Energy Storage And Civilization: A Systems Approach 3030330923, 9783030330927, 9783030330934

Fossil fuels comprise the accumulation of prehistoric biomass that was energised by sunlight, and formed by earth system

531 27 5MB

English Pages 186 Year 2020

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Energy Storage And Civilization: A Systems Approach
 3030330923,  9783030330927,  9783030330934

Table of contents :
Preface......Page 6
Introduction......Page 8
Contents......Page 11
1.1 The Argument for the Primacy of Energy Storage......Page 15
1.2.2 The Crucial Advantage of Grains and Cereals......Page 17
1.2.3 Why Farm?......Page 18
1.2.4 The Paradoxes of the Neolithic Transition......Page 20
1.3.1 Energy Storage as a Pivotal Factor in Post-Palaeolithic Evolution......Page 21
1.3.2 Surplus Energy and Energy Storage......Page 23
1.3.3 Counter-Examples......Page 24
1.4.1 Salt as Food Preservative and Commodity Money......Page 25
1.4.2 Rice......Page 26
1.4.5 Metallic Commodities......Page 27
1.5 The First Complex State Societies......Page 28
1.6.1 Coal and Steam Power......Page 30
1.6.2 British Empire......Page 33
1.7.1 Crude Oil......Page 35
1.7.3 Oil and Currencies......Page 36
1.7.4 Oil and Economic Growth......Page 38
1.8.1 Electricity Compared to Gasoline......Page 40
1.9 Summary......Page 41
2.2 Photosynthesis and Oxygen......Page 43
2.3 The Formation of Fossil Fuels......Page 45
2.4.1 Introduction......Page 48
2.4.2 Hydrogen and Oxygen Combustion......Page 49
2.4.3 Methane Combustion......Page 51
2.5 Properties of Fossil Fuels......Page 52
2.6 Summary......Page 53
3.1.2 Forces and Work......Page 55
3.2 Energy in Nature......Page 57
3.3 End-Use Energy Services from a Human Perspective......Page 59
3.4 Energy Density......Page 60
3.5.2 Functional Equivalence......Page 65
3.5.3 Functional Units......Page 66
3.6.1 Commercial Fuels......Page 68
3.6.3 Interpretation of Primary Energy Results......Page 69
3.7 Summary......Page 70
4.2.1 Price Systems for Non-buffered Electricity Flows......Page 71
4.2.2 Historical Evolution of Electricity Markets......Page 72
4.3.1 Competing Objectives......Page 74
4.3.4 Misalignment of Biophysical and Monetary Value......Page 75
4.3.6 Congestion Management......Page 76
4.3.7 Storage Scale......Page 77
4.4.1 Introduction......Page 78
4.4.2 Biophysical Economics in Context......Page 79
4.4.3 Objectivist Versus Marginalist Conceptions of Value......Page 80
4.4.4 EROI as a Specialized Energy Productivity Metric......Page 82
4.5 Summary......Page 84
5.2 Approaches to Analysing EROI for Systems with IntegratedStorage......Page 85
5.3 Energy Costs of Storage......Page 88
5.5 Reversibility and Irreversibility......Page 90
5.6.1.1 Context......Page 93
5.6.1.3 P2G System......Page 94
5.6.1.4 P2G Conversion Efficiency Summary......Page 95
5.6.2.1 Context......Page 96
5.6.2.2 PHS and Battery Conversion Efficiency Summary......Page 97
5.6.3.2 Solar PV as a Consumer Product......Page 98
5.6.3.3 Dynamic EROI......Page 99
5.7 The Concept of a `Minimum EROI for Society'......Page 100
5.8 Summary......Page 101
6.2 Global Storage Context......Page 103
6.3 Historical Role of Electrical Energy Storage......Page 105
6.4.1 Dispatchability......Page 108
6.4.2 Capacity Firming......Page 109
6.4.3 Frequency Regulation......Page 110
6.4.4 Spinning Reserve......Page 112
6.5 Supply Continuity......Page 115
6.6.1 Forced Outage Rate......Page 116
6.6.2 Loss-of-Load-Expectation (LOLE)......Page 118
6.7.1 Inertia......Page 119
6.7.2 Reactive Power......Page 120
6.8 Summary......Page 122
7.1 Introduction......Page 123
7.2 Scenario Modelling Approaches......Page 124
7.3 Strategies for Balancing Supply and Demand......Page 125
7.4 Fossil Fuel Energy Storage as a Reference for Required Capacity......Page 129
7.5.1 Introduction......Page 131
7.5.2 Macroeconomic Modelling Is Not Compatible with Thermodynamic Laws......Page 132
7.5.3 Models Are Too Narrow......Page 133
7.5.4 Reliability......Page 134
7.5.5 A Layered Approach to Modelling......Page 135
7.6 Summary......Page 138
8.1.1 Introduction......Page 139
8.1.2 A Moonshot......Page 140
8.1.3 Hydrogen Carriers......Page 141
8.2.1 Water, Hydrogen and Oxygen......Page 142
8.2.2 Hydrogen Fuel Cell......Page 144
8.3 A Parallel Energy Supply Network Based on Hydrogen......Page 145
8.4 Challenges Faced in Adopting Hydrogen as a Large-Scale Energy Carrier......Page 146
8.5 EROI for Hydrogen as an Energy Carrier......Page 148
8.6 Reframing Our Understanding of Efficiency......Page 149
8.7 Summary......Page 150
9.2 The Magnitude of the Problem We Face......Page 152
9.2.1 Electric Vehicle Batteries Replacing Petroleum Storage......Page 153
9.2.2 Pumped Hydro Storage......Page 155
9.2.3 Reconciling the Storage Challenge......Page 156
9.3 History Through an Energy Storage Lens......Page 157
9.4 The Quest for a Universal Energy Storage and DistributionMedium......Page 158
9.5.1 A Global Grid as a Solution to Storage......Page 159
9.5.2 A General Purpose Technological Storage Device......Page 161
9.5.3 Large Scale Substitution of Petroleum with Biofuels......Page 162
9.5.4 Hydrogen Economy......Page 164
9.5.5 Renewable Overbuild and Demand Management......Page 165
9.5.6 Energy Descent......Page 166
9.6 Conclusions......Page 168
Bibliography......Page 170
Index......Page 184

Citation preview

Lecture Notes in Energy

Graham Palmer Joshua Floyd

Energy Storage and Civilization A Systems Approach

Lecture Notes in Energy

Lecture Notes in Energy (LNE) is a series that reports on new developments in the study of energy: from science and engineering to the analysis of energy policy. The series’ scope includes but is not limited to, renewable and green energy, nuclear, fossil fuels and carbon capture, energy systems, energy storage and harvesting, batteries and fuel cells, power systems, energy efficiency, energy in buildings, energy policy, as well as energy-related topics in economics, management and transportation. Books published in LNE are original and timely and bridge between advanced textbooks and the forefront of research. Readers of LNE include postgraduate students and non-specialist researchers wishing to gain an accessible introduction to a field of research as well as professionals and researchers with a need for an up-to-date reference book on a well-defined topic. The series publishes single- and multi-authored volumes as well as advanced textbooks. **Indexed in Scopus and EI Compendex** The Springer Energy board welcomes your book proposal. Please get in touch with the series via Anthony Doyle, Executive Editor, Springer ([email protected]).

More information about this series at http://www.springer.com/series/8874

Graham Palmer • Joshua Floyd

Energy Storage and Civilization A Systems Approach

Graham Palmer Monash University Melbourne, VIC, Australia

Joshua Floyd Melbourne, VIC, Australia

ISSN 2195-1284 ISSN 2195-1292 (electronic) Lecture Notes in Energy ISBN 978-3-030-33092-7 ISBN 978-3-030-33093-4 (eBook) https://doi.org/10.1007/978-3-030-33093-4 © Springer Nature Switzerland AG 2020 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

This is a book about energy storage in modern civilization, beginning from the Neolithic through the early modern and contemporary periods. The origin of the book traces back several years to exchanges about the inclusion of storage in netenergy analysis of renewable energy systems. At its simplest, net-energy analysis is the field of research that seeks to evaluate how much energy an energy supply process, such as crude oil production, returns after taking into account the energy needed to extract, process, upgrade, and deliver that energy. By definition, an energy supply system, whether it includes storage or not, that uses more energy than it delivers is energetically unviable—it is not worth implementing. The basis of what we present in this book is directly, or indirectly, tied in with net-energy analysis. Before starting the book, we set ourselves the goal of answering several questions–What role does energy storage play in economic systems and the ways of life they enable? How do we value the benefits and costs of storage? Is it sufficient to consider storage solely in technological terms? And can storage technologies substitute for the unique performance characteristics of petroleum fuels, with their large inherent storage built-in, at large scale? There are already many texts on the technical, engineering, and market-based aspects of energy storage. But most of these place storage within the domain of technology and assess the value of storage from a techno-economic “micro” perspective. Although this is important, we argue that a macroscale assessment of the role of storage needs to be undertaken from a biophysical economic perspective— price dynamics and market behavior alone are insufficient for understanding the roles and limits of storage at the macroscale. The book starts from fundamentals and aims to provide a foundation for a deeper understanding of the role of storage in economic systems and the societies that they support. We begin the exploration with a historical perspective, starting with an anthropological overview. How did energy storage enable particular human evolutionary pathways and forms of social organisation, and what is its relationship with the cultural norms that emerged? The focus shifts forward to the beginning of industrialisation, where we seek to provide a multidisciplinary perspective that identifies various questions for v

vi

Preface

subsequent exploration. This includes questions relating to biophysical economics, the chemistry of fossil fuel combustion, electrical engineering concepts, and lifecycle concepts. Our objective is to encourage a broader appreciation of energy storage in scenario analyses, EROI studies, and energy-economic modelling. But more importantly, we seek to challenge the dominant frame of reference that storage can be understood in technological terms alone, with its value defined solely by techno-economic analysis. We thank David Packer from Springer, and especially Charlie Hall, for encouraging us to write this book, editing the manuscript, and the valuable contributions. We are very grateful to the community of researchers and participants in the International Society of Biophysical Economics (ISBPE) and researchers from the University of Melbourne, including the Melbourne Sustainable Society Institute (MSSI) and Energy Transition Hub, for the many conversations that have influenced our thinking. Graham would especially like to thank Josh for helping to bring to life many of the ideas explored in this book through his clear writing style and capability of bridging a deep historical perspective with insights from futures research. Josh in turn wishes to express his appreciation to Graham for his generosity in inviting him along on the inquiry and learning process that lies at the heart of an endeavour such as this. Graham would also like to give special thanks to Kylie, Lachlan, and Sarah for their unwavering support and Sarah for her reviews and important feedback. Melbourne, VIC, Australia Melbourne, VIC, Australia

Graham Palmer Joshua Floyd

Introduction

Fossil fuels are derived from the accumulation and storage of prehistoric biomass. The energy embedded in biomass comprises a small portion of ancient sunlight that was initially captured by green plants, which were subsequently protected from oxidation by anaerobic conditions and sedimentation, and concentrated by Earth system dynamics. Less than 0.1% of the total organic activity was eventually converted to oil and gas. Fossil fuels can be conceptualized as stored energy stocks that can be readily processed, transported, and converted to power flows on demand. However, there are two strong drivers of a long-run transition away from fossil fuels. Firstly, the release of millions of years of stored carbon within a short time is perturbing the earth’s natural carbon cycle and enhancing the greenhouse effect with probably severe adverse climate impacts. Secondly, although significant resources remain in place, fossil fuels are of course nonrenewable and finite. The era of easily accessible oil has passed, and all fossil resources, excepting perhaps unconventional oil and gas, are near or past their peak production rates. Pre-industrial societies lived within the natural cycle, governed by the seasons, the sun’s rising and setting, and the constraints of organic agriculture. In industrialized societies, the last few generations have become accustomed to copious power on demand regardless of the season. A post-fossil fuel society will need to rely on energy released by nuclear fusion and/or fission, either directly or indirectly. By far, the most important primary energy source on a planetary scale is solar, derived ultimately from hydrogen and helium fusion reactions within the Sun. Much of the focus of this book is on how humans can utilize solar fusion indirectly, via Earth-based solar power technologies and wind power, to replicate the roles of fossil fuels. Direct use of fission, as nuclear power, is introduced but is not the main focus of this book, primarily because nuclear is not widely projected to expand significantly for the foreseeable future. A transition from energy sources comprising accumulated stocks of past sunlight to sources derived from present flows will require a replication of the energy storage services provided by geologically sequestered fossil fuels. Replicating these

vii

viii

Introduction

services has both economic and energetic costs. In light of these costs, human societies will very likely need to re-examine demand expectations, and indeed the wants and needs fulfilled by current energy demand, that have taken shape in the context of copious on-demand power. Most analyses of energy storage adopt a relatively narrow technical or economic focus. This book broadens the scope of the study by placing it within broader historical and biophysical frameworks. The focus is on the underlying physical properties of storage rather than the role of energy markets and economics, which are treated in detail in other publications. Much of the discussion refers to electricity systems, but the scope includes the provision of energy services via liquid and gaseous fuels. The discussion is grounded in the principles of biophysical economics, which adopts various tools including net-energy analysis (NEA). NEA, including the energy return on investment (EROI) metric, has the potential to supplement conventional economic and environmental models by providing an energetic valuation of energy supply. The layout of the chapters is as follows: Chapter 1 begins by exploring the role and value of storage from first principles, taking into account lessons from the historical trajectories of human societies. We explore three key cultural transitions in human societies and argue that energy storage played a defining role in all three. We identify the transitions as (1) the Neolithic transition, manifesting in the shift from foraging and hunting to agriculture and settlement; (2) the first industrial revolution, manifesting in the rise of coal-fired steam power; and (3) the Age of Oil, manifesting in the emergence petroleumfuelled mass mobility. Chapter 2 examines the origin and role of fossil fuels and explores their importance in enabling contemporary forms of social organization. The focus is on the ways in which shifts from flow-based to stock-based energy sources supported economic changes previously unavailable. Chapter 3 is a primer on energy and the fundamental forces of nature and an exploration of the implications for energy storage. Chapter 4 introduces the net-energy concept and explains how net-energy analysis and the EROI metric can contribute insights that are not obvious from market-based economic analysis alone. Chapter 5 reconceptualizes efficiency as it applies to storage. It introduces the concepts of direct and indirect efficiency in order to provide context for the importance of embodied energy and EROI. Chapter 6 considers electricity supply from an electrical engineering perspective. Electricity faces storage limitations that are unusual compared with other energy carriers, and it warrants special attention due to the system-level implications of transitioning to high penetrations of variable renewable generation. The discussion draws attention to the role of existing electricity system services, with an emphasis on how these might be replicated with renewable generation and storage. It also discusses the functions of the most common forms of storage presently deployed in modern electrical grids.

Introduction

ix

Chapter 7 considers the strengths and limitations of scenario analyses reliant on quantitative modelling for investigating the storage magnitude that may be required in high penetration renewable energy futures. Chapter 8 explores the role of hydrogen as an energy carrier and the concept of the “hydrogen economy.” Finally, Chap. 9 synthesizes the various strands of inquiry pursued over the book’s course.

Contents

1

History as a Guide to Understanding the Future of Storage . . . . . . . . . . . . 1.1 The Argument for the Primacy of Energy Storage . . . . . . . . . . . . . . . . . . . . 1.2 The Neolithic Transition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.1 Human Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.2 The Crucial Advantage of Grains and Cereals . . . . . . . . . . . . . . . . 1.2.3 Why Farm? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.4 The Paradoxes of the Neolithic Transition . . . . . . . . . . . . . . . . . . . . 1.3 A Broader Conception of Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.1 Energy Storage as a Pivotal Factor in Post-Palaeolithic Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.2 Surplus Energy and Energy Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.3 Counter-Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 The Relation Between Energy Storage and Currencies . . . . . . . . . . . . . . . 1.4.1 Salt as Food Preservative and Commodity Money . . . . . . . . . . . . 1.4.2 Rice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.3 Cattle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.4 Land and Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.5 Metallic Commodities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 The First Complex State Societies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6 The First Industrial Revolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6.1 Coal and Steam Power. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6.2 British Empire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7 The Age of Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7.1 Crude Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7.2 US Empire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7.3 Oil and Currencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7.4 Oil and Economic Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.8 Electricity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.8.1 Electricity Compared to Gasoline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.8.2 Electricity and Nuclear Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 3 3 3 4 6 7 7 9 10 11 11 12 13 13 13 14 16 16 19 21 21 22 22 24 26 26 27 27 xi

xii

Contents

2

Storage with Fossil Fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Photosynthesis and Oxygen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 The Formation of Fossil Fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Oxidation via Combustion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.2 Hydrogen and Oxygen Combustion . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.3 Methane Combustion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Properties of Fossil Fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

29 29 29 31 34 34 35 37 38 39

3

Energy Primer for Storage Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Energy and Force. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.2 Forces and Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Energy in Nature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 End-Use Energy Services from a Human Perspective. . . . . . . . . . . . . . . . . 3.4 Energy Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Energy and Power as Stocks and Flows. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5.1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5.2 Functional Equivalence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5.3 Functional Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6 Primary Energy Equivalence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6.1 Commercial Fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6.2 LCA Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6.3 Interpretation of Primary Energy Results . . . . . . . . . . . . . . . . . . . . . . 3.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

41 41 41 41 43 45 46 51 51 51 52 54 54 55 55 56

4

Comparing Market and Biophysical Approaches to Evaluating Electricity Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Electricity Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1 Price Systems for Non-buffered Electricity Flows . . . . . . . . . . . . 4.2.2 Historical Evolution of Electricity Markets . . . . . . . . . . . . . . . . . . . 4.3 Markets and Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.1 Competing Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.2 Baseload-Storage Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.3 Markets for a Storage Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.4 Misalignment of Biophysical and Monetary Value . . . . . . . . . . . 4.3.5 Storage Cannibalizes its Own Value . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.6 Congestion Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.7 Storage Scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 A Biophysical Approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.2 Biophysical Economics in Context . . . . . . . . . . . . . . . . . . . . . . . . . . . .

