This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: 54.39.50.9

This content was downloaded on 10/07/2018 at 18:52

Please note that terms and conditions apply.

You may also be interested in:

Organic batteries move a step closer

Michel Armand

Inductor-converter superconductive magnetic energy storage for electric utility usage

R.W. Boom and R.F. Bischke

New measuring system of transient response of advanced batteries for abrupt load changes

M Futamata and S Takahashi

Long-term deuterium release from CFC NB31 in the air atmosphere

K Sugiyama, V Kh Alimov and J Roth

Hydrogen generation by laser irradiation of colloids of iron and beryllium in water

I.A. Sukhov, G.A. Shafeev, E.V. Barmina et al.

Evolution of porous silicon crystal structure during storage in ambient air

L M Sorokin, V V Ratnikov, A E Kalmykov et al.

Hydrogen Generation by Koh-Ethanol Plasma Electrolysis Using Double Compartement Reactor

Nelson Saksono, Johannes Sasiang, Chandra Dewi Rosalina et al.

Thin film separators with ion transport properties for energy applications

Zhongyuan Li

Catalysts for hydrogen generation from water vapor

K. Nakayama, M.J. Nishi, T. Taniguchi et al.

Energy Storage Systems

Energy Storage Systems

David ElliottOpen University, UK

IOP Publishing, Bristol, UK

ª IOP Publishing Ltd 2017

All rights reserved. No part of this publication may be reproduced, stored in a retrieval systemor transmitted in any form or by any means, electronic, mechanical, photocopying, recordingor otherwise, without the prior permission of the publisher, or as expressly permitted by law orunder terms agreed with the appropriate rights organization. Multiple copying is permitted inaccordance with the terms of licences issued by the Copyright Licensing Agency, the CopyrightClearance Centre and other reproduction rights organisations.

Permission to make use of IOP Publishing content other than as set out above may be soughtat permissions@iop.org.

David Elliott has asserted his right to be identified as the author of this work in accordance withsections 77 and 78 of the Copyright, Designs and Patents Act 1988.

ISBN 978-0-7503-1531-9 (ebook)

DOI 10.1088/978-0-7503-1531-9

Version: 20170701

Physics World DiscoveryISSN 2399-2891 (online)

British Library Cataloguing-in-Publication Data: A catalogue record for this book is availablefrom the British Library.

Published by IOP Publishing, wholly owned by The Institute of Physics, London

IOP Publishing, Temple Circus, Temple Way, Bristol, BS1 6HG, UK

US Office: IOP Publishing, Inc., 190 North Independence Mall West, Suite 601, Philadelphia,PA 19106, USA

Contents

Abstract vi

Preface vii

About the author viii

1 Introduction: the uses of storage 1

2 Energy storage criteria: size, cost and utility 2

3 What’s on offer: current and new developments 7

3.1 Electro-chemical storage 8

3.2 Mechanical systems 9

3.3 Thermal systems 10

3.4 Hydrogen options 11

4 Outlook: the issues ahead 13

5 Additional resources 13

References 14

v

Abstract

As renewable energy use expands there will be a need to develop ways to balance itsvariability. Storage is one of the options. Presently the main emphasis is for systemsstoring electrical power in advanced batteries (many of them derivatives of paralleldevelopments in the electric vehicle field), as well as via liquid air storage,compressed air storage, super-capacitors and flywheels, and, the leader so far,pumped hydro reservoirs. In addition, new systems are emerging for hydrogengeneration and storage, feeding fuel cell power production. Heat (and cold) is also astorage medium and some systems exploit thermal effects as part of wider energymanagement activity. Some of the more exotic ones even try to use gravity on a largescale.

This short book looks at all the options, their potentials and their limits. There areno clear winners, with some being suited to short-term balancing and others tolonger-term storage. The eventual mix adopted will be shaped by the pattern ofdevelopment of other balancing measures, including smart-grid demand manage-ment and super-grid imports and exports.

vi

Preface

This book explores the way in which energy storage systems are evolving and thedevelopment of a system for storing the energy produced by power stations andother power sources. While some of the new storage techniques and systems aremainly the province of electrical engineers, and some storage devices (e.g. advancedbatteries) are the result of progress made by chemists, the underlying principles ofhow these systems operate and can be integrated together relies on understandingand applying basic physics. As new ideas emerge and research progresses it seemsclear that physicists will continue to play key roles, both in terms of technology andpolicy making.

