanosheets for sodium storage with high performance. ACS applied materials & interfaces 12: 54732–54741. 142 Ma, X., Xiong, X., Zeng, J. et al. (2020). Melamine-assisted synthesis of Fe3 N featuring highly reversible crystalline-phase transformation for ultrastable sodium ion storage. Journal of Materials Chemistry A 8: 6768–6775. 143 Li, N.N., Sheng, Z.M., Huang, H. et al. (2019). Fe2 O3 nanoparticles encapsulated with N-doped porous graphitic shells approached by oxidizing Fe3 C@C precursor for high performance sodium-ion batteries. Journal of Alloys and Compounds 792: 25–31. 144 Sreedhara, M.B., Santhosha, A.L., Bhattacharyya, A.J., and Rao, C.N.R. (2016). Composite of few-layer MoO3 nanosheets with graphene as a high performance anode for sodium-ion batteries. Journal of Materials Chemistry A 4: 9466–9471. 145 Xu, Y., Zhou, M., Wang, X. et al. (2015). Enhancement of sodium ion battery performance enabled by oxygen vacancies. Angewandte Chemie 54: 8768–8771. 146 Wang, B., Ang, E.H., Yang, Y. et al. (2020). Interlayer engineering of molybdenum trioxide toward high-capacity and stable sodium ion half/full batteries. Advanced Functional Materials 30: 2001708. 147 Dong, Z., Wu, X., Cai, D.-K. et al. (2022). Interlayer-expanded MoS2 @C hollow nanorods for enhanced sodium storage. Chemical Engineering Science 262: 117976. 148 Zhu, C., Mu, X., van Aken, P.A. et al. (2014). Single-layered ultrasmall nanoplates of MoS2 embedded in carbon nanofibers with excellent electrochemical performance for lithium and sodium storage. Angewandte Chemie-International Edition 53: 2152–2156. 149 Bai, J., Zhao, B., Wang, X. et al. (2020). Yarn ball-like MoS2 nanospheres coated by nitrogen-doped carbon for enhanced lithium and sodium storage performance. Journal of Power Sources 465: 228282. 150 Bai, L. and Liang, F. (2022). Hierarchical MoS2 /carbon composites as superior anode for advanced sodium-ion battery. Ionics 28: 3341–3345. 151 Chang, K., Chen, W., Ma, L. et al. (2011). Graphene-like MoS2 /amorphous carbon composites with high capacity and excellent stability as anode materials for lithium ion batteries. Journal of Materials Chemistry 21: 6251. 152 Chen, Z., Chen, S., Zhang, H. et al. (2019). Exfoliated MoS2 @C nanosheets as anode for sodium/potassium storage. Ionics 26: 1779–1786. 153 Choi, S.H., Ko, Y.N., Lee, J.-K., and Kang, Y.C. (2015). 3D MoS2 -graphene microspheres consisting of multiple nanospheres with superior sodium ion storage properties. Advanced Functional Materials 25: 1780–1788. 154 Han, M., Lin, Z., and Yu, J. (2019). Ultrathin MoS2 nanosheets homogenously embedded in a N,O-codoped carbon matrix for high-performance lithium and sodium storage. Journal of Materials Chemistry A 7: 4804–4812.

References

155 Yan, J., Li, Q., Hao, Y. et al. (2020). MoS2 /SnS2 nanocomposite as stable sodium-ion battery anode. Functional Materials Letters 13. 156 Shi, N., Xi, B., Huang, M. et al. (2020). Hierarchical octahedra constructed by Cu2 S/MoS2 subset of carbon framework with enhanced sodium storage. Small 16. 157 Zhou, Y., Zhang, M., Han, Q. et al. (2022). Hierarchical 1 T-MoS2 /MoOx @NC microspheres as advanced anode materials for potassium/sodium-ion batteries. Chemical Engineering Journal 428. 158 Ma, K., Dong, Y., Jiang, H. et al. (2021). Densified MoS2 /Ti3 C2 films with balanced porosity for ultrahigh volumetric capacity sodium-ion battery. Chemical Engineering Journal 413. 159 Ma, J., Xing, M., Yin, L. et al. (2021). Porous hierarchical TiO2 /MoS2 /RGO nanoflowers as anode material for sodium ion batteries with high capacity and stability. Applied Surface Science 536. 160 Gui, D., Wei, Z., Chen, J. et al. (2021). Boosting the sodium storage of the 1T/2H MoS2 @SnO2 heterostructure via a fast surface redox reaction. Journal of Materials Chemistry A 9: 463–471. 161 Ge, P., Hou, H., Banks, C.E. et al. (2018). Binding MoSe2 with carbon constrained in carbonous nanosphere towards high-capacity and ultrafast Li/Na-ion storage. Energy Storage Materials 12: 310–323. 162 Zhang, Y., Deng, S., Shen, Y. et al. (2020). Construction of 1T-MoSe2 /TiC@C branch-core arrays as advanced anodes for enhanced sodium ion storage. Chemsuschem 13: 1575–1581. 163 Zeng, L., Fang, Y., Xu, L. et al. (2019). Rational design of few-layer MoSe2 confined within ZnSe-C hollow porous spheres for high-performance lithium-ion and sodium-ion batteries. Nanoscale 11: 6766–6775. 164 Ma, L., Wang, X., Zhao, B. et al. (2018). Novel metal chalcogenide supported on three-dimensional graphene foam for enhanced lithium storage. Journal of Alloys and Compounds 762: 149–156. 165 Fan, H.-N., Zhang, Q., Gu, Q.-F. et al. (2019). Exploration of the sodium ion ordered transfer mechanism in a MoSe2 @Graphene composite for superior rate and lifespan performance. Journal of Materials Chemistry A 7: 13736–13742. 166 Yin, Y., Fan, L., Zhang, Y. et al. (2019). MoP hollow nanospheres encapsulated in 3D reduced graphene oxide networks as high rate and ultralong cycle performance anodes for sodium-ion batteries. Nanoscale 11: 7129–7134. 167 Yin, Y., Zhang, Y., Liu, N. et al. (2020). Metal-organic frameworks-derived porous yolk-shell MoP/Cu3 P@carbon microcages as high-performance anodes for sodium-ion batteries. Energy & Environmental Materials 3: 529–534. 168 Asayesh-Ardakani, H., Yao, W., Yuan, Y. et al. (2017). In situ TEM investigation of ZnO nanowires during sodiation and lithiation cycling. Small Methods 1: 1700202. 169 Li, J., Zheng, Y., Bao, X. et al. (2022). Ultrasmall ZnO nanocrystals confined in honeycombed N-doped carbon for high-performance and stable lithium/sodium ion batteries. Energy Technology 10: 2200446.

157

158

4 Anode Materials of Sodium-ion Batteries

170 Jing, M., Li, F., Chen, M. et al. (2019). Binding ZnO nanorods in reduced graphene oxide via facile electrochemical method for Na-ion battery. Applied Surface Science 463: 986–993. 171 Teng, Y., Mo, M., Li, Y. et al. (2018). Amorphous carbon-coated ZnO porous nanosheets: facile fabrication and application in lithium- and sodium-ion batteries. Journal of Alloys and Compounds 744: 712–720. 172 Zhang, X., Gao, X., Li, D. et al. (2020). Flower-like NiO/ZnO hybrid coated with N-doped carbon layer derived from metal-organic hybrid frameworks as novel anode material for high performance sodium-ion batteries. Journal of Colloid and Interface Science 563: 354–362. 173 Li, J., Fu, Y., Shi, X. et al. (2017). Urchinlike ZnS microspheres decorated with nitrogen-doped carbon: a superior anode material for lithium and sodium storage. Chemistry-a European Journal 23: 157–166. 174 Qin, W., Li, D., Zhang, X. et al. (2016). ZnS nanoparticles embedded in reduced graphene oxide as high performance anode material of sodium-ion batteries. Electrochimica Acta 191: 435–443. 175 Jing, M., Chen, Z., Li, Z. et al. (2018). Facile synthesis of ZnS/N,S Co-doped carbon composite from zinc metal complex for high-performance sodium-ion batteries. Acs Applied Materials & Interfaces 10: 704–712. 176 Zhang, R., Wang, Y., Jia, M. et al. (2018). One-pot hydrothermal synthesis of ZnS quantum dots/graphene hybrids as a dual anode for sodium ion and lithium ion batteries. Applied Surface Science 437: 375–383. 177 Su, D., Kretschmer, K., and Wang, G. (2016). Improved electrochemical performance of Na-ion batteries in ether-based electrolytes: a case study of ZnS nanospheres. Advanced Energy Materials 6. 178 Zhou, P., Zhang, M., Wang, L. et al. (2019). Synthesis and electrochemical performance of ZnSe electrospinning nanofibers as an anode material for lithium ion and sodium ion batteries. Frontiers in Chemistry 7. 179 Dong, C., Wu, L., He, Y. et al. (2020). Willow-leaf-like ZnSe@N-doped carbon nanoarchitecture as a stable and high-performance anode material for sodium-ion and potassium-ion batteries. Small 16. 180 Wang, X., Zhao, W., Zhang, W. et al. (2021). Ultrafine ZnSe encapsulated in nitrogen-doped porous carbon nanofibers for superior Na-ion batteries with a long lifespan and low-temperature performance. Acs Sustainable Chemistry & Engineering 9: 11705–11713. 181 Xie, X., Yin, Z., Li, Y. et al. (2022). Zn-O-C bonds for efficient electron/ion bridging in ZnSe/C composites boosting the sodium-ion storage. Journal of Materials Chemistry A 10: 3732–3742. 182 Li, Y., Zhang, M., Qian, J. et al. (2019). Freestanding N-doped carbon coated CuO array anode for lithium-ion and sodium-ion batteries. Energy Technology 7. 183 Lu, Y., Zhang, N., Zhao, Q. et al. (2015). Micro-nanostructured CuO/C spheres as high-performance anode materials for Na-ion batteries. Nanoscale 7: 2770–2776. 184 Fang, Y., Guan, B.Y., Luan, D., and Lou, X.W. (2019). Synthesis of CuS@CoS2 double-shelled nanoboxes with enhanced sodium storage properties. Angewandte Chemie-International Edition 58: 7739–7743.

References

185 Zhai, X., Zuo, Z., Xiong, Z. et al. (2022). Large-scale CuS nanotube arrays@graphdiyne for high-performance sodium ion battery. 2d Materials 9. 186 Zhao, D., Yin, M., Feng, C. et al. (2020). Rational design of N-doped CuS@C nanowires toward high-performance half/full sodium-ion batteries. Acs Sustainable Chemistry & Engineering 8: 11317–11327. 187 Xiao, Y., Yue, F., Wen, Z. et al. (2022). Elastic buffering layer on CuS enabling high-rate and long-life sodium-ion storage. Nano-Micro Letters 14. 188 Lin, H., Li, M., Yang, X. et al. (2019). Nanosheets-assembled CuSe crystal pillar as a stable and high-power anode for sodium-ion and potassium-ion batteries. Advanced Energy Materials 9. 189 Yu, D., Wei, X., Zhao, D. et al. (2022). Ultrathin CuSe nanosheets as the anode for sodium ion battery with high rate performance and long cycle life. Electrochimica Acta 404. 190 Li, Z., Xiao, S., Wang, Y. et al. (2022). Fast and stable Na insertion/deinsertion in double-shell hollow MnO nanospheres. Journal of Alloys and Compounds 920. 191 Li, H., Liu, A., Zhao, S. et al. (2018). In situ growth of a feather-like MnO2 nanostructure on carbon paper for high-performance rechargeable sodium-ion batteries. Chemelectrochem 5: 3266–3272. 192 Zheng, S., Li, D., Li, W. et al. (2022). MnO(2) nanosheets on a carbon nanofiber freestanding film by electrospinning and in situ spraying for lithium and sodium storage. Acs Applied Energy Materials 5: 3587–3594. 193 Lan, D., Zhao, Y., Liu, Y. et al. (2022). Nano-MnS@ N doped lignite derived carbon composites as superior anode material for sodium-ion batteries. Journal of Alloys and Compounds 912. 194 Gao, X., Zhang, X., Jiang, J., and Chen, J. (2018). Rod-like carbon-coated MnS derived from metal-organic frameworks as high-performance anode material for sodium-ion batteries. Materials Letters 228: 42–45. 195 Liu, Q., Zhang, S.-J., Xiang, C.-C. et al. (2020). Cubic MnS-FeS2 composites derived from a prussian blue analogue as anode materials for sodium-ion batteries with long-term cycle stability. Acs Applied Materials & Interfaces 12: 43624–43633. 196 Zhu, J., Wei, P., Zeng, Q. et al. (2020). MnS@N,S Co-doped carbon core/shell nanocubes: sulfur-bridged bonds enhanced Na-storage properties revealed by in situ Raman spectroscopy and transmission electron microscopy. Small 16. 197 Hartmann, F., Liers, L., Etter, M. et al. (2022). Is there a common reaction pathway for chromium sulfides as anodes in sodium-ion batteries? A case study about sodium storage properties of MCr2 S4 (M = Cr, Ti, Fe). Journal of Solid State Electrochemistry 26: 2501–2514. 198 Ali, G., Akbar, M., and Chung, K.Y. (2022). Electrochemical investigations of a high-capacity Na2 CrO4 /C nanocomposite anode for sodium-ion batteries. International Journal of Energy Research 46: 1803–1812. 199 Dong, X., Liu, Y., Zhu, S. et al. (2022). Architecting hierarchical WO3 agglomerates assembled with straight and parallel aligned nanoribbons enabling high capacity and robust stability of lithium storage. Frontiers in Chemistry 9.

159

160

4 Anode Materials of Sodium-ion Batteries

200 Liu, Y., Zhang, N., Kang, H. et al. (2015). WS2 nanowires as a high-performance anode for sodium-ion batteries. Chemistry-a European Journal 21: 11878–11884. 201 Liu, W., Wei, M., Ji, L. et al. (2020). Hollow carbon sphere based WS2 anode for high performance lithium and sodium ion batteries. Chemical Physics Letters 741. 202 Wu, H., Chen, X., Qian, C. et al. (2020). Confinement growth of layered WS2 in hollow beaded carbon nanofibers with synergistic anchoring effect to reinforce Li+/Na+ storage performance. Small 16. 203 Kang, B., Wang, Y., He, X. et al. (2021). Facile fabrication of WS2 nanocrystals confined in chlorella-derived N, P co-doped bio-carbon for sodium-ion batteries with ultra-long lifespan. Dalton Transactions 50: 14745–14752. 204 Li, J., Han, S., Zhang, J. et al. (2019). Synthesis of three-dimensional free-standing WSe2 /C hybrid nanofibers as anodes for high-capacity lithium/sodium ion batteries. Journal of Materials Chemistry A 7: 19898–19908. 205 Kang, B., Chen, X., Zeng, L. et al. (2020). In situ fabrication of ultrathin few-layered WSe2 anchored on N, P dual-doped carbon by bioreactor for half/full sodium/potassium-ion batteries with ultralong cycling lifespan. Journal of Colloid and Interface Science 574: 217–228. 206 He, G., Li, L., and Manthiram, A. (2015). VO2/rGO nanorods as a potential anode for sodium- and lithium-ion batteries. Journal of Materials Chemistry A 3: 14750–14758. 207 Zhao, L., Wei, Q., Huang, Y. et al. (2019). Pseudocapacitive graphene-wrapped porous VO2 microspheres for ultrastable and ultrahigh-rate sodium-ion storage. Chemelectrochem 6: 1400–1406. 208 Wu, F., Jiang, Y., Ye, Z. et al. (2019). A 3D flower-like VO2 /MXene hybrid architecture with superior anode performance for sodium ion batteries. Journal of Materials Chemistry A 7: 1315–1322. 209 Chao, D., DeBlock, R., Lai, C.-H. et al. (2021). Amorphous VO2 : a pseudocapacitive platform for high-rate symmetric batteries. Advanced Materials 33. 210 Putungan, D.B., Lin, S.-H., and Kuo, J.-L. (2016). Metallic VS2 monolayer polytypes as potential sodium-ion battery anode via ab initio random structure searching. ACS Applied Materials & Interfaces 8: 18754–18762. 211 Li, W., Huang, J., Feng, L. et al. (2018). VS2 nanoarchitectures assembled by single-crystal nanosheets for enhanced sodium storage properties. Electrochimica Acta 286: 131–138. 212 Chen, J., Tang, Z., Pan, Z. et al. (2020). Template-free growth of spherical vanadium disulfide nanoflowers as efficient anodes for sodium/potassium ion batteries. Materials & Design 192. 213 Fan, R., Zhao, C., Ma, J. et al. (2022). Boosting reaction kinetics and improving long cycle life in lamellar VS2 /MoS2 heterojunctions for superior sodium storage performance. Journal of Materials Chemistry A 10: 939–949. 214 Yi, Y., Du, X., Zhao, Z. et al. (2022). Coupling of metallic VSe2 and conductive polypyrrole for boosted sodium-ion storage by reinforced conductivity within and outside. Acs Nano 16: 7772–7782.

References

215 Kim, H., Lim, E., Jo, C. et al. (2015). Ordered-mesoporous Nb2O5/carbon composite as a sodium insertion material. Nano Energy 16: 62–70. 216 Yang, L., Zhu, Y.-E., Sheng, J. et al. (2017). T-Nb2 O5 /C nanofibers prepared through electrospinning with prolonged cycle durability for high-rate sodium-ion batteries induced by pseudocapacitance. Small 13. 217 Tong, Z., Yang, R., Wu, S. et al. (2019). Surface-engineered black niobium oxide@graphene nanosheets for high-performance sodium-/potassium-ion full batteries. Small 15. 218 Huang, Y., Wang, Z., Jiang, Y. et al. (2018). Conductivity and pseudocapacitance optimization of bimetallic antimony-indium sulfide anodes for sodium-ion batteries with favorable kinetics. Advanced Science 5. 219 Joshi, A.C., Halankar, K.K., Dutta, D.P., and Ravuri, B.R. (2022). Improved electrochemical properties of In2 S3 /g-C3 N4 nanocomposite for application as anodes in lithium and sodium ion batteries. Materials Letters 320. 220 Luo, W., Ren, J., Feng, W. et al. (2021). Engineering nanostructured antimony-based anode materials for sodium ion batteries. Coatings 11: 1233. 221 Liu, Z., Song, T., and Paik, U. (2018). Sb-based electrode materials for rechargeable batteries. Journal of Materials Chemistry A 6: 8159–8193. 222 Hu, L.-N., Li, X.-Z., Lv, Z.-C. et al. (2022). Design of Sb2 Se3 -based nanocomposites for high-performance alkali metal ion batteries driven by a hybrid charge storage mechanism. Chemical Engineering Journal 440: 135971. 223 Hou, H., Jing, M., Yang, Y. et al. (2015). Sb porous hollow microspheres as advanced anode materials for sodium-ion batteries. Journal of Materials Chemistry A 3: 2971–2977. 224 Hou, H., Jing, M., Zhang, Y. et al. (2015). Cypress leaf-like Sb as anode material for high-performance sodium-ion batteries. Journal of Materials Chemistry A 3: 17549–17552. 225 Liu, Y., Zhou, B., Liu, S. et al. (2019). Galvanic replacement synthesis of highly uniform Sb nanotubes: reaction mechanism and enhanced sodium storage performance. ACS nano 13: 5885–5892. 226 Wu, P., Zhang, A., Peng, L. et al. (2018). Cyanogel-enabled homogeneous Sb-Ni-C ternary framework electrodes for enhanced sodium storage. ACS nano 12: 759–767. 227 Li, Q., Zhang, W., Peng, J. et al. (2021). Metal-organic framework derived ultrafine Sb@Porous carbon octahedron via in situ substitution for high-performance sodium-ion batteries. ACS nano 15: 15104–15113. 228 Li, C., Wei, G., Wang, S. et al. (2018). Two-dimensional coupling: Sb nanoplates embedded in MoS2 nanosheets as efficient anode for advanced sodium ion batteries. Materials Chemistry and Physics 211: 375–381. 229 Wang, N., Bai, Z., Qian, Y., and Yang, J. (2016). Double-walled Sb@TiO2-x nanotubes as a superior high-rate and ultralong-lifespan anode material for Na-ion and Li-ion batteries. Advanced Materials 28: 4126–4133. 230 Ma, W., Wang, J., Gao, H. et al. (2018). A mesoporous antimony-based nanocomposite for advanced sodium ion batteries. Energy Storage Materials 13: 247–256.