57 57 57 57 58 60 60 61 61 61 62 62 63 64 64 65

Contents

5

xiii

4.4.3 Objectivist Versus Marginalist Conceptions of Value . . . . . . . . . 4.4.4 EROI as a Specialized Energy Productivity Metric . . . . . . . . . . . 4.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

66 68 70

Electricity: A New Challenge for Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Approaches to Analysing EROI for Systems with Integrated Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Energy Costs of Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Implications of System Boundary for Evaluating Efficiency . . . . . . . . . 5.5 Reversibility and Irreversibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6 Three Energy Storage Case Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6.1 Case Study 1: Power to Gas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6.2 Case Study 2: Pumped Hydro and Grid-Scale Battery Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6.3 Case Study 3: Off-Grid Solar PV and Batteries . . . . . . . . . . . . . . . 5.7 The Concept of a ‘Minimum EROI for Society’ . . . . . . . . . . . . . . . . . . . . . . 5.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

71 71 71 74 76 76 79 79 82 84 86 87

6

The Role of Storage in Management of Electricity Grids . . . . . . . . . . . . . . . 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Global Storage Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Historical Role of Electrical Energy Storage . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4 Functions Essential to the Operation of Electricity Grids . . . . . . . . . . . . . 6.4.1 Dispatchability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.2 Capacity Firming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.3 Frequency Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.4 Spinning Reserve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.5 Interconnection with Adjacent Grids . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5 Supply Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6 Grid Reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6.1 Forced Outage Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6.2 Loss-of-Load-Expectation (LOLE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.7 Storage Contribution to System-Level Performance . . . . . . . . . . . . . . . . . . 6.7.1 Inertia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.7.2 Reactive Power. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

89 89 89 91 94 94 95 96 98 101 101 102 102 104 105 105 106 108

7

The Use of Scenario Analyses to Estimate the Magnitude of Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Scenario Modelling Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3 Strategies for Balancing Supply and Demand . . . . . . . . . . . . . . . . . . . . . . . . . 7.4 Fossil Fuel Energy Storage as a Reference for Required Capacity . . .

109 109 110 111 115

xiv

8

9

Contents

7.5 The Limits of Model-Based Energy Futures Investigation . . . . . . . . . . . . 7.5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5.2 Macroeconomic Modelling Is Not Compatible with Thermodynamic Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5.3 Models Are Too Narrow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5.4 Reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5.5 A Layered Approach to Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

117 117

Hydrogen as an Energy Carrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1 The Promise of the Hydrogen Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1.2 A Moonshot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1.3 Hydrogen Carriers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 The Physical Basis for Hydrogen as an Energy Carrier . . . . . . . . . . . . . . . 8.2.1 Water, Hydrogen and Oxygen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.2 Hydrogen Fuel Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3 A Parallel Energy Supply Network Based on Hydrogen . . . . . . . . . . . . . . 8.4 Challenges Faced in Adopting Hydrogen as a Large-Scale Energy Carrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.5 EROI for Hydrogen as an Energy Carrier. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.6 Reframing Our Understanding of Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

125 125 125 126 127 128 128 130 131

Synthesis and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2 The Magnitude of the Problem We Face . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2.1 Electric Vehicle Batteries Replacing Petroleum Storage. . . . . . 9.2.2 Pumped Hydro Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2.3 Reconciling the Storage Challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3 History Through an Energy Storage Lens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4 The Quest for a Universal Energy Storage and Distribution Medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.5 Scenarios for Replicating Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.5.1 A Global Grid as a Solution to Storage . . . . . . . . . . . . . . . . . . . . . . . . 9.5.2 A General Purpose Technological Storage Device . . . . . . . . . . . . 9.5.3 Large Scale Substitution of Petroleum with Biofuels . . . . . . . . . 9.5.4 Hydrogen Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.5.5 Renewable Overbuild and Demand Management . . . . . . . . . . . . . 9.5.6 Energy Descent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

139 139 139 140 142 143 144

118 119 120 121 124

132 134 135 136

145 146 146 148 149 151 152 153 155

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

Chapter 1

History as a Guide to Understanding the Future of Storage

1.1 The Argument for the Primacy of Energy Storage Human energy use is derived from sources that can be characterized as either stocks or flows. In this view, the solar energy reaching the earth is an energy flow, but the energy embodied in wood that was derived from solar energy, via photosynthesis, is an energy stock. Energy storage deals with the relationship between stocks and flows: storing energy, whether by natural or anthropic processes, involves the accumulation of flows as stocks; exploiting stored energy involves the conversion of stocks to flows. The concepts of energy flows and energy stocks are important for understanding the pivotal roles that energy carriers play in enabling economic activity. An energy carrier is any physical phenomenon that can be harnessed to transfer energy from one time and place to another. Gasoline, run-of-mine coal, and electricity are all widely recognizable energy carriers. Depending on context, water flowing in a river, compressed air or hydraulic fluid can also play the roles of energy carriers. The energy associated with gasoline can be quantified in terms of a flow rate, for example the production of an oil refinery per some time period, distributed by truck, ship or pipeline. It can also be quantified in terms of a stock—for instance, the same time period’s production from the refinery, now stored as part of a nation’s strategic fuel reserve. In other cases, the energy flow associated with a carrier is not readily accumulated as a stock. This is exemplified most prominently by electricity, but in the early years of the seventeenth century industrial revolution in England, run-of-river mills faced a similar situation: available power, being the rate at which hydraulic energy could be put to mechanical use, fluctuated with rainfall and with the seasons. In the case of electricity, accumulation of energy flows as energy stocks can be achieved on a large scale only by first converting the energy to another form, associated with a physical phenomenon of different type. These ideas will be developed further as the book proceeds.

© Springer Nature Switzerland AG 2020 G. Palmer, J. Floyd, Energy Storage and Civilization, Lecture Notes in Energy, https://doi.org/10.1007/978-3-030-33093-4_1

1

2

1 History as a Guide to Understanding the Future of Storage

In energy research, energy storage is usually discussed with reference to technological devices such as batteries, or fuels such as gasoline. Energy storage is also essential in nature. In biology, adenosine triphosphate (ATP) production during photosynthesis can be viewed as an energy storage process. The role of ATP in energizing cellular function involves continuous processes of storage and discharge. Similarly in ecology, energy storage in the form of biomass is studied in terms of the differential energy flow rates between trophic levels. All energy flows in ecology are ultimately derived from sunlight via photosynthesis. Although there are many contexts for energy storage, we want to argue that energy storage, as both a technological and natural phenomenon, has been much more significant to the development of human civilizations than usually understood. In this chapter, we will explore three key historical transitions in the ways that human societies have organized, and argue that energy storage was a defining factor of critical importance in all three. Energy storage need not have been the primary causative factor, but was an essential enabling factor. In other words, the emergence of a new energy storage process was necessary but not sufficient for allowing those transitions. We identify the transitions as (1) the Neolithic transition, manifesting in the shift from foraging and hunting to agriculture and settlement; (2) the first industrial revolution, manifesting in the rise of coal-fired steam power; and (3) the Age of Oil, manifesting in the emergence petroleum-fueled mass mobility. Although all three transitions are some of the most important cultural events in human history, there is a remarkable divergence of causative explanations, especially for both the first and second. Even the role of petroleum is often underestimated in economic history—the key drivers of modernity are generally given as technological, cultural or institutional. Given the declining fossil fuel base and adverse effects of global climate change, the aim of this book is to explore what energy storage strategies or technologies may come next. A post-fossil fuel society will need to rely on fusion and/or fission, either directly or indirectly. By far the most important primary energy source on a planetary scale is solar, derived ultimately from hydrogen and helium fusion reactions within the Sun. But substituting solar energy for all the energy services currently provided by fossil fuels will require incorporating storage into energy systems, on a very large scale. This leads to two questions—what type of storage, and what is the economic and energetic cost? We believe that starting from first principles reveals deeper insights about the past, present and future roles for energy storage. The aim of this chapter is to provide an introductory ‘macro’ perspective, before moving on to a more detailed ‘micro’ perspective on energy storage.

1.2 The Neolithic Transition

3

1.2 The Neolithic Transition 1.2.1 Human Evolution The precise evolution of modern humans continues to be debated, especially the relationship between Homo sapiens and archaic hominin species (Galway-Witham and Stringer 2018). An ancient ancestor of modern humans, Homo erectus (or upright man) lived from about 2 million years ago. Following the adaption of tree dwelling anthropoids to walking on two feet in a drying, more open, grassland environment, it is believed that primitive stone tools were developed, fire was harnessed, along with early advances in social organization, art and perhaps religion (Diamond 2005). Anatomically modern humans appeared roughly 200,000 to 150,000 years ago (Galway-Witham and Stringer 2018; Scott 2017), and dispersed out of Africa somewhere between 100,000 and 60,000 years ago (Galway-Witham and Stringer 2018; Weaver 2015). There is no definitive answer as to when fire was harnessed for cooking, but Wrangham (2009) argues that it could have been as early as the emergence of Homo erectus. Fire increased the digestible organic matter in both vegetables and meat (increasing energy availability from a given food mass), and significantly reduced the time spent on chewing foods. The Neolithic transition occurred at different times and in different places, and began roughly 10,000 years ago. It refers to the passage of human tribes from a nomadic life of foraging and hunting, to one of agriculture and settlement. Up until the transition, Palaeolithic societies lived a subsistence life of hunting and gathering, governed by the diurnal and seasonal cycles.

1.2.2 The Crucial Advantage of Grains and Cereals In the Neolithic transition, humans began domesticating plants and animals. Competition for land increased as populations increased and bands were pushed to less productive land. At the beginning of this development, protocultivation was applied to wild plants, leading to domestic species. The same process was occurring with animal raising and breeding. Initially, domestication would have occurred alongside traditional foraging, hunting, and fishing. It gradually led to specialized crop cultivation, land clearing and basic irrigation (Mazoyer and Roudart 2006; Hibbs and Olsson 2004). In the next stage of crop development, fruit and nut trees were cultivated, such as olives, figs and grapes. These reached maturity much more slowly and would have required settled village life but were a source of seasonal nutrition and variety. Further developments required advanced cultivation techniques, such as crosspollination and grafting.

4

1 History as a Guide to Understanding the Future of Storage

The crucial development was the adoption of grain and cereal farming. Cereal types differed between regions—in the Fertile Crescent, wheat and barley dominated; in China, millet and rice; in Mesoamerica, corn (Diamond 2005). The organized cultivation of grains and cereals enabled production surpluses, which in turn permitted inter-seasonal storage. In principle, inter-seasonal storage increases the food available for consumption during the leanest season. The implication of this is sometimes framed in terms of Liebig’s ‘law of the minimum’—in the absence of such inter-seasonal storage, the population level or survival capacity of a community is regulated not by the annual resources, but by the smallest quantity of food available during the leanest season (Testart et al. 1982). But in the early stage of farming, the calorific return from consuming grains, relative to the overall calories expended for sowing, harvesting and preparing, was much less than traditional foraging and hunting and it is not obvious that farming would have been worth the effort. More precisely, the energy returned on (energy) investment (EROI) is the ratio of the calorific return to the calorific expenditure for food procurement. Agriculture requires intense effort over long periods, often with variable results. In contrast, the energy surplus of gathering (Lee 1969) and hunting was often very high—contemporary studies of basic hunters find an EROI of greater than 26:1 (Glaub and Hall 2017).

1.2.3 Why Farm? Given that the returns on investment from foraging and hunting were probably greater than early agriculture, it is not obvious why early humans adopted and persisted with agriculture, and why those people survived while others were displaced. Indeed, the Neolithic transition is one of the defining events in human cultural history, yet the underlying cause is still the subject of intense debate. The prevailing view until at least the twentieth century, was that ‘agriculture was simply a practice waiting to be discovered’ (Weisdorf 2005). Reflecting the orthodox nineteenth century view, Darwin (1868, p. 309) believed in the idea of the inevitability of agriculture, noting— The savage inhabitants of each land, having found out by many and hard trials what plants were useful, or could be rendered useful by various cooking processes, would after a time take the first step in cultivation by planting them near their usual abodes.

However, by the second half of the twentieth century, anthropologists found evidence that foragers and hunters may have been better fed and healthier than comparable agriculturists, eventually leading Harlan (1992, p. 27) to pose the question— Why farm? Why give up the 20 hour work week and the fun of hunting in order to toil in the sun? Why work harder for food less nutritious and a supply more capricious? Why invite famine, plague, pestilence and crowded living conditions? Why abandon the Golden Age and take up the burden?

1.2 The Neolithic Transition

5

Many hypotheses have been proposed for the adoption of agriculture, including population pressure, overkill of large fauna, environmental factors, climatic changes, the end of the last glaciation, and others, however no explanations have proven complete (Weisdorf 2005). Perhaps it is because barley could not be ‘hunted out’ as human populations increased. Deevey (1960) compiled several population sources and plotted human population estimates over the last million years on a log-log graph. Log-log axes uncover changes that are significant in relative magnitude but may be insignificant in absolute magnitude. He argued that three population surges are evident in the historical record, as a result of: (1) tool-making; (2) agriculture; and (3) the scientific-industrial revolution. One explanation for increased population due to agriculture may have been that the end of nomadism meant less stress during pregnancy, although a grain-based diet also led to lower general health (Angel 1984). The role of food storage as an explanation has been widely explored in the anthropological literature, including James Scott’s (2017) recent book: Against the grain: a deep history of the earliest states. Where storage has been identified as an explanation for the Neolithic, it is usually identified as a straightforward response to gaps in food supply (Rowley-Conwy and Zvelebil 1989). Mumford (1967, p. 139–40) drew a connection between food and the means of storage, arguing that ‘the radical Neolithic inventions were in the realm of containers’, pointing to the development of baked clay vessels for the storage of grain, oil, wine and beer. Testart et al. (1982) connected the ideas of seasonal food supply, settlement and storage, arguing that ‘whenever resources are highly seasonal, sedentism and large-scale storage imply each other: storage brings forth sedentism, and sedentism presupposes storage. Which historically precedes the other is a chicken-and-egg question.’ Halstead and O’Shea (1989) summarized the role of storage as one of four basic strategies for responding to food shortages due to either the seasonal cycle, short term scarcity, or other natural variability in the environment. The four strategies include: 1. diversification, as a strategy to counteract scarcity of one resource by sourcing others, especially through sourcing a wider range of plant and animal species; 2. mobility, as a strategy to even out spatial discrepancies in resource availability by movement between areas of resource abundance; 3. storage, as a strategy to even out temporal discrepancies in resource availability, by ‘saving it for later’; and 4. exchange, sharing and reciprocity, as a group of strategies for playing off temporal variability in resource availability, against spatial variability in resource abundance. Exchange functions in a fashion similar to storage, in that present abundance is converted, via social transactions, into a future obligation in time of need. The idea of ‘negative reciprocity’, or theft, might also be treated as belonging in the same category as exchange.

6

1 History as a Guide to Understanding the Future of Storage

The key insight from Halstead and O’Shea is that food availability is a function of the spatial and temporal supply and demand of food. Human societies, whatever their form, depend for their viability on the availability of food for immediate consumption at sufficient rates. A society can remain viable if enough food is available right here and right now. The four strategies provide a general framework for thinking about the role of storage in regulating energy supply, which we will return to.