vii

About the author

David Elliott

Professor David Elliott BSc PhD has worked in the powerengineering industry and in academia and has written extensivelyon sustainable energy system development and linked energypolicies, including two books on renewable energy for IoPPublishing. He is Emeritus Professor of Technology Policy at theOpen University, where he worked for many years developingcourses and research on sustainable energy innovation issues.

viii

Physics World Discovery

Energy Storage Systems

David Elliott

1 Introduction: the uses of storageThe energy system is changing rapidly, with renewable energy sources nowproviding around 25% of both global and UK electricity. Projections are that thiswill expand to at least 50% by 2050 and maybe much more (Elliott 2013). For that tohappen, attention will have to be given to energy storage and other forms of gridsystem balancing, since some renewable sources are variable. This will require thedevelopment of new storage and energy management technologies. They areexplored in this book, which focuses on systems for storing energy from powerstations and other power sources, including domestic-scale generators.

This book does not look at battery or other systems specifically used for vehicles:that is a separate topic, although there can be overlaps. Indeed, as we shall see, someof the technologies now being used in the so-called ‘stationary’ field emerged fromthe push for electric vehicles. Batteries and other types of small-scale storage are alsoa part of the mobile electronic device field, which is not explored directly here,although again there are overlaps, with lithium ion batteries beginning to populateall fields, including the power engineering field.

There are some general introductory points that apply to all fields. Perhaps themost obvious is that, whatever the end use, energy storage is not cheap. Yet, we areprepared to pay a lot for it when it offers specific services that cannot be providedeasily otherwise. For example, the ubiquitous small 1.5 V alkaline batteries used in awide range of portable consumer electrical and electronic devices, radios and torchesonly deliver a tiny amount of electrical energy, but we are willing to pay severaldollars for them even though the cost/kWh is vastly more than the cost of mainselectricity from the grid. We want portability. It is the same for laptops and electricvehicles, although with increasing size, power output duration and weight then oftenplay more of a role.

In the power system contexts that are the focus of this book, costs are importantbut, as we shall see, they are not the only criteria, in that, although storage systemsare costly, we may want to use them nevertheless to meet system needs. Certainlylarge-scale storage can make economic sense if the cost of the input energy is verylow and/or the price that can be charged for the energy output is high. The latterwould be the case when variable renewable supplies are low but energy demand ishigh and no other sources are available. However, in most cases other sources are

doi:10.1088/978-0-7503-1531-9ch1 1 ª IOP Publishing Ltd 2017

available, and are often low-cost, such as natural gas, which can be used in back-upplants when renewable inputs are low. So, the argument goes, electricity storage willnever compete with cheap gas turbines, which already exist on the grid, especially ifthere is a need to store energy over significant lengths of time—many hours or daysor even weeks, for example, due to long lulls in the wind. Gas can be stored for longperiods, at low cost and with low losses. Indeed, gas pipelines act as an energystorage buffer.

Nevertheless, new types of storage systems are beginning to make their mark inthe wider energy system. They can offer useful system services, such as, for example,making it possible to use a wider range and type of energy sources and avoidingemissions from fossil fuel-fired backup plants. The most developed and widely usedsystem is large-scale pumped hydro storage. Electricity is used to pump water uphillinto hydro plant reservoirs for later use to generate electricity via the hydro plants’turbines. If the energy used for this pumping were surplus output from wind or solarplants, then this approach could be seen as a way to balance the variability ofrenewables—energy from them is still available for a while, indirectly, when currentinputs from them are low. As we shall see, there are many other forms of storage thatcan be used in this manner, at a variety of scales, including advanced types ofbattery.

It is a rapidly developing field, with storage systems being seen as a major newgrowth area. In what follows we will look at existing and emerging energy storagesystems, and at the constraints on, and opportunities for, their widespread use.

2 Energy storage criteria: size, cost and utilityIn the power sector, as noted above, energy is sometimes stored so that it can be usedlater when supplies are not available or demand is high. Electricity cannot easily bestored directly, but the energy in the electrical output from power stations can bestored in a variety of ways including, most familiarly, electro-chemical batteries, butalso gravitationally, as noted above, in large ‘pumped storage’ hydro reservoirs. Thelatter are already quite widely used, in order to protect power grid systems fromsupply failures or demand surges, but, as variable renewables expand, they are likelyto play more of a role. However, there are limited sites available for large reservoirs,and additional storage systems will also have to be developed. This section sets thescene by looking at the general characteristics of, and criteria for, energy storagesystems such as this one, scale included.