161

162

4 Anode Materials of Sodium-ion Batteries

231 Wang, Z., Zeng, F., Zhang, D. et al. (2021). Equably-dispersed Sb/Sb2 O3 nanoparticles in ionic liquid-derived nitrogen-enriched carbon for highly reversible lithium/sodium storage. Electrochimica Acta 395: 139210. 232 Gandi, S., Chinta, S.R., Ghoshal, P., and Ravuri, B.R. (2019). SnO-GeO2 -Sb2 O3 glass anode network mixed with different Ba2+ fractions: investigations on Na-ion storage capacity and stability. Journal of Non-Crystalline Solids 506: 80–87. 233 Lakshmi, K.P., Janas, K.J., and Shaijumon, M.M. (2018). Antimony oxychloride/ graphene aerogel composite as anode material for sodium and lithium ion batteries. Carbon 131: 86–93. 234 Lin, Y., Feng, W., Li, Z. et al. (2017). Facile synthesis of phase-pure Sb8 O11 Cl2 microrods as anode materials for sodium-ion batteries with high capacity. Ionics 23: 3197–3202. 235 Hou, H., Jing, M., Huang, Z. et al. (2015). One-dimensional rod-like Sb2 S3 -based anode for high-performance sodium-ion batteries. ACS applied materials & interfaces 7: 19362–19369. 236 Ge, P., Cao, X., Hou, H. et al. (2017). Rodlike Sb2 Se3 wrapped with carbon: the exploring of electrochemical properties in sodium-ion batteries. ACS applied materials & interfaces 9: 34979–34989. 237 Zhao, Y. and Manthiram, A. (2015). Amorphous Sb2 S3 embedded in graphite: a high-rate, long-life anode material for sodium-ion batteries. Chem Commun (Camb) 51: 13205–13208. 238 Shi, Y., Li, F., Zhang, Y. et al. (2019). Sb(2)S(3)@PPy coaxial nanorods: a versatile and robust host material for reversible storage of alkali metal ions. Nanomaterials 9. 239 Wen, S., Zhao, J., Zhao, Y. et al. (2019). Reduced graphene oxide (RGO) decorated Sb2 S3 nanorods as anode material for sodium-ion batteries. Chemical Physics Letters 716: 171–176. 240 Choi, J.-H., Ha, C.-W., Choi, H.-Y. et al. (2016). Sb2 S3 embedded in amorphous P/C composite matrix as high-performance anode material for sodium ion batteries. Electrochimica Acta 210: 588–595. 241 Zhou, X., Zhang, Z., Yan, P. et al. (2020). Sulfur-doped reduced graphene oxide/Sb2 S3 composite for superior lithium and sodium storage. Materials Chemistry and Physics 244: 122661. 242 Peng, Q., Fu, J., Chen, G. et al. (2022). Boost the cyclability and Na+ diffusion kinetics of Sb2 S3 anode by CNTs cross-linking N-doped carbon matrix. Journal of Alloys and Compounds 898: 162855. 243 Li, W., Zhou, M., Li, H. et al. (2015). Carbon-coated Sb2 Se3 composite as anode material for sodium ion batteries. Electrochemistry Communications 60: 74–77. 244 Zhao, W. and Li, C.M. (2017). Mesh-structured N-doped graphene@Sb2 Se3 hybrids as an anode for large capacity sodium-ion batteries. Journal of colloid and interface science 488: 356–364. 245 Nam, K.-H. and Park, C.-M. (2019). 2D layered Sb2 Se3 -based amorphous composite for high-performance Li- and Na-ion battery anodes. Journal of Power Sources 433: 126639.

References

246 Guo, L., Cao, L., Huang, J. et al. (2019). Design of an ultra-stable Sb2 Se3 anode with excellent Na storage performance. Journal of Alloys and Compounds 810: 151930. 247 Man, Q., Hou, Q., Liu, P. et al. (2018). Cube-like Sb2 Se3 /C constructed by ultrathin nanosheets as anode material for lithium and sodium-ion batteries. Ionics 25: 1551–1558. 248 Xie, F., Zhang, L., Gu, Q. et al. (2019). Multi-shell hollow structured Sb2 S3 for sodium-ion batteries with enhanced energy density. Nano Energy 60: 591–599. 249 Yang, J., Wang, T., Guo, X. et al. (2021). Flexible sodium-ion capacitors boosted by high electrochemically-reactive and structurally-stable Sb2 S3 nanowire/Ti3 C2 Tx MXene film anodes. Nano Research. 250 Li, K., Liu, X., Qin, Y. et al. (2021). Sb2 S3 -Bi2 S3 microrods with the combined action of carbon encapsulation and rGO confinement for improving high cycle stability in sodium/potassium storage. Chemical Engineering Journal 414: 128787. 251 Wang, D., Cao, L., Luo, D. et al. (2021). Chain mail heterostructured hydrangea-like binary metal sulfides for high efficiency sodium ion battery. Nano Energy 87: 106185. 252 Jia, R., Li, L., Shen, G., and Chen, D. (2022). Hierarchical Sb2 S3 /SnS2 /C heterostructure with improved performance for sodium-ion batteries. Science China Materials 65: 1443–1452. 253 Kim, D.S., Bae, J., Kwon, S.H. et al. (2020). Synergistic effect of antimony-triselenide on addition of conductive hybrid matrix for high-performance lithium-ion batteries. Journal of Alloys and Compounds 828: 154410. 254 Ge, P., Zhang, L., Zhao, W. et al. (2020). Interfacial bonding of metal-sulfides with double carbon for improving reversibility of advanced alkali-ion batteries. Advanced Functional Materials 30. 255 Stratford, J.M., Mayo, M., Allan, P.K. et al. (2017). Investigating sodium storage mechanisms in tin anodes: a combined pair distribution function analysis, density functional theory, and solid-state NMR approach. Journal of the American Chemical Society 139: 7273–7286. 256 Wang, J.W., Liu, X.H., Mao, S.X., and Huang, J.Y. (2012). Microstructural evolution of tin nanoparticles during in situ sodium insertion and extraction. Nano letters 12: 5897–5902. 257 Choi, Y.S., Byeon, Y.W., Park, J.H. et al. (2018). Ultrafast sodiation of single-crystalline Sn anodes. ACS applied materials & interfaces 10: 560–568. 258 Zhang, B., Rousse, G., Foix, D. et al. (2016). Microsized Sn as advanced anodes in glyme-based electrolyte for Na-ion batteries. Advanced Materials 28: 9824-+. 259 Wang, L., Ni, Y., Lei, K. et al. (2018). 3D porous tin created by tuning the redox potential acts as an advanced electrode for sodium-ion batteries. Chemsuschem 11: 3376–3381. 260 Ma, J. and Prieto, A.L. (2019). Electrodeposition of pure phase SnSb exhibiting high stability as a sodium-ion battery anode. Chemical Communications 55: 6938–6941.

163

164

4 Anode Materials of Sodium-ion Batteries

261 Qin, J., Wang, T., Liu, D. et al. (2018). A top-down strategy toward SnSb in-plane nanoconfined 3D N-doped porous graphene composite microspheres for high performance Na-ion battery anode. Advanced Materials 30. 262 Liu, Y., Zhang, N., Jiao, L. et al. (2015). Ultrasmall Sn nanoparticles embedded in carbon as high-performance anode for sodium-ion batteries. Advanced Functional Materials 25: 214–220. 263 Liu, Z., Zhang, S., Qiu, Z. et al. (2020). Tin nanodots derived from Sn2+ /graphene quantum dot complex as pillars into graphene blocks for ultrafast and ultrastable sodium-ion storage. Small 16. 264 Li, Y., Ou, C., Zhu, J. et al. (2020). Ultrahigh and durable volumetric lithium/sodium storage enabled by a highly dense graphene-encapsulated nitrogen-doped carbon@Sn compact monolith. Nano letters 20: 2034–2046. 265 He, W., Chen, K., Pathak, R. et al. (2021). High-mass-loading Sn-based anode boosted by pseudocapacitance for long-life sodium-ion batteries. Chemical Engineering Journal 414. 266 Sato, F., Honma, T., Komatsu, T. et al. (2022). Formation of highly dispersed tin nanoparticles in amorphous silicates for sodium ion battery anode. Journal of Physics and Chemistry of Solids 161: 110377. 267 Ja´skaniec, S., Kavanagh, S.R., Coelho, J. et al. (2021). Solvent engineered synthesis of layered SnO for high-performance anodes. npj 2D Materials and Applications 5. 268 Kim, H., Sadan, M.K., Kim, C. et al. (2021). Electrochemical properties of flexible anode with SNO2 nanopowder for sodium-ion batteries. Archives of Metallurgy and Materials 66: 931–934. 269 Wang, J., Zhu, G., Liu, X. et al. (2019). In situ synthesis of tin dioxide submicrorods anchored on nickel foam as an additive-free anode for high performance sodium-ion batteries. Journal of colloid and interface science 533: 733–741. 270 He, F., Xu, Q., Zheng, B. et al. (2020). Synthesis of hierarchical Sn/SnO nanosheets assembled by carbon-coated hollow nanospheres as anode materials for lithium/sodium ion batteries. RSC advances 10: 6035–6042. 271 Wang, Y., Jin, Y., Zhao, C. et al. (2018). 1D ultrafine SnO2 nanorods anchored on 3D graphene aerogels with hierarchical porous structures for high-performance lithium/sodium storage. Journal of colloid and interface science 532: 352–362. 272 Yuan, J., Hao, Y., Zhang, X., and Li, X. (2018). Sandwiched CNT@SnO2 @PPy nanocomposites enhancing sodium storage. Colloids and Surfaces A: Physicochemical and Engineering Aspects 555: 795–801. 273 Sun, Y., Yang, Y., Shi, X.-L. et al. (2022). N-doped silk wadding-derived carbon/SnO @reduced graphene oxide film as an ultra-stable anode for sodium-ion half/full battery. Chemical Engineering Journal 433: 133675. 274 Tian, Y., Hu, P., Zhu, T. et al. (2020). Three-layer structured SnO2 @C@TiO2 hollow spheres for high-performance sodium storage. Energy & Environmental Materials 4: 428–433.

References

275 Gandi, S., Chinta, S.R., Katari, N.K., and Ravuri, B.R. (2018). Electrochemical performance of SnO–V2 O5 –SiO2 glass anode for Na-ion batteries. Materials for Renewable and Sustainable Energy 7. 276 Kondo, H., Honma, T., and Komatsu, T. (2014). Synthesis and morphology of metal Sn particles in SnO–P2 O5 glasses and their battery anode performance. Journal of Non-Crystalline Solids 402: 153–159. 277 Wang, Y., Zhou, J., Wu, J. et al. (2017). Engineering SnS2 nanosheet assemblies for enhanced electrochemical lithium and sodium ion storage. Journal of Materials Chemistry A 5: 25618–25624. 278 Xu, X., Zhao, R., Chen, B. et al. (2019). Progressively exposing active facets of 2D nanosheets toward enhanced pseudocapacitive response and high-rate sodium storage. Adv Mater 31: e1900526. 279 Cheng, L., Zhang, Y., Chu, P. et al. (2021). Heterostructure enhanced sodium storage performance for SnS2 in hierarchical SnS2 /Co3 S4 nanosheet array composite. Journal of Materials Chemistry A 9: 1630–1642. 280 Ma, Z., Lyu, Y., Yang, H. et al. (2018). Systematic investigation of the Binder’s role in the electrochemical performance of tin sulfide electrodes in SIBs. Journal of Power Sources 401: 195–203. 281 Gao, S., Liu, Z., Yang, L. et al. (2021). Fabrication of S,N-doped carbon-coated SnS2 /SnS heterostructures supported by hollow carbon microspheres for sodium-ion storage. Journal of The Electrochemical Society 168: 050527. 282 Sun, H., Ren, Y., Chen, Z., and Ding, J. (2018). One-step addition of Ag nano-ions to improve the sodium storage capacity of SnS2 /rGO. Solid State Ionics 326: 5–10. 283 Xia, J., Jiang, K., Xie, J. et al. (2019). Tin disulfide embedded in N-, S-doped carbon nanofibers as anode material for sodium-ion batteries. Chemical Engineering Journal 359: 1244–1251. 284 Zhang, S., Zhao, H., Wang, M. et al. (2018). Low crystallinity SnS encapsulated in CNTs decorated and S-doped carbon nanofibers as excellent anode material for sodium-ion batteries. Electrochimica Acta 279: 186–194. 285 Jiang, Y., Feng, Y., Xi, B. et al. (2016). Ultrasmall SnS2 nanoparticles anchored on well-distributed nitrogen-doped graphene sheets for Li-ion and Na-ion batteries. Journal of Materials Chemistry A 4: 10719–10726. 286 Zheng, J., Xiong, X., Wang, G. et al. (2018). SnS2 nanoparticles anchored on three-dimensional reduced graphene oxide as a durable anode for sodium ion batteries. Chemical Engineering Journal 339: 78–84. 287 Chen, R., Li, S., Liu, J. et al. (2018). Hierarchical Cu doped SnSe nanoclusters as high-performance anode for sodium-ion batteries. Electrochimica Acta 282: 973–980. 288 Park, G.D., Lee, J.H., and Kang, Y.C. (2016). Superior Na-ion storage properties of high aspect ratio SnSe nanoplates prepared by a spray pyrolysis process. Nanoscale 8: 11889–11896. 289 Zhao, W., Ma, X., Li, Y. et al. (2020). Achieving ultrastable cyclability and pseudocapacitive sodium storage in SnSe quantum-dots sheathed in nitrogen doped carbon nanofibers. Applied Surface Science 504: 144455.

165

166

4 Anode Materials of Sodium-ion Batteries

290 Zhou, X., Ding, S., He, H. et al. (2020). Encapsulated SnSe in carbon nanofibers as anode of sodium ion batteries with improved properties. Ionics 26: 3937–3946. 291 Wu, X., Xiong, X., Yuan, B. et al. (2022). Understanding the phenomenon of capacity increasing along cycles: in the case of an ultralong-life and high-rate SnSe-Mo-C anode for lithium storage. Journal of Energy Chemistry 72: 133–142. 292 Khan, R., Xing, Y., Wang, J. et al. (2021). Core-shell SnSe@TiO2 /C heterostructure high-performance anode for Na-ion batteries. Journal of Alloys and Compounds 880: 160469. 293 Yuan, H., Jin, Y., Lan, J. et al. (2018). In situ synthesized SnSe nanorods in a SnOx @CNF membrane toward high-performance freestanding and binder-free lithium-ion batteries. Inorganic Chemistry Frontiers 5: 932–938. 294 Ren, X., Wang, J., Zhu, D. et al. (2018). Sn-C bonding riveted SnSe nanoplates vertically grown on nitrogen-doped carbon nanobelts for high-performance sodium-ion battery anodes. Nano Energy 54: 322–330. 295 Usui, H., Domi, Y., Yamagami, R., and Sakaguchi, H. (2019). Degradation mechanism of tin phosphide as Na-ion battery negative electrode. Green Energy & Environment 4: 121–126. 296 Mogensen, R., Maibach, J., Naylor, A.J., and Younesi, R. (2018). Capacity fading mechanism of tin phosphide anodes in sodium-ion batteries. Dalton transactions 47: 10752–10758. 297 Jung, S.C., Choi, J.-H., and Han, Y.-K. (2018). The origin of excellent rate and cycle performance of Sn4 P3 binary electrodes for sodium-ion batteries. Journal of Materials Chemistry A 6: 1772–1779. 298 Pan, E., Jin, Y., Zhao, C. et al. (2018). Dopamine-derived N-doped carbon encapsulating hollow Sn4 P3 microspheres as anode materials with superior sodium storage performance. Journal of Alloys and Compounds 769: 45–52. 299 Choi, J., Kim, W.-S., Kim, K.-H., and Hong, S.-H. (2018). Sn4 P3 –C nanospheres as high capacitive and ultra-stable anodes for sodium ion and lithium ion batteries. Journal of Materials Chemistry A 6: 17437–17443. 300 Zhao, W., Ma, X., Gao, L. et al. (2020). Engineering carbon-nanochain concatenated hollow Sn4 P3 nanospheres architectures as ultrastable and high-rate anode materials for sodium ion batteries. Carbon 167: 736–745. 301 Zhang, J., Wang, W., and Li, B. (2020). Enabling high sodium storage performance of micron-sized Sn4 P3 anode via diglyme-derived solid electrolyte interphase. Chemical Engineering Journal 392: 123810. 302 Liu, Z., Wang, X., Wu, Z. et al. (2020). Ultrafine Sn4 P3 nanocrystals from chloride reduction on mechanically activated Na surface for sodium/lithium ion batteries. Nano Research 13: 3157–3164. 303 Liu, J., Kopold, P., Wu, C. et al. (2015). Uniform yolk–shell Sn4 P3 @C nanospheres as high-capacity and cycle-stable anode materials for sodium-ion batteries. Energy & Environmental Science 8: 3531–3538. 304 Palaniselvam, T., Mukundan, C., Hasa, I. et al. (2020). Assessment on the use of high capacity “Sn4 P3 ”/NHC composite electrodes for sodium-ion batteries

References

305

306

307

308

309

310

311

312

313

314

315

316

317

318

with ether and carbonate electrolytes. Advanced Functional Materials 30: 2004798. Wang, W., Zhang, J., Yu, D.Y.W., and Li, Q. (2017). Improving the cycling stability of Sn4 P3 anode for sodium-ion battery. Journal of Power Sources 364: 420–425. Chen, J., Xiao, J., Li, J. et al. (2022). Nano-engineering induced Bi dots in situ anchored into modified porous carbon with superior sodium ion storage. Journal of Materials Chemistry A 10: 20635–20645. Cai, S., Yan, F., Zhao, Y. et al. (2022). Hierarchical micro-composite assembled from Bi spheres and expanded graphite flakes as anodes for sodium-ion half/full cells with excellent comprehensive electrochemical performance. Chemical Engineering Journal 430: 132938. Yin, H., Li, Q., Cao, M. et al. (2017). Nanosized-bismuth-embedded 1D carbon nanofibers as high-performance anodes for lithium-ion and sodium-ion batteries. Nano Research 10: 2156–2167. Hwang, J., Park, J.-H., Yoon Chung, K., and Kim, J. (2020). One-pot synthesis of Bi-reduced graphene oxide composite using supercritical acetone as anode for Na-ion batteries. Chemical Engineering Journal 387: 124111. Hu, C., Zhu, Y., Ma, G. et al. (2021). Sandwich-structured dual carbon modified bismuth nanosphere composites as long-cycle and high-rate anode materials for sodium-ion batteries. Electrochimica Acta 365: 137379. Hou, T., Fan, A., Sun, X. et al. (2021). Improving cycling stability of Bi-encapsulated carbon fibers for lithium/sodium-ion batteries by Fe2 O3 pinning. Chinese Chemical Letters 32: 2459–2462. Pan, J., Yu, K., Mao, H. et al. (2020). Crystalline Sb or Bi in amorphous Ti-based oxides as anode materials for sodium storage. Chemical Engineering Journal 380: 122624. Ding, X. and Liu, Y. (2020). Hollow bismuth ferrite combined graphene as advanced anode material for sodium-ion batteries. Progress in Natural Science: Materials International 30: 153–159. Xi, Y., Wang, D., Peng, Y. et al. (2021). Bark shaped structure BiPO4 @rGO applying to high efficiency of sodium ion battery. Journal of Alloys and Compounds 857: 157547. Zhu, H., Liu, T., Peng, L. et al. (2021). A compact Bi2 WO6 microflowers anode for potassium-ion storage: taming a sequential phase evolution toward stable electrochemical cycling. Nano Energy 82: 105784. Xu, X., Xu, Y., Xu, F. et al. (2020). Black BiVO4 : size tailored synthesis, rich oxygen vacancies, and sodium storage performance. Journal of Materials Chemistry A 8: 1636–1645. Zhao, Z., Ye, Y., Zhu, W. et al. (2018). Bismuth oxide nanoflake@carbon film: a free-standing battery-type electrode for aqueous sodium ion hybrid supercapacitors. Chinese Chemical Letters 29: 629–632. Demir, E., Hayat Soytas, S., and Demir-Cakan, R. (2019). Bismuth oxide nanoparticles embedded carbon nanofibers as self-standing anode material for Na-ion batteries. Solid State Ionics 342: 115066.