1.2.4 The Paradoxes of the Neolithic Transition In the contemporary world, mechanized agriculture delivers a labor productivity that is roughly two orders of magnitude greater than for pre-industrial agriculture. But prior to the mechanized age and without the benefit of hindsight, the Neolithic transition seems paradoxical at face value. Firstly, most food caught or collected by hunter-gatherers was eaten in its raw state or cooked over a fire, but grains were much more labor intensive and delivered only moderate nutritional benefit. Farming was difficult, and even following harvest, grains were exceptionally difficult to turn into something useful and digestible. They required several processing and conversion steps, starting with threshing and winnowing. Then the grains needed to be ground, but when ground by hand, required a laborious shearing motion to extract the flour by breaking the husk. Finally, the flour could be mixed with water and baked to produce bread. Prior to mechanization, these steps would have taken considerable skill and practice. Bread would have probably added some variety to traditional diets but it seems unlikely that early agriculturists pursued farming because of the nutritional value or appeal to the palate of grains. Secondly, humans are genetically hard-wired to prefer proteins, fats and sugars. On the African savanna, there was a strong evolutionary advantage to being able to conserve and store energy between meals. Fat has twice the energy density of proteins and carbohydrates, and so is the most effective biological mass to store energy. Meat is rich in proteins and fats, and humans have a natural preference to eat animal protein when it is available. Sugar was mainly found in ripe fruits, is easily digestible and rapidly converted to fat. Humans and primates evolved to have three different types of light-sensitive cone cells in their eyes rather than two, allowing us to see colorful, ripe fruit more easily against a background of green foliage. When we see ripe fruit, we have a natural desire to pick and eat it. On the other hand, there seems to be no hard-wired instinct for humans to select for grain. We have no innate tendency to reap and thresh grasses and consume them immediately. Thirdly, the substitution of a wide variety of fish, game, and foraged foods with grains and cereals significantly reduces diet diversity and leads to worse dietary outcomes. Skeletal analyses of human remains from early agricultural regions indicate declines in health associated with nutritional deficiencies and increased physical stress (Angel 1984; Latham 2013). Although grains and cereals can be

1.3 A Broader Conception of Storage

7

beneficial as part of a modern diet, they do not provide a balanced selection of amino acids (protein) and vitamins, nor do they contain the fatty acids derived from fish, game and certain plant foods. The limited anthropological evidence of farmers that lived close in time and place to hunter-gatherers suggests that farmers were generally malnourished in comparison to their hunter-gather contemporaries (Scott 2017). In an archaeological site in the Levant, in what is modern day Syria, Moore et al. (2000, pp. 369–373) identified 142 species of plant and concluded that ‘probably well over 250’ types of plant foods were consumed by hunter-gatherers. In contrast, near contemporary farmers from the same region had a much less diverse and nutritious diet. Fourthly, sedentism led to the emergence of many of the now familiar diseases of crops, people and livestock. Insects and parasites thrived in mono-cultures. Measles, mumps and diphtheria were unknown before large settlements. Several millennia after the first farming was practiced, the first encounters between peoples of the Old World and New World were often cataclysmic for New World peoples precisely because the diseases of sedentism and crowding were new to them. Sedentary populations eventually develop immunity, and diseases may circulate in sub-clinical form for many carriers. Lastly, exploration and discovery seems to be a natural aspect of the human condition—we are profoundly curious about our world. Palaeolithic societies relied on small and relatively mobile parties of advance scouts both to forage for food and to establish the likeliest routes of seasonal migration (Weaver 2015). During the conquests of the New World, there was often determined resistance by native peoples, including Native Americans, to sedentism and permanent settlements by their conquerors (Scott 2017). Mobility and dispersal seems more instinctive than settlement. The several arguments against a biological tendency towards grains and agriculture suggest that there must have been some very strong countervailing reason why humans tried, and persisted, with agriculture, and why agricultural, by displacing prior forms of social organization, came to predominate as a way of life.

1.3 A Broader Conception of Storage 1.3.1 Energy Storage as a Pivotal Factor in Post-Palaeolithic Evolution We propose here that the storage potential for its products played a decisive role in agriculture’s displacement of previous ways of life, despite its many disadvantages. Grains are hard and dry, and therefore resist spoiling. Since they are dry, they have a high calorific to weight ratio, making them both storable and readily transportable (Laudan 2015). Despite requiring much more effort per unit of nutritional gain, grains, and later, other storable food, allowed humans to do something that wasn’t

8

1 History as a Guide to Understanding the Future of Storage

practically possible before—accumulate, transport and control the dispersal of energy at large scale. Control implies the capability of using energy when and where required—regulating the availability of energy in time and space. The distinction between regulating temporal and spatial availability, and how it relates to energy storage will be drawn out in later chapters. The four strategies laid out earlier by Halstead and O’Shea (1989), including diversification, mobility, and exchange also contributed to large-scale control over the temporal and spatial availability of energy. However stored energy, in the form of non-perishable and transportable food, went further than simply supporting survival during periods of low food production. It made possible human evolutionary pathways, and particular forms of social organization, cultural norms and technological suites, that were unavailable prior to the Neolithic transition. A popular view here holds that the availability of surplus food enabled humans to pursue, as desirable ends in their own right, ways of life characterized by increasingly elaborate forms of social and political organization, larger population, expanded territorial scale, more diverse material possessions and so on. Tainter (2011) argues that this view of how large-scale, socio-politically complex societies emerge and grow is misguided. He proposes instead that these forms of organization emerge in the course of solving immediate existential problems. More importantly though, they remain viable only to the extent that their resource costs—most prominently, their energetic costs—can be met on an ongoing basis. Tainter (1988) defines socio-political complexity in terms such as the diversification of specialist social roles, and just as importantly, elaboration of the means for coordinating the diversified roles. This is characterized by features such as increasingly hierarchical structures of organization and people occupying leadership positions of extended duration. Central to his thesis is the insight that humans will implement solutions involving greater socio-political complexity, and seek to maintain such complexity, only if the benefits entailed are perceived to outweigh the costs of funding those arrangements. This always entails a trade-off. Human existence, though, is characterized by the encounter of new problems. We understand our places in the world in relation to individual and collective histories, and plan and act in the context of envisaged futures. This necessarily brings us into contact with countervailing currents of natural and other human origin. Where such currents are perceived to conflict with collective and individual desires, our situations take on a problematical character. As such, our responses to problematical situations, however necessary they may be, are always of short-term efficacy. The responses themselves help to establish the contexts for future problematical situations. According to Tainter’s (1988) central thesis, wherever humans choose to implement collective responses to collective problems, socio-political complexity will proliferate, providing that there is also collective willingness to meet the energetic costs entailed. The availability of ready means to accumulate and store surplus energy has profound implications here, for it weights the problem solving cost-benefit equation in favor of increased potential for socio-political complexity. By removing a prior

1.3 A Broader Conception of Storage

9

Fig. 1.1 Stylized graph of energy surplus (y-axis) versus storage (x-axis). The value of energy supply is determined not just by the surplus (or net) energy, but by the capacity to store that energy. In this vector-based depiction, coupling a source of surplus energy with a means of storage adds to its underlying value

constraint to exploring certain areas of socio-political ‘state-space’, evolutionary pathways that would not be viable in the absence of such storage are now on the table.

1.3.2 Surplus Energy and Energy Storage Figure 1.1 qualitatively depicts the magnitude of the ‘state-space’ opened in this way in terms of the vector of energy surplus potential and energy storage potential. Energy surpluses were necessary, but not sufficient, for supporting Neolithic sociopolitical complexification. In relation to Palaeolithic societies, surplus energy implied the availability of food to support the band or tribe in excess of the energy costs of hunting and gathering. The emergence and persistence of early agricultural societies demonstrates that the benefits of greater control over energy use outweighed the short-run costs of accumulating and storing surpluses.

10

1 History as a Guide to Understanding the Future of Storage

1.3.3 Counter-Examples Numerous examples of societies that did not adopt sedentary agriculture provide counter examples. The role of potatoes also provides an interesting contrast between nutritional and storage properties. The !Kung of the African Kalahari Desert were readily able to produce food surpluses, but lacked the means and motivation for food storage; nearly all food was consumed within two days of acquisition (Lee 1969). Instead, the !Kung ‘spent’ their surplus on rest, social life, and visiting other groups. Similarly, many Aboriginal Australians lived an abundant nomadic life, and practiced an advanced form of what Gammage (2011, p. 281) termed ‘farming without fences’. Gammage collated a staggering list of farming and conservation methods including burning, herding, hunting, weir construction, seeding, and other activities that would now be recognized as natural resource management. But none of these ‘farming’ activities were conducted as part of a tradition of sedentary agriculture. The American anthropologist Marshall Sahlins studied both the Kalahari !Kung and Aboriginal Australians, and noted that they were easily able to satisfy all basic needs and enjoyed a great deal of leisure time (Sahlins 1974; Harlan 1992). Sahlins’ anthropological studies led him to coin the phrase ‘the original affluent society’, referring to healthy and well nourished hunter-gatherers. Interestingly, Aboriginal Australians cooked a type of bush bread known as damper, but collected the grains as part of a foraging tradition (Roth 2015). A variety of native seeds, and sometimes nuts and roots, were collected and crushed, and made into a dough that was baked in the coals of a fire. Seeds varied depending on the time of year and the particular area. Despite a surplus exceeding basic needs and a rich cultural life, Aboriginal culture didn’t replicate the complex city-state cultures of classical Europe, Asia or Central and South America. Aboriginal Australians relied more on Halstead and O’Shea’s strategy of diversification and mobility, rather than seasonal storage. One reason for not adopting sedentary agriculture is that the long occupation of Australia had given Aboriginal Australians a profound understanding of the El Niño climate phenomenon (Clarke 2002). El Niño is notoriously difficult to forecast and delivers multi-year droughts, which would have spelled disaster for pre-industrial societies overly dependent on cropping. The role of potatoes as a later staple crop provides another counter-example. Potatoes belong to the root vegetable group, including carrots, parsnips and turnips. They are nutritionally superior and are much easier to prepare for consumption than grains, but they have inferior storage characteristics. Root vegetables were traditionally stored in root cellars or storage rooms, where they could be preserved in cool, dry, dark places, insulated from winter frosts and summer heat. Stored properly, they can have a shelf life of several months (Nourian et al. 2003). From a nutritional standpoint, potatoes are generally better than wheat and other grains because they are richer in vitamins and nutrients. The two vitamins that are lacking,

1.4 The Relation Between Energy Storage and Currencies

11

vitamins A and D, can be provided by dairy products. Historically, they also provided more calories per unit of land than grains (Nunn and Qian 2011). But potatoes can be stored only for a period of months after harvesting. They are also more fragile than grains, and not as amenable to bulk handling.

1.4 The Relation Between Energy Storage and Currencies 1.4.1 Salt as Food Preservative and Commodity Money The earliest forms of planned energy storage involved reserving surplus grain and cereal production for later consumption. Initially this provided societies that mastered food storage techniques with a survival advantage relative to other groups. In time though, food storage took on a more strategic role in the planning and management of social organization. This included the facilitation of exchange and trade. With the emergence of trade in surplus food between social groups, grains and cereals came to be used also as forms of commodity money—that is, money consisting of material that has value in its own right. A biophysical perspective may help to shed light on why grains, which are perishable and therefore subject to decline in value with time, could nonetheless perform effectively as a form of money. Consider here the proposition that money represents a claim on future work (Hagens 2014). That is, money that I hold today may be exchanged at some point in the future for goods or services, the production of which requires physical work and hence energy use. From a biophysical perspective, the current value of the money that I hold relates to the expected claim that I can make on future work. Considered in this light, grains and cereals may have been recognized by sellers of goods and services as having monetary value due not only to their direct nutritional value for ‘funding’ physical work, but also for what we now know as their embodied energy value. That is, a quantity of grain or cereal ‘in the hand’ also represents a quantity of past labor investment. When grain is paid to me in exchange for my labor in some other area of economic activity, I am in effect acquiring the benefit of the labor that went into the grain’s production. A notable feature of food-based commodity money comes to light here: it constitutes an abstract representation of the past work required to produce it, and, as digestible and nutritious organic matter, a concrete representation of future work potential. In accepting grain or cereal as a form of money, I am taking on the risk that its future value may decline due to spoilage. But I am also gaining, in an immediate and direct manner as a result of the exchange, effective control over a quantity of past agricultural labor that can now be directed towards ends of my choosing, including work that does not itself need necessarily to be nutritionally productive. Perhaps the most important early preservation method was salting. Salt can be used for curing foods such as beef, pork, fish, and butter. In meats, salt acts as an inhibitor of spoiling by drawing water out of the microbial cells via the process of

12

1 History as a Guide to Understanding the Future of Storage

osmosis. Such was the importance of salt for food storage, that it became a strategic commodity and was later adopted as an early form of currency (Cowen 2005). Salt was explicitly recognized for its exchange value from as early as biblical times. In the Book of Ezra from the Hebrew Bible, the act of accepting salt from a person was associated with being in that person’s service. From the Roman period, the English word salary derived from the Latin salarium, which has the root sal, or ‘salt’. Applying preservation techniques had the dual benefits of reducing the spoilage rate for storable foods, and expanding the range of foods that could be stored for extended periods. Preservation therefore had potential to amplify the value of labor invested in food production, by reducing the risk that spoilage would occur before the nutritional value of the food was realized through consumption. Preservation techniques therefore stood to increase the energy return on investment for food production, even where they required an initial increase in investment.

1.4.2 Rice As a durable food source, rice shares many of the qualities of cereals. In some parts of the Philippines, rice was extensively used as a form of currency (Einzig 1949). A manojo (handful) of palay (unthreshed rice) was a basic unit of account. Unthreshed rice could be tied up below the fruit heads. The palay served all three of the basic functions of money—the palay was a convenient unit of account, or numéraire for the valuation of a variety of goods and services; the rice could act as a store of value because it kept for 8 years or more without deterioration; and rice was convenient as a medium of exchange because it was an important daily food staple. Palay wasn’t the only form of rice that was used as currency. The Tinguian used cleaned rice instead, with the basic unit of account being a coconut shell full. Rice was also able to serve the fourth function of money, being a method of deferred payment. The existence of money is usually considered to be a precondition for the division of labor (Greenspan 1967). By granting loans in rice, repayable with interest out of the next crop, a relatively stable form of valuing labor was established. This allowed the equivalence of labor to be compared across diverse economic activities. In theory, a debtor could agree in advance to provide a certain amount of labor to fulfil an obligation. But fulfilling an obligation with a storable and transportable commodity is much more convenient for both debtor and creditor. In settling the debt, the debtor is effectively providing both embodied labor and a direct source of nutritional energy derived from sunlight. Rice also had two additional useful attributes. Since it was a food staple, demand was relatively constant for a given population. Moreover, as production required a constant amount of labor and land to produce, rice was a stable unit of account, and therefore complied with one of the attributes of modern currency—there is general agreement as to what a unit is worth, and the value is stable from day to day. The second useful attribute of rice was that it was a consistent product that could not be counterfeited.

1.4 The Relation Between Energy Storage and Currencies

13

The most sophisticated system of rice currency occurred in Japan. At the beginning of the seventeenth century, Japan estimated that the country’s wealth was equivalent to 28 million kokus, a koku being the quantity of rice to feed a person for 1 year, or about 150 kg. Other metallic currencies circulated alongside rice, including copper, gold, and silver. Debts, taxes, and wages were often denominated in rice. Rice was also accounted for and traded via a form of paper money, with large landholders issuing ‘rice notes’.

1.4.3 Cattle The place of cattle in certain societies can be usefully considered through the lens of food, value, and hence energy storage. Most food was hunted or gathered, but herding was a form of ‘autonomous’ food transport, sometimes called ‘storage on the hoof’. In regions with difficult travel routes, the food source could be autonomously transported and slaughtered when required. Many regions adopted cattle as an early form of currency. Cattle had been used in Asiatic Russia, South Africa, Columbia, ancient Germany, Ireland, Iceland, and Sweden. The English word percuniary meaning of or related to money comes from the Latin peculium for private property and is related to the Latin pecus meaning herd or flock.

1.4.4 Land and Energy Access to land is a necessary condition for harvesting the sun’s energy via crops, pasture or forests. Such was the significance of land that the eighteenth-century French Physiocrats believed that land and sunlight were the ultimate origins of wealth. Manufacturing and commerce were deemed to be ‘unproductive’ or ‘sterile’ (Murphy 1993). Given the persistence of land, it provided a basis for securing loans. Loans could be repaid in commodities or money, with or without interest. Loans backed by land as collateral were in effect secured by that land’s productive potential. The first fruits, or best produce, may have served as the basis for loan interest (Homer and Sylla 1963, p. 19).

1.4.5 Metallic Commodities The development of debt, money and storable foods enabled economic specialisation. Non-farming specialists could concentrate on making a single or group of products and trading the product of their labor. Such specialisation was only possible, though, where agricultural labor provided a sufficient surplus.

14

1 History as a Guide to Understanding the Future of Storage

Other commodity currencies emerged, especially gold, silver, and copper. These metallic commodities provided a durable and long lasting form of stored work that was much more convenient than storable foods, particularly due to the effective concentration of its embodied energy. Furthermore, they permitted individual and private accumulation that was not practical with earlier and more visible forms of currency. Producing a given unit of mass of metal typically required more energy input than producing the same unit of mass of food. The metal also occupies a smaller volume for a given mass. All three of the common currency metals are ‘group 11’ elements from the periodic table. They occur naturally in elemental form, and both gold and copper are the only transition metals that possess a distinctive non-silvery color. Their melting points are also within the temperature ranges achievable by early civilizations’ metallurgical technologies, thus permitting pyrometallurgical refinement of ores, and forming of coins by casting. Gold has the additional attribute that, traditionally, it had few practical functions other than for jewellery. Even today, only a fraction of gold production is used for industrial purposes. Gold’s economic value derives from the durability of its embodiment of past energy expenditure, termed its ‘embodied energy’. Provided a market for the metal is available, a unit of energy invested in its production today can therefore, in effect, be put aside indefinitely (Rickards and Stanczyk 2017). If I have a surplus source of energy available today, and wish to realize the value of this surplus in the future, using it to produce gold may under appropriate circumstances be a more reliable way of effectively ‘storing’ that energy than storing the energy source itself. Within the communities that created them, the gold, silver or copper that was shaped into coins, could take on a greater value than the embodiment of past energy expenditure. Both the material, and the symbol, could be communicative abstractions. Seaford (2004, p. 8) argued that ‘The connection of money with ritual is no accident. Both money and ritual mediate social relations by providing a detached, easily recognized, symbolic paradigm that depends on collective confidence and persists through everyday vicissitudes, bringing order to numerous potentially uncontrollable transactions.’