For large-scale relatively long-term storage (hours, days and maybe longer), inaddition to pumped hydro, the options include compressed air energy storage(CAES) in large underground caverns and liquid air storage in cryogenic containers.Some advanced batteries, including new types of flow battery (in which two separatechemicals interact), also look promising for bulk power storage. There are somelocational pros and cons. Liquid air stores and large battery stores can be sitedwherever it is convenient, but in the same way that sites for hydro reservoirs aregeographically determined, so are the sites for cavern-based CAES. Such large costly

Energy Storage Systems

2

systems also have to compete with medium sized gas turbines, which can offer cheapbackup power when needed, and can be sited near energy-demand centres.

There is also a range of generally medium-scale fast-response technologies suitedto dealing with day-to-day grid balancing, including the short-term variations inwind and solar inputs. The grid system deals with some variations in demand andsupply by adjusting voltages and frequency slightly over short timescales (seconds,minutes, sometimes hours), within tight regulatory limits, but some storage systemscan help to limit this. Flywheel systems are one option, and some types of super-capacitors and batteries can also be used.

At the smaller end of the scale range, battery systems can also play useful roles,such as, for example, storing the surplus outputs sometimes generated by rooftopphotovoltaic (PV) solar arrays on houses and other buildings. Indeed, some look toan energy system in which both generation and storage are widely distributed. Localgeneration directly by users avoids the energy losses involved with transmittingelectricity long distances from larger power plants. However, in general, smallstorage systems are more costly per kWh stored than large systems, so there aredebates about the optimal approach (Elliott 2016a).

To put the cost issues in a specific context, the 7 kWh Tesla Powerwall homebattery unit is retailing at $3000 in the USA, and that excludes installation costs andthe cost of an electrical DC-to-AC inverter to link it into the domestic mains supply.If fully charged, for example, from a domestic roof-top PV solar array during thedaytime, a unit like this could run a one bar electric fire overnight, i.e. for 7 h. Butthen it would need recharging and that may not be possible during winter, when PVsolar performance may be very poor for some while heating needs are at a premium.It would be far cheaper in such circumstances to import (increasingly green) energyfrom the grid. Indeed, that may often be the case. And it may also be moreeconomic, in system terms, to export any surplus for storage in larger more efficientenergy stores of the type mentioned earlier. Independent home battery storage maythus not be that helpful in optimal system terms, but with the cost of both PV solarand batteries falling, self-gen/home storage is nevertheless becoming popular (Blueand Green 2016, Moixa 2017).

Battery storage may make more sense on a larger scale, but there are limits. Theretail costs of grid electricity can at present range from roughly $0.1–0.2/kWh,sometimes less, sometimes more, depending on the sources and the country. Forcomparison, recent improvements in lithium ion batteries have brought their costdown to around $3–4000 for a 10 kWh rated storage system, such as that developedby Tesla, with stacks of these being offered for utility-scale use (Tesla 2015). Ifrecharged, they each could deliver 10 kWh regularly, for maybe 1000 cycles over theirlifetime (before their performance degraded), so the electricity would cost $0.3–0.4/kWh (Naan 2015).

At root, this large cost difference compared with generation costs is because thegenerally high capital cost of storage has to be recouped from, in most cases,relatively short, although hopefully multiple, periods of sometimes limited output.See table 1, which also indicates some other key characteristics of storage systems:some have limited storage capacity/energy storage densities.

Energy Storage Systems

3

Clearly cost is a major factor, and with all storage systems the emphasis is on costreduction. However, that is only one factor. The others include size, storage densityand capacity, as well as charge and discharge times. These characteristics vary witheach type of system, and will determine which system is best for a specific use. So willround-trip energy conversion losses, which defines the overall energy conversionefficiency, and in turn, the economics of the system. Figure 1 illustrates some of thetrade-offs that are available between discharge time and capacity. Some systemshave high storage capacity but cannot deliver output over long periods and viceversa. There are similar trade-offs with the other characteristics.

As just noted, a key issue is the round-trip efficiency, which can be very high forsimple storage systems. Batteries can become warm when charged, but otherwisethere are no energy losses in converting electrical energy into battery charge andthen getting electricity out when needed—with maybe 90% or greater efficiency. Insystems with several conversion stages it will not be as efficient, depending on thenumber and types of conversions involved.