167

168

4 Anode Materials of Sodium-ion Batteries

319 Sun, W., Rui, X., Zhang, D. et al. (2016). Bismuth sulfide: a high-capacity anode for sodium-ion batteries. Journal of Power Sources 309: 135–140. 320 Yang, W., Wang, H., Liu, T., and Gao, L. (2016). A Bi2 S3 @CNT nanocomposite as anode material for sodium ion batteries. Materials Letters 167: 102–105. 321 Liang, H., Ni, J., and Li, L. (2017). Bio-inspired engineering of Bi2 S3 -PPy yolk-shell composite for highly durable lithium and sodium storage. Nano Energy 33: 213–220. 322 Zhang, Y., Fan, L., Wang, P. et al. (2017). Coupled flower-like Bi2 S3 and graphene aerogels for superior sodium storage performance. Nanoscale 9: 17694–17698. 323 Lu, C., Li, Z., Yu, L. et al. (2018). Nanostructured Bi2 S3 encapsulated within three-dimensional N-doped graphene as active and flexible anodes for sodium-ion batteries. Nano Research 11: 4614–4626. 324 Sun, X., Wang, L., Li, C. et al. (2021). Dandelion-Like Bi2 S3 /rGO hierarchical microspheres as high-performance anodes for potassium-ion and half/full sodium-ion batteries. Nano Research 14: 4696–4703. 325 Yang, Q., Gao, W., Zhong, W. et al. (2020). A synergistic Bi2 S3 /MXene composite with enhanced performance as an anode material of sodium-ion batteries. New Journal of Chemistry 44: 3072–3077. 326 Guan, H., Du, X., Yi, Y. et al. (2021). Minimal TiO2 coupled with conductive polymer-stimulated synergistic effect on fast and reversible sodium-ion storage for bismuth sulfide. ACS applied materials & interfaces 13: 55051–55059. 327 Qin, Y., Zhang, Y., Wang, J. et al. (2020). Heterogeneous structured Bi2 Se3 /MoS2@NC nanoclusters: exploring the superior rate performance in sodium/potassium ion batteries. ACS applied materials & interfaces 12: 42902–42910. 328 Fan, R., Zhao, C., Ma, J. et al. (2022). Rich self-generated phase boundaries of heterostructured VS4 /Bi2 Se3 @C nanorods for long lifespan sodium-ion batteries. Small e2205175. 329 Huang, Y., Zhu, X., Cai, D. et al. (2021). Bi2S3 spheres coated with MOF-derived Co9S8 and N-doped carbon composite layer for half/full sodium-ion batteries with superior performance. Journal of Energy Chemistry 59: 473–481. 330 Xie, L., Yang, Z., Sun, J. et al. (2018). Bi2 Se3 /C nanocomposite as a new sodium-ion battery anode material. Nano-micro letters 10. 331 Gan, Y., He, H., Mu, M. et al. (2022). Fabrication of Bi2 Se3 /Mo3 Se4 composite for efficient sodium storage. Journal of Alloys and Compounds 923. 332 Han, M., Zhou, Z., Li, Y. et al. (2022). Constructing Bi2 Se3 /Bi2 O3 heterostructure as promising anode for efficient sodium-ion storage. Journal of Alloys and Compounds 892. 333 Lahiri, A., Olschewski, M., Gustus, R. et al. (2016). Surface modification of battery electrodes via electroless deposition with improved performance for Na-ion batteries. Physical chemistry chemical physics : PCCP 18: 14782–14786.

References

334 Gavrilin, I.M., Smolyaninov, V.A., Dronov, A.A. et al. (2018). Electrochemical insertion of sodium into nanostructured materials based on germanium. Mendeleev Communications 28: 659–660. 335 Hu, L., Shang, C., Huang, L. et al. (2019). Cu3 Ge coated by nitrogen-doped carbon nanorods as advanced sodium-ion battery anodes. Ionics 26: 719–726. 336 Tang, D., Yu, H., Zhao, J. et al. (2020). Bottom-up synthesis of mesoporous germanium as anodes for lithium-ion batteries. Journal of colloid and interface science 561: 494–500. 337 Lebedev, E.A., Gavrilin, I.M., Kudryashova, Y.O. et al. (2022). Effect of vinylene carbonate electrolyte additive on the process of insertion/extraction of Na into Ge microrods formed by electrodeposition. Batteries 8: 109. 338 Shen, K., Lin, N., Xu, T. et al. (2018). Amorphous mesoporous GeOx anode for Na-ion batteries with high capacity and long lifespan. Royal Society Open Science 5. 339 Li, M., Zhang, Z., Ge, X. et al. (2018). Enhanced electrochemical properties of carbon coated Zn2 GeO4 micron-rods as anode materials for sodium-ion batteries. Chemical Engineering Journal 331: 203–210. 340 Wang, K., Liu, M., Huang, D. et al. (2020). Rapid thermal deposited GeSe nanowires as a promising anode material for lithium-ion and sodium-ion batteries. Journal of colloid and interface science 571: 387–397. 341 Zhou, Y., Zhao, M., Chen, Z.W. et al. (2018). Potential application of 2D monolayer beta-GeSe as an anode material in Na/K ion batteries. Physical chemistry chemical physics : PCCP 20: 30290–30296. 342 Sun, Q., Fu, L., and Shang, C. (2017). A novel open-framework Cu-Ge-based chalcogenide anode material for sodium-ion battery. Scanning 2017: 3876525. 343 Li, F., Qu, Y., and Zhao, M. (2016). Germanium sulfide nanosheet: a universal anode material for alkali metal ion batteries. Journal of Materials Chemistry A 4: 8905–8912. 344 Hao, K.-R., Fang, L., Yan, Q.-B., and Su, G. (2018). Lithium adsorption and migration in group IV–VI compounds and GeS/graphene heterostructures: a comparative study. Physical Chemistry Chemical Physics 20: 9865–9871. 345 Wasalathilake, K.C., Hu, N., Fu, S. et al. (2021). High capacity and mobility in germanium sulfide/graphene (GeS/Gr) van der Waals heterostructure as anode materials for sodium–ion batteries: a first-principles investigation. Applied Surface Science 536: 147779. 346 Fu, L., Shang, C., Ma, J. et al. (2018). Cu2 GeS3 derived ultrafine nanoparticles as high-performance anode for sodium ion battery. Science China Materials 61: 1177–1184. 347 Wen, N., Chen, S., Feng, J. et al. (2021). In situ hydrothermal synthesis of double-carbon enhanced novel cobalt germanium hydroxide composites as promising anode material for sodium ion batteries. Dalton transactions 50: 4288–4299.

169

170

4 Anode Materials of Sodium-ion Batteries

348 Li, W., Ke, L., Wei, Y. et al. (2017). Highly reversible sodium storage in a GeP5 /C composite anode with large capacity and low voltage. Journal of Materials Chemistry A 5: 4413–4420. 349 Kulova, T.L., Skundin, A.M., Gavrilin, I.y.M. et al. (2022). Binder-free Ge-Co-P anode material for lithium-ion and sodium-ion batteries. Batteries 8: 98.

171

5 Electrolyte, Separator, Binder and Other Devices of Sodium Ion Batteries Mingguang Yi 1 , Mingjun Jing 1 , Wentao Deng 2 , Guoqiang Zou 2 , Hongshuai Hou 2 , Tianjing Wu 1 , and Xiaobo Ji 2 1 Xiangtan University, National Base for International Science & Technology Cooperation, National Local Joint Engineering Laboratory for Key Materials of New Energy Storage Battery, Key Laboratory of Environmentally Friendly Chemistry and Application of Ministry of Education School of Chemistry, Xiangtan 411105, P. R. China 2 Central South University, College of Chemistry and Chemical Engineering, Key Laboratory of Powder Metallurgy, Changsha 410083, P. R. China

5.1 Introduction Sodium ion batteries (SIBs) are regarded as one of the most promising nextgeneration rechargeable batteries due to their plentiful advantages, such as low production cost, unlimited availability of sodium resources (a content of 2.83% in the Earth’s crust), and high security [1]. As an indispensable component of SIBs, the electrolytes play a decisive influence on the electrode performance, and an in-depth understanding and comprehensive analysis of the electrolytes is beneficial for developing state-of-the-art SIBs techniques. Similar to LIBs, the electrolytes in SIBs can be commonly divided into organic liquid electrolytes (OLEs) and solid-state electrolytes (SSEs). In general, OLEs exhibit higher ionic conductivity and better compatibility with the electrodes and separators than SSEs as a result of their preferable fluidity, which contributes to the migration of Na+ . To be specific, OLEs include three components of organic solvents, Na salts, and additives. According to the kind of solvents, OLEs can be divided into ester-based electrolytes, ether-based electrolytes, as well as ionic liquids (ILs) electrolytes. In addition, the use of multifunctional additives is the most efficient and cost-effective way to improve the overall performance of SIBs. In fact, the presence of additives not only is conducive to inhibiting electrolytic decomposition, bringing about resultant ameliorations such as reversible capacity and cyclic life, but also diminishes the safety hazards by preventing the thermal runaway induced by the interior short circuit or excessive heating in the battery system. Nevertheless, considering the high inflammability and the leakage risk of OLEs, researchers pay more attentions on the investigation

Sodium-Ion Batteries: Technologies and Applications, First Edition. Edited by Xiaobo Ji, Hongshuai Hou and Guoqiang Zou. © 2024 WILEY-VCH GmbH. Published 2024 by WILEY-VCH GmbH.

172

5 Electrolyte, Separator, Binder and Other Devices of Sodium Ion Batteries

of SSEs at present owing to their higher safety, larger energy density, and lower cost. Compared to OLEs, SSEs exhibit more prominent thermal stability and do not have the risk of leakage, which can fundamentally solve the security issues of SIBs. Moreover, SSEs have broader voltage windows and superior mechanical properties than OLEs. Yet, it is imperative to mention that SSEs demonstrate intrinsically lower ionic conductivity and inferior interface compatibility with active materials in comparison to OLEs, which severally hinders their large-scale application in the domain of SIBs. Typically, SSEs can be classified into inorganic solid electrolytes (ISEs), polymer electrolytes (PEs), and composite solid electrolytes (CSEs). ISEs possess higher ionic conductivity at RT and preferable chemical/thermal stabilities than PEs. The conduction of ions in ISEs obeys the models of Schottky and Frenkel point defects, namely that Na+ ions move from the local sites to the adjoining sites, whose migration process needs to overcome the energetic barrier. The most diffusely studied ISEs for SIBs incorporate β-Al2 O3 , Na superionic conductors (NASICON), as well as sulfides. PEs can be further divided into solid polymer electrolytes (SPEs) and gel polymer electrolytes (GPEs). PEs deliver better interfacial contact with electrodes and matchless processibility with regard to ISEs. In general, SPEs are comprised of polymer matrix and Na salts, and their ionic transmission mechanism is different from OLEs. For instance, Na+ ions experience the process of solvation with the aid of polymer chains and migrate along the polymer segments, while dissociated ions in OLEs allodially move in the whole liquid range. The combination of polyethylene oxide (PEO) matrix and Na salts is the most widely adopted configuration for SPEs. GPEs have a similar structure to SPEs but exhibit higher ionic conductivity than SPEs on account of the addition of organic solvents. In order to accelerate the application of SSEs, the concept of CSEs, which integrate the separate virtues of ISEs and PEs, is raised, and CSEs are prepared by adding inorganic passive/active fillers into the polymer host. Moreover, a phase interface will inevitably be generated on the surface of anode and cathode, respectively, which is derived from the side reactions between the electrodes and electrolytes. The interphase is separately called the solid electrolyte interphase (SEI) and cathode electrolyte interphase (CEI), and both of them determine the morphology of sodium deposits, capacity retention, cycle life, and safety of SIBs to a great extent. Hence, understanding the formation mechanism, the evolution process, as well as the composition and structure of the SEI and CEI are vital in developing high-efficiency and stabilized SIBs. So far, the great mass of studies in the area of SIBs have focused on adjusting the chemical components and microstructure of electrolytes, which assists the electrodes obtain better electrochemical performance. By contrast, the other components of SIBs, such as separator, binder, conductive agent, and current collector, which are as important as the electrodes and electrolytes, are mostly neglected. Separators are used to serve as the physical barrier preventing direct contact between the anode and cathode and the media of Na+ migration, and the composition of separators is mainly polyolefin materials and glass fiber (GF). Binders act as the cement holding together the active material, conductive agent, and the current collector. As for conductive agents and

5.2 Organic Liquid Electrolytes

el

O rg liq a n i ec ui c tr o d lyt es

Solid electrolyte interphase Cathode electrolyte interphase

e el

Good mechanical property

Inorganic solid electrolytes Polymer electrolytes Composite solid electrolytes

Stability Component Structure

SIBs

lid So te tes a st roly ct

High safety

e as Ph face er int

High conductivity Prominent compatibility Solvents Na salts Additives Ionic liquids

Great variety r Proper price he ents t O on p Binder m co Separator Conductive agent Current collector

Figure 5.1 The main ingredients and their specific compositions of SIBs excepting electrodes.

current collectors, the former are used to improve the electronic conductivity of the electrodes, while the latter function as a channel to transmit electrons between the electrodes and external circuit, respectively. In this review, at first, we elaborately elucidate the physical-chemical properties of OLEs and review the characteristics, basic requirements, research progress, and optimization tactics for solvents, salts, additives, and new electrolyte systems. Furthermore, in addition to sum up the indispensable physical-chemical properties of SSEs, we also summarize the latest advances in developing ISEs, PEs, and CSEs, including the generalization of advantages and disadvantages, challenges, and solutions. What is more, thanks to the importance of the interphase in improving the electrochemical performance and maintaining the battery safety, we illustrate the formation mechanism of the SEI and CEI, discuss the influence of stability, components, structures, etc. on the electrochemical performance, and introduce some advanced characterization techniques to analyze the interphase. Finally, the discussion will be done in the range of separator, binder, conductive agent, and current collector. Figure 5.1 summarizes the main conformations of SIBs apart from the anode and cathode. We hope this review can provide a new insight and help the researchers comprehend the development of SIBs.

5.2 Organic Liquid Electrolytes 5.2.1

Physical and Chemical Properties

As an indispensable constituent in SIBs, electrolytes not only serve as an important bridge to link the anode and cathode but also play a key influence on electrochemical performances (such as reversible capacity, rate capability, and cycling performance). Organic liquid electrolytes (OLEs), mainly made up of solvent, electrolyte

173

174

5 Electrolyte, Separator, Binder and Other Devices of Sodium Ion Batteries

salt, and additives, are the most extensively investigated in various electrolytes, and they are the mainstream electrolyte materials used in commercial SIBs because of their outstanding advantages. For the purpose of reducing the production cost and expediting the practical application of SIBs, it is imperative to conduct a systematic and comprehensive study of OLEs. Besides, it is well-known that the selection of electrode materials, the electrochemical properties, and operation range of batteries are indirectly determined by the physical and chemical properties of OLEs. Therefore, in this part, we will elaborately introduce the physico-chemical properties of OLEs and briefly indicate the current deficiency in OLEs. Generally speaking, viscosity, whose values will reflect the impact on the solvation kinetics of electrolytes, is an important parameter to evaluate the conductivity of electrolytes. Compared with solid electrolytes, OLEs demonstrate better fluidity due to their lower viscosity, which is conducive to ion diffusion, and thus OLEs show higher ionic conductivity than solid electrolytes, expediting the reaction dynamics. Not only that, the dielectric constant is also a significant factor for selecting the proper electrolytes because it can influence solution structure and the electrostatic interactions between Na salts and the solvent, further determining the conductivity of electrolytes. Intriguingly, OLEs manifest a large dielectric constant (𝜀 > 15), which facilitates the dissociation of Na salt and inhibits ion pairing, and the value of dielectric constant decreases with increasing temperature derived from improved molecules mobility, as proved by Zhang and his groups [2]. What is more, benefiting from the relatively low melting points (M p ) and high boiling points (Bp ), OLEs possess a wider liquid range with regard to aqueous electrolytes. Moreover, the electrochemical stability windows (ESWs) of electrolytes are an important factor that cannot be ignored for fulfilling high-performance SIBs, which are related to the component and concentration themselves, as well as the electron acceptor/donor capacity of solvents. Practically, ESWs of OLEs are wider than that of aqueous electrolytes, which helps SIBs obtain higher power density and more stable electrochemical behaviors, but narrower than that of SSEs. Currently, a convenient method to extend the ESWs of OLEs is increasing Na salt concentration or ameliorating interface stability. According to previous investigations, the risk of thermal reactions in SIBs is higher than in LIBs due to the higher activity of sodium metal [3], and hence higher requirements on the thermal stability of OLEs are put forward. In fact, the thermal stability of OLEs is highly associated with the flash points of solvents and the concentration of Na salts. The adoption of solvents with high flash points (e.g. nitriles, fluorinated-based compounds, and nitriles) and increasing the concentration of salts have an obvious promotion towards the thermal stability and safety of OLEs. Besides, OLEs possess outstanding compatibility with different electrode materials and superior wettability with porous separators, anodes, cathodes. However, although OLEs exhibit many preferable advantages, they still have some intrinsic disadvantages as well. On one hand, they easily trigger off safety problems because of the high flammability, bad mechanical properties, and high volatility. On the other hand, OLEs are often associated with high toxicity, which has a bad influence on the human body and environment.

5.2 Organic Liquid Electrolytes

5.2.2

Organic Solvents

As a needful element of OLEs, the option of solvents is a crucial factor in deciding the electrochemical behavior of OLEs. An excellent electrolyte solvent should meet the following four requirements: (i) sufficient solubility to Na salt; (ii) relatively low viscosity; (iii) high chemical and electrochemical stability; and (iv) no pollution to the environment and low production costs. In fact, no solvent can simultaneously meet the aforementioned requirements, and organic solvents are still best suited as electrolyte solvents on the whole. Generally, organic solvents can be divided into two categories of ester-based and ether-based solvents. 5.2.2.1 Ester-based Solvents

As we know, esters are the most adopted electrolyte solvents in SIBs because they exhibit a higher oxidation potential (up to 4.3 V) than that of ether-based electrolytes (under 4.0 V), and alkyl carbonates are most commonly used in ester-based solvents, primarily including cyclic carbonates and linear carbonates. The former incorporate propylene carbonate (PC) and ethylene carbonate (EC), and the latter cover ethyl methyl carbonate (EMC), dimethyl carbonate (DMC), and diethyl carbonate (DEC). The following Table 5.1 summarizes some basic properties of the usual ester-based solvents used in SIBs. Among multifarious organic solvents, EC presents some eye-catching advantages. For instance, EC possesses the highest dielectric constant amid PC, EC, DEC, EMC, and DMC, indicating a great solubility for Na salts, which is helpful to increase the Table 5.1

Solvents

Physical and chemical properties of the ester-based solvents for SIBs.