1.5 The First Complex State Societies Complex and socially stratified societies were made possible through the accumulation and control of the staple foods that represented their principal primary energy sources. Eventually tribal chiefdoms agglomerated to spawn complex state societies (see Fig. 1.2). The Bronze Age (3000 to 1200 BC) marks the beginning of the first complex states, which emerged in the Fertile Crescent and China. Scott (2017) makes the striking observation that virtually all classical states were based on grain—there is no historical account of any sago, yam or banana states. He argued that ‘wealth in the form of an appropriable, measurable, dominant grain crop’ and a ‘population that can be easily administered and mobilized’ was a defining

1.5 The First Complex State Societies

15

Fig. 1.2 Evolution of social organization. Based on Bejan et al. (2018)

feature of early states. Labor coercion was inextricably linked with the emergence of such states since members of a lower class will not automatically produce a surplus that can be appropriated by the ruling class. Slavery took on the same role in trade as other energy-based commodities. Frankopan (2015, p. 130) noted that the base currency for trading along the Silk Road during the eighth to tenth centuries had been slaves, and it was only the influx of silver into Europe that shifted the base currency. For foreign trade, it is important to distinguish between local money, such as shell or bead money, and money that was a durable embodiment of energy. Graeber (2014, p. 60) makes the point that the ‘primitive currencies’ that were used within local communities often served as a way to re-arrange relations between people, such as arranging marriages or settling disputes. These local currencies, or tokens, only needed agreement within communities in order to serve the functions of a unit of value and medium of exchange. In Egypt, the state assumed the role of storage management for wheat and other food commodities, including provision of communal silos and granaries (Janick 2014). The goddess Isis was said to have ‘discovered the fruit of both wheat and barley’. The central importance of the harvest and handling of grain is apparent in their place as favored themes in Egyptian art (see Fig. 1.3). At the same time, less complex societies emerged throughout modern day Europe and Central Asia, in the northern Indian subcontinent, and in parts of Mesoamerica. In the New Testament, written during the period roughly 50 to 100 CE, the word ‘bread’ appears in 72 verses, such as ‘Give us each day our daily bread’ (Luke 11:3), and ‘I am the bread of life’ (John 6:48).

16

1 History as a Guide to Understanding the Future of Storage

Fig. 1.3 Workers carry grain into silos while scribes register the amount. Tomb of Antefoker at Thebes, Middle Kingdom. Source Janick (2014)

Boserup (1965, p. 63) argued that all the ancient communities that applied intensive agriculture used servile labor, and that ‘where population is sparse, and fertile land abundant and uncontrolled, a social hierarchy can be maintained only by direct, personal control over the members of the lower class.’ Slavery can be conceptualized as a primitive means of converting agricultural produce into useful and directed work. Since it was difficult to prevent the lower class from finding a subsistence living beyond the boundaries of the state, the earliest states required a steady source of coerced labor to replenish and maintain a servile population. These were typically sourced from captives of war, and descendants of such captives. Boserup (1965, p. 63) noted that ‘When population becomes so dense that the land can be controlled, it becomes unnecessary to keep the lower class in personal bondage; it is sufficient to deprive the working classes of the right to be independent cultivators.’ Hence state rulership and control of energy were two sides of the same coin. Similarly, the role of usury and money lending demonstrates the relation between money, as loan interest, and a call on energy. The poor were sometimes the victims of debt peonage—the practice of lending money at interest to those who could not afford it, and then demanding that the loans be repaid with work (Graeber 2014, p. 319).

1.6 The First Industrial Revolution 1.6.1 Coal and Steam Power The dynamics that drove early agriculture were broadly paralleled in the emergence of steam power. Just as early agriculture was relatively inefficient, early steam engines were extremely inefficient in converting chemical energy to mechanical work. The first commercial steam engine was invented by Thomas Savery, and

1.6 The First Industrial Revolution

17

followed by Thomas Newcomen’s atmospheric engine. Newcomen’s early engines were an improvement, but were still only around 1% efficient. John Smeaton built many large Newcomen engines, and is credited with nudging the efficiency up to 1.4% by the late eighteenth century. James Watt’s innovations, beginning with an external condenser, raised the efficiency to 4% by the beginning of the nineteenth century (Buenstorf 2004). Improvements in cast steel and manufacturing tolerances in the second half of the nineteenth century allowed for steam boilers with fewer seams and higher temperature operation. During the nineteenth century, steam pressures increased from below 50 pounds per square inch (psi), to around 200 psi, giving a four-fold increase in power. These and other incremental improvements lifted the frontier efficiency to 16% by 1850 and to 21% by the close of the nineteenth century. The first use of steam power was to pump water from coal mines. The ready availability of coal in proximity to the steam prime mover was an important use of power in the formative period. Despite the low conversion efficiency, coal, as ‘stored sunlight’, was able to provide copious quantities of power on demand. This entailed a major leap in the control of work in both time and space. Coal, and wood, were already providing energy in the form of heat. But steam engines converted heat into useful work, initially as linear motion, and eventually as rotary motion, for provision of various energy services including pumping and transport. Prior to steam, access to inanimate sources of mechanical power relied on water and wind. Coal and steam power evolved within a virtuous cycle (see Fig. 1.4), breaking the prior ‘organic limit’ imposed by reliance on crops for capturing and concentrating solar energy flows (Wrigley 2010). Deforestation as a result of agriculture, construction, and industry had already led, by the middle of the sixteenth century, to wood shortages in Britain, and commensurate high prices. Hall and Klitgaard (2018, p. 392) used a Swedish study to estimate the EROI (see Sect. 1.2.2) of wood harvesting for charcoal production, arriving at an estimate of 4:1 when all of the

Fig. 1.4 The coal-steam-steel synergy. These developments mutually reinforced each other via positive feedback loops, most notably in Great Britain

18

1 History as a Guide to Understanding the Future of Storage

energy to feed, warm and support a foresting family was included. This was based on the availability of a sustainable forest that could be harvested in perpetuity. Despite the low conversion efficiency, coal probably surpassed the EROI of wood from the early stages of the wood-coal transition. This account of the improvement in prime mover thermal efficiency portrays the situation from a micro-economic perspective. With each incremental improvement, fuel costs for a given service reduced. At the same time, the range of economically viable applications for steam power expanded, driving macro-economic change. This shifting macro-economic picture can also be viewed in terms of the energy required to provide the fuel itself. At the advent of steam power around the start of the seventeenth century, EROI for coal was likely in the single digits only. By 1950, the EROI of US coal at the ‘mine-mouth’ was estimated to be 80:1 (Cleveland et al. 1984). Improved steam technology led to improved coal extraction productivity, which led to higher labor productivity, and so on. In this causal-loop structure, there were no natural limits to growth except eventual resource decline. Humans had discovered an extraordinary positive feedback loop that drove exponential economic growth. Technological progress in steam and steel delivered improved resource efficiency, which drove unit costs down. The subsequent lower costs drove increased demand for the end-use products. This was the original basis for the phenomenon now known as the Jevons paradox, following William Stanley Jevons’ account in The Coal Question (1866). The latent demand for cheap power was so great that the scale effects of declining cost swamped the efficiency effects of lower coal use per unit of useful energy, resulting in ever-increasing aggregate energy use. Likewise, coal steamships had largely displaced the sailing ship by the end of the nineteenth century. Sailing ships were highly developed and effective, and incurred no fuel costs, but their speed and course were subject to prevailing winds. Steam ship engineers had to resolve difficult problems with iron hulls, marine engines, and screw propellers. Early steam ships used inefficient engines and incurred high fuel requirements. Bulk handling of coal was cumbersome. Nonetheless, the benefits of continuous and rapid propulsion were so decisive that steam almost completely displaced sail for fishing and transport vessels. Ironically, in the later period of displacement, a consolation for sailing ships was as the cheapest carrier for replenishing coal refuelling depots (Rosenberg 1972). On land, British coal was centred in rural areas. The most economic and practical means of hauling coal in mines continued to be horses in many situations. The early exploitation of coal could be conceptualized, metaphorically, as an agricultural or forest harvest that didn’t require sowing or growing, and that was essentially unconstrained by the productive capacity of the land. Coal mining was immune to the seasons and the vagaries of the climate. Harvesting coal required an investment for prospecting and development, but once developed, produced a guaranteed, continuous and reliable harvest. Coal provided a source of power that rendered the need for seasonal fodder storage obsolete. Such was the impact of continuously available power that the agrarian conception of ‘nature’s time’, governed by the seasons and daily rising and setting of the sun,

1.6 The First Industrial Revolution

19

was replaced with the notion of hurried industrial time (Thompson 1967). The industrial revolution went on to fundamentally alter the organization of work, and led to what Szeman (2007) termed ‘steam capitalism’—‘the first great subsumption of agricultural labor into urban factories’. For the first time, most people became dependent on food energy supplied by others (Mitchell 2011). It is interesting to compare early industrial England with early Neolithic farming. In London, the health impacts due to coal smoke had been identified as early as 1658, and air pollution significantly worsened up until the mid-1880s (Brimblecombe 1978). From the early nineteenth century, Manchester became known as the ‘chimney of the world’ and the ‘blackening of London’ was captured by poets such as Wordsworth and Blake. But despite this, the view of the ruling class, and mostly supported by the general community, was that a ‘factory or domestic chimney belching out black smoke symbolized the creation of wealth and personal wellbeing’. Smoke was ‘the inevitable and innocuous accompaniment of the meritorious act of manufacturing’ (Mosley 2001). Whatever the health costs and loss of amenity, the benefits of industrialization evidently outweighed the costs to the industrializing English. Even the minority anti-pollution activists wanted to avoid threatening the nation’s economic prosperity, preferring to argue for ‘technological fixes’ and ‘tighter legislative controls’ (Mosley 2001, p. 117). Of course, the story of the relationship between smoke and development was more complex than a simple costbenefit calculus. But it was apparent that the Victorian community understood the role and importance of coal in economic development. The English story reflects a path dependency that would have been difficult to shift once put into motion. Population increases, urban development, and the loss of traditional ways of life tend to be one-way processes. They are also exemplars of Lotka’s (1922) ‘maximum power principle’. Natural selection operates as a maximum power organizer of systems—those sub-systems that maximize power throughput, as opposed to efficiency, come to dominate ecological niches (Hall 2004). Applied to the sphere of human economic systems, the implication is that those corporations and institutions that maximize power throughput come to dominate society (Boyle 2016). In other words, once the process of industrialization has started, it is difficult to voluntarily alter course. Similarly, it is hard to understand why Neolithic farmers subjected themselves to the intense work of agriculture for little immediate gain, yet persist they did. With coal, we see the same pattern replicated throughout history. In contemporary China, the choice is not between using, or not using, coal, but between using coal or a substitute primary energy source that can deliver the equivalent economic growth but with less pollution.

1.6.2 British Empire Just as the first complex city states emerged as a consequence of control over agricultural stocks, the British Empire emerged as a consequence of its pre-eminence

20

1 History as a Guide to Understanding the Future of Storage

in coal and steam. Britain was the undisputed leader in coal resources and steam technology. The period between 1815 and 1914 is referred to as Britain’s ‘imperial century’ and marks the periods of the first and second industrial revolutions. At its peak, Britain ruled or administered 23% of the world’s population. Britain became the leading exporter of manufactured goods and services, and the largest importer of food and industrial raw materials (World Bank 2011). The gold standard was formalized in 1816, and complemented by a token silver coinage for smaller denominations (Redish 1990). During the ‘imperial century’ the British pound sterling was the dominant international currency. In 1870, around 60% of global trade was settled in pounds sterling. Though dominating the Atlantic slave trade and sugar production during the eighteenth century, it is not altogether surprising that Britain also led the abolition of the Atlantic slave trade from 1807 until final success in 1867. Sugar, as a form of luxury food energy storage, had been the most valuable commodity produced by the Atlantic slave trade. Where slavery could be considered as a means of converting agricultural produce into useful work, coal and steam were able to substitute for much of the motive power that would otherwise have relied on animate power. The abolition was estimated to have cost Britain roughly 2% of national income annually for 60 years (Kaufmann and Pape 1999). Beyond the immorality of slavery, it is now obvious that the benefits of coal and steam exceeded any economic losses resulting from abolition. Steam became the first ‘general purpose technology’ that propelled a suite of technologies, innovations and institutions. Coal, as a source of effectively unlimited stored energy, was clearly not the only causal reason for British dominance, but we would argue, was the defining ingredient that distinguished Britain’s path from the trajectories of earlier empires. While periods of significant economic dynamism were apparent in Europe from the thirteenth century up to the first industrial revolution (Fouquet 2014), none of these translated into strong and sustained economic development. Agricultural yields in Italy doubled between the tenth and the beginning of the fourteenth century. Florence, Venice and Genoa developed strong trade links with the Near, Middle and Far East, and were crucial to the development of banking. Spain had strong trade, wool and textile production, and between 1490 and 1590, colonized Latin America. Holland had a strong maritime tradition, a large increase in urbanization, and a significant proportion of the workforce in industrial production. They were greatly subsidized by Swedish forests that were used to make their iron cannons. But none of these developments were capable of driving the strong and continuous increase in economic development that coal and steam were to provide. By the time the Carnegie Bessemer steel plant had opened in Pittsburgh, Pennsylvania in 1875, both the USA and Germany had successfully replicated the British model, and were propelled into economic powerhouses.

1.7 The Age of Oil

21

1.7 The Age of Oil 1.7.1 Crude Oil Each of the three transitions identified at the beginning of this chapter were enabled by a revolutionary advance in energy storage. The first by grains and cereals, and the second by coal. The third was the development of an oil industry that could supply a vast and increasing quantity of petroleum fuels on demand (see Fig. 1.5). Crude oil is the supreme store of energy because of its liquid state at room temperature and atmospheric pressure, high gravimetric and volumetric energy density, and wide variety of petroleum products that can be produced from a single feedstock. It provides the ultimate degree of control over the spatial and temporal distribution of energy. One of the clearest examples illustrating the superiority of petroleum fuels over ‘King Coal’ was Winston Churchill’s pre-World War I decision to convert the British battleship fleet from coal to oil fuel (Yergin 2011; Dahl 2001). At that time, Britain possessed an overwhelming surplus of coal but no oil. In the mature stage of the Age of Oil, it is difficult to appreciate the strategic risk that the conversion represented, but testifies to Churchill’s vision. As a solid fuel, coal was much more demanding to handle, both during fuelling, and on board. On the other hand, as a liquid, petroleum could be pumped and more easily managed and stored. Petroleum possess a higher gravimetric energy density, is much easier to inject into engines, and enables higher efficiency. At the time, the higher power densities achievable with oil ensured a speed advantage could be maintained over the expanding German navy. In order to secure a supply of oil, the British Government acquired a 51% share of the Anglo-Persian Oil Company and negotiated a 20 year supply contract.

Fig. 1.5 Energy timeline: Paeolithic era to twenty-first century. Dates approximate

22

1 History as a Guide to Understanding the Future of Storage

1.7.2 US Empire Just as the control of grains and cereals spawned complex city states including the Egyptian Empire, and pre-eminence in coal and steam enabled the British Empire, the forerunner in crude oil resource exploitation and related technological development was the United States. Once again, an intimate connection between energy and currency was established. The British ‘imperial century’ had ended with the outbreak of World War I. The demands of war financing had forced Britain to spend much of its financial assets overseas and to liquidate its gold reserves. By the end of World War II, the US had supplied around 6 out of the 7 billion barrels of oil consumed by the Allies over the period of the war (Miller 2001). Nearing the end of the war, the Bretton Woods agreement shifted the dominant trading currency from the pound sterling to the US dollar. National currencies were fixed to the US dollar giving convertibility to a fixed amount of gold. In Cohen’s (2018) conception of the ‘currency pyramid’, the US dollar took on the rarefied status of the ‘top currency’. The collapse of Bretton Woods in 1971 occurred at the same time as US hegemony in oil production passed its peak. Once the US became a major oil importer, it was liable for claims on its gold stores by oil exporters. By 1973, local oil price regulation, now no longer practical, also ended, marking the beginning of the modern era of oil markets (Baumeister and Kilian 2016). Already aware of its impending dependence on imported oil, the US militarized the Middle East through a Grand Bargain—the US and other Western nations would arm dictatorships in return for a guaranteed oil supply (Bove et al. 2018). Furthermore, Saudi Arabia agreed to recycle petro-dollars back into US Treasuries to finance America’s spending. Such is the importance of oil production to the global economy that ownership of oil generally complies with the principle of ‘might makes right’—regardless of the moral implications of its acquisition, even where this entails virtual theft, the global oil trading system essentially treats the possessor of the oil as the legal owner of the commodity (Wenar 2015). Britain could afford to take a moral stand on slavery because it had found an improved substitute. Yet the global economy is still to find an improved substitute for oil.