Here is where an understanding of basic physics and chemistry helps. Forexample, electricity can be converted into hydrogen gas by the electrolysis of water,which also produces oxygen (most school science classes will have done that withtwo electrodes linked to a battery in a slightly acidic water-filled jar). The oxygenmay be captured and used for some other purpose, but hydrogen gas can be collectedand stored for later conversion back to electricity by combustion in a rotary car-typeengine or in a gas turbine, or directly in a fuel cell.

Table 1. Electricity storage options—typical costs and key characteristics.

Based on ESA data (ESA 2017)

Pumped storage:Very high capacity, low cost, but site specific $100/kWCompressed air storage:High capacity, low cost, but site specific <$100/kWFlow Batteries:High capacity, but low energy density $100–1000/kWMetal–air batteries:Cheap, very high energy density, but not rechargeable ∼$50/kWNaS, Li-ion, Ni-Cd batteries:Expensive, but Li ion now getting cheaper ∼$1000/kWLead-acid batteries:cheaper, but bulky and limited deep cycling life < $1000/kWFlywheels:High power, but low energy density > $1000/kWCapacitors:High efficiency, cheap but low energy density >$100/kWHigh-power capacitors:Very expensive <$10 000/kW

Energy Storage Systems

4

At each stage there will be energy losses. Standard water-based electrolysis isaround 50%–60% efficient, more in advanced systems, including high-temperaturesystems and solid state variants: a PEM (proton exchange membrane) system canhave efficiencies of 70% or more. The lost energy emerges as heat. If some of that canbe captured and used, the overall energy conversion efficiency can be higher—thePEM electrolysis cell developed by UK company ITM Power is claimed to have anoverall efficiency, with heat recovery, of 86% (ITM 2017).

Storing hydrogen gas once produced, possibly by compressing it or chilling it tobecome a cryogenic fluid, requires energy. When the pressurised gas is released foruse, heat will be absorbed as it expands (that is how a fridge works) and heat is alsoneeded to re-vaporise the cryogenic fluid. That is why liquid nitrogen ‘burns’ skin—it sucks out heat. So again there are energy losses, perhaps 10%–20%. The nextstage, converting back to electricity, will also have losses. Typical engines andturbines are under 50% efficient, though advanced designs, with heat recovery, candemonstrate significantly higher efficiencies, maybe 70% or more. Fuel cell efficien-cies vary depending on the type and whether heat is recovered. Without that, 40%–

50% may be typical for standard cells. Being optimistic, let us say we use one at 70%for our very best case, with a high-temperature advanced fuel cell and full heatrecovery.

Multiplying all these separate efficiencies together gives us the overall systemefficiency. As can be seen, it might be 70 × 80 × 50 in the worst case, with no heatrecovery and using an engine, i.e., 28.0% overall. In a good, near-best case, with a

Figure 1. Graphical representation of several types of energy storage technologies in terms of their dischargetime and total capacity. Source: U.S. Energy Information Administration (EIA 2011).

Energy Storage Systems

5

very good fuel cell and high heat recovery, the efficiency may be 80 × 90 × 80, or57.6%,—essentially double. Of course, the specific numbers used here are somewhatarbitrary, but it should be clear from this simple exercise that, firstly, with multiplestages, the overall efficiency can be quite low, and secondly, it is important to makeeach stage as efficient as possible, for example, with heat recovery.

The efficiencies of some of the conversion stages in the example above werelimited by the laws of thermodynamics, which apply to all heat-based systems, butnot to fuel cells or mechanical storage and associated electrical conversion systems,such as electrically driven flywheels, for example. They rely on the kinetic energy ofmovement, or more accurately, rotational angular momentum. They are powered byhigh-efficiency electric motors, spinning them up to speed, with energy beingextracted when needed using shaft-linked high efficiency dynamos or alternators.The net result can be low losses, especially if magnetic bearings are used and theflywheels are run in a vacuum.

Systems like pumped hydro storage rely on the potential energy of the waterpumped up into the reservoir, which is converted into kinetic energy when the headof water is released from the reservoir, and then rotational energy in the hydro plantturbines, and finally into electricity in the generators they drive. This is quite a fewstages, some involving friction in large-scale fluid movement, so there are likely to besome losses.

Table 2 summarises some quite conservative estimates of round-trip efficiencies,along with system capacity. Some of the figures used above are more optimistic.