Chemical structure

PC

O

O

Viscosity (25 ∘ C) (cP)a)

Dielectric constant (25 ∘ C)

2.53

64.92b)

2.1

89.78b) (48 ∘ C)

0.59 (20 ∘ C)

3.107

−53

0.65

2.958

−74.3

0.75

2.805

Molecular weight (g mol−1 )

Density (g cm−3 )

Melting point (∘ C)

102.09

1.204

−49.2

88.06

1.321

36.4

90.08

1.069

4.6

104.11

0.997

118.13

0.975

H3C O

EC

O O O

DMC

O H3C

CH3 O

O

EMC

O CH3 H3C

O

DEC

O

O

H3C

O

O

CH3

a) The viscosity and b) Dielectric constant values of ester-based solvents are from the title [4].

175

176

5 Electrolyte, Separator, Binder and Other Devices of Sodium Ion Batteries

transfer number of Na+ . More importantly, EC can also facilitate the formation of a stable and compact SEI layer on the surface of the anode because it possesses a high reduction potential and a low energy barrier for ring-opening [5]. On the other hand, EC has the advantages of a wide electrochemical window and superior thermal stability. However, the high M p of EC determines that it cannot serve as a single solvent appended to electrolytes for room temperature (RT) SIBs. Nowadays, the main strategy to overcome this disadvantage is to organize binary or ternary mixed solvents. However, to the best of our knowledge, the same ratio of mixed solvents suitable for LIBs is not equal for SIBs, and finding the proper proportion of mixed solvents is worth exploring further. Take an example, EC/DEC and EC/DMC are deemed as the optimum solvent formation in LIBs. Yet, according to the early investigation of Ponrouch and co-workers [6], they found the binary PC/EC solvent mixtures were described as the most appropriate composition added to the Na/hard carbon (HC) cell. In detail, binary solvent-based electrolytes had a higher ionic conductivity than these single solvent-based electrolytes, and the ionic conductivity followed the diminishing trend: EC/DME > EC/DMC > EC/PC > EC/Triglyme > PC > Triglyme > DMC. Nay, the abovesaid research team [7] made tremendous efforts in searching for the optimal ternary solvent mixtures for SIBs. They evidenced the electrolytes based on ternary co-solvents exhibit higher ionic conductivity than binary mixed solvents by reducing the viscosity, following the sequence: EC/PC/DME > EC/PC/DMC > EC/PC > PC. Besides, Nagmani’s research group [5] recently studied the electrochemical performance of three kinds of binary solvents involving EC for SIBs, and they also obtained the conclusion that EC : PC exhibited the best rate performance and cyclic stability. By means of density functional theory (DFT) calculation, Chen et al. found FeS@N,S-C cycled in EC/PC/NaClO4 delivered a discharged capacity of 205 mAh g−1 after 2450 cycles because EC/PC-based electrolyte possessed high electron density and weak adsorption, effectively facilitating the formation of stable SEI layer (Figure 5.2a,b) [8]. Similar to EC, PC is also regarded as an excellent organic solvent because it shows a high dielectric constant (only lower than EC) and high ionic conductivity. Moreover, PC has a wide work temperature range owing to its low M p and high Bp , and does not have the corrosivity and toxicity. In contrast to EC, PC has the potential to be the pure solvent applied in SIBs. Nonetheless, PC will generate detrimental side reactions with anode materials because it lacks the ability to form a stable and impermeable SEI layer. Hence, in order to overcome the disadvantages of pure solvents, the commercial electrolyte solvents are generally binary or ternary mixtures consisting of solvents characterized by high viscosity and high Bp and solvents with low viscosity and low Bp . To uncover the electrolyte solvation mechanism, Liu et al. conducted a comparative study on the Na+ solvation behavior in ester-based and ether-based solvents by adopting DFT theory, and they found ester-based solvents exhibit stronger solvation ability towards Na+ than their ether-based counterparts. More

(a)

NaClO4 EC/PC

NaCF3SO3 EC/PC

NaClO4 DGM

0

Solvated Na+

ΔG (kcal mol–1)

Weak adsorption

–40

NaCF3SO3 DGM

NaClO4 DGM NaClO4 EC/PC

Strong adsorption

–80 –120

31.4

–75.3

5.2 Organic Liquid Electrolytes

NaCF3SO3 EC/PC –97.615

Electrode surface –124.375

–125.878

–125.860

–160 N,S-doped carbon with various solvated Na+ structures (b)

NaCF3SO3 DGM

Figure 5.2 (a) Electrostatic potential maps (EPM) of different Na+ -solvation structures. (b) ΔG values between N,S-doped carbon and different solvated Na+ structures. Source: Reproduced from Nagmani et al. [5]/with permission of Elsevier.

specifically, in contrast to linear ester-based solvents, solvation-Na+ complex was more toilless to be constructed between Na+ and cyclic ester-based solvents. All ester-based solvents could establish 4sol-Na+ complex while the ether-based solvents such as DOL and DME only formed 3sol-Na+ complex, excepting THF, which proved Na+ can achieve a high concentration in ester-based solvents [9]. In contrast, linear carbonates exhibit preferable wettability than the cyclic carbonates towards hydrophobic polyethylene (PE) separator. 5.2.2.2 Ether-based Solvents

With regard to ester-based electrolytes, ether-based electrolytes show much less application potential in commercialization, which is attributed to inferior initial Columbic efficiency (ICE) and inferior oxidation stability. As we know, graphite cannot be employed as an anode for SIBs because Na+ cannot embed into the graphite layer because the ionic radius of Na+ is larger than the lattice spacing of graphite. Nevertheless, ether electrolytes have drawn intensive attention again in the field of SIBs due to the outstanding ability that ether-based solvents can form a co-intercalation layer with Na+ in graphite and generate a stable SEI between the anode and electrolytes. For example, Zhu et al. [10] found graphite electrodes combining tetraglyme-based electrolytes could exhibit an ultralong cycle life span (95% capacity retention after 6000 cycles) and excellent rate capability (110 mAh g−1 at 10 A g−1 ). Most ethers are generally characterized by higher chemical stability against reduction in terms of ester-based electrolytes. However, ether-based solvents

177

178

5 Electrolyte, Separator, Binder and Other Devices of Sodium Ion Batteries

Table 5.2

Solvents

Physical and chemical properties of the ether-based solvents for SIBs. Molecular Viscosity weight Density (25 ∘ C) (g mol−1 ) (g cm−3 ) (cP)

Chemical structure

DME

CH3

O H3C

TEGDME

H3C

O

O O

CH3

O O

O

O H3C

CH3

O

H3C

Triglyme

90.12

0.867

0.46a)

7.18

222.28

1.009

3.39a)

7.53b)

134.18

0.937

1.06a)

7.4

178.23

0.986

—

—

74.08

1.046

0.531

7.13

72.11

0.89

0.53 (20 ∘ C)

7.58

O

O

DEGDME

Dielectric constant (25 ∘ C)

O O

CH3 O

DOL

O

O

THF

O

a) The viscosity and b) dielectric constant values of ether-based solvents are from the title [4].

are easily oxidated at high potentials, which means they are unsuitable for most cathode materials unless adopting modifying strategy. The ether-based solvent is mainly classified into two categories of linear ethers and cyclic ethers. The former include 1,2-dimethoxyethane (DME), diethylene glycol dimethyl ether (DEGDME), tetramethylene glycol dimethyl ether (TEGDME), and triglyme, while the latter involve 1,3-dioxolane (DOL) and tetrahydrofuran (THF). Table 5.2 summarizes some basic characteristics of ether-based solvents for SIBs. In general, the thermal stability of ether-based solvents is not as good as these ester-based solvents and follows the decreased order, namely DME < DMC < DEC < EC < PC. Just as ester-based solvents, ether-based solvents also play a significant role in improving the electrochemical properties of SIBs. Wang et al. [11] demonstrated an ultralong cycle life of SIBs combining undissolved cathode, N,N ′ -bis(glycinyl) naphthalene diimide (Na2 BNDI), with NaClO4 -DEGDME electrolytes, whose capacity retention remained 57.3% after 70 000 cycles at 10 C. The excellent cycle performance was ascribed to a thin, robust, and ionic conducive inorganic-abundant SEI formed in the ether-based electrolytes via the reduction decomposition of anions of Na salt prior to solvent (Figure 5.3a,c). In contrast, the Na2 BNDI cathode with ester-based electrolyte revealed inferior rate performance and cyclic life owing to the formation of permeable and broken SEI layers, as displayed in Figure 5.3b. Li et al. [12] found 1 M NaOTf-DEGDME exhibited the best cyclic stability with the lowest overpotential in eight kinds of investigated

(a)

F 1s

Pristine

(d)

0.5

Voltage (V)

5.2 Organic Liquid Electrolytes

0.0

–20 °C

0.5 mA cm–2, 0.5 mAh cm–2

C-F

DEGDME, before etching Na-F

Intensity (a.u.)

P-F

DEGDME, etching 120s

–0.5

0

200

400 Time (h)

(e)

EC/DEC, before etching

600

800

(f)

5 GPa

50 nm

EC/DEC, etching 120s

692

690

688 686 684 682 Binding energy (eV)

680

(b)

–40 °C

1 µm

–50 nm –40 °C

(c) EC/DEC electrolyte

DEGDME electrolyte

1 µm

–1 GPa

NaF Na2CO3

Decompostion

Polyether Stable ROCO2Na

Na2BNDI electrode

Na2BNDI electrode

Polyester

Figure 5.3 (a) X-ray photoelectron spectroscopy (XPS) spectra of the initial and cycled Na2 BNDI electrodes after 50 cycles in two kinds of electrolytes. Schematic plot of the SEI formation and decomposition in (b) EC/DEC-based electrolytes and (c) DEGDME-based electrolytes. Source: Reproduced from Wang et al. [11]/with permission of John Wiley & Sons. (d) Galvanostatic cycling of symmetric Na||Na cells in eight kinds electrolytes dissolved 1 M Na salts at –20 ∘ C. (e) Atomic force microscopy (AFM) topography and (f) Young’s modulus of the SEI generated in 0.5 M NaOTf-DEGDME/DOL (2 : 8, V%) after the first cycle at –40 ∘ C. Source: Reproduced from Wang et al. [12], © 2022/Springer Nature/CC BY 4.0.

electrolytes at both +20 and –20 ∘ C (Figure 5.3d). Furthermore, they designed a DEGDME/DOL-based binary solvent that dissolved 0.5 M NaOTf, and discovered it exhibited good thermal stability at 150 ∘ C and excellent low-temperature electrochemical performance because this electrolyte could promote the formation of a robust and smooth SEI layer by means of generating more NaF and decreasing surface roughness (Figure 5.3e,f). Besides, the ester-based electrolytes such as PC, EC, and DEC are usually not the appropriate candidates for Na/FeS2 batteries for the reason that a detrimental

179

180

5 Electrolyte, Separator, Binder and Other Devices of Sodium Ion Batteries

side reaction will take place between ester-based solvents and the anionic groups of polysulfide in the process of initial discharge [13]. On the contrary, ether-based solvents exhibit good dissolving capacity towards polysulfide and reduce the polysulfide more thoroughly. Thus, Chen’s group [14] obtained a high-rate capability and an ultralong cycle life of Na/FeS2 cells by choosing diglyme-based electrolytes and controlling the cut-off voltage. The capacity retention could maintain about 90% after undergoing 20 000 cycles at 1 A g−1 , and the cell acquired prominent high-rate performance, whose capacity achieved 170 mAh g−1 at an ultrahigh current density of 20 A g−1 . The splendid electrochemical performance was attributed to the reversible phase transition from FeS2 to layered Nax FeS2 and the pseudocapacitance reaction. Although HC is widely used as the work electrode in SIBs, it exhibits some dissatisfactory electrochemical performances such as inferior rate capability and low ICE in some ester-based electrolytes. Hirsh et al. [15] detailedly investigated the springhead of poor electrochemical performance of HC electrodes via testing them in two different electrolytes containing PC and TEGDME solvent. They discovered HC with TEGDME-based electrolyte possessed higher ICE and preferable rate performance than PC-based electrolyte, which was assigned to a robust, thin, and uniform SEI layer induced by an ether-based electrolyte.

5.2.3

Electrolyte Salt

Electrolyte salt, composed of Na+ and anion groups, is an important ingredient in electrolytes. In fact, the composition and thickness of the SEI layer are mainly determined by the anion and cation of the electrolyte salt. Up to now, numerous researchers are devoted to seeking proper Na salts that can meet the required functions of SIBs. Among various Na salts, NaPF6 and NaClO4 are the most widely added in OLEs owing to their excellent compatibility with electrode materials and high ionic conductivity. In addition, other Na salts containing fluorocomplex anions, such as sodium bis(trifluoromethanesulfonyl)imide (NaTFSI), sodium bis(fluorosulfonyl)imide (NaFSI), sodium fluorosulfonyl-(trifluoromethanesulfonyl) imide (NaFTFSI), and sodium trifluoromethanesulfonate (NaOTf), also occupy a certain share in consequence of the advantages of non-toxic and satisfactory thermostability. Moreover, inspired by the successful application of Li borate salts in LIBs, the Na borate salts, namely sodium-difluoro(oxalato)-borate (NaDFOB), sodium bis(oxalate)-borate (NaBOB), and sodium bis(salicylato)borate (NaBSB), also get a lot of attention from the scientific community. Table 5.3 lists some basic properties of the Na salts for SIBs. In general, the thermal stability is a significant parameter to evaluate the electrolyte salt, and it can be measured by M p and decomposition temperature (T decom ) gained from differential scanning calorimetry (DSC) and the thermogravimetric analysis (TGA) measurements, respectively. As early as in 2012, Ponrouch and his

5.2 Organic Liquid Electrolytes

Table 5.3

Na salt

Basic properties of the most commonly adopted Na salts for SIB electrolytes.

Anion chemical structure

NaPF6

F F

Molecular weight (g mol−1 )

Density (g cm−3 )

Decomposition temperature (∘ C)

Conductivity (mS cm−1 )a)

167.95

2.369

300

7.98

122.44

2.52

482

6.4

109.79

2.47

384

—

211.9

—

—

—

258.74

3.38

—

—

209.84

—

345

—

306.01

—

353

—

159.81

—

—

—

203.11

—

118

—

303.13

—

257

6.2

253.12

—

—

—

172.05

—

248

—

F – P F

F F

NaClO4

– O O

Cl

O

O

NaBF4

F

B– F

F F

NaAsF6

F F

F

– As F

F F

NaSbF6

F F

F

– Sb F

F F

NaBOB

O

O

O

O B–

O

O

O

O

NaBSB

O O

O B–

O

O

O

NaDFOB

F

F

O

O

B – C

C

O

NaFSI

O

O

– N

O

S

S

F

NaTFSI

F

O

O O

O F

– N

F S

S C

C O

F

O

F F

F

NaFTFSI

F F

O

O

C

F

F

S

S

– N

O

O

NaOTf

O

– O S

F C

O

F F

a) The conductivity values of Na salts are from the title [16].

181

182

5 Electrolyte, Separator, Binder and Other Devices of Sodium Ion Batteries

team [6] found that a mixture of EC and PC containing NaPF6 -based electrolytes possessed high safety as a result of the high exothermic onset temperature. Eshetu et al. [17] conducted an exhaustive evaluation of the thermal stability of different Na salts, and they found that the thermal stability increased in the order NaFSI < NaFTFSI < NaPF6 < NaTFSI < NaClO4 . Nevertheless, although NaClO4 displays the best thermal stability, the risk of explosion in a dry atmosphere restricts its large-scale application. Therefore, more efforts are paid to seek the desired substitutes for NaClO4 . Geng et al. [18] performed an exhaustive investigation of different kinds and concentrations of electrolyte salts on the physical properties and electrochemical performance of half-cells using Nax Ni0.22 Co0.11 Mn0.66 O2 (NaNCM) as cathode material. This led to the conclusion that 1 M salt concentrations (NaPF6 and NaTFSI) added to a PC-based electrolyte revealed the highest ionic conductivity among 0.1, 1, and 3 M salt concentrations in the whole temperature range from − 20 to 60 ∘ C. Furthermore, cells with electrolytes containing 0.1 M NaPF6 and NaTFSI, respectively, unfolded good responses only at low currents while experiencing a serious capacity decay at high currents. Conversely, compared to low salt concentrations, 1 M NaPF6 and NaTFSI showed better electrochemical performance regardless of the current density. In a word, 1 M salt concentrations achieve a trade-off between physical and electrochemical properties. However, electrolytes added to these Na salts have their inherent drawbacks. For instance, NaPF6 -based electrolytes are highly sensitive to moisture because they will react with H2 O and release the corrosive gas HF, which results in the dissolution of positive materials. Besides, it is speculated that NaOTf salts deliver relatively low ionic conductivity due to terrible solubility in the solvents. Both NaSbF6 and NaAsF6 are poisonous and have a bad influence on organisms and the natural world. Although NaBOB has been successfully prepared by Zavalij and co-workers [19] for nearly 20 years, the insolubility of NaBOB in routine organic solvents critically hinders its adhibition in SIBs. Until 2020, Younesi et al. [20] firstly used NaBOB as Na salt dissolved in TMP and investigated its physical character and electrochemical performance in SIBs. The results indicated NaBOB salt with high heat stability exhibits a comparable ionic conductivity with NaPF6 in TMP, and the TMP solvent combining NaBOB had a negligible impact on the flammability of TMP (Figure 5.4b). More encouragingly, NaBOB/TMP-based electrolytes possessed better sodiation/desodiation ability than their equivalent adopting NaPF6 , and thus the cell incorporating NaBOB-TMP showed higher initial Coulombic efficiency (CE) than the equivalent adopting NaPF6 , indicating NaBOB salt has the capacity employed in OLEs (Figure 5.4a). Besides, the same group [21] further confirmed that NaBOB salt possesses a higher solubility in NMP than in TMP, and the ionic conductivity of the NMP-TMP-NaBOB-based electrolyte displayed a positive correlation with the content of NMP (Figure 5.4c). The test results indicated TMP-rich electrolytes revealed higher capacity retention and CE, while NMP-rich electrolytes had better rate performance (Figure 5.4e). Not only that, the authors conducted a comprehensive assessment of NaBOB salt in different ratios of NMP-TMP solvent, as shown in Figure 5.4d.

5.2 Organic Liquid Electrolytes

(a)

(b)

(c)

(d)

(e)

Figure 5.4 (a) Schematic diagram of the formed SEI layer in NaBOB/trimethyl phosphate (TMP)-based electrolytes and NaPF6 /TMP-based electrolytes. (b) The flammability contrast of TMP dissolved 0.5 M NaBOB (Left) and PC (Right), “T” represents time measuring in seconds. Source: Reproduced with permission from Mogensen et al. [20], © 2020/American Chemical Society/CC BY 4.0. (c) Ionic conductivity of electrolytes as a function of NaBOB concentration in the binary mixtures with different ratios of N-methyl pyrrolidone (NMP) and TMP (NMP vs. TMP, V%). (d) Comprehensive comparison of NaBOB-TMP, NaBOB-NMP, and mixed TMP-NMP (1 : 1)-NaBOB electrolytes. (e) Average CE of the cell using HC anode and Prussian white cathode in different ratios of NMP and TMP-based electrolytes. Source: Reproduced with permission Mogensen et al. [21]. Copyright 2021, American Chemical Society.