1.7.3 Oil and Currencies Since the collapse of Bretton Woods, oil and gold prices have tracked remarkably closely—see Fig. 1.6. One explanation for this close correlation is that in the absence of a metallic monetary base, oil has become a monetary commodity, taking on the role of the primary physical backing for the current monetary system (Sager 2016).

1.7 The Age of Oil

23

Fig. 1.6 Prices of oil (blue) and gold (red), 1970–2019. Source: Federal Reserve Bank of St. Louis, https://fred.stlouisfed.org/graph/?g=lHLb

Fig. 1.7 Oil-gold ratio, monthly, 1970–2019. Average of 15.8 for entire period shown as dashed line. Source: Federal Reserve Bank of St. Louis, https://fred.stlouisfed.org/graph/?g=n9Or

Using the same data from Fig. 1.6 but plotting the ratio of the gold and oil price gives the number of barrels of oil that one Troy ounce of gold will buy (see Fig. 1.7). For the period 1970 to 2019, the ratio has averaged 15.8, and remained in a band between 9 and 28 for 90% of the time. A high ratio indicates that the gold price is historically high or the oil price historically low, and vice-versa. The benefit of using a ratio in which both numerator and denominator are expressed in dollars is that the dollars cancel, circumventing price inflation and measures of CPI. The financial industry uses the ratio as a measure of economic volatility due to its sensitivity to significant political and economic events. Gold, the US dollar, and oil can be conceptualized as three legs of the same stool. Nearly all global oil is denominated in US dollars, and the economic value of gold derives from the durability of its embodiment of past energy expenditure (Rickards and Stanczyk 2017). The relation between the ‘three legs’ is underpinned

24

1 History as a Guide to Understanding the Future of Storage

by the SWIFT network (the principal international network for exchanging financial transactions). Although SWIFT is a ‘neutral global cooperative’ based in Belgium, the US can exercise its strategic authority to blacklist countries, as it has done with respect to Iran as part of a strategy of sanctions. Acquiring and hoarding gold bullion confers benefits similar to that of possessing oil reserves. The US still holds the largest gold reserve, at just over 8000 tonnes (US Bureau of the Fiscal Service 2018). But since the 2008 financial crisis, China, Russia, Turkey, Iran and India have substantially increased central bank gold holdings (Stoeferle and Valek 2018). Their motivation is likely related to deprecating the dominant role of the US dollar. In emerging markets, gold is becoming an increasingly important component of the official currency reserve, reflecting growing distrust in the dominance of the US dollar and the global monetary and credit system associated with it (Stoeferle and Valek 2018). The value of a fiat currency is contingent upon trust in the issuing authority. Grant (2014) argues that the gold price is the ‘reciprocal of the world’s faith in fiat currencies’. Unlike paper money, which can be re-valued by central bank activities, gold’s value is closely tied to the physical effort required to extract and produce the commodity. Regardless of prevailing market conditions, gold production is dependent on energy flows governed by immutable physical laws. Gold is extremely energy intensive to mine—the ore grade of gold mining is typically from about 1 to 8 grams gold per tonne of ore (i.e. 0.0001 to 0.0008%) (Mudd 2007, fig.7). Should an alternative energy storage medium to petroleum emerge that also possesses the essential characteristics of money, it may have potential to disrupt the entire international architecture that has underpinned the global financial system. After the Neolithic transition and the emergence of coal and steam, the petroleum age marks a key civilizational juncture, a departure point for a new regime of path dependency. As Tverberg (2018) has observed, if an energy carrier is viable in physical economic terms—if it delivers a net energy return after all energy investments are comprehensively accounted for—then governments will be able to tax those who produce it, rather than incentivize its production via subsidies. Following this logic, petroleum’s role as the primary energy source, and carrier, of principal global economic importance could be considered superseded when a major energy carrier of alternative form becomes the subject of taxation rather than subsidy.

1.7.4 Oil and Economic Growth Beinhocker (2007, pp. 316–7) argued that ‘all wealth is created by thermodynamically irreversible processes’ and that ‘the act of creating wealth is an act of creating order’. By definition, the organization of physical systems and processes depends fundamentally on free (or ‘available’) energy, in addition to suitable material and information flows.

1.7 The Age of Oil

25

Fig. 1.8 Index of the ratio of crude oil spot price USD (WTI) and US GDP implicit price deflator. Shaded areas indicate US recessions. In the US, oil price spikes preceded 10 of the recessions since World War II (Hamilton 2013). Source: Federal Reserve Bank of St. Louis, https://fred.stlouisfed. org/graph/?g=lHKY

The effects of oil scarcity or price shocks on broader economic performance are profound. A large empirical literature has explored the connection between oil prices and economic growth (van de Ven and Fouquet 2017; Aucott and Hall 2014). In the US, oil price spikes preceded 10 of the recessions since World War II (Hamilton 2013)—see Fig. 1.8. Despite expenditure on oil only accounting for 2–5% of national GDP throughout the global economy, every dollar of GDP is dependent on oil in some way. If that 2–5% expenditure on oil was to stop, the resulting decline in GDP would be close to 100% (Hall and Klitgaard 2018, p. 96). From a biophysical perspective, this observation seems non-controversial. Yet economic theory has long struggled to explain how a factor of production with a low factor share can contribute such a large impact. The answer is readily available from biophysical economics—energy is valuable not because it is expensive (which it is not) but because it is cheap. For the same financial outlay, you get an enormously greater quantity of work (and hence of economic production) from oil compared to, for example, hiring labor. None of the other energy sources seem to have this intimate relationship with the global economy or financial system. The global coal and liquefied natural gas (LNG) markets do not transmit price shocks to national economies the same way as oil price shocks do, even though coal and gas comprise around 45% of global primary energy combined.1

1 Historically

though, coal price shocks were significant in Britain during its ‘imperial century’ (van de Ven and Fouquet 2017).

26

1 History as a Guide to Understanding the Future of Storage

Just as money ‘embodies abstract value as a general means of payment, of exchange, of the measurement and storage of value’ (Seaford 2004, p. 2), petroleum embodies a general means for controlling the provision of energy services across space and time. Petroleum’s universality is linked to its physical properties, ubiquity, and storage attributes. Petroleum is fungible within global or national markets; it is easily traded with low transaction costs; highly divisible (retail outlets will sell you 1 litre of gasoline without a price penalty); entails low international shipping costs; and is easily stored for indefinite periods without special treatment. Although the possession of ‘money-like’ attributes may not be an essential pre-requisite for a storage substitute, these attributes provide a useful reference point for assessing a potential contender’s universality and ubiquity.

1.8 Electricity 1.8.1 Electricity Compared to Gasoline Electricity is perhaps the most difficult energy carrier to place into the evolutionary framework we outline here. In common English language usage, both gasoline and electricity are discussed as energy sources. Furthermore, the use of common energy units, such as mega-joules or kilowatt-hours, implies a functional equivalence. But electricity is a flow and energy carrier, rather than a primary energy stock. The use of common energy units belies the radical differences between gasoline and electricity with respect to storage and the organization and control of work in space and time. Whereas gasoline is an energy stock, the concept of a ‘stock of electricity’ is restricted to the limited case of DC current, stored as an electrical field in capacitors. Furthermore, as a precisely regulated flow, grid electricity is underpinned by a highly complex and integrated system of generation and distribution. Electricity systems have been described as the ‘largest machines in the world’, and subject to cascading faults which can black out entire regions or countries (see Chap. 6). On the other hand, energy conversions from electricity to end-use services achieve the highest efficiency and flexibility. Where readily and reliably available, electricity is often the most highly valued energy source. A thought experiment might illustrate the difference between oil and electricity (or equivalently, between gold and Bitcoin) more clearly. Say there is a global catastrophe and essential services are lost. But you have a farm and self-sufficiency for the short term. You have some fresh produce to trade, preferably for a store of value that can be spent later. What would you prefer to receive in exchange—a jerry can of gasoline (or a quantity of gold), or a promise to deliver the jerry can’s equivalent energy quantity via electricity (or Bitcoin equivalent in market value to the gold)? Which is more fungible? Which is more convenient as a unit of account? Which can be stored for later use at a convenient time? Which can be counted on?

1.9 Summary

27

What complex systems are needed to deliver it? It is clear that, as a physically tangible energy stock, the gasoline’s value to the end-user is far less likely to be compromised by factors outside the user’s control.

1.8.2 Electricity and Nuclear Power The role of nuclear power in this framework is that uranium, formed by nucleosynthesis, is analogous to fossil fuels in that it represents a geological stock of energy stored by natural processes, over periods vastly longer than human existence. However, nuclear fuels are currently used exclusively to power the supply of electricity and therefore do not offer a universal pathway for addressing the broader energy storage needs of human societies. To the extent that a general trend towards electrification of final energy supply continues, then uranium-235 (or uranium-238, thorium-232, or other fissile or fertile materials) could increasingly substitute for fossil fuels. This would, however, entail much greater complexity with respect to engineering, regulatory oversight, and waste management. Nuclear electricity or high temperature heat derived from fission of nuclear fuels can also be used to produce liquid fuels, which we discuss in Chap. 8. On the basis that nuclear is not widely projected to increase its share of electricity generation significantly for the foreseeable future, we give it only passing attention in this book. Both the US Energy Information Administration (EIA 2016, fig ES-6) and International Energy Agency (IEA 2017b, fig 1.12) are supportive of nuclear, but they forecast nuclear contributing only a slightly higher share of global electricity by 2040 or 2060 respectively. The high capital cost (particularly in Western Europe, North America and Japan) and intense social and political opposition are key constraints (Moriarty and Honnery 2019; Alexander and Floyd 2018). This doesn’t preclude a greater long-term role if the key challenges of cost, waste, and safety were to be overcome.

1.9 Summary We argue that energy storage, in both naturally occurring and technologically mediated forms, has played a much more significant role in the development of human civilizations than is typically recognized. Beginning with the Neolithic transition, the availability of a ready means to accumulate and store surplus energy had profound implications—it enabled socio-political pathways to evolve that would not have otherwise been viable. The Neolithic revolution was the first of three pivotal transitions widely recognized as marking the advent of new eras in human socio-economic organization. The second of these was the industrial revolution of the eighteenth and nineteenth centuries, marked by the rise to economic prominence of coal-fired steam power;

28

1 History as a Guide to Understanding the Future of Storage

and the third was the emergence of the Age of Oil, in which petroleum fuels enabled the rapid expansion and democratisation of mass mobility. While each era is marked by a distinctly different principal primary energy source, viewing the overall historical arc through an analytical lens attuned to energy storage considerations can help make sense of why changes unfolded as and when they did. Smil (2008, p. 380) argues that contemporary civilization, marked by megacities, global trade, intensive transportation, unprecedented affluence, and mass consumerism, could not have arisen without the specific properties of fossil fuels. Szeman (2013) observes that ‘despite being a concrete thing, oil animates and enables all manner of abstract categories, including freedom, mobility, growth, entrepreneurship, and the future in an essential way’. The contribution of petroleum and other fossil fuels to contemporary societies is widely appreciated, but is usually attributed principally to the enormous increase in the magnitude of energy resources that they make available. We argue that their contribution derives from two related, but distinct, factors: the surplus energy flows that are available after accounting for their extraction and processing costs; and their natural occurrence as geological stocks, or as ‘stored sunlight’. This book is devoted to exploring and evaluating fossil fuel and non-fossil fuel energy storage from an energetic and biophysical perspective. Transition to economies reliant on contemporary solar flows will require that the energy storage function of fossil fuels be reproduced at a massive scale. Over the course of the chapters that follow, we derive an estimate for this scale that implies expansion of current technologically-mediated storage capacity by three orders of magnitude, if industrialized societies are to maintain their current levels of socio-political complexity. Deploying storage at such scale will entail prodigious energy and material flows, the implication of which cannot be evaluated solely in terms of price dynamics and market behavior. We argue that substituting technological devices for the storage function inherent in fossil fuels will be much harder than commonly appreciated.

Chapter 2

Storage with Fossil Fuels

2.1 Introduction This chapter explores the origin of fossil fuels and their roles in powering modern industrial economies. The purpose of this chapter is not to examine the multitude of geopolitical misadventures and the environmental destruction due to fossil fuel extraction, nor to promote a discourse on ‘strategic realism’. Indeed, the very reason for these outcomes is the critical importance of oil to civilization. Rather, the purpose is to seek to understand why the biophysical properties of fossil fuels have been so critical to enabling contemporary society. From an energy harvesting perspective, fossil fuels and atmospheric oxygen are two sides of the same coin. Both share the same origins in earlier life on Earth. Exploring the origin and unique properties of fossil fuel energy is important for appreciating the contribution of fossil fuels to human civilization, and understanding just how difficult they are to replace. We begin with the role of photosynthesis, and the important role of oxygen (and in some cases, its absence).

2.2 Photosynthesis and Oxygen Fossil fuels and atmospheric oxygen are connected to the origins of life. All current life on Earth evolved from a single organism known as the Last Universal Common Ancestor (LUCA), a single celled form of life. LUCA was not the first cellular life, and would have been part of a larger population of early complex life, but is the last ancestor common to all life today. LUCA is believed to have lived 3.5– 4 billion years ago, and essentially seeded our planet (Lane 2002, pp. 147–148). Crucially, LUCA had already evolved the capability of using oxygen as an energy source before there was free oxygen in the atmosphere, and had developed a defence against oxidative stress generated by ultraviolet radiation. Geochemical records © Springer Nature Switzerland AG 2020 G. Palmer, J. Floyd, Energy Storage and Civilization, Lecture Notes in Energy, https://doi.org/10.1007/978-3-030-33093-4_2

29

30

2 Storage with Fossil Fuels

indicate that molecular oxygen evolution must have taken place in the precursors to cyanobacteria before around 2.8 billion years ago and that a photocatalyst capable of driving photosynthesis evolved around 3 billion years ago (Dismukes et al. 2001). Photosynthesis is the process by which sunlight (photons) splits water into its oxygen and hydrogen (proton) constituents, and uses carbon dioxide from the atmosphere to produce carbohydrates to energize plant function. The hydrogen atom bonds with carbon and oxygen to form carbohydrate molecules within the organism, and releases oxygen as a waste product. At standard temperature and pressure, two oxygen atoms will bind to form an oxygen molecule (O2 ). The respective hydrogen, carbon and oxygen atoms can recombine in different ways to release the energy that was initially captured by sunlight. The key development of photosynthesis was that by using water as a reductant, a virtually unlimited source of electrons and protons was made available (Dismukes et al. 2001). All biomass, and therefore fossil fuel, is ultimately derived via photosynthesis. Chemically, animal respiration is almost exactly the opposite of photosynthesis. It has been estimated that 99.99% of the oxygen released by plants is used up by animals, fungi and bacteria (Lane 2002). The tiny 0.01% residual represents the organic matter that is not recycled back into the ecosystem, and is instead buried under sediments. The reason for the high concentration of oxygen in the earth’s atmosphere is the roughly three-billion-year mismatch between primary production and consumption. For around the first billion years, free atmospheric oxygen was taken up by aerobic life or chemically reduced by iron, producing iron oxides that eventually precipitated to the ocean floor. These led to the deposition of banded ironstone formations between 2.4 and 1.8 billion years ago. Eventually the oxidizable minerals reached saturation, and could not capture any more oxygen. Buried organic matter and atmospheric oxygen are two sides of the same coin. Hence, the energy that is available from the combustion of fossil fuels can be attributed to two pathways, both of them driven by sunlight and photosynthesis—the carbon-hydrogen pathway that led to the formation of fossil fuels; and the molecular oxygen pathway that produced atmospheric oxygen. Their recombination allows very energy dense reactions to occur. Berner (1989, table 1) estimated that there is roughly 54,000 times the biogeochemical carbon stored as carbonate and organic carbon in rocks compared to that present in the atmosphere, terrestrial biosphere, and marine biosphere combined. The total quantity of atmospheric oxygen is ∼34 × 1018 mol O2 , equating to 1,088,000 Gt (Petsch 2013). Annual primary production of oxygen is 294 Gt, being nearly matched by autotroph, heterotroph, and soil respiration. Global fossil fuel combustion is currently responsible for removing 0.0019% of oxygen molecules every year (Scripps 2009). Broecker (1970) estimated that if all known fossil fuel reserves were burned, less than 3% of available oxygen would be depleted. Hence fossil fuel burning is not an immediate environmental concern, from the perspective of atmospheric oxygen depletion (Keeling and Shertz 1992). On the other hand, carbon dioxide is roughly 600 times less abundant in the atmosphere than oxygen, so the same processes that produce such small changes in oxygen have a major impact on atmospheric carbon dioxide concentration.