Clearly it would help if some of the efficiencies could be improved and costsreduced. So there is plenty to challenge physicists and engineers. That is also the casewhen it comes to making optimal choices amongst systems. As indicated, there is awide range of options, even without looking at systems in which electricity or motionare converted into heat for storage, as in some of the heat churn systems we will belooking at in the next section. Hopefully though, the above tour around the physicalcharacteristics and limits of storage systems and processes will have set the scene fora review of the existing and emerging hardware options.

Table 2. Storage technology key characteristics; source: IEA (2005).

Storage technology Typical round trip efficiency Typical capacity

Pumped hydro ∼80% >100–1000 MWCompressed air ∼75% >50–100 MWFlywheel ∼90% >1–50 kWConventional batteries ∼50–90% >1–>10 MWFlow battery ∼70% ∼15 MWHydrogen fuel cell ∼40% >50 kW–>1 MW(complete system)

Energy Storage Systems

6

3 What’s on offer: current and new developmentsAt present the principle emphasis is on systems for storing electrical power inadvanced batteries (many of them derivatives of parallel developments in theelectric vehicle field), as well as via liquid air storage, compressed air storage,super-capacitors and flywheels, and the well-established option of pumped hydrostorage. New systems are also emerging for hydrogen generation and storage,feeding fuel cell power production. Heat (and cold) is also a storage medium andsome systems make use of heat energy as part of a wider energy managementactivity.

While scale is an obvious issue, it is helpful to classify these various systems on thebasis of the basic type of energy conversion process used:

1. Electro-chemical systems, e.g. batteries.2. Mechanical systems, e.g. flywheels, pumped hydro.3. Thermal systems, e.g. heat storage.4. Hydrogen-based options, e.g. electrolysis with cryogenic storage.

This classification can be rendered in very general terms on the basis of physicalprocesses. Electro-chemical systems essentially involve shifting electrons around,stacking them chemically for later use, or just as a charge in a capacitor. Electro-magnetic systems might be seen as an extension: energy can be stored in magneticfields, which can be created using electric currents, which are flows of electrons.

Mechanical systems involve mass and weight, moving very large assemblies ofmolecules around to create potential or rotational energy, with Newton’s laws ofmotion describing what can be done. That also applies to compressed air storage,with its stored energy being released in a high-pressure gas flow, kinetic energy, todrive a turbine, though fluid flow processes are subject to additional physical laws.

Thermal systems make use of heat energy, which is fundamentally about the fastbut (very) microscopic movement of molecules in solids, liquids or gases, and mayinvolve conversions between these forms. These conversions are subject to the lawsof thermodynamics. Heat, in whatever form, is an efficient way to store energy. So is‘cold’: the energy needed to cool something is in effect stored in it, and it can be usedto power devices.

Hydrogen is an atom made up of one proton and one electron (occurringnaturally as a diatomic molecule H2) which can be made by splitting water (H2O)thermally, or by electrolysis. It can be stored in various forms (as a gas, as acryogenic liquid or chemi-absorbed in suitable solids) and then recombined withoxygen through combustion to produce heat, or in a fuel cell, which in effect iselectrolysis in reverse, to produce electricity. Essentially, you could say that in thislater process we are dealing with protons as charge carriers.

This type of analysis may be abstract, but, in addition to indicating how andwhere physics relates, it should place the engineering reality that we will now look atinto context, as we review some examples of what has been developed, and of what isbeing developed, in each case.

Energy Storage Systems

7

3.1 Electro-chemical storage

There are some purely electrical systems, such as super-capacitors and magneticinduction devices, that can provide short-term storage, but electro-chemical batteriesare the most familiar option. Conventional lead acid batteries are expensive andbulky and not very suited to bulk energy storage. The more advanced lithium ioncells were initially expensive, though costs have fallen dramatically due to their usein laptops and increasingly in electric vehicles (Nykvist and Nilsson 2015).

Some batteries are now being used in large utility projects, such as, for example,the Zhangbei project in China, with 36 MWh of output capacity linked to 40 MW ofwind and solar plants. In the UK, a 6 MW/10 MWh utility scale lithium ion batterysystem has been on trial near Leigthon Buzzard and community-scale projects arealso emerging (Coyne 2016, Mis 2017)

Sodium sulphur (NaS) batteries are also used at the multi-MW scale, e.g. Japanhas a 50 MW/300 MWh unit. However, lithium ion batteries dominate, althoughnew options are emerging, such as, for example, aluminum ion batteries (Lin et al2015). Metal–air batteries are high-energy density and low cost, but are not directlyrechargeable electrically, although some new liquid metal variants may be(LaMonica 2013), while lithium–air batteries are sometimes seen as a major newoption (Hoster 2015).