5.2.4

Electrolyte Additives

As an important part of the electrolyte, the total content of additives is relatively small (generally lower than 10% by weight or volume). However, the presence of additives will remarkably improve the electrochemical properties and safety of SIBs because they can be preferentially reduced on the anode surface and oxidized on the cathode surface before electrolytes, respectively. In this regard, the reduction potential of additives should be higher than that of the electrolyte solvents corresponding to the lowest unoccupied molecular orbital (LUMO) lower than that of these electrolyte solvents. Meanwhile, the additives meet a standard where the highest occupied molecular orbital (HOMO) energy levels are higher than the electrolyte solvents, correspondingly having a lower oxidation potential compared to the electrolyte solvents. According to different functions, additives can be mainly divided into the following categories: (i) film formation additives; (ii) flame retardant additives; (iii) overcharge protection additives; and (iv) other types such as overall performance enhancers, Al corrosion inhibitors, and salt stabilizers. Table 5.4 provides a few basic details about the additives.

183

184

5 Electrolyte, Separator, Binder and Other Devices of Sodium Ion Batteries

Table 5.4

Additive

Basic properties of the additives for SIBs electrolytes.

Chemical structure

FEC

O

Molecular weight (g mol−1 )

Density (20 ∘ C) (g cm−3 )

Bp (∘ C)

Mp (∘ C)

Viscosity (25 ∘ C) (cP)

106.05

1.41

249.5

18

2.35

86.05

1.42

74

20.5

2.23

F O

O

VC

O O O

PS

O

O S

EFPN

OCH2CH3

F

122.14

1.393 (40 ∘ C)

180

31

—

274.01

1.392

125

—

1.25

140.07

1.215

196

−46

—

124.08

1.1735

182

100% increase in nanowire diameter; upon desodiation, the Nax Sn phase transforms into Sn nanoparticles with the formation of nanopores, leading to the increase of electrical impedance and the deterioration of cycle stability of SnO2 [15]. In summary, in situ TEM techniques are good at detecting dynamic structure and chemical information of SIBs in terms of phase transformation, structural evolution, reaction kinetics, and mechanical properties.

6.1.3

Synchrotron X-Ray Imaging Studies of SIBs

The interactions between X-rays and matter generate a series of phenomena such as X-ray diffraction (XRD) determined by the Bragg equation, X-ray fluorescence determined by Moseley’s law, and X-ray absorption determined by the attenuation coefficient [3]. Darkness in an X-ray image corresponds to the amount of matter that X-rays pass through. As far as synchrotron X-ray imaging is concerned, it can be mainly grouped into two categories, namely lens-based methods (full-field transmission X-ray microscopy [TXM], scanning transmission X-ray microscopy (STXM), and X-ray fluorescence [XFM]) and lensless-based methods (coherent diffraction imaging [CDI]). The characteristics of these X-ray imaging methods are compared in Table 6.1 [26]. A full-field synchrotron TXM operates similarly to a conventional visible light microscope, which is firstly focused down the X-ray beam to micrometers by a capillary condenser, followed by a beam stop and a

6.1 Imaging and Microscopy

pinhole blocking out unfocused X-ray beams, transmitted photons illuminating the specimen, and generating a magnified image with fine resolution by a Fresnel zone plate. In STXM mode, the X-ray beam is focused on a small spot by using a micrometer zone plate, and subsequently rasterized over the region of interest in the object. XFM can be performed by spectrally analyzing emitted secondary photons as the sample is raster-scanned in front of a small-X-ray focal spot, which is advantageous to map trace elements down to ppm levels in a specimen. In comparison with lens-based imaging techniques above, the lensless X-ray imaging technique, also known as coherent diffractive imaging, can provide 2D or 3D reconstruction of the image of nanostructures when the X-ray beam scattered by the object produces a diffraction pattern downstream, which is then collected by a detector. It enables structure determination of nanocrystals with resolutions of as high as 1–2 nm, which are only theoretically limited by the wavelength of the incident X-ray [26]. Based on different imaging principles (transmission, diffraction, and fluorescence), a series of synchrotron X-ray imaging methods have been used to provide insightful views for SIBs in terms of their multidimensional morphologies, bulk/surface structural and chemical information (Table 6.2) [26]. Various electrode materials for SIBs have been studied by synchrotron X-ray imaging in terms of structural evolution and degradation mechanisms, including Sn [27], Bi [28], Na [29], CuO [30], Fe3 S4 [31], Ni3 S2 [32], NaNi1/3 Fe1/3 Mn1/3 O2 [33], NaNiO2 [34]. In situ synchrotron TXM technique was used to visualize the 3D structural and chemical evolution of Sn anodes in SIBs, demonstrating a superior (de)sodiation equilibrium and the failure mechanism of Sn anodes during repeated electrochemical cycling. The 3D morphological evolution, surface curvature, and the corresponding electrochemical performance provide insights into understanding the failure mechanisms of Sn anodes. It is noted that two critical sizes (0.5 and 1.6 μm) for the fracture of Sn particles and the volume expansion of ∼326% after the initial sodiation were also determined by TXM for SIBs [27]. Besides, the structural evolution of the NaNi1/3 Fe1/3 Mn1/3 O2 cathode for SIBs was investigated by in operando TXM-XANES technique. The in operando TXM-XANES mapping reveals a few mixed-phase zones in the cathode material during cycling, demonstrating the existence of a substitutional solid solution in Na-poor phase Na1–𝛿 Ni1/3 Fe1/3 Mn1/3 O2 , whose composition spans over two thermodynamic phases NaNi1/3 Fe1/3 Mn1/3 O2 and Ni1/3 Fe1/3 Mn1/3 O2 [33]. In summary, in situ X-ray imaging techniques enable the real-time 3D observation of dynamic intermediates and chemical information of SIBs in relatively high spatial resolution during battery operation (Figure 6.2).

6.1.4

Neutron Imaging Studies of SIBs

Neutron microscopes use neutrons to create images by nuclear fission of lithium-6 using small-angle neutron scattering [35]. Neutrons carry no electric charge and interact with atomic nuclei through the strong force, which empowers them to

251

Table 6.2

Characteristics of various synchrotron X-ray imaging techniques. Acquired information

Spatial and temporal resolution

TXM

Element; chemistry; 2D/3D imaging

STXM

Technique

Energy range

Sample thickness

Advantage

Limitation

20–30 nm; A few seconds to a few minutes

5–11 keV

Tens of nanometers to tens of micrometers

Chemical analysis; Atmosphere condition

Weak image contrast for light elements

Element (including light ones); chemistry; 2D/3D imaging

12–40 nm; Hundreds of seconds to a few hours

0.5 Na) is extracted from the structure. Coupled with the gas evolution analysis data, the researchers concluded that the reduction of Fe during de-sodiation can be ascribed to the

6.3 Synchrotron Radiation X-ray Absorption Spectroscopy Technique

Top cap (+) Electrode Separator Li foil Spacer

static

Galvano

Spring Bottom cap (–)

e–

Synchrotron light source

Monochromator

It

ber

Ion cham

I0 am X-ray be

ber Ion cham In-situ coin-cell

sorption

X-ray ab

Figure 6.7 Schematic illustration of in situ/operando XAS experiments conducted in the transmission mode for SIBs. Source: Reproduced with permission Wu et al. [62], © 2021, John Wiley & Sons, Inc.

oxygen evolution from the host lattice and that the oxygen redox activity should be responsible for charge compensation [77]. In addition, XAS can also be used to investigate the electrochemical reaction mechanisms of new electrode materials. A new high-voltage cathode material Na3 V(PO3 )3 N with a NASICON-type structure was detected by in situ XAS, revealing that the cathode presents high reversibility with V3.2+ /V4.2+ redox reaction [78]. Regarding layered oxides, it is important to determine the exact oxidation states and local atomic structure of transition metal elements, thus gaining in-depth understanding of the variable layered structure. Regarding polyanions and Prussian blue analogs, using XAS techniques to detect the spin state and the local coordination information is important for tuning the voltage profiles and improving electrochemical activity [76]. Beyond cathode materials for SIBs, XAS was also used to probe the electrochemical reaction mechanisms of anode materials for SIBs, including insertion-type, alloying-type, and conversion-type anode materials. The sodiation process of the Bi0.5 Sb0.5 alloy was studied by in situ Bi L3 -edge and Sb K-edge XAS, which were collected simultaneously on the same cycling cell [79]. It provides insightful views of the amorphous intermediate phases, that is, Nay Bi1−x Sbx and Nax Sb, of the

267

6 Advanced Characterization Techniques and Theoretical Calculation

1 A3 A1

A2

264

0.8

200

200

20

B

184

250

0.6

150

10

5

0.2

50

(a)

100

0.4

100

0 4950

15

4960

4970

4980

4990

5000

Energy (eV)

5010

5020

0

Time (h)

300

25

366

1.2 300

350

12

Spectra (#)

268

2.5

(b)

2

1.5

1

0.5

0

0

Potential vs. Na+/Na (V)

Figure 6.8 Evolution of operando Ti K-edge XANES of TiO2 anatase anodes for SIBs. Source: Reproduced from Fehse et al. [81]/MDPI/CC BY 4.0.

Bi0.5 Sb0.5 anode for SIBs, which can be used as a complementary explanation for the XRD test. Furthermore, the SnS2 anode for SIBs was studied by in situ Sn K-edge EXAFS experiments, indicating the formation of Na–Sn alloy intermediate phases during discharge [80]. The sodiation/desodiation mechanism of TiO2 anatase was studied via operando XAS during discharge/charge (Figure 6.8) [81]. During the initial sodiation, a shift of the Ti K-edge position toward lower photon energies can be observed, revealing the reduction of Ti4+ with Ti4+ /Ti3+ as the redox couple. A slight increase of the A2 pre-edge peak intensity and a general fading of the other features can be observed, which can be owing to the aggravating distortion of the Ti–O polyhedrons with respect to the crystalline TiO2 . The main edge position tends to be largely reversed upon desodiation, while the changes in the pre-edge features are not, demonstrating that nonreversible structural rearrangements occurred during initial sodiation. Based on the results of PCA and Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS), the formation of a reactive intermediate via the amorphization of the original crystalline anatase structure was identified here in while the formation of metallic titanium upon conversion of the initial anatase can be excluded. As for anodes, XAS can be a particular tool to understand the amorphous phases, charge transfer, and redox reactions involved in the sodiation/desodiation process.

6.3.4

Challenges and Opportunities

Although significant progress has been made in the preparation and in situ measurement of SIBs in the past few years, there are still challenges in understanding their complex electrochemical storage behavior and proposing feasible solutions, and major breakthroughs can still be made in the future. In situ electrochemical experiments based on synchrotron XAS can provide unique insights into the electrochemical reaction mechanisms of rechargeable batteries on the high-resolution time and length scale, but there are also challenges, namely: (i) under high current densities or high-quality loads, in situ cells may experience electrochemical imbalances that may affect the accurate understanding of the in situ XAS spectrum; (ii) in situ cells can sometimes differ from the manipulation

6.3 Synchrotron Radiation X-ray Absorption Spectroscopy Technique

of traditional (unmodified) cells owing to the possible side effects originating from the poor contact below the wells and imposed windows; (iii) additional and/or unusual phenomena can be hidden since XAS measures the averaged local atomic information of selected elements; (iv) in situ soft XAS experiments are challenging, especially for light elements such as O and N, due to the high vacuum environment and susceptibility to electrolyte and matrix; (v) difficulties in data analysis and fitting pose challenges to the extraction of fruitful results from XAS; (vi) the intense competition and limited beamtime for synchronous XAS can inhibit the development of in situ XAS research [62]. In fact, the electrochemical reaction in the battery is under nonequilibrium state when electrodes charge/discharge at relatively large currents. The polarization effect has an important influence on the operation of the battery. The actual reaction mechanism is crucial to consider the polarization effects, which correlate to the electrochemical reaction kinetics and the mass transfer inside the battery. In addition, parasitic reactions, different from the main redox reactions of the battery, will occur, which are a terrible source of harmful effects on capacity and Coulombic efficiency. XAS has the unique ability to determine the chemical and oxidation states of elements, and the selectivity of elements can be a unique function to solve these problems. Important information can be obtained when the experimental constraints (penetration depth, light absorption concentration, etc.) are met. The potential development of operando XAS for battery applications includes obtaining more detailed details about the electronic structure and chemical state, which are theoretically included in the XANES spectrum. These details are attractive because they can indicate the reaction mechanism by revealing the local atomic and electronic structure of the target element. However, the extraction of this information is complex because it is inherently difficult to simulate XANES spectrum from the first principle and in operando experiments, as it is inevitable to detect various chemical environments at the same time. The first problem can be solved by using model compounds, advanced computing tools, and/or their combinations in principle, while the second problem requires advanced experimental probes, such as nanobeams, to map the spatial distribution of different chemical environments [82]. In the soft X-ray environment, the application of operation level XAS (soft XAS) is greatly limited, mainly due to the low penetration depth of X-ray with energy less than 1 keV and severe vacuum restrictions. However, soft XAS will provide valuable information for fully understanding the mechanism of phenomena occurring on the surface and interface of materials and will have basic applications in battery science. In fact, light elements (O, N, and C) have absorption edges within the energy range of sub 1 keV, while the L2,3 edges of transition metals, where the 3d valence shows that strong dipoles are allowed to be excited within the same range. Therefore, the development of operando batteries for experiments in soft X-ray environment is very desirable for completely monitoring all elements of most battery materials. It is also expected to have a more comprehensive understanding of the electrochemical reaction mechanism by combining several advanced in situ

269

270

6 Advanced Characterization Techniques and Theoretical Calculation

characterizations. With the rapid development of synchronous light sources, we believe that the shortage of XAS beam time will be alleviated in the future, which also provides an exciting opportunity to reveal the unprecedented details of updated synchronous light sources. Based on reasonable in situ battery design and comprehensive in situ multi-edge XAS measurement, the research on battery materials based on XAS in the future will flourish on the basis of this latest characterization technology.

6.4 Solid-state Nuclear Magnetic Resonance Spectroscopy The nuclear magnetic resonance phenomenon was discovered by Purcell and Bloch in 1946. It is based on the interaction between magnetic moment and magnetic field of various atomic nuclei. The magnetic moment of the nucleus is related to the nuclear spin, and the value of the nuclear spin is defined by the spin number. It can be said that the nuclear magnetic moment related to nuclear spin depends on the properties of the atomic nucleus and its spin number. Even atomic numbers and neutrons have zero nuclear spin and magnetic moment, while odd atomic numbers or neutrons have nonzero nuclear spin and magnetic moment [83]. Odd numbers of protons or neutrons with nonzero nuclear spin have magnetic moments, which can be obtained by NMR spectroscopy in principle. The technology is based on the interaction between the nuclear magnetic moment and the electromagnetic field in the RF range when the strengthened magnetic field B0 is applied. The local magnetic field of the surrounding atomic nucleus and electrons affects these interactions, which are basically divided into external and internal interactions in the nuclear magnetic resonance spectrum. External interactions include nuclear spin and magnetic field interactions (the Pieter Zeeman effect; the difference between the energy levels corresponding to the so-called La Môle frequency Ω0 ) and the manipulation of nuclear spin by RF fields. NMR signal shifts and line shapes are determined by internal interactions, such as chemical, Knight, and Fermi contact shifts, and dipole and quadrupole coupling [84]. In addition to liquid electrolytes, most of the internal components of secondary batteries are solid, which can also be replaced by solid electrolytes to improve the energy density and safety of current secondary batteries. Therefore, the advanced role-shaping technology suitable for solid-state detection is essential for the diagnosis of secondary ssNMR [85]. ssNMR is a powerful tool to detect the local atomic environment in crystalline and amorphous materials. XAS, PDF, and ssNMR techniques have been used to characterize the short-range structure of materials. Compared with XAS and PDF technologies that require synchrotron radiation sources, the laboratory-scale ssNMR technology shows experimental convenience. In addition, abundant ssNMR active nuclei cover most of the components of battery materials, which makes ssNMR highly compatible with battery technology. In particular, 6 Li and 23 Na, as carriers, directly participate in the electrochemical

6.4 Solid-state Nuclear Magnetic Resonance Spectroscopy

Figure 6.9 The disordered nuclear spin forms a net magnetic moment under the action of an external magnetic field. Source: Reproduced from Pecher et al. [86]/American Chemical Society/CC BY 4.0.

B0 z

Nuclear spins x

Net magnetic moment

y

processes of LIB and SIB, respectively, and become the widely studied atoms in the field of energy storage technology.

6.4.1

Principles of ssNMR

In the static magnetic field, the odd protons or neutrons of the nonzero magnetic moment tend to migrate to the direction of the lowest energy due to the interference of heat energy. Therefore, we can imagine that the direction of nuclear torque in the magnetic field is almost random in space [86]. We can think of the nuclear spins of odd protons or neutrons as tiny compass needles, rearranged under the effect of an external magnetic field, resulting in the formation of a net magnetic moment along the B0 direction (usually defined as the z direction) (Figure 6.9). Magnetization enhancement is driven by the so-called spin-lattice relaxation process (T1 relaxation), in which paramagnetic materials are generally enhanced very quickly, while diamagnetic materials are generally enhanced very slowly. Once the maximum longitudinal magnetization is reached, the spin system is in thermal equilibrium. In the semi-classical image of NMR, the (very simplified) description of the nuclear spin as a compass needle must be enhanced by describing it as a spinning gyroscope precession under the influence of a gravitational field (in this case, a magnetic field). The frequency of this precession is what’s called the La Môle Frequency omega-zero, which is a characteristic of each nucleus relative to b-zero. This frequency is related to the difference in energy between nuclear spin states, such as −1/2 something +1/2, in which, in the classical quantum mechanical description of NMR, the sum of the energy states determines the magnetization intensity. The internal NMR interaction interferes with these different energy levels, thus interfering with the frequency distribution. The longitudinal magnetized thermal balance is now controlled by RF pulses applied to the sample via a RF coil that is used both to manipulate the spin system and for signal detection. The net magnetization is thus transferred to the x–y plane (transverse magnetization), where it rotates about the z-axis. The rotation of the magnetic moment in the coil produces a voltage. Due to the internal NMR interaction and the so-called spin–spin relaxation process (T2 relaxation), this coherence is lost relatively quickly (transverse magnetization) and is also slowly restored to longitudinal magnetization (back to equilibrium). This results

271

6 Advanced Characterization Techniques and Theoretical Calculation

Perturbations Internal interactions Acquisition Shift NMR signal

Free induction decay (FID)

Intensity

Voltage

272

FT Broadening

Time

ω0

Isotope's N.A. Number of sites Temperature B0 strength Sample size/fraction Probe characteristics

Frequency

Figure 6.10 The measurement is free induction decay (FID), which converts a voltage-time signal into an intensity-frequency signal via an FT. Source: Reproduced from Pecher et al. [86]/American Chemical Society/CC BY 4.0.

in a decay of the induced voltage, which is measured as free induction decay (FID). Voltage-time signals are converted to strength-frequency signals by FT (Figure 6.10). Here, the displacement and signal broadening (linear shape) are determined by internal interactions and can become quite complex. In addition, the strength of the NMR signal depends on the natural abundance (N.A.) of the isotopes studied, the number of sites/chemical species, temperature, magnetic field strength, sample size/fraction, and probe properties of the samples. Species in locally different chemical environments, such as structural defects, atomic disorder, and dynamic processes, usually have different NMR signals. Thus, NMR spectroscopy can distinguish these species at the local atomic level.