2.3 The Formation of Fossil Fuels

31

The oxygen molecule (O2 ) is unusual because, thermodynamically, oxygen is highly reactive and reactions with hydrocarbons are strongly exothermic. Oxygen can form compounds with the entire periodic table except the inert elements (Ho et al. 1995). But kinetically, oxygen is quite stable at standard temperature. The reason for the unusual kinetic stability is that O2 has two unpaired electrons whereas virtually all organic compounds have only paired electrons. Thus reactions with O2 would require at least one of the unpaired electrons to change their spin to comply with the Pauli Exclusion Principle, requiring a large amount of energy, termed the spin barrier. Iron, copper and other metals that can exist in two or more stable oxidation states bypass the spin barrier at standard temperature, forming the familiar metal oxides. The properties of molecular oxygen are a thermodynamic quirk of nature that prevent the self-combustion with carbon-hydrogen molecules that would otherwise occur in an atmosphere of 21% oxygen. This gives oxygen a unique set of properties: 1. 2. 3. 4.

Oxygen is highly reactive under certain conditions. Reactions with oxygen are strongly exothermic. Yet oxygen is kinetically stable at standard conditions. Oxygen is pervasive—the atmosphere is composed of 21% oxygen by volume. We inhabit a virtual ‘tank’, in which oxygen is freely available. 5. When we see wild fires, we identify the trees as the source of fuel but don’t think about the reactivity of oxygen. We inhabit a remarkably reactive lower atmosphere, but since air (including gaseous oxygen) is transparent, we don’t notice it.

2.3 The Formation of Fossil Fuels From the Carboniferous (350 million years ago), autotrophs produced complex organic compounds using photosynthesis. Most coals are remnants of terrestrial higher plants, and are found as solids at their site of deposition. They formed from accumulations in peat swamps, and over geological time with heat and pressure, underwent coalification. The peat was gradually transformed into brown coal (lignite and sub-bituminous), and finally to hard coal (bituminous and anthracite). In contrast, the kerogen of petroleum source beds is generally dominated by marine phytoplankton and bacteria. Being a liquid, it migrates from its place of origin into porous reservoir rocks (Tissot and Welte 1984). The organic carbon eventually contained in oil and gas is estimated to be 0.01–0.1% of primary organic production (Tissot and Welte 1984, pp. 9–10). From a human perspective, fossil fuels and atmospheric oxygen can be conceptualized as millions of years of ‘stored sunlight’ pressure-cooked and concentrated by geological processes. Dukes (2003) estimated that 1 L of gasoline required 24 ton of ancient plant matter. Using the metaphor of an electrochemical battery, Schramski et al. (2015) conceptualized the extraction and combustion of fossil fuels as the

32

2 Storage with Fossil Fuels

Fig. 2.1 Metaphor of fossil fuel use as an Earth-Space battery. Based on Schramski et al. (2015). Arrows depict conventional current

discharge of the ‘Earth-Space battery’ (see Fig. 2.1). The earth is metaphorically said to represent the cathode, and the low-temperature (2.7 K) space represents the anode. There are several important characteristics of fossil fuel combustion as a source of energy: 1. Fossil fuels represent highly concentrated energy resources. 2. Fossil fuel resources are a one-time-only inheritance. Their anthropogenic use is extraordinarily fast when considered in relation to the geological time scale of their production. 3. The primary energy is available as stored stocks, the exploitation rate of which is limited only by the means of anthropogenic appropriation. 4. The effect of anthropogenic energy conversions on the atmospheric concentration of oxygen is small except on a multi-millennial time scale. However, the contribution of carbon dioxide emissions is much more significant. 5. Oxygen comprises a substantial proportion of the molar mass of the reactants of combustion. Since only the fossil fuels need to be extracted and stored, the apparent energy density of fossil fuels is much higher than if the oxygen mass required for their complete combustion was also included. 6. Despite the expansion of hydro, nuclear, and more recently, wind and solar, fossil fuels still dominate world primary energy supply (see Figs. 2.2, 2.3, and 2.4).

Fig. 2.2 Stacked graph of global primary energy 1850–2018. Hydro, nuclear, wind and solar shown with primary energy substitution (quality) factor of 2.6 based on Grubler. Data from Grubler (1998), BP (2019)

Fig. 2.3 Log-linear graph of the global penetration of primary energy sources 1850–2018, graphed by the ratio F/(1-F) where F is share of total primary energy by fuel—see Marchetti (1977). A logistic S-curve is shown as a straight line when plotted on log-linear axes (Fisher and Pry 1971). All fuels except wind plus solar shown as 5-year smoothed average. Note vertical axis begins at 1% share of total primary energy. Hydro, nuclear, wind and solar shown with primary energy substitution (quality) factor of 2.6 based on Grubler (quality factor is discussed in Sect. 3.6). Data from Grubler (1998), BP (2019)

34

2 Storage with Fossil Fuels

Fig. 2.4 Share of primary energy for first 50 years after reaching 1% of global primary energy. Hydro, nuclear, wind and solar shown with primary energy substitution (quality) factor of 2.6 based on Grubler. Data from Grubler (1998), BP (2019)

2.4 Oxidation via Combustion 2.4.1 Introduction This section explores some basic organic chemistry in order to draw out some of the important qualities of fossil fuels. Although not a fossil fuel, hydrogen is discussed in the first example because it’s combustion involves a simple reaction that demonstrates the essential principles. Hydrogen is also explored in more detail in Chap. 8. In an engine or boiler, fossil fuels undergo a redox (reduction-oxidation) reaction with oxygen to produce gaseous products and heat. Although chemical reactions are understood as single-step processes, combustion can involve a sequence of elementary radical reactions, sometimes involving phase changes in the reactants. This applies particularly to the oxygen molecule, discussed earlier, due to the limitations of electron pairing. Furthermore, since reactions may occur on a time scale comparable with the time scale of flow processes, reactions may be complicated by chemical kinetics.

2.4 Oxidation via Combustion

35

2.4.2 Hydrogen and Oxygen Combustion One of the simplest exothermic (heat releasing) reactions is given by the combustion of hydrogen and oxygen. The oxidation of hydrogen provides a useful starting point for examining fossil fuels, and exploring the difference between heat release measured in terms of change in enthalpy of reactants and products, and the energy that is available via electrochemical reactions. During oxidation, the chemical bonds in two H-H molecules and one O=O molecule are broken, and two H-O-H molecules, comprising a total of four single covalent bonds, are made. The breaking of the H-H and O=O bonds is endothermic (needing an energy input), while the making of the H-O-H bonds is exothermic, shown as follows: 2H2 + O2 −→ 2H2 O(g)

2xH H + O O Hydrogen Oxygen

(2.1)

2x

O H

H Water

    H 0 = (2 × 432) + 494 − 4 × 460 = −482 kJ or H 0 = −241 kJ /mol of H2 In a practical heat engine, such as a steam or gasoline engine, the heat output from the combustion reaction is converted to mechanical work, with an upper limit of conversion efficiency defined by the Carnot limit. The Carnot limit arises as a consequence of the second law of thermodynamics and defines the maximum efficiency of a heat engine operating in a continuous cycle between a given hot (i.e. higher temperature) heat ‘source’ and cold (lower temperature) heat ‘sink’. The Carnot limit means that losses are unavoidable when fossil fuels are burnt to provide mechanical work. We can never make all of the chemical energy associated with these energy sources available to us in useful forms. Waste heat must be discarded in order to gain useful work. At best, we may be able to direct this waste heat towards some useful task, such as providing industrial or space heating. Atkins (2010, p. 41) puts it this way—‘Nature exerts a tax on the conversion of heat into work. Some of the energy supplied by the hot source must be paid into the surroundings as heat.’ Some practical implications of this include: 1. In addition to the heat source, all heat engines also require an effective cold sink. In some applications, the cold sink is as important to the operation and efficiency as the heat source. For instance, coal power plants typically require massive

36

2 Storage with Fossil Fuels

Fig. 2.5 Loy Yang coal-fired power station in the Latrobe Valley, Australia. The three large cooling towers with the distinctive hyperboloid shape are in the foreground. Source: authors

cooling towers (see Fig. 2.5). Although fuelled by uranium, nuclear power plants utilize the same Rankine thermodynamic cycle. Nuclear power plants are often located on the coast or adjacent to a river in order to ensure that a high capacity cold sink is available. 2. The Carnot limit determines the upper limit on efficiency, but practical efficiencies are always less than this. Enormous R&D investments are made with the aim of closing the gap between the actual efficiency achieved in practice and the theoretical maximum Carnot efficiency. Gains are usually small and incremental. In the aerospace industry, the need for higher efficiency air travel has driven jet engine manufacturers to develop higher temperature combustion, placing greater reliance on high-temperature materials and enhanced blade cooling. From Eq. 2.1, and Table 2.1, most of the mass of the reactants is accounted for by the heavier oxygen atoms. Remarkably, oxygen comprises 89% of the reactant mass. The characterization of petroleum fuels as having ‘high energy density’ implies that the energy liberated by their combustion is initially associated specifically with chemical bonds within the fuels themselves (Schmidt-Rohr 2015). From the example though, it is evident that the fuel, in this case hydrogen rather than petroleum, comprises only a minor share of the total reactant mass: most of the reactant mass is extracted from free air. The energy liberated by combustion originates from chemical bonds not only in the fuel, but in the oxygen also. It is

2.4 Oxidation via Combustion

37

Table 2.1 Hydrogen oxygen reaction Element Atomic weight Number of atoms Total molecular weight Proportion of total mass Hydrogen 1.0 4 4 11% Oxygen 16.0 2 32 89%

merely a matter of convention that specific energy values (or gravimetric energy densities) are specified in terms of the mass of fuel, or reductant. This provides a part explanation why hydrogen fuel has such high apparent gravimetric energy density.

2.4.3 Methane Combustion The general chemical equation for the complete combustion of a hydrocarbon fuel Cx Hy is given by  y y O2 = xCO2 + H2 0 Cx Hy + x + 4 2

(2.2)

where x denotes the number of carbon atoms, and y denotes the number of hydrogen atoms. Hydrocarbons in which all carbon-to-carbon bonds are only simple single bonds are called alkanes. The simplest alkane is methane which is the largest component of natural gas. During the oxidation of methane, for each methane molecule four single C-H bonds and two pairs of double O=O bonds are broken, while two pairs of double C=O bonds and four single O-H bonds are made, shown below. H H

C

H

+ 2xO

O

C

O

O

+ 2x

O H

H Methane

Carbon dioxide

H Water

Oxygen CH4 + 2O2 −→ CO2 + 2H2 O

    H 0 = (4×413)+(2×498) − (2×799)+(4×467) = −818 kJ /mol

(2.3) (2.4)

The energy released by fossil fuel combustion is commonly assumed to derive predominantly from the enthalpy difference between hydrocarbons and carbon

38

2 Storage with Fossil Fuels

Table 2.2 Methane-oxygen molar mass balance Element Hydrogen Carbon Oxygen

Atomic weight 1.0 12.0 16.0

Number of atoms 4 1 4

Total molecular weight 4 12 64

Proportion of total mass 5% 15% 80%

dioxide. However, close inspection of the enthalpy1 balance in Eq. 2.4 indicates that the majority of the energy released derives not from the enthalpy difference between methane (4 × 413) and carbon dioxide (2 × 799), as might be expected, but from the difference between oxygen (2 × 498) and water (4 × 467). The large difference between oxygen and water enthalpies is due to the relatively weak double bond of the oxygen molecule (Schmidt-Rohr 2015) (Table 2.2).

2.5 Properties of Fossil Fuels A notable feature of fuels derived from, or associated with, fossil organic compounds is the diversity of forms that they take. This diversity encompasses: states of matter (solid, liquid, gas); complexity (short chain versus long chain molecules); classes (aromatic versus aliphatic compounds); and groups (hydrocarbons versus alcohols). This enables a vast range of products that possess distinct properties that can be tailored to meet specific end-use functions. For instance, diesel fuel is combusted in compression ignition engines. These more efficient but less responsive engines dominate heavy vehicles and shipping, while gasoline-fuelled spark ignition engines dominate light duty vehicles. While the generic term ‘liquid fuels’ is applied to all the energy products of petroleum, this belies the substantial variation in properties between those products. In principle, substitution between liquid fuels is often possible, however there are multiple factors that determine the effectiveness of any given fuel in a particular end-use context. The case of jet fuel is illustrative here. Jet fuel (kerosene, including additives) needs to be sufficiently volatile to combust under the full range of conditions encountered by aircraft engines that operate from sea level to well into the stratosphere, but also sufficiently stable during refuelling and flight. It requires a freezing-point of −47 ◦ C to meet the low-temperature requirements of long, highaltitude flights, but equally, the boiling point needs to be sufficiently high to prevent vapor locks within the confines of a hot engine. The energy density of jet fuel is a fundamental determinant of aircraft performance dynamics and range (see Chap. 3).

1 Bond

enthalpy refers to average values, assuming all reactants and products are in the same physical state, data from Silberberg (2009, p. 360).

2.6 Summary

39

2.6 Summary This chapter sketched some of the important properties and characteristics of fossil fuels. These include: 1. Hydrocarbon molecules are geologically stable, representing millions of years of accumulated sunlight. 2. Earth’s oxygen rich atmosphere was generated by oxygenic photosynthesis, the biological process by which water molecules are split using the energy of sunlight. Photosynthesis maintains the concentration of atmospheric oxygen at a relatively stable level. 3. Despite oxygen’s high thermodynamic reactivity and high atmospheric concentration, the oxygen molecule is kinetically stable at standard conditions. When we see wild fires, we identify the trees as the source of fuel but don’t think about the reactivity of oxygen. We inhabit a remarkably reactive lower atmosphere, but since air (including gaseous oxygen) is transparent, we don’t notice it. 4. Hydrocarbon oxidation consumes free oxygen from the atmosphere, and therefore represents a generally available ‘free gift’ that is not available to other energy sources or converters. In the case of the combustion of methane, oxygen comprises 80% of the reactant mass. The equivalent estimates for diesel and gasoline are 77 and 78% respectively. The earth’s atmosphere metaphorically represents a planetary scale ‘tank’ to which oxygen can be freely fed and from which it can be freely extracted. The impact of anthropogenic energy conversions on atmospheric concentration of oxygen is small except on a multimillennial time scale. 5. Any other storage device or system that uses atmospheric oxygen, such as hydrogen energy or lithium-oxygen batteries, may be able to exploit this gift of nature. 6. Likewise, most of the combustion product mass is disposed of to the atmosphere as gaseous CO2 and water vapor. Both of these are relatively environmentally benign on a local scale. CO2 is a problem because of the global scale of fossil fuel use. 7. The double bond of the O2 molecule is much weaker than other double bonds or pairs of single bonds, and the formation of stronger bonds in H2 O and CO2 results in strongly exothermic reactions. Therefore, the large enthalpy change associated with combustion reactions is mostly due to the weak bond strength of the O2 molecule. 8. Under typical atmospheric conditions, hydrocarbon oxidation proceeds to completion. On combustion the chemical energy associated with the entire fuel mass is liberated as heat, and only the mass of the fuel tank remains as residual mass. This gives a comparatively higher specific energy since the mass of the fuel container and related equipment is a relatively small proportion of the total system mass including the fuel itself. This can be contrasted with batteries and other methods of storing energy, for which chemical reactants account for only a minor share of the total system mass.

40

2 Storage with Fossil Fuels

9. The conversion of chemical energy to work via heat engines is subject to a maximum efficiency ceiling, the Carnot efficiency, which is a function of the difference between the temperature at which input heat is supplied to the engine, and the temperature at which waste heat is disposed of to the engine’s surrounding environment. 10. Despite the illusion of plenty, fossil fuel resources are a one-time-only inheritance. The pace of depletion is extraordinary when considered over the geological time scale required to accumulate these resources, and the EROI of global fossil fuel supply is low and declining (Cleveland et al. 1984; Hall and Klitgaard 2018; Brockway et al. 2019).

Chapter 3

Energy Primer for Storage Analysis

3.1 Energy and Force 3.1.1 Introduction In order to appreciate the prospects for different types of storage, some basic grounding in the underlying physics relating to energy is essential. Energy ultimately derives from the forces of nature, so we take this as our starting point.

3.1.2 Forces and Work There are four fundamental forces in the universe. They are all non-contact forces, and include: the strong nuclear force, weak nuclear force, electromagnetic force, and gravitational force (see Table 3.1). On a macroscopic scale, electromagnetism and gravity are the basis for most of the physical phenomena we encounter in everyday life. Towards the other end of the scale spectrum, for understanding chemical interactions it is the electrostatic force, a sub-type of electromagnetic force, that is important—chemical bonds are formed from the transfer or sharing of electrons between atomic nuclei. They rely on the electrostatic attraction between the protons in nuclei and the electrons in the orbitals, and repulsion between the respective protons, and between the respective electrons. The concept of work provides the link between force and energy. Work is done whenever a force is applied to an object over some distance. This results in a transfer of energy from the system applying the force, to the object to which the force is applied and that moves under its application. The actions resulting from the application of forces are described by Newton’s Laws. In mechanics, the change in kinetic energy of an object is equal to the net work done on the object.