6.4.2

NMR Interactions and Shift Ranges for Battery Materials

The key to NMR experiments on battery materials is to obtain the signal displacement information of different chemical substances and the interface information of electrochemical batteries. Since all the components have the signal investigated by the NMR active isotopes, the NMR spectra are usually less sensitive. In order to reduce the impact of this situation, the approach we can take is to extract the NMR coupling parameters by analyzing the range of displacements and various effects on the line shape of the different interactions and using the NMR signal line shape analysis; thus, the various internal interactions are quantified. Many paramagnetic properties of these systems are actually an advantage rather than an additional complexity. NMR can be used to extract a large amount of information about these systems. There are four main types of NMR interactions of battery materials 6.4.2.1 Shift Interactions (Nuclear Spin−Electron Spin)

The spins of the valence, conduction, and/or unpaired electrons cause different local magnetic fields and, therefore, a shift of the NMR signal from its Larmor frequency. The shielding of the nucleus by its surrounding valence electrons is called chemical shift. While the chemical shift is due to the orbital angular momentum of the paired

6.4 Solid-state Nuclear Magnetic Resonance Spectroscopy

electrons, there are two additional shift interactions due to spin magnetic moment of unpaired electrons, namely Knight and Fermi contact shift. The interaction with conduction electrons in metals or metallically conductive samples causes the so-called Knight shift, which is generally outside the range of shifts in diamagnetic materials. Because it is related to the conduction electrons, the Knight shift is a measure of the density of states at the Fermi level. The through-bond interaction of nuclear spins with time-averaged magnetic moments causes a Fermi contact shift, which is a measure of unpaired electron spin density that is transferred from the paramagnet to the nucleus under investigation. 6.4.2.2 Dipolar Coupling (Nuclear Spin−Nuclear Spin)

The magnetic dipole-dipole interaction provides a direct spectral route to determine the interatomic/internuclear distance and thus to study the crystal structure. This is beneficial if no other NMR coupling governs the shape of the signal line. Dipole coupling may occur between two identical nuclei (homonuclear) or two different NMR active nuclei (heteronuclear). This interaction also makes many multi-dimensional NMR experiments possible. 6.4.2.3 Quadrupolar Coupling

The interaction of the nuclear quadrupole moment Q (nonzero for nuclei with I > 1/2) with the electric field gradient (EFG) at the nucleus is known as quadrupole coupling. This coupling only occurs for atoms/ions in noncubic symmetry [87]. Hence, structural information on local distortions is accessible. The aforementioned NMR interactions are referred to as “Anisotropy” because the different directions of the material’s crystal can cause different local magnetic fields, resulting in different displacements [88]. This leads to the characteristic “Powder mode” in solid-state NMR measurements of powders. For chemical shifts and Knight shifts, they are caused by chemical shift anisotropy (CSA) [89] and Knight shift anisotropy (KSA), respectively. Dipole coupling can lead to the so-called Pake bistate. The multiple energy transitions possible for I > 1/2 are affected differently by the quadrupolar coupling, resulting in characteristic line shapes for the central transition (−1/2 ↔ +1/2) and satellite transition (e.g., −3/2 ↔ −1/2 and +1/2 ↔ +3/2, etc.) NMR signals. As long as the underlying coupling parameters can be extracted and the contribution of the signal analyzed, the shape of the characteristic lines generated by various NMR interactions becomes an important information source.

6.4.3

ssNMR Studies of SIBs

ssNMR is very sensitive to the detection of local environment and the dynamic information of atoms/ions, so it has received extensive attention in the characterization of alkali ion battery materials. ssNMR can be used to directly follow the workhorses of rechargeable batteries, lithium and sodium ions, and determine their mobility over a wide range of time scales. As such, ssNMR is well poised to address the three aspects that are crucial for our understanding of the EEI:

273

274

6 Advanced Characterization Techniques and Theoretical Calculation

the composition of the interphases, their 3D architecture, and their effect on ion transfer [90]. The ssNMR technology has also been applied to the research of battery cathode materials. 17 O NMR is a very valuable technology to determine the local structure of battery cathode materials. It can characterize the changes of chemical structure, phase transition, and electronic structure during charging and discharging. In the original material, O starts from oxide ions and usually combines with at least one highly paramagnetic transition metal (TM) center, leading to strong hyperfine interaction, large displacement, and fast nuclear relaxation. The concentration or broadening degree of resonance mainly depends on the intensity of hyperfine interaction, the relaxation time of 17 O nucleus, and the distribution of local environment [91]. In terms of anode research, ssNMR spectroscopy has made great progress in understanding the alloying mechanism of high-capacity antimony anodes in SIBs [92]. Phoebe K. Allan et al. [92] studied the alloying mechanism of high-capacity antimony anode for SIB by using Operando PDF analysis and super in situ 23 Na magic-angle spin solid-state nuclear magnetic resonance (MAS ssNMR) spectroscopy. In this experiment, 23 Na solid-state NMR data were collected on the Bruker avii-700 spectrometer with the operating frequency of 185.2 MHz. The rotor is spun at magic angle spinning (MAS) rate of 5∼10 kHz. The spectrum is obtained by pulse acquisition sequence, and the 90∘ pulse duration is 2.75 μs. The cycle interval is 5 seconds. A total of 2000 to 10 000 transients were added to each spectrum based on the mass and sodium level of the sample in the rotor. Ex situ 23 Na MAS ssNMR provides insight into the sodium local environments as a function of (de)sodiation (Figure 6.11). At a low sodiation level of 0.6 Na per Sb, the 23 Na NMR spectra are dominated by resonances between −40 and 20 ppm from sodium within the binder, the conductive carbon, and the SEI as discussed in Section 3.1. On further sodiation (>1 Na per Sb), an additional broad peak is observed in the 23 Na NMR, centered around 37 ppm. It is clear that this shift is different from those observed for both NaSb and Na3 Sb and indicates that the sodium exists in environments where the antimony connectivity is intermediate between NaSb and Na3 Sb. 23 Na NMR spectroscopy highlights the anomalously high sodium mobility within the Na3 Sb final sodiation product, a likely contributing factor to the exceptional rate performance of antimony compared to other alloying anodes. ssNMR often corroborates with other characterization techniques. Na2 FePO4 F is a promising cathode material for SIBs owing to its relatively high discharge voltage and excellent cycling performance. Using in-situ high-energy XRD, ectopic ssNMR, and first-principle density functional calculations, Qi Li et al. [93] investigated the long- and short-range structural evolution of Na2 FePO4 F during the cycling process. They determined the structure of mesophase Na1.5 FePO4 F by density functional theory (DFT) calculation, calculated the hyperfine shift by 23 Na NMR experimental spectrum, and mixed DFT was compared. In order to further study the structure evolution of cathode Na2 FePO4 F, the samples prepared under different charge states were characterized by ex situ ssNMR. At the same time, the seriously overlapping isotropic resonance in the rotating sideband was eliminated

6.4 Solid-state Nuclear Magnetic Resonance Spectroscopy

55 ppm

27 ppm

SEI/carbon

#Na per Sb EOS 3 2.3 1.9 1.2 EOD 0.6 0.9 1.8 2.4 EOS 3.4 3.0 2.4 1.8 1.2 0.6 Pristine electrode

42

33 ppm

Super P --> 0V 200

150

100

50

0

–50

–100

–150

δ(23Na)/ppm

Figure 6.11 Ex situ 23 Na NMR spectra (normalized) of cycled Sb electrodes at the states of charge. Spectra were recorded at 10 kHz MAS with an external field of 16.4 T. Chemical shifts of major isotropic resonances are marked. The shaded region marks where resonances from sodium within the CMC binder, the SEI, and the conductive carbon are dominant. *mark spinning sidebands. Number (#) of sodium per antimony is labeled next to each spectra, based on the calculations outlined in the supplementary information, EOS = end of sodiation, EOD = end of desodiation. Alternate lines are dashed for clarity. Source: Reproduced from Allan et al. [92]/American Chemical Society/CC BY 4.0.

by measuring two spectra at 60 and 55 kHz. During charging (Figure 6.12a–c), the resonance intensity of Na2′ in sample C6 is half of the resonance intensity of Na2 in the original sample, which means that half of the Na ions on the Na2 position are deinked in this process. At C8, the Na2′ resonance is attenuated until it disappears, indicating that at the end of charging, the Na ions in the Na2′ position are completely extracted. The discharge process (Figure 6.12d–f) showed reversible behavior compared to the charging process, which suggests that Na2 FePO4 F has

275

6 Advanced Characterization Techniques and Theoretical Calculation 60 kHz 320 Na1aʹ 345 Na1ʺ

55 kHz

–130 Na2ʹ

–180 Na2

615 Na1bʹ 1.0 C8

0.8

Na2–xFePO4F

C7

0.6 C6

0.4

C5 C4 C3

0.2

C2 C1

0 4

3

Voltage (V)

(a)

600 400 200 23

(b)

0 –200 –400

Na shift (ppm)

60 kHz

600 400 200 23

(c)

–180 Na2

55 kHz

345 Na1ʺ 615 Na1bʹ

0 –200 –400

Na shift (ppm) –130 Na2ʹ

320 Na1aʹ 440 Na1

0 D8

0.2 D7

Na2–xFePO4F

276

0.4

D6

D5

0.6

D4 D3 D2

0.8 D1 D0

1.0

4 3 2

Voltage (V)

(d)

600 400 200

(e)

23Na

0 –200 –400

shift (ppm)

600 400 200

(f)

23Na

0 –200 –400

shift (ppm)

Figure 6.12 (a) Selected different states of charge and the normalized 23 Na MAS NMR spectra of Na2 FePO4 F during first charge process at (b) 60 kHz and (c) 55 kHz, respectively. (d) Selected different states of charge and the normalized 23 Na MAS NMR spectra of Na2 FePO4 F during first discharge process at (e) 60 kHz and (f) 55 kHz, respectively. Spinning sidebands are marked with an asterisk (*). Source: Reproduced from Li et al. [93]/with permission of John Wiley & Sons.

6.4 Solid-state Nuclear Magnetic Resonance Spectroscopy

a high reversible capacity when used as a SIB cathode. The experiment found that the two crystallographically unique Na sites in the structure of Na2 FePO4 F behave differently during cycling, where the Na ions on the Na2 site are electrochemically active while those on the Na1 site are inert. This study determines the structural evolution and the electrochemical reaction mechanisms of Na2 FePO4 F in SIB. For high-rate battery materials at room temperature, traditional DFT cannot perform accurate spectral allocation. Min Lin et al. [94] combined the DFT calculation of the paramagnetic displacement with the simulation of the depth potential molecular dynamics (DPMD) to achieve a convergent distribution of Na+ in a few hundred nanoseconds, resulting in a statistically average paramagnetic displacement, the results were in good agreement with ssNMR measurements. In the rapid chemical exchange spectra of P2 -type Na2/3 (Mg1/3 Mn2/3 ) O2 , the two 23 Na shifts caused by different stacking orders of transition metal layers were revealed for the first time. This DPMD simulation assistant protocol can be widely used for ssNMR analysis in rapid chemical exchange material systems. In the 23 Na MAS NMR spectrum of P2 -Na2/3 (Mg1/3 Mn2/3 )O2 , in Figure 6.13, two unexpected isotropic shifts at 1665 and 1522 ppm were observed experimentally, which may be due to Fermi contact interaction between columnar sites in the Na interlayer and rapid chemical exchange of Na+ . The displacement of different sites was calculated by using the all electron DFT method of the mixed functional, and further combined with the 200 ns DPMD simulation analysis of the Na+ distribution, the calculation of the fast chemical exchange displacement in the paramagnetic battery

Experimental Fit Peak1 Peak2

1665 1522 298 K 1645 1506 308 K 1632

1494 318 K

1611

1475 328 K

3000

2500

2000 23Na

1500

1000

500

0

shift (ppm)

Figure 6.13 Variable-temperature 23 Na NMR VOCS spectra and fitting curves, in which the peaks of isotropic shifts are labeled with “+” and the intensities of spectra are plotted at absolute scale. Source: Reproduced from Lin et al. [94]/with permission of John Wiley & Sons.

277

278

6 Advanced Characterization Techniques and Theoretical Calculation

material was realized for the first time. As a result, the experimental 23 Na shifts of P2 -Na2/3 (Mg1/3 Mn2/3 )O2 are assigned to two stacking sequences that correspond to the space groups P63/mcm (1665 ppm) and P6322 (1522 ppm). Furthermore, the NMR deconvoluted intensity gives 41.3% of P6322, which is in excellent agreement with PXRD refinement of mixed stacking, which gives a probability of 39% for P6322. It is anticipated that the combination of ssNMR experiment, chemical shift calculation, and DPMD simulation as demonstrated in this work can be applicable for interpreting the NMR spectra of high-rate paramagnetic battery materials and other widely dynamic systems.

6.4.4

The Challenge of NMR Detection

When we conduct NMR experiments, if we change the sample conditions, the performance of the battery material will change during the electrochemical cycle. For example, nonmetallic materials can become metals or semiconductors, or new microstructures may be formed that affect the susceptibility of the electrode. These changes can affect the optimal NMR measurement conditions that have been established for blast cells. As a result, signal detectability and linear measurement may be affected, and the NMR circuit needs to be recalibrated. In addition to the shift interactions (chemical/Knight shift), the signal line shape of the battery materials can be influenced by quadrupole coupling and paramagnetic broadening. In particular, the hyperfine interaction with unpaired electrons in paramagnetic materials often gives rise to very broad resonances since many battery components are paramagnetic. Furthermore, BMS effects influence the resonance shift and line width; sample shape, packing of the material (particles), and sample orientation with respect to the static magnetic field give rise to BMS effects. The implementation of an electrochemical cell connected to an EC inside an NMR coil can cause interferences between the alternating current of the NMR and direct current of the EC circuits. In worst-case scenarios, this influences both the electrochemical performance and NMR detectability. Hence, the EC circuit needs filtering, e.g. by using low-pass filters, to prevent it from acting as an antenna that brings all the RF noise of the environment into the NMR circuit. There are still many challenges in the detection of SEI films [85]: (i) low sensitivity. Because SEI film is a micro-interface phase, ssNMR is not a highly sensitive technology. In order to obtain a spectrum with good signal-to-noise ratio, it usually requires a long time of data accumulation (≈ days), which may lead to extremely low time resolution of field experiments. Therefore, how to improve the detection sensitivity of ssNMR is an important research direction in the future. In this regard, DNP technology should be a very useful technology. (ii) Low resolution. The chemical shift of diamagnetic SEI components is generally about ±10 ppm. Even under the high-speed rotation of MAS, their signals are always overlapped and difficult to distinguish. Reasonable peak assignment through multi-core and multi-dimensional spectra under ultrahigh magnetic field will help to quantitatively understand SEI components and their post-cycle evolution.

6.5 Electrochemical Test Techniques

6.5 Electrochemical Test Techniques We generally need to evaluate the comprehensive performance of specific electrode materials according to energy/power density, conversion efficiency, and calendar lifespan. The determination of these characteristics requires the use of voltage, current, capacity and test duration, and other characteristic parameters. We need to rely on electrochemical measurements to collect these characteristic parameters directly. Current cyclic voltammetry (CV) and galvanostatic charge–discharge (GCD) are two of the most popular electrochemical testing techniques. CV records the current response at different applied voltages (potential technique), while GCD monitors the voltage evolution at a given forced current (current technique). Both methods provide information about the voltage and energy distribution of the material being measured. It is worth noting that due to the high interference (voltage or current) applied, data collection in CV and GCD measurements is usually carried out in an unbalanced state. When the disturbance is small enough, the system can be in a quasi-steady state, and the relationship between voltage and current is simplified as a linear relationship. Electrochemical impedance spectroscopy (EIS) is a general tool for understanding the complex processes of batteries. The technology can study the role of battery components such as electrodes and electrolytes, electrochemical reactions, interfaces, and interphase formation in electrochemical systems [95]. In the following sections, we will briefly introduce the most important electrochemical concepts relevant to battery research and introduce the basics of the three most commonly used electrochemical techniques-potential, current, and impedance.

6.5.1

Cyclic Voltammetry

CV is one of the most important techniques in electrochemical measurement, and linear scan/sweep CV (linear scan/sweep CV) is the most typical mode. As far as batteries are concerned, there are many complex chemical reactions involved, including solid–liquid interface, ion diffusion in electrolyte and electrode, and multiple reactions in/on electrode. In practice, most researchers can only finally identify some specific values collected from the CV curve that are helpful to understand the electrode reaction. In order to further understand CV, we need to understand its basic working principle. CV is based on linear sweep voltammetry (LSV), which is a technology to measure the current change when the potential is linearly scanned with time. Here, we define the slope of voltage change with time as the scanning rate (mV s−1 ). According to the following factors: (i) the rate of electron transfer process, (ii) the electrochemical reactivity of the material, and (iii) the measured scanning rate can show different results. The result of LSV is expressed as one of the anode or cathode peaks. Although the CV is very similar to the LSV, the difference is that it does not scan at a fixed voltage range. For example, if the voltage range is from E1 to E2, when E2 reaches the end point, it reverses again to E1. Because of this property, it is called “Cyclic,” and the current and voltage diagrams are called CV diagrams. In CV

279

6 Advanced Characterization Techniques and Theoretical Calculation

measurement, the most important parameter is the scanning rate (v). The voltage is scanned from E1 to E2, and the scanning rate (v) represents the slope of the linear voltage change during the measurement. By repeating several cycles under the same conditions, information about the voltage, reversibility, and cyclicity of redox reactions can be obtained. At the same time, the kinetics of electrochemical redox reaction can be studied by changing the scanning rate. Cometto, C et al. [96] showed how to use CV to quickly find signs of electrolyte decomposition and determine whether the resulting species are dissolved or adsorbed, leading to the formation of SEI. They identified a new electrolyte, which consists of a solution of 1M NaPF6 in EC-DMC (1 : 1 v/v ratio) to which we added three additives namely vinylene carbonate (VC), sodium (oxalate) difluoro borate (NaODFB), and tris (trimethylsilyl) phosphite TMSPi. CV results are shown in Figure 6.14. Figure 6.14a shows the CV of 100 mV/s toward low potential (black) and high potential value (red). Both CVs correspond to the first scan after electrode polishing. The inset shows the scaling of the restored wave at 20 mV/s. The CVs scanning rates are 100 mV/s (black), 200 mV/s (red), 500 mV/s (blue), 750 mV/s (magenta), 1000 mV/s (green), 1500 mV/s (navy blue) and 2000 mV/s (purple) in Figure 6.14b. The inset shows the relationship between the oxidation peak at 3.2 V and the ip and V plot of Na+ /Na. The CV of the solvent mixture shows some small reduction waves in the range of 2.2 to 1.6 V, with the main peak located at approximately V vs. Na+ /Na, becoming more pronounced when the scanning rate decreases from 100 to 20 mV/s. If the peak intensity at 3.2 V vs. Na+ /Na is plotted as a function of scanning rate (V), linear correlation is achieved, which means that the oxide is adsorbed (or simply 10 2000 mV s–1

60

–1 5 v = 100 mV s

40 Increasing v 0

0.4

Current (μA)

–10 –15

0.0

100 mV s–1

0 –20 12

–0.4

–40

10

ip (μA)

–5

Current (μA)

20 Current (μA)

280

–0.8

–20

0.5

–25 0 (a)

1

–60

v = 20 mV s–1

–1.2

2

3

6 4 2

1.0 1.5 2.0 2.5 E (V vs Na+/Na)

E (V vs. Na+/Na)

y = 1.25 + 5.28 x 2 R = 0.992

8

–80

0 0.0

0

4 (b)

1

0.5

2

1.0 1.5 v (V/s)

3

2.0

4

E (V vs. Na+/Na)

Figure 6.14 CVs of an EC/DMC (1 : 1 in volume +1 M NaPF6 ) solution at 25 ∘ C [96]/with permission of Elsevier.

6.5 Electrochemical Test Techniques

precipitated) on the electrode surface (at least in the short term) and effectively promotes the formation of anode SEI. From the above introduction, we need to determine two test parameters in CV: channel terminal voltage and scanning rate. How do we know whether terminal voltage and scanning rate are appropriate? Two standards are provided here: The electrode reaction takes place within the selected voltage range (also called electrochemical window), and the terminal current almost drops to zero; 2) The integral capacity at the selected scanning rate is close to the theoretical capacity of the electrode material. During the experiment, the slower the scanning speed is, the closer the integration capacity is to the theoretical capacity of the electrode material [97]. Therefore, the scanning speed of the battery is usually slow (within the range of 0.1–10 mV s−1 ), but in fact, it cannot be too slow. Because the equipment has detection limits, researchers have limited patience, and the battery life is limited. Most importantly, when the scanning speed is small enough, the diffusion of the charging carrier will not become a limiting step. In regards to the CV test of a coin cell, it is based on a two-electrode model where the as-prepared electrode serves as the working electrode while Na foil acts as both reference electrode and counter electrode. When redox reactions occurred at the electrode/electrolyte interface, the corresponding anodic and cathodic peaks can be observed from the CV curve. Therefore, CV can be used as a basic and versatile electrochemical tool to reveal the reaction kinetics, probe the redox couples and describe the electrochemical processes in the electrochemical cell.

6.5.2

Galvanostatic Charge–Discharge

GCD measurements record the voltage response under constant applied current. This is the most practical method to evaluate the capacity, reversibility, stability, and rate performance of electrode materials, and it is also one of the most important methods to study the electrochemical performance of materials. In the charge–discharge experiment under constant current conditions, when the control signal of current is applied, the potential is the response signal of measurement. The experiment mainly studies the change rule of potential with time. Its basic working principle is to charge and discharge the measured electrode under constant current conditions, record the change rule of its potential with time, study the charge and discharge performance of the electrode, and calculate its actual specific capacity. The charge/discharge capacity (mAh g−1 ) of coin cells can be estimated by multiplying the applied constant current density (mA g−1 ) by the charge/discharge time (hours). A repetitive loop of charge and discharge is called as a cycle. The cycle life of coin cells can be evaluated by cyclic galvanostatic charge-discharge while the rate performance can be evaluated by setting different current densities at different cycles for a coin cell. The current charging or discharging procedure is very flexible, and can be used to determine the actual capacity, evaluate the cycle stability and rate capability, study the self-discharge characteristics, and even quantify the electrode reaction kinetics.

281

282

6 Advanced Characterization Techniques and Theoretical Calculation

In order to determine the actual capacity and give the voltage distribution of some electrodes, the current is usually set to a small value to reduce polarization. By repeatedly charging/discharging the battery, we can evaluate the cycle life, which usually refers to the number of cycles when the reversible capacity decays to 80% of the initial value. Life is probably the most important parameter of rechargeable batteries, as their service life is expected to be very long. From an economic perspective, the longer the service life, the lower the average time cost, and the stronger the competitiveness of the battery. The cycle stability is related to the inherent characteristics of the electrode material and the test process, mainly charge or discharge depth. But in practice, because the low-speed cycle is very time-consuming. For batteries that cycle at 0.1 ∘ C, a full cycle takes 20 hours, and 500 cycles of testing take nearly 14 months, which is generally unrealistic for many researchers. Typically, the charge or discharge cycle is cut off when the battery reaches a cut-off voltage, and most battery test systems allow researchers to customize the termination conditions. Jayashree Pati et al. [98] used GCD to analyze the electrochemical study of honeycomb Na2 Ni2 TeO6 material as the cathode of sodium ion battery. The constant current charging and discharging multiples at different current rates are shown in Figure 6.15a, The results indicate that Na2 Ni2 TeO6 is a high-voltage positive electrode material for SiB in the range of 3.6–3.75 V. In addition, to investigate capacity retention over long cycles, they measured GCD characteristics at 0.1 and 0.5 C, respectively, as shown in figures Figure 6.15b,c. It is proved that the material is very stable.

6.5.3

Electrochemical Impedance Spectroscopy

EIS or AC impedance spectroscopy is a very useful technique for studying any electrochemical system, including electrochemical energy storage devices. EIS can be employed to diagnose the impedance of an electrochemical cell by recording the current response to an applied potential (usually between 1 and 10 mV) at varying frequencies (usually between 1 mHz and 1 MHz). A Nyquist plot is commonly used to present the spectrum for rechargeable batteries. The Nyquist plot usually consists of one or more semicircles at high frequencies (referring to charge-transfer processes) and a linear tail at low frequencies (referring to diffusion of alkali ions at the interface involving the “Warburg” impedance). Therefore, it is a useful tool to identify the charge transfer resistance, ohmic resistance, ion diffusion, double layer capacitance, etc. for the electrode [99]. AC impedance is used to study the effective resistance of linear circuit to AC excitation signal. When applied to electrochemical systems, AC impedance or EIS can reveal the kinetics of rapid interfacial reactions close to equilibrium potential. Symmetrical AC signals are applied to the electrochemical system. When the frequency is high enough, the half-wave time is short enough, and no obvious polarization occurs in the electrochemical system. In addition, cathodic and anodic reactions occur repeatedly under AC signals. There is no accumulation effect on the electrode. The electrochemical system is modulated by low-amplitude

6.5 Electrochemical Test Techniques

Potential vs Na/Na+ (V)

4.4 4.2

0.05C 0.1C 0.3C 0.5C 1C 0.05C

4.0 3.8 3.6

1 10 20 30 40 50 60 70 80 90 100 110 150 200 250 320

320 cycles @ 0.1C

3.4 3.2 3.0

(a)

320

1

0 10 20 30 40 50 60 70 80 90 0 10 20 30 40 50 60 70 80 (b) Specific capacity (mAhg–1) Specific capacity (mAhg–1)

500 cycles @ 0.5C

560

61 110 160 210 260 310 360 410 460 510 560

61

0 10 20 30 40 50 60 70 80 90 (c)

Specific capacity (mAhg–1)

Figure 6.15 GCD profiles of the Na2 Ni2 TeO6 cathode in a voltage window of 3–4.45 V (a) at different current rates, (b) at 0.1C current rate, (c) at 0.5C current rate. Source: Reproduced from Pati et al. [98]/Royal Society of Chemistry.

AC signal and can be regarded as a linear system. Then, each electrochemical process can be represented by a corresponding linear element. Therefore, the entire electrochemical system can be simulated by an equivalent circuit, which is the logical arrangement of these linear elements. By comparing the input AC signal with the output AC signal, the electrochemical kinetic parameters can be derived. The advantage of EIS is that it is a frequency-domain measurement method. In the study of electrochemical systems, impedance spectroscopy can cover a wide frequency range. Compared with other DC electrochemical methods, this method can obtain more kinetic and interface structure information. Although AC impedance has always been a traditional electrochemical analysis technology, it has only been widely used in the past few decades, thanks to revolutionary progress in fast computers, electronics, and computing algorithms. For example, using a digital frequency response analyzer and FT, a full-frequency scan can be completed in minutes. The progress of electronic technology and computing technology has promoted the application of AC impedance technology in electrochemical system analysis. However, the automation of the process and low-cost instruments have produced unexpected results for electrochemical

283

6 Advanced Characterization Techniques and Theoretical Calculation

research: the causal application of impedance technology. Unlike traditional (ancient) AC Wheatstone Bridge and X–Y oscilloscope technologies, modern commercial impedance spectrometers can provide a group of so-called data and result maps for electrochemical systems, even without being in a stable state. A new vanadium-based orthophosphate NaV3 (PO4 )3 was prepared by a simple sol-gel method and used as anode material for SIBs. To better understand the influence of interfacial property on capacity retention, EIS was performed with different cutoff voltages during the first discharge, as shown in Figure 6.16 [100]. Each spectrum contains two semicircles at high- and medium-frequency regions, which could be ascribed to the impedance of surface film (Rf ) on the electrode and charge

Discharge to 1.0 V Discharge to 0.5 V Discharge to 0.05 V Fitting results

Zʺ (Ω)

–100

–50

0 0

50

(a)

100 Zʹ (Ω) Re

Discharge to 1.0 V Discharge to 0.5 V Discharge to 0.05 V

R (Ω)

284

150

CPE1

Rr

CPE2

200

W

Rct

8

8

200

6

6

150

4

4

100

2

2

50

0

(b)

Re

0

Rf

Rct

0

Figure 6.16 (a) EIS of battery under different initial discharge voltages, (b) histogram of EIS and equivalent circuit fitting results. Source: Reproduced from Hu et al. [100]/with permission of Elsevier.

6.5 Electrochemical Test Techniques

transfer (Rct ) [100]. The EIS was simulated by a possible equivalent circuit presented in Figure 6.16, and the results were displayed in histogram of Figure 6.16b, in which Re refers to the uncompensated ohmic resistance of the working electrode. The fitting results showed that the values of Re and Rct were almost constant at various discharged states. One misconception about EIS is that it is wrongly regarded as an unsteady state technology. The disturbance signal in the unsteady state experiment is large enough to take the electrochemical system far away from the initial steady state. For example, when the battery is discharged with large current, the electrochemical system polarizes to the Tafel zone far away from the equilibrium state. However, the EIS experiment is carried out under steady state, and small amplitude disturbance signals are superimposed on the steady-state potential. Because the amplitude of the excitation signal is very small, the electrode never leaves the vicinity of the steady state. This is different from the unsteady state experiment, in which the polarization signal will inevitably lead to irreversible changes in the electrochemical system.

6.5.4

Other Electrochemical Testing Techniques

The galvanostatic intermittent titration technique (GITT) has become a widely applied electroanalytical method for the kinetic analysis of SIBs. The GITT measures the transient voltage change and open-circuit voltage (OCV) change during the charging and discharging processes using only a constant current supply and specified cut-off intervals. This procedure retrieves both thermodynamic and kinetic parameters and was first developed to examine the lithium-ion diffusion coefficient in host materials in battery electrodes [101]. The GITT can calculate diffusivity values at various states of charge (SOC) simply by voltage change, unlike the CV, which calculates the overall average diffusivity in the material by the sweep rate technique, and EIS analysis, which requires the fitting of Warburg impedance. Besides, sodium-ion diffusion rate of the coin cell can be determined by measuring CV curves at different scanning rates. The reversibility of the coin-cell can be judged by peak potential separation. Another complementary and novel technique for studying ion mobility in energy-related materials is muon spin rotation and relaxation (μ+ SR). Here, the muon is an elementary particle with a spin (S = 1/2) that is very susceptible to magnetic fields due to the muon’s exceptionally large gyromagnetic ratio, 𝛾 𝜇 = 135 MHzT−1 . This makes μ+ SR a unique and extremely sensitive (fractions of a Gauss) probe of both electronic as well as nuclear magnetic moments and fields [102]. Traditionally, the μ+ SR technique has been used extensively for studies of correlated electron physics, e.g., magnetism and superconductivity. However, lately, the scientific scope of this technique has broadened significantly. About a decade ago, the targeted application for studies of ion diffusion in energy-related materials using μ+ SR was initiated through a collaboration between academia and industry when the first systematic study of Li-ion diffusion in the archetypal

285

286

6 Advanced Characterization Techniques and Theoretical Calculation

battery cathode material Lix CoO2 was presented. Since then, μ+ SR has been utilized to study charge dynamics within a broad range of Li/Na/K-ion battery cathodes, anodes, and solid electrolytes, as well as hydrogen diffusion/desorption in H-storage materials, along with ion movement in photovoltaic materials. Hence, μ+ SR has become a novel and unique method in the area of sustainable energy materials.

6.5.5

Electrochemical Analysis of SIBs

Based on a variety of electrochemical techniques, such as galvanostatic cycling, CV, and the GITT, Le Anh Ma et al. [101] discuss the effects of Mg doping on the long-term cycle life and the diffusion coefficient, and hence the Na-ion mobility. The systematic and comprehensive study reveals that even a very small amount of Mg substitution has a strong effect on the overall performance of P2-type Na0.5 Mgx Ni0.17−x Mn0.83 O2 . Notably, the Mg substitution of x = 0.02 is the most favorable composition due to its highest capacity retention, the highest redox reversibility based on the peak potential differences, and the highest Na-ion mobility based on GITT measurements. Xiaowen Zhan et al. [103] performed extensive battery testing and characterization on Na-MH batteries, Na–NiCl2 , Na–NiBr2 , and Na–NiI2 , to investigate the cathode reaction mechanisms and their correlations with battery performance. The NaBr/Ni cathode demonstrated the best capacity retention at higher C-rates among the three by delivering the highest energy density of 174 Wh kg−1 at 100 mA (∼0.8 C), which is 2.5 and 1.9 times those of NaCl/Ni and NaI/Ni cells, respectively. The superior rate capability of NaBr/Ni is ultimately attributed to the faster sodium bromide salt dissolution kinetics in the melts at 190 ∘ C, which enables easier Br− anion transfer from the salt to the nickel particles via the melts. The findings of this work, particularly on the NaBr/Ni cathode, provide opportunities to revisit the reaction mechanism of Na-MH batteries and can inspire novel design strategies of cathode materials for high-power and high-energy Na-MH batteries. Giovanna Maresca et al. [104] studied the electrochemical stability of sodium ion electrolytes based on EMIFSI, EMITFSI, N1114FSI, N1114TFSI, N1114IM14, PIP13TFSI, and PIP14TFSI ionic liquids by voltammetry. The influence of pollutants such as water and/or molecular oxygen on the electrochemical robustness of electrolytes was studied. Preliminary CV and charge–discharge tests were carried out in Na/hard carbon and Na/α-NaMnO2 half cells using selected ionic liquid electrolytes. Preliminary electrochemical investigations, run through CV and charge–discharge measurements, have shown good compatibility between the EMIFSI and N1114FSI electrolytes and hard carbon anodes and α-NaMnO2 cathodes, matching and/or overcoming the performance obtained in standard organic solutions. Therefore, EMIFSI- and N1114FSI-based electrolytes are of interest for the realization of highly safe, reliable, and advanced sodium-ion electrochemical energy storage devices, which will be addressed in further studies.

6.6 Other Characterization Techniques

6.6 Other Characterization Techniques 6.6.1

Neutron Diffraction Technique

X-ray scattering amplitude is dependent on the number of electrons, and therefore excels at detecting atoms of elements with higher atomic numbers. The scattering amplitude of neutrons, on the other hand, has no dependence on the atomic number of the element since neutrons interact with matter via nuclear forces. For these reasons, neutron diffraction can also be used to distinguish between light atoms, isotopes, and elements of similar atomic number, making it an effective technique for the study of the movement of Li+ and Na+ ions within electrode materials [105]. It can be seen from Figure 6.17 that the nearby elements can have significantly different neutron scattering lengths, while their X-ray scattering lengths are very similar [106], such as the transition metal elements Mn, Fe, Co, and Ni. This is why neutron diffraction is often used to distinguish those nearby elements. Neutrons have no charge, their electric dipole moment is either zero or too small to measure, and they are more penetrating than charged particles. Unlike X-rays, these inherent characteristics allow neutrons to travel long distances without being scattered or absorbed. The destructive effect of neutrons on the analyzed material is also much lower than that of X-rays. The energy of an X-ray photon with a wavelength of 1.5 Å is more than 105 times that of a neutron with the same wavelength. However, neutron beam has low intensity, so neutron scattering is usually a signal-limited technology, which requires careful data 4

X-ray scattering at 10 keV X-ray scattering at 100 keV

3

3

2

2

1

D C Si O P Al S

Fe

Pb

Ni Cu Nb Ag

Gd

Au

U

W

1

Co 0

0

V H

Li

Neutron scattering length

Mn

–1 0

20

40

60

80

Neutron scattering length (10–12 cm)

X-ray scattering length (10–12 cm)

4

–1 100

Atomic number (Z)

Figure 6.17 X-ray and neutron scattering lengths of some selected elements. X-ray scattering length at photon energies of 10 and 100 keV is also plotted for comparison. Source: Ren and Zuo [106]/John Wiley & Sons.

287

288

6 Advanced Characterization Techniques and Theoretical Calculation

reduction, refinement, and fitting procedures. Nevertheless, neutron characterization technology is very powerful in providing in-depth understanding of material structure, and can provide additional information for X-ray research of materials [105, 107]. Neutron diffraction is the simplest type of coherent neutron scattering. Scattered waves can interfere with each other. Constructive interference occurs when the scattered waves reinforce each other, as a result of their being in phase (whereas waves that are out of phase cancel each other out and cause destructive interference). The conditions for constructive interference are captured by Bragg’s law, as shown in equation:n𝜆 = 2d sin 𝜃, where d is the distance between adjacent planes, 𝜆 is the wavelength of neutron radiation, and 𝜃 is the angle between the incident beam and the plane of atoms. The scattered waves from adjacent planes must travel a path difference equal to 2d sin 𝜃 to remain in phase. This occurs when the path difference is equal to an integer, n, of the wavelength of radiation, 𝜆. For single crystals, the sample must be correctly oriented to the incident neutron beam to obtain Bragg diffraction peaks for specific planes of atoms. Typically, any orientation of a polycrystalline powder sample will give Bragg diffraction peaks for all d-spacings within the crystal, as there will always be grains in the correct orientation. This method is neutron powder diffraction (NPD) and is commonly used for battery electrode materials, which consist of many randomly oriented single-crystal grains [108]. Once neutron diffraction data is retrieved, there are different methods that can be used to extract specific information about the sample. For example, Rietveld refinement is mostly used to refine lattice parameters, atomic parameters, and atomic site occupancies. It employs a method of least squares to fit a calculated diffraction pattern (from FTs) to the experimentally obtained diffraction pattern. The neutron pair distribution function (nPDF) method requires a wide range in Q-space to be measured in order to obtain an accurate FT to real space [109]. This is optimally achieved by specific instruments at pulsed neutron sources where a wide range of neutron wavelengths (thermal and epithermal), coupled with a wide solid angle coverage, afford a wide simultaneous Q range. The resulting total PDF, G(r), gives a view of the local structure with materials allowing structural analysis of liquid, amorphous, semi-crystalline, and highly defective systems. Both Rietveld refinement and nPDF analyses can identify distinct phases of SIBs electrode materials and exactly where Na+ ions reside; therefore, they can provide information about Na storage and how the structures may be improved to increase the Na+ ion occupancy. Zhaohui Ma et al. [110] reported that the diffusion path of sodium ions in Na3 [Ti2 P2 O10 F] was first shown by high-temperature NPD experiment. Temperaturedriven Na displacements indicate that sodium ions follow well-established diffusion paths within the ab plane. Furthermore, the feasibility of Na3 [Ti2 P2 O10 F] as an anode for SIBs was examined as well. Structures of the as-prepared Na3 [Ti2 P2 O10 F]. The structural refinement was performed from room-temperature (RT) NPD data. Figure 6.18 (a) shows a good agreement between the observed and calculated

6.6 Other Characterization Techniques

8000 7000

Neutron counts

6000 5000 4000 3000 2000 1000 0 –1000 –2000 20

40

60

80

100

120

140

100

120

140

2θ (°)

(a) 5500 5000 4500

Neutron counts

4000 3500 3000 2500 2000 1500 1000 500 0 –500 –1000 20

(b)

40

60

80

2θ (°)

Figure 6.18 Observed (crosses), calculated (line), and difference (bottom) neutron-diffraction patterns of the tetragonal Na3 [Ti2 P2 O10 F] at room temperature (a) and 600 ∘ C (b). Source: Ma et al. [110]/MDPI/CC BY 4.0.