© Springer Nature Switzerland AG 2020 G. Palmer, J. Floyd, Energy Storage and Civilization, Lecture Notes in Energy, https://doi.org/10.1007/978-3-030-33093-4_3

41

42

3 Energy Primer for Storage Analysis

Table 3.1 Characteristics of the four basic forces Force Gravitational Weak nuclear Electromagnetic Strong nuclear

Approximate relative strengths 6 × 10−39 10−6 7 × 10−3 1

Range ∞ 300,000 GWh). This would support electrical energy storage equivalent to several weeks or more supply at Germany’s annual average rate (Sterner 2009, p. 105). The conversion pathway from electricity, through gaseous carriers, and back to electricity, incurs a substantial energy efficiency and financial cost penalty (see Chap. 8).

7.4 Fossil Fuel Energy Storage as a Reference for Required Capacity The capacity of fossil fuel storage facilities offers a handy reference with which to compare the scale of currently deployed electrical energy storage assets, and to consider possible future requirements. The European gas network and the US strategic oil reserve each hold energy stocks a little over 100-times the combined capacity of global pumped hydro storage and secondary battery (lead-acid and Liion) capacity (see Fig. 7.1). Replicating the storage capacity of European gas or US oil stores via electrical energy storage technologies would entail prodigious energy and material flows, with costs—including of a biophysical nature—that remain invisible to market-mediated financial pricing mechanisms. And yet these fossil fuel storage facilities that provide a scale reference here represent only a small fraction of the global storage capacity likely required by a global energy system dominated by solar and wind inputs.

116

7 The Use of Scenario Analyses to Estimate the Magnitude of Storage

30 days global primary energy supply assuming half current demand 6,600 TWh

European gas network 1,131 TWh US strategic oil reserve 649 million barrels 1,027 TWh Fuel in tanks of global vehicle fleet 1.2 billion vehicles at 45 litres each 537 TWh

Global capacity of pumped hydro (6,000 GWh), lead acid (3,800 GWh) and lithium ion battery (1,000 GWh) storage

Fig. 7.1 Illustration of the comparative scale of currently deployed energy storage capacity compared to 30 days of World energy use, assuming primary energy demand at half of current annual rate. The largest existing energy storage system is probably the European gas network, which holds roughly 75 days of average European use (GIE 2018). This is followed by the US strategic petroleum reserve, which holds roughly 35 days of US oil use (US DOE 2019b). Also shown is the estimated stock of gasoline and diesel that would be stored in the global passenger and freight vehicle fleet, if each vehicle’s tank was filled simultaneously. The square in the bottom left corner represents the global combined capacity of pumped hydro storage, lead-acid batteries and lithium ion batteries for transport, stationary energy supply, consumer devices and other applications. Author estimates for 2018 from Austrade (2018); IRENA (2017); ITRI (2017)

World primary energy consumption in 2018 was 580 EJ (BP 2019). To estimate the storage magnitude required for a global energy system based on solar and wind flows, we could assume, as a starting point, a halving of primary energy use due to reduced thermal losses. This would reduce the annual primary energy requirement to 290 EJ. Assuming total storage capacity equivalent to 30 days of all demand (i.e. including stationary and transport energy use) equates to around 24 EJ or 6,600,000 GWh on a primary energy basis. This is approximately 600 times the current installed capacity for all forms of electrical energy storage. It should be noted that, with reference to long historical experience, 30 days represents only a

7.5 The Limits of Model-Based Energy Futures Investigation

117

modest buffer. It is much less than the inter-seasonal storage that underpinned the development of post-Neolithic societies (see Chap. 1). In comparison, a coal mine or an oil field can be considered, in effect, to represent years to decades of energy storage.

7.5 The Limits of Model-Based Energy Futures Investigation 7.5.1 Introduction Scenario exercises based on quantitative models can provide first-pass estimates of the storage requirements that arise at high VRE penetrations. They can offer insights into sensitivities to changes in important parameters, providing valuable input into decision making processes. They can also play an important role in establishing the physical and economic performance boundaries for proposed future energy systems. But the widespread role that model-based knowledge claims play in shaping public perceptions of energy futures often involves blurred lines between empirically established facts and abstract speculations. Counter to Alfred Korzybski’s cautionary advice that ‘the map is not the territory’, model-based long-range scenarios are often treated as representing futures with ‘empirical validity’, in the sense that the envisaged worlds are physically and techno-economically realisable, if only the right institutional conditions fall into line. We argue that relations between energy, society, economy, and environment are sufficiently complex, and subject to sufficient irreducible uncertainty, that confident projections relating to largescale, global energy transitions stretching decades into the future, must always be treated as speculative. That is, findings must be viewed, and should be presented, as model relative—they relate to ‘model worlds’, rather than the actual world that we inhabit day-to-day (Floyd 2017). These exercises are better understood as relating to the practice of post-normal science (Ravetz 1999; Tainter et al. 2006; Sorman and Giampietro 2013; Saltelli and Funtowicz 2017), rather than treating them as contiguous with the practice of normal science grounded in empirical data relating to experience of the actual physical world. We contend that accurately determining the storage capacity required in practice for a wholesale transition away from reliance on fossil fuels lies beyond the scope of abstract quantitative modelling conducted in isolation from real-world experimentation. Nonetheless, relatively simple mind-sized models have great utility for exploring the boundaries of the storage challenge, and for investigating the implications of adopting different assumptions (Bardi 2013). A map, or model, should include detail sufficient for generating novel insight into the situation it depicts, while remaining computationally tractable. One of the present authors has developed and made publicly available a system dynamics model specifically for the purpose of such ‘boundary exploration’ relating to questions of storage capacity and related costs for a global energy transition (see https://tinyurl.com/y4bmkoum).

118

7 The Use of Scenario Analyses to Estimate the Magnitude of Storage

7.5.2 Macroeconomic Modelling Is Not Compatible with Thermodynamic Laws An underling challenge for model development is that there is no unified theory of energy and the economy, although there has been some recent success in deriving an economic production function that is compatible with thermodynamic laws (Keen et al. 2019). Physical scientists understand the primacy of energy for circumscribing ‘performance limits’, both in nature and society, and the seemingly self-evident truth that ‘the economy’ is embedded in the physical world. Yet mainstream economics ascribes no such weight to energy considerations in macroeconomic modelling. Energy is treated as a substitutable resource that is limited only by the human imagination (Simon 1981). Nearly all of the large scale energy-economy models relied upon by the IPCC and international energy organizations are grounded in orthodox macroeconomic theory. In the climate change mitigation scenario literature, economic and productivity growth are treated as exogenous inputs, predefined on the basis of assumptions made outside the context of the scenario models themselves. Productivity is assumed to be a function of ‘technical innovation’ (Solow 1957), and therefore largely immune from the effects of resource scarcity. Scenarios developed on this basis are intended to represent plausible future worlds. The parameters with which these worlds are characterised include population and GDP per-capita. These two central parameters, along with the energy intensity of the global economy, and emission intensity of energy supply, are combined in the Kaya identity (7.1), used to calculate current and projected future global carbon dioxide emissions. CO2 = population ×

energy CO2 GDP × × population GDP energy

(7.1)

The assumption of modest compound economic growth over a multi-decadal time horizon leads to average per-capita income projections several multiples of current levels. In the IPCC literature, baseline scenarios project a three- to eight-fold increase in (real) GDP-per-capita by 2100 (Palmer 2018). It is worth re-emphasising that these multiples are not derived from the models themselves but result from an a priori growth rate assumption. Collapse and recovery scenarios are explicitly ruled out in Solow growth models (Sherwood et al. 2017). These models assume and require that macroeconomic outcomes follow a smooth trajectory. The Kaya identity takes the general form of the mathematical identity given in Eq. (7.2), which states that the left hand side equals the right hand side because all the X s cancel, leaving Y = Y . The X s can be any non-zero parameters, that may be related directly (or not) to Y . As such, an identity function describes neither a causal relationship, nor a correlation. Y = X1 ×

X2 X3 Xn Y × × ... × × X1 X2 Xn−1 Xn

(7.2)

7.5 The Limits of Model-Based Energy Futures Investigation

119

The utility of the Kaya identity is that it allows the decomposition of the global carbon dioxide emission rate into a set of widely measured metrics that are commonly available in national economic and energy statistics. The identity is useful for organizing the way we think about the factors that influence global emissions. However, as a result of the prominence given to them, in the policy literature those factors are often viewed as though they are the primary drivers of emissions, and hence treated as ‘policy levers’ (e.g. Pielke 2009) for emission reduction. The early IPCC scenario literature highlighted the pitfalls of Kaya-based modelling, with Nakicenovic et al. (2000, p. 105) noting: While the Kaya identity can be used to organize discussion of the primary driving forces of CO2 emissions and, by extension, emissions of other greenhouse gases, there are important caveats. Most important, the four terms on the right-hand side of [the Kaya] equation should be considered neither as fundamental driving forces in themselves, nor as generally independent from each other.

Despite these caveats, the Kaya identity continues to underpin analysis in the emission scenario literature. It’s prominent use illustrates the general problem faced in simplifying the characterization of complex systems sufficiently to make them tractable in models, while ensuring that model findings remain credible, and moreover, useful, in the face of deep uncertainties.

7.5.3 Models Are Too Narrow Large-scale energy-economy modelling is now so complex that its practice is very difficult to conduct, and its products similarly difficult to interpret, outside of highly specialist research groups. In common with many professional fields, specialisation improves the depth of expertise within a narrowly defined area, but can also excessively narrow the contexts within which its research questions are formulated. In the field of medicine for example, the debate around generalists versus specialists is widely appreciated. For symptoms of the musculoskeletal system, an orthopaedic surgeon might recommend a surgical intervention, a general physician might prescribe physiotherapy, a pharmacist might supply an anti-inflammatory, and a naturopath might recommend a ‘vitalistic’ solution. The tendency for specialists to view problems through a particular lens is encapsulated is the so-called ‘law of the instrument’, or more colloquially, the ‘law of the hammer’. The saying ‘I suppose it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail.’ is attributed to Maslow (1966). More generally, the law states that when we acquire a tool or learn a skill, we tend to be influenced by its utility both in terms of the world that we perceive, and how we then interact with the perceived world (Kaplan 1964). Modellers and coders are similarly subject to these human traits and dispositions, finding ‘solutions’ that make sense within their particular domain of modelling practice, but that may make less sense in relation to (or be only weakly connected with) the ‘real world’.

120

7 The Use of Scenario Analyses to Estimate the Magnitude of Storage

It is possible to demonstrate mathematical and internal consistency within energy-economy models relating to envisaged futures, or to internally verify that the model’s code is consistent with the modeller’s conceptual understanding of the situation under investigation, but it is impossible to validate that the model outputs represents the ‘truth’ (Oreskes et al. 1994). In the scientific literature, the peer review process generally ensures that there is a lower bound to numerical verification, but the process says nothing about the ‘truthfulness’ of the model. It is only through the process of actually building and operating the proposed infrastructure and plant within the envisaged socio-economic context that the ‘truth’ can emerge. A useful rule of thumb is that claims based on conceptual modelling should be treated with a healthy dose of scepticism (Alexander and Floyd 2018).

7.5.4 Reliability Optimization modelling, by its very nature, can only optimize ‘known-knowns’. In relation to performance investigation for projected high-penetration VRE systems, the ‘known-knowns’ relate principally to observed variability of supply, and natural resource availability. Engineers who design and maintain real-world energy systems must also account for so-called ‘credible contingency events’, or ‘known-unknowns’—events that are known to occur but with uncertainty as to when. Typical examples here are unanticipated breakdowns and other events that prevent the ordinary operation of machines and systems. Engineers also need to build sufficient redundancy to account for so-called ‘noncredible contingency events’, such as multiple concurrent failures. This includes events that are not precluded physically or institutionally, but that have not previously been observed. These could be classified as ‘unknown-unknowns’— experience shows that unusual events can occur but the how, when and where of such events, and possibly even the ‘what’, is unknown and often unpredictable. An example of such an event was the South Australian blackout of 2016. Storm damage to transmission infrastructure initiated a cascading failure that resulted in a state-wide blackout. South Australia is noteworthy for its high penetration of VRE and reliance on interconnectors with the adjoining state of Victoria. The blackout was significant for demonstrating the vulnerability of the South Australian grid to infrequent, but high significance events (AEMO 2016). The system operator was criticized for failing to foresee the incident, but it is now also clear that the ‘optimal’ operation of the grid was outside the bounds of established best-practice for managing supply reliability. Electricity system optimization studies often adopt the overall generation reliability standard specified for the grid as a whole as a target for matching VRE supply with projected demand, rather than correctly applying this standard as an engineering constraint for detailed system design. Such studies essentially assume that renewable energy plants and transmission infrastructure will operate with 100%

7.5 The Limits of Model-Based Energy Futures Investigation

121

technical reliability. For example, Ziegler et al. (2019) used the grid regulator’s specified equivalent availability factor (EAF—a particular reliability indicator, see Sect. 6.6) as the basis for setting an allowable level of unserviced demand in an electricity grid supplied by 100% VRE plus storage. Ziegler et al. (2019) noted that ‘some maintenance downtime is likely for any power plant in operation. In the case of renewables, it may be possible to plan for maintenance during hours of low resource availability or off-demand hours. Some unexpected maintenance during in-demand hours may still be required, and we do not model the effects of this.’ Although Ziegler et al. did not actually apply an engineering constraint in determining the required supply capacity, their study is noteworthy for at least identifying reliability as a real-world factor. The way that ‘optimality’ is generally conceptualized in the modelling literature gives little consideration to the multiple factors materially relevant to the design, operation and overall effective performance of real-world storage facilities. For instance, as discussed previously in Sect. 7.3, Europe has strengthened its natural gas security by expanding LNG import terminals and maintaining gas storage equal to 76 days of annual average demand (GIE 2018), and even the US, which is a major oil producer, typically holds at least 35 days of petroleum stocks. Clearly all-renewables energy systems would be subject to different security-of-supply considerations than apply in relation to oil and gas. Nonetheless, all real world systems are subject to ‘non-credible contingency events’. Accounting for these in practice will ensure that real systems diverge sharply from the conditions that accord with highly abstract views of optimality in the world of quantitative modelling.

7.5.5 A Layered Approach to Modelling The pyramid in Fig. 7.2 and Table 7.1 depicts a ‘layered’ approach to the analysis of energy (and energy-economy) systems. It describes the set of considerations that require attention for comprehensive investigation of energy-society futures. Model-based studies typically take into account a subset only of the layers—or more specifically, one or more of the layers will typically be present in the model as exogenous assumptions, either explicitly stated, or implicit in the investigator’s world view and habits of thought. The pyramid’s top layer, termed the ‘simulation layer’, depicts the symbolic representation of key inputs, parameters and processes related to techno-economic energy system performance characteristics and behavior. When the situation being investigated involves incremental changes at the margins of existing energy systems, and where consequential implications for the energy system’s economic and social contexts are plausibly small, techno-economic modelling conducted within this top layer can reasonably be assumed to adequately represent real system behavior. Ringkjøb et al. (2018) identified 75 modelling tools currently used for analyzing energy and electricity systems with large shares of variable renewables. Most of these are developed for use with a specific electricity grid, although many models,

122

7 The Use of Scenario Analyses to Estimate the Magnitude of Storage Narrowly defined desktop simulation.

Market and policy feasibility, feasibility of carbon taxes.

Technical and engineering feasibility.

Social and economic feasibility, community participation or opposition, impact of energy expenditures on economic growth.

Feasibility based on underlying drivers, Jevon's Paradox, EROI, material scarcity.

Fig. 7.2 Depiction of model layers

in principle, could be ported for use in different contexts. An example is the use of the PLEXOS simulation software, which is widely used by electricity system operators for operational planning. Many proprietary models additional to those identified by Ringkjob are also in use, often developed and deployed by commercial consultancies. Moving down the pyramid, deeper levels of physical, economic, social, cultural, political (etc.) contextual complexity come into play. The changes envisaged within the contexts of major energy transitions go well beyond incremental adjustments at the margins of current systems. Implementing such changes can be expected to have second-order feedback effects that impact the very contextual foundations upon which current knowledge of system behavior is established. We argue that the complexity involved when such contextual feedback becomes relevant implies that pursuing energy transitions of the nature now widely envisaged will lead to inherently uncertain, even indeterminate outcomes. The investigation of such futures lies well beyond the capability of techno-economic modelling solutions alone. Hodges anb Dewer (1992) argued that models that are intended to be used for prediction2 need to be validated, and set out four prerequisites for demonstrating validity. One prerequisite is that the situation being modelled must exhibit ‘constancy of structure’ in time, meaning that the laws and rules underlying the model are fixed and invariable. An example of a situation exhibiting structural constancy

2 Prediction

is a term that the authors defined precisely, but that applies here in the broad sense of ‘anticipating plausible future behavior’—we recognise that most energy system modellers do not regard their model findings as concrete accounts of what will happen at future point in time.