NPD patterns. In order to clarify the Na distribution and how Na-ion is transported in the framework, they have undertaken a high-temperature neutron diffraction (HTND) of Na3 [Ti2 P2 O10 F] from RT to 600 ∘ C. No structural transition was observed in the temperature range studied; the crystal structures can be refined within the same tetragonal structural model. Figure 6.18b illustrates the goodness of the fit at 600∘ C. The preliminary results demonstrate Na3 [Ti2 P2 O10 F] is a promising anode material for SIBs with a high capacity, good cycling stability, and rate capability.

289

290

6 Advanced Characterization Techniques and Theoretical Calculation

6.6.2

Fourier Transform Infrared Spectrometry

The vibrational energy levels of molecular bonds are detected by Fourier transform infrared spectrometry (FTIR) spectroscopy. This technology can provide qualitative and quantitative information of solid, dissolved, and volatile analytes under non-in situ and operating conditions [111]. The IR spectrum is the direct absorption of photons in near-infrared or middle-infrared spectrum, in which there is a transition to the excited vibration mode. For it to be infrared active, vibration must involve a change in the dipole moment of the molecule or structure. FTIR can be used to measure the state of gases, liquids, and solids. The information revealed by FTIR can help to understand the binding strength of functional groups and molecular structures, especially at different charging states in battery operation, which is essential for the design of novel electrode materials with ideal electrochemical properties [112]. There are four main operating modes of FTIR spectrum: transmissivity, reflectivity, attenuated total reflection, and diffuse reflection [113]. In fact, the information that can be inferred from FTIR spectrum is complementary to that of other methods, and provides valuable information that cannot be obtained by other methods [112b]. In general, the specific capacity, cycle life, and multiplying capacity are the most important performance indicators of secondary batteries, which are closely related to the bonding performance of electrode materials. Therefore, it is of great significance to study the evolution of the binding properties of electrode materials in the electrochemical process. FTIR can reveal the bonding environment, that is, the coordination of metal ions in inorganic structures and the state of functional groups such as carbonyl groups in organic solid electrodes. Application of FTIR spectroscopy in electrolytes Cations and anions of solvates form contact ion pairs, especially in concentrated solutions. This can be reflected by the change in FTIR vibration frequency of solvent carbonyl. Understanding the electrode–electrolyte interface reaction, including the formation/ change of SEI layer, the solvation/desolvation of cations, and the insertion/ extraction of cations, is very important for improving the cycle stability and safety of secondary batteries. However, due to the inevitable secondary reaction between air gas and electrode surface substance or active material itself, the use of non-in situ FTIR may lead to the loss of important information at the electrode/electrolyte interface. Here, we need to try to study the electrochemical interface with in situ FTIR spectroscopy [112a]. In situ FTIR measurement cells were constructed in three-electrode system (Figure 6.19). Reflection spectra of samples were collected using FTIR spectrometer (FT/IR-670, JASCO) with a mercury cadmium telluride (MCT) detector. A trapezoidal CaF2 crystal with a measurable frequency range of more than 1100 cm−1 was used as an IR window. The prepared LiFePO4 thin film electrode as a working electrode was placed on the IR window. A nickel wire was used as a counter electrode, and lithium metal was used as a reference electrode [114]. Poly(hexaazatrinaphthalene) (PHATN), an environmentally benign, abundant, and sustainable polymer, is employed as a universal cathode material for

6.6 Other Characterization Techniques

Figure 6.19 Schematic illustration of in situ FTIR measurement cell. Source: Akita et al. [114]/with permission of Elsevier.

WE (LiFePO4 thin film)

Micrometer RE

CE Electrolyte solution

IR window (CaF2)

IR beam

Figure 6.20 FTIR Spectral Study of PHATN in SIBs. Source: Mao et al. [115]/John Wiley & Sons.

MCT detector

2.3 V

Transmittance (%)

C=N

Benzene ring

2000

0.5 V

Pristine

1500 Wavenumbers

1000

500

(cm–1)

these batteries. In SIBs, PHATN delivers a reversible capacity of 220 mAh g−1 at 50 mA g−1 , corresponding to the energy density of 440 Wh kg−1 , and still retains 100 mAh g−1 at 10 Ag−1 after 50 000 cycles, which is among the best performances in SIBs. Minglei Mao et al. [115] showed by DFT calculation, X-ray photoelectron spectroscopy, Raman, and FTIR results that the electron-deficient pyrazine site in PHATN is the redox center that can reversibly react with metal ions. In this study, FTIR showed that there was a strong vibration absorption peak at around 1495 cm−1 assigned to the —C=N of imine groups (Figure 6.20). Compared to benzene ring at 1615 cm−1 , the intensity of —C=N peak remarkably decreases when discharged to 1 V, which recovers after being recharged to 2.3 V, indicating the -C=N sites of imine groups are the redox-active centers of PHATN. IR is a powerful tool to enter the microscale, but it is not easy to convert from laboratory test batteries to commercial-grade batteries. In this case, nondestructive diagnostic techniques, such as optical sensing, allow real-time characterization through the spatial resolution provided by placing sensors directly inside cells. Based on recent fiber/battery innovations, C Gervilli é-Mouravieff et al. [116] reported

291

292

6 Advanced Characterization Techniques and Theoretical Calculation

the use of operando infrared fiber evanescent wave spectroscopy to monitor the electrolyte evolution of 18 650 SIBs and LIBs under real operating conditions. This method can identify chemical species and reveal the decomposition mechanism of electrolytes and additives during circulation, thus providing important insights into the growth and properties of solid electrolyte interphases, solvation kinetics, and their complex interrelationships. Since silicon dioxide (SiO2 ) optical fibers commonly used for telecommunications are limited to 0.8−2 μm transmission region, they use chalcogenide (sulfide, selenide, telluride) glass fiber instead, and its transmission window is from 3 to 13 μm [117]. This has opened up a key way for molecular recognition and monitoring of chemical kinetics occurring in cells during circulation. Therefore, this method can provide evidence of the steps related to SEI growth after cell manufacturing, as well as means to track Li (Na) inventory during the cycle (for example, calculate the charge state and health state) and corresponding ion insertion dynamics. These findings highlight the emergence of a new era of rechargeable batteries, both in terms of chemical selection and cycling conditions.

6.6.3

Raman

Raman effect was discovered by Raman and Krishnan in 1928. They discovered this effect through sunlight. This observation stems from the interaction between photons and matter. When the energy of light matches the energy difference between the two vibrational states of molecules, photons can be absorbed, and photons will be scattered when they are not absorbed. The scattering of monochromatic light under molecular irradiation can be divided into elastic scattering and inelastic scattering. In the case of elastic scattering, light does not exchange any energy with the molecule, so the scattered light exhibits the same energy or frequency as the incident light, that is, Rayleigh scattering. In inelastic scattering, there is an energy transfer between the incident light and the irradiated molecule, where the frequency of the scattered light is not equal to that of the incident light. If the energy is transferred from the photon to the vibrational energy level of the molecule, the frequency of the scattered photon will be lower than the frequency of the incident photon called Stokes Raman scattering. If a photon gains energy from the vibration of a molecule, known as a phonon, then the scattered photon will have a higher frequency than the incident photon, which is called anti-Stokes Raman scattering [118]. Raman spectroscopy is usually based on Stokes scattering because it is stronger than anti-Stokes scattering. In the Raman spectrum, the intensity of the scattered radiation is plotted as a function of its frequency shift relative to the incident light (Raman frequency shift, cm−1 ). Vibrating molecules and structures can lead to changes in polarizability, and they are Raman active. Therefore, Raman spectroscopy can be used to obtain important information about the atomic environment and local structure of molecules or crystals, such as structural disorder and strain Raman spectroscopy, which can provide a set of information about

6.7 Theoretical Calculation

structural changes caused by electrochemical reactions [112a]. New Raman bands represent new species or new phases of electrodes. The peak shift indicates the hardening or softening of vibrational phonons; The intensity of Raman peaks is often related to the existence of some structures or the polarization of phonons. The peak width is affected by the degree of structural disorder. When the analyte is adsorbed on the rough metal surface, the Raman signal will be enhanced, which is caused by the electromagnetic and charge transfer mechanisms. The enhancement of the electromagnetic field, called surface-enhanced Raman scattering (SERS) [119], is caused by the surface plasmon produced by laser on the metal surface. The charge transfer mechanism is related to the charge transfer between the metal surface and the tested sample. The charge transfer enhances the vibration. Unlike SERS, which is limited to metal substrates with nanostructures, shell isolated nanoparticles enhanced Raman spectroscopy (SHINERS) is applicable to surfaces with different materials and nanostructures [120]. SHINERS uses shell-isolated nanoparticles composed of plasma gold or silver cores and inert shells (such as SiO2 and Al2 O3 ), which can be easily manufactured, Then it is covered on the surface of analytes with different compositions and morphologies. These shell-isolated nanoparticles were used to enhance Raman vibration signals of nearby molecules [121]. Over the past years, great attention has been paid to the development of Raman spectroscopy as an in situ/operando probe of electrode materials or electrolyte in an operating Li/Na-ion battery [122]. Jia and co-workers synthesized ultrathin MoS2 -xSex nanocomposite vertically aligned on the graphene-like carbon foam (MoS2 -xSex /GF) for SIBs anode [123]. In situ Raman results displayed in Figure 6.21 demonstrated that MoS2 -xSex /GF went through a reversible intercalation rather than decomposition reaction in the initial discharge/charge process in the potential window of 0.5–3 V, which preserved the crystal structure well during cycling.

6.7 Theoretical Calculation According to the different methods of dealing with the atomic interaction force, we can divide the molecular dynamics simulation into classical molecular dynamics (CMD), ab initio molecular dynamics (AIMD), and machine-learning molecular dynamics (MLMD). The classical potential energy function describes the interaction between atoms in CMD in a specific mathematical form, while AIMD uses the ab initio quantum chemistry methods to calculate the interaction force. MLMD is a new MD method, which is produced by training ML model based on experimental data or ab initio data. Although different, the principles behind atomic motion and statistical analysis are derived by simulating the physical chemistry properties of systems that, in most cases, are the three-way molecular dynamics. From the perspective of classical mechanics, in MD simulation, the atom as the smallest unit constitutes a multi-particle system. The mass of the atom is almost in the nucleus, and the contribution of the electron cloud is small. Thus, each atom

293

6 Advanced Characterization Techniques and Theoretical Calculation

1

E2g and A1g (Mo–S) (Mo–Se) 1

A1g and E2g Intensity (a.u.)

Discharge (OCV) Initial 2.55 V 1.9 V 1.3 V 1.1 V 0.9 V 0.7 V 0.5 V

200

300

400

500

600

–1

Raman shift (cm )

(a)

(Mo–Se) 1

A1g and E2g

(Mo–S) 1 E2g and A1g

Charge

3.0 V 2.7 V 2.4 V 1.9 V 1.5 V 1.1 V 0.9 V 0.7 V 0.5 V

Intensity (a.u.)

294

200

300

400

500

600

–1

(b)

Raman shift (cm )

Figure 6.21 In situ Raman spectra of the MoS2−x Sex /GF electrode at different cut-off potentials during (a) discharge and (b) charge processes (0.5–3.0 V). Source: Jia et al. [123]/with permission of Elsevier.

can be thought of as a particle whose position is governed by Newton’s second law of motion: F = ma

(6.1)

where F, m, and a are the force, mass, and acceleration of the atom, respectively. And since F is equal to the potential gradient, a can be described further in terms of atomic velocity and coordinates, and the formula can be rewritten as: −

dv d2 r dU =m =m 2 dr dt dt

(6.2)

6.7 Theoretical Calculation

where U, R, and V represent the potential energy, coordinates, and velocity of the atom, respectively, and t is time. As long as the potential energy and initial coordinates of the atom are defined, the position and velocity of the atom over time can be derived from Eqs. (6.1) and (6.2) above. The potential energy is a function of the positions of all the atoms in the system. Because of the complexity of the function, the equation of motion has no analytic solution, so it can only be solved by numerical method. We have discussed the details of a single particle system, such as the position and velocity of particles, and we can get MD simulation. However, we usually need to judge the quality of materials by their macroscopic properties. In order to obtain macroscopic properties, we often need to use statistical mechanics. Generally speaking, the macro properties can be obtained by taking the average of the whole. The whole is a very large thermodynamic system with the same macro or thermodynamic state, but different micro states. ∑ ⟨A⟩ = Ai pi all state

where A is the value of a certain property, Ai is the value of this property at the microstate i, and pi is the probability of the corresponding microstate. According to the identical macroscopic state, ensembles can be classified into microcanonical (constant NVE), canonical (constant NVT), isothermal−isobaric (constant NPT), and grand canonical (constant 𝜇VT) ensembles, where N, V, E, T, P, and 𝜇 represent the number of particles, volume, total energy, temperature, pressure, and chemical potential, respectively. The NVE ensemble refers to an isolated system with no exchange of matter or energy with the surrounding environment, and the model box has no volume change during the simulation. The NVT ensemble has the same characteristics as the NVE ensemble, except that the system needs to exchange energy with the thermostat to maintain a constant temperature. Unlike the two sets of devices with a fixed volume and fluctuating pressure, the NPT device allows for volume changes to keep the pressure constant. μVT integration has a material exchange between the attribute system and the surrounding environment. All of these combinations are considered to be selected or combined to suit different experimental conditions. For example, a typical liquid electrolyte MD simulation requires an initial equilibrium in the NPT ensemble to obtain the size of the equilibrium model, followed by equilibrium and production run in the NVT ensemble. However, MD simulation and any experiment cannot sample all possible attribute values of the infinite system contained in the system set. The method we adopted is to determine the property of system (A) from the average value of sufficient time. 𝜏

1 A(t)dt n→∞ 𝜏 ∫0

A = lim

where 𝝉 is the average time interval and A(t) is the value of this property at a certain time. The system can evolve (or arbitrarily approach) all microscopic states over a

295

296

6 Advanced Characterization Techniques and Theoretical Calculation

long period of time, which results in a time average equal to the ensemble average as described in statistical mechanics. A can represent many properties, such as the total energy E of a system. The isochoric heat capacity CV can be further derived according to the first law of thermodynamics: ( ) 𝜕E CV = 𝜕T V In addition, assuming the absolute zero of temperature as the reference state, the entropy S: [124] T0

S=

∫0

CV dT T

From our discussion above, in order to ensure the accurate prediction of the macro properties, the simulation system should contain enough particles and simulate for enough time. However, the number of particles in the system is inevitably limited by computing power. In addition, particles near the boundary in the model may be subjected to different forces than particles inside the model. To avoid this problem, we introduce periodic boundary conditions (PBC) to avoid these obstacles. Its principle is to copy the original model box, also known as the basic unit, into the entire space to form an infinite system. When a particle moves in primitive units, all periodic objects in its copy box move in the same way. Once a particle leaves the original cell, its corresponding copy will enter through the opposite boundary. When considering a sufficiently large system, this method eliminates the surface effect. It should be noted that the use of PBC depends on the specific problem. When it is used to study the molecular behavior at the electrode−electrolyte interface, PBC can be discarded in a certain direction.

6.7.1

Classical Molecular Dynamics

According to Formula 1 and Formula 2, the motion of particles is determined by the force or potential energy function acting on the particles, which shows the interaction between the particles themselves and all other substances in the system. In CMD simulation, the potential energy function follows a fixed pattern. The force field describes the intramolecular and intermolecular potential energy of atoms as mathematical function: U = Unb + Ubond + Uangle + Utorsion where U nb represents a non-bond interaction between an atom and a U bond , U angle, and U torsion represent the bond interaction between atomic pairs, the angular interaction between atomic triplets, and the torsion interaction between atomic quadruplets, respectively. U bond , U angle, and U torsion describe the intramolecular potential energy. U nb includes not only the intermolecular potential energy of atoms in different molecules but also the intramolecular potential energy of atoms that are not covered

6.7 Theoretical Calculation

by the bond, angle, and torsion interactions. Detailed expressions of some widely used force fields, such as OPLS (optimized potentials for liquid simulations) [125], CHARMM (chemistry at Harvard macromolecular mechanics) [126], and AMBER (assisted model building with energy refinement) [127], follow the format below with some variations: ⎧ ⎡( )12 ( )6 ⎤ ⎫ qi qj ⎪ 𝜎ij ⎪ ⎢ 𝜎ij ⎥ + − Unb = ⎨4𝜀 ⎢ r rij ⎥ 4𝜋𝜀0 rij ⎬ ij ⎪ nonbonded ⎪ ⎣ ⎦ ⎩ ⎭ ∑ 2 Ubond = Kr (rij − r0 ) ∑

bonds

Uangle =

∑

K𝜃 (𝜃ijk − 𝜃0 )2

angles

Utursions =

∑ ∑

( ) K𝜑,n [1 + cos n𝜙ijkl + 𝛿n ]

tursions n

where 𝜺ij is LJ well depth, 𝝈 ij is the LJ radius, r ij is the distance between two atoms, qi and qj are the charges of two atoms, 𝜺0 is the dielectric constant of vacuum, K r , K 𝜽 , K 𝝓,n are the force constants, r 0 is the balance key value, 𝜽ijk is the equilibrium bond angle, 𝜽0 is the equilibrium angle, n is the multiplicity, 𝝓ijkl is a dihedral angle, 𝜹n is the phase. Subscripts i, j, k, and l denote different atoms. For different atomic types and interaction modes, the force field and corresponding parameters in the function can be obtained either by experiment or by quantum mechanics (QM) method. Developing empirical force fields is an ancient but still active topic in current medical research, although more than 30 force fields have been developed in the past decades. As early models evolved, new ones emerged to improve their efficiency, accuracy, and versatility. However, although many force fields have been developed, their universality and accuracy often conflict with each other. Choosing the right force field and the right model is also important to obtain reasonable MD simulation results.

6.7.2

Ab Initio Molecular Dynamics

Different from using empirical force fields in CMD, QM calculations are applied to generate the potential energy and force during AIMD simulations. Therefore, AIMD is independent of empirical parameters but depends on the accuracy of the corresponding QM method. After solving the Schrodinger equation through QM calculation, the motion of atoms and molecules can be solved according to Newton equation (Eqs. (6.1) and (6.2)) to obtain the molecular trajectory, which is the same as CMD simulation. ̂ Here is Hthe Hamiltonian of the system, 𝝋 is the wave function, and E is the energy. Including the kinetic energy and potential energy of the system. ̂ = E𝜑 H𝜑

297

298

6 Advanced Characterization Techniques and Theoretical Calculation

Despite its concise expression, the Schrödinger equation is difficult to solve, and some approximations are required for practical calculations. The Born−Oppenheimer (BO) approximation, also known as the adiabatic approximation, is widely used. It assumes that the electronic motion and nuclear motion in molecules can be separated based on much more massive nuclei than electrons. The BO approximation enables the separation of the wave function ˆ into nuclear and electronic ones, that is 𝜑 = 𝜑n 𝜑e and 𝜑 and Hamiltonian H ̂ =H ̂ n + He H For a given set of nuclear (atomic) coordinates, the original Schrödinger equation is simplified into the electronic Schrödinger equation: Ĥ e 𝜑e = Ee 𝜑e ∑ ∑ ∑ ̂ e = − i 1 𝛻2 − i,a Za + i