7.5 The Limits of Model-Based Energy Futures Investigation

123

Table 7.1 Considerations requiring attention in comprehensive investigation of energy-society futures, represented as layers of increasing contextual depth Layer Simulation

Market and policy

Physical infrastructure

Society and economy

Energy and the environment

Notes Narrowly defined technical feasibility based on wind, solar and hydro resources, balance demand and generation at aggregated levels, assume most optimal performance, assumption of energy efficiency, demand management Market mechanisms subject to political and practical constraints, availability of capital financing, optimal policy design, carbon pricing Broader technical feasibility based on engineering reliability, system inertia, frequency control, reactive power, transmission expansion, land acquisition, availability of finance Social and economic feasibility, community participation or opposition to infrastructure, impact of energy expenditures on economic growth and taxation, social implications of energy hardship and redistribution of wealth, energy security implications Feasibility based on macro-scale factors including environmental and resource limits. Feedback effects expected due to finite scale of sources and sinks. The role of Jevon’s Paradox, contribution of embodied energy, EROI, material scarcity, and broader environmental impacts

Energy-related quantitative modelling typically addresses a subset of these layers, with lower layers accounted for via explicit or implicit assumptions

is celestial mechanics—Newtonian mechanics can be relied upon to provide a basis for predicting the pathways of planets. On the other hand, situations and projects that are affected by changes in the economic, social, and political domains do not exhibit constancy of structure in time (Scher and Koomey 2011). Hodges anb Dewer (1992) noted that validity is not necessarily subject to binary determination. A predictive model might be assessed as having greater or lesser relative validity depending on its perceived sensitivity to structural changes that lie outside (or are exogenous to) the model itself. Furthermore, even models for which validation is not possible may still be useful for formulating hypotheses. Nonetheless, a model that can’t be validated should not be relied upon for policy, planning, or forecasting. We argue that the situations with which the energy modelling literature deals are not defined solely by the parameters of interest in optimization models. Instead, they encompass techno-economic, social, political and other perspectives, which cannot be understood as complying with rules or relationships that are fixed and invariable across space and time. Economic, political and social systems are influenced by non-linear and abrupt changes that cannot be known with full—and often not with sufficient—precision or confidence. Sherwood et al. (2017) argued that ‘highly interdependent industrialized economies may behave more like a complex adaptive

124

7 The Use of Scenario Analyses to Estimate the Magnitude of Storage

system with non-linear, path-dependent, and unexpected growth trajectories.’ All statements about what is known via a model-derived perspective must therefore be statements relative to the model, and not relative to the ‘real world’ situation that such treatment purports to represent.

7.6 Summary A review of the literature relating to model-based scenario analysis of future energy systems suggests that abstract quantitative methods, in the absence of real-world experimentation, are unlikely to provide concrete guidance on the magnitude of electrical energy storage required for renewably-powered economies. These techniques can support the investigation and comparison of various storage options, but we should not expect them to converge on consensus answers. Unlike conventional generation, which is fuelled by energy stocks that could be treated as effectively unlimited for short- to mid-term operational planning, variable renewable energy is based on natural flows that vary independently of society’s demand for electricity, and that must therefore be buffered by technological and infrastructural means. By their very nature, quantitative energy system modelling techniques can only take into account ‘known-knowns’. Real-world systems function in highly complex contexts, with multiple constraints in the regulatory, social and community, technical, security, and political spheres. Real-world systems must be resilient to external and internal shocks. In engineering and nature, the goals of resilience and optimal efficiency are often at odds with one another. An idealized ‘least cost’ energy supply solution represents a lower theoretical bound that offers invaluable guidance for investigation of what might actually be required in practice to achieve a given transition objective. A review of the optimization literature suggests that 2–7 weeks of storage and substantial renewable overbuild would be required for an all-renewable electricity system. Taking into account the unavoidable need for redundancy would axiomatically increase this. Geo-strategic considerations and responses to ‘long-tail’ events can be expected to push the storage requirement well beyond cost-optimal levels. The examples of European gas, at 76 days storage, and the US strategic petroleum reserve of 35 days, give some sense of the scale of currently operating real world systems. Even so, these examples relate to supply systems servicing only a part of overall energy demand. The societies and economies that they support have other energy options to call on, and so we can infer that if all energy supply is integrated under a single suite of variable energy sources, as is widely envisaged with a transition to 100% renewable energy supply, the storage requirement will be much larger. We note also that most of the optimization modelling applies only to electricity and does not resolve the more difficult challenge of presently non-electrified enduse services, especially mobility and transport. If we set a modest target of 30 days storage, a complete shift to solar and wind flows would require roughly three orders of magnitude greater storage than the total currently deployed globally.

Chapter 8

Hydrogen as an Energy Carrier

8.1 The Promise of the Hydrogen Economy 8.1.1 Introduction The earlier chapters focused largely on energy storage as it relates to electricity supply, including the role of storage in supporting a transition to high-penetration (perhaps even 100%) variable renewable generation for established grid systems. Electricity, though, comprises between 18% and 38% of global energy use, depending on the stage of the conversion process from primary energy source to final carrier at which its share is measured, and the energy accounting methodology employed. According to the IEA, in 2016 electricity comprised 18.8% of world total final consumption by fuel-type, having risen from 9.4% in 1973 (IEA 2018, p. 16). If we also account for the primary energy used to produce electricity, the 18% ‘final consumption’ translates to 38% (Palmer and Floyd 2017), or roughly a third of global primary energy use. There is a long run trend in energy supply towards increased electrification, albeit unfolding at a slow pace. However, many end-uses will not easily be electrified except through energy intensive electricity-to-liquid-fuel processes. Furthermore, for many industrial processes that employ fossil fuels as both energy sources and chemical reagents, and others that use them specifically as chemical feedstocks, there is no practical pathway to electrification without wholesale development of novel processes and plant redesign. Two brief examples illustrate the extent to which a perspective on energy transition and energy futures centered on ‘wind plus solar plus electrical energy storage’ inadequately characterizes the nature and scale of the challenge faced in an across-the-board planned departure from fossil-fuelled economies.

© Springer Nature Switzerland AG 2020 G. Palmer, J. Floyd, Energy Storage and Civilization, Lecture Notes in Energy, https://doi.org/10.1007/978-3-030-33093-4_8

125

126

8 Hydrogen as an Energy Carrier

1. World supply of ammonia was estimated as 172 million tonnes in 2017 (FAO 2015). Ammonia (NH3 ) is essential for the production of nitrogen fertilizers, ammonium nitrate explosives, nitric acid, and many other industrial chemical compounds. Nearly all ammonia is currently produced via the Haber–Bosch process in which atmospheric nitrogen is reacted with hydrogen gas in the presence of a catalyst at high temperature and pressure. Smil (2004, chpt. 8) has estimated that by 2050, Haber–Bosch fixation of nitrogen could account for 60% of global nutrition. The hydrogen for this process is currently produced via steam-methane reforming (for which the methane source is natural gas), or coal gasification (Weger et al. 2017). 2. World steel production was 1689 million tonnes in 2017 (World Steel Association 2018). Nearly all reduction of iron ore to pig iron, to enable further processing to produce steel, is carried out using coal as a chemical reductant, with a minor share attributable to natural gas. This chapter explores hydrogen’s use as an energy carrier, chemical feedstock, and storage medium. Hydrogen represents perhaps the only universal energy storage alternative to fossil fuels. It can be produced using energy from all of the major primary sources, including renewable electricity. However, a ‘hydrogen economy’ is not inevitable and the realization of an alternative to energy storage based on fossil fuels faces significant, and perhaps insurmountable, challenges. Even if successful, a hydrogen-based energy system will be costlier than the fossil fuel-based energy system it replaces. Depending on the degree to which national economies can lower their energy intensity of production, hydrogen will likely require greater expenditures on energy, as a proportion of GDP, than has been usual since the second half of the twentieth century. However, as discussed in Chap. 1, the historical record shows that societies are likely to embrace energy storage solutions even where the costs may appear prohibitive.

8.1.2 A Moonshot The prospect of using hydrogen as the principal energy carrier in large-scale economies has a long history. NASA adopted liquid-hydrogen technology in the early 1960s for the upper stages of the Apollo space program’s Saturn V rocket (see Fig. 8.2). NASA also chose the then cutting-edge hydrogen fuel cell as the primary source of electricity for the Apollo Command Module. Despite needing to overcome many difficult technical challenges, the decision to pursue hydrogen was later seen as a major competitive advantage for the United States’ space program over its Soviet rival, and a key contribution to the overall success of the Apollo ‘moonshot’ (Dawson and Bowles 2004). By the 1970s, hydrogen’s characteristics as an energy carrier were being explored for general transport, chemical synthesis, and metallurgical uses. John Bockris coined the term ‘hydrogen economy’ to describe an energy distribution system capable of supporting all economic sectors via a universal ‘medium of

8.1 The Promise of the Hydrogen Economy

127

energy transport’ that could meet energetic, ecological and economic performance requirements in an integrated way (Bockris 1972). Bockris envisaged the use of low-cost electricity, originally to be supplied via nuclear generation and later from renewable primary sources (Bockris 2013), to produce hydrogen via electrolysis. The hydrogen would then be distributed for use in trucks, cars, ships, trains, and aircraft, along with industrial chemical synthesis and metallurgical production. The physicist Marchetti (1973) published a more detailed account of a prospective solar- and nuclear-powered (fission or fusion) economy, for which energy would be distributed via hydrogen. Marchetti coined the term ‘SOLFUS’—a portmanteau of solar and fusion—in which some combination of solar-based technologies, nuclear fission, and fusion would supplant fossil fuels. Marchetti (1982) also observed the historical trend towards fuels with a higher hydrogen-to-carbon ratio, of which pure hydrogen is the natural end point. Despite the initial enthusiasm with which these visions were greeted, progress towards practical implementation was hindered by numerous technical challenges, high costs, and the inertia associated with fossil fuel incumbency. Interest in hydrogen as an energy carrier recurred periodically in response to policies such as California’s Zero Emission Vehicle Law of 1990 and the US Hydrogen Posture Plan of 2006 (US DOE 2006). But critical scrutiny of ‘hydrogen hype’ generally revealed the ‘rhetorical visions’ to be lacking in substance. The ideal of a universal energy carrier molecule remained elusive, due in part to ongoing technical challenges and the high costs faced in reconfiguring economies to suit (Eisler 2009; Sovacool and Brossmann 2010).

8.1.3 Hydrogen Carriers Hydrogen-based energy conversion pathways can be based on molecular hydrogen, and hydrogen-rich synthetic fuels, and hydrogenated compounds. These gaseous or liquid chemicals intermediate primary electricity generation, and the provision of energy services including heating, and stationary and transport work. Proposed hydrogen carriers include methane (CH4 ), ammonia (NH3 ), methanol (CH3 OH), and methylcyclohexane (CH3 C6 H11 ). Energy systems based on these compounds have been referred to respectively as the ‘hydrogen economy’ (Bockris 1972), ‘methane economy’ (Gloor 2004), ‘ammonia economy’ (Avery 1988), ‘methanol economy’ (Olah 2005), and ‘MCH supply chain’ (Obara 2019). Germany has been the leading proponent of methane (Moore and Shabani 2016). Ammonia is often associated with North American (NH3FA 2018) and Australian research groups including one based at Monash University (Mott 2018). Methanol is usually associated with the Hungarian-American chemist George Olah (Olah 2005). Hydrogen-based pathways potentially address some of the most intractable and long-term challenges related to fossil fuel depletion and carbon dioxide emissions. The solutions offered here include substitution of coal with hydrogen as a reductant for iron production, inter-seasonal electricity storage, provision of transport fuels for

128

8 Hydrogen as an Energy Carrier

heavy vehicles, ships and aircraft, and production of ammonia for fertilizer production. Electrification of energy supply is a key energy transition and decarbonization strategy, however providing the full suite of desired end-use energy services on a global scale will still require a diverse range of energy carriers. Hydrogen, like electricity, is an energy carrier and not a primary energy source. Pursuit of a hydrogen economy would involve a strategy of integrating electrification with hydrogen storage and distribution. A transition towards increased reliance on hydrogen for energy storage and distribution is therefore likely to involve complex interactions between incumbent primary energy sources and enduses, rather than a swift and complete substitution of hydrogen for one or more incumbent fuels. Furthermore, such a transition will necessarily involve changes to both supply-side and demand-side technologies. Historical experience suggests that such a process is likely to diverge from what might be considered an optimal change trajectory. A disjointed and protracted process could be expected, with inevitable missteps along the way. At the same time, a feature of hydrogen-based energy distribution and conversion is its versatility, the potential for large-scale storage, and its ability to both substitute for and complement incumbent energy systems. Versatility refers to two characteristics—(1) the potential multiple roles of hydrogen as a storable energy carrier; and (2) the multiple pathways both into and out of molecular hydrogen. As such, economies in which hydrogen plays a major energy system role may offer significant scope for adaptive response to new challenges as they emerge.

8.2 The Physical Basis for Hydrogen as an Energy Carrier Hydrogen energy is frequently discussed as a technology, but a prospective hydrogen economy is better understood as an ecosystem encompassing a suite of technologies, conversions and linkages between primary energy sources and enduses. This section briefly explores the characteristics of hydrogen as an energy carrier, including its strengths and weaknesses.

8.2.1 Water, Hydrogen and Oxygen The rationale for a hydrogen-based energy system arises as a consequences of the physical and chemical properties of hydrogen, oxygen and water. 1. Water, hydrogen and oxygen are abundant, and for the scale of human use envisaged for a global hydrogen economy, can be considered essentially unlimited. Earth’s atmosphere is composed of 21% oxygen and the oceans cover 71% of Earth’s surface. Elemental hydrogen is the most common element in the universe, but almost all of the hydrogen on Earth exists in compound form, including as water.

8.2 The Physical Basis for Hydrogen as an Energy Carrier

129

O2 photosynthesis

respiration

O2 atmospheric oxygen

power

89 % of molar mass

O2

renewable

e-

H2 Electrolyser

Hydrogen storage

O2

H2 11 % of molar mass

Fuel cell

e-

electricity

~34% pathway efficiency

other electricity

H2O

H2O w

Fig. 8.1 Biophysical flows associated with hydrogen energy storage and distribution. The atmosphere can be conceptualized metaphorically as a planetary scale tank of oxygen, and the hydrosphere as a tank of water. Of the chemical reactants and products involved in the overall cycle, only the hydrogen component, comprising just 11% of the total molar mass, needs to be isolated, transported and stored. Oxygen cycles from the hydrosphere to the atmosphere via the electrolyzer, then returns to the hydrosphere via the fuel cell. The atmospheric concentration of oxygen is stabilized via the rate balance between photosynthesis and respiration

2. The natural combination of hydrogen and oxygen to produce water, and the reverse decomposition, are relatively environmentally benign in comparison to most other human energy use. The earth’s atmospheric oxygen was produced almost entirely by the photosynthetic decomposition of water, energized by sunlight (see Fig. 8.1). 3. There are several processes that can drive the reactions between hydrogen (and hydrogen carriers) and oxygen (or other reactants necessary for energy conversion processes), enabling several synthesis and use pathways. Hydrogen can be oxidized via combustion to provide heat and work, or via fuel cells to produce electricity. Hydrogen can be carried by carbon, for example as methane, or by nitrogen, as ammonia. For some processes, it may be desirable to use carbon as the carrier, and to then capture, store and potentially re-use carbon dioxide produced as part of the fuel cycle. In other cases, it may be desirable to eliminate carbon entirely from the cycle. 4. The oxidation of hydrogen, and hydrogen carriers, is strongly exothermic. The double bond of the O2 molecule is much weaker than comparable bonds, and the formation of the stronger bonds in H2 O results in a strongly exothermic reaction. 5. Although provision of ‘hydrogen energy’ is dependent on oxygen and water, only the hydrogen component needs to be isolated, stored, and transported. Hydrogen is the lightest element in the periodic table, and comprises only 11% of the molar mass of water. The specific energy value for molecular hydrogen takes into account only the hydrogen itself and ignores the mass of oxygen required to

130

8 Hydrogen as an Energy Carrier

Fig. 8.2 Saturn V rocket for the Apollo 4 mission, launched Nov 9, 1967. Stage 2 had five hydrogen-fueled Rocketdyne J-2 engines and stage 3 had one J-2 engine. Image: NASA

liberate that quantity of energy. This results in the apparent specific energy being nearly an order of magnitude greater than if the oxygen mass was also accounted for. Nonetheless, even where the oxidizer must be stored and transported along with the hydrogen, as is the case with rocket motors, hydrogen remains a very attractive fuel—for instance, the Saturn V rocket that propelled the first people to the Moon used hydrogen to fuel its second and third stages (Fig. 8.2).

8.2.2 Hydrogen Fuel Cell An advantage of hydrogen as an energy carrier is that it can readily be oxidized in a fuel cell to generate electricity without the need for a thermal conversion step via a heat engine. This allows fuel cells to bypass the Carnot efficiency limit discussed in Chap. 2 (Haseli 2018). Furthermore, since fuel cells possess no moving parts, they are free of mechanical noise, vibration, and significant waste heat. At room temperature, a catalyst is required to facilitate the reaction shown in Eq. (8.1), where the letter g indicates that the reaction product is in the gaseous state. 2H2 + O2 −→ 2H2 O(g)

(8.1)

8.3 A Parallel Energy Supply Network Based on Hydrogen

131

Table 8.1 Practical hydrogen fuel cell efficiencies based on lower heating value (LHV) of water

Fuel cell type Polymer Electrolyte Membrane (PEM) Alkaline (AFC) Phosphoric Acid (PAFC) Molten Carbonate (MCFC) Solid Oxide (SOFC)

Operating temperature