The field is clearly developing rapidly, driven in part by the electric vehiclemarket, but with the stationary power market also benefitting. Interestingly, hybridsystems combining ultra- capacitors (which have fast, high-output response times)and lithium ion batteries (for the longer haul) are gaining some traction for use onlarger scales (Reid 2016). Moreover, some see super-capacitors as winning out insome power markets (IDTechEx 2014). However, many new ideas have emergedfrom research laboratories, some claiming large cost savings or performance gains.It is important to be wary of media hype, but some batteries may yet change thewhole picture (Edie 2015, Robertson 2016, Cassey 2016). So might market changes,e.g. the widespread adoption of ‘vehicle to grid’ (V2G) charging, using the batteriesof electric cars.

In a parallel development, the various types of advanced flow batteries, withround trip efficiencies of 70% or so, only slightly lower than those for lead acidbatteries, show promise for applications on larger scales. They mix separatechemical electrolytes to create a charge, in a reversible process (PD Energy 2013).Zinc–bromine and vanadium redox systems are some top contenders, but the USSandia National Laboratories have been looking at electrochemically reversiblemetal-based ionic liquids, which are non-toxic (Sandia 2012). So has Harvard, usingorganic (carbon based) materials (Harvard 2014, Burrows 2017), though thisremains a laboratory-based project. For now, vanadium seems to dominate(Conca 2016) and there are some very large projects envisaged, including a200 MW/800 MWH scheme in China (Lombardo 2016).

In addition to large, utility-scale systems, domestic-scale lithium ion batteries arealso now being widely used with rooftop PV solar. For example, it has been claimedthat around 41% of all new domestic PV projects installed in 2015 in Germany

Energy Storage Systems

8

included battery storage, and, with costs falling, take up can be expected to continueto grow there and elsewhere.

The end result of all this is that the battery market, especially for lithium ion, isbooming (Maloney 2016). This has been driven by dramatically falling prices forlithium ion, but also for flow batteries (Spector 2016). This progress is likely tocontinue as new battery concepts emerge, some of them quite exotic (Radford 2015,Hsu 2016, Flynn 2017).

3.2 Mechanical systems

As noted above, electro-mechanical storage technologies include advanced flywheelsand pumped hydro storage. Advanced high-energy flywheels are available at a rangeof scales (Stornetic 2017, Beacon 2017) and can be used for short-term gridbalancing/frequency stabilization, as well as for energy storage. Some projects ofup to 20 MW are in use.

As indicated, pumped hydro is already used for storing excess electricity from thegrid and there are also some non-hydro pumped reservoir schemes (using off-peakpower), e.g. a 1.9 GW plant on the shore of Lake Michigan (Ludington 2013). TheUK has nearly 3 GW of pumped storage capacity, including the Dinorwic scheme inWales, and across Europe there are many large hydro schemes, some of which havepumped storage capacity (JRC 2013). Pumped storage, using existing or new hydroreservoirs, or just free-standing reservoirs, has significant potential for helping tocompensate for the variable output of some renewables.

There are novel pumped storage ideas, such as the Green Power Island proposedoff the coast of Denmark using an artificial lagoon built in the sea and linked to a150 MW offshore wind farm (GPI 2013). A similar idea is being studied in Belgium,where a 3 km wide donut-shaped island with a 30 meter deep reservoir at its centerand 300 MW of turbine pump/generator units is being used. Excess wind-derivedelectricity would be used to pump water out of the reservoir into the sea. Whendemand is high, the water would be let back into the reservoir through the turbines.Similar pumped storage ideas have been suggested for tidal lagoons and barrages.

Pumped storage is based on raising water against the force of gravity to createpotential energy, but even more energy can be stored if large masses of heaviermaterial can be raised. In one such scheme a huge plug is raised hydraulically in avast piston, and squeezes water through turbines when it is allowed to fall (GravityPower 2014). Gravitricity offers another version: suspended heavy masses such asscrap iron on a cable hanging down an old mineshaft, which is raised by winches andthen allowed to fall to generate power (Fraenkel 2016).

Another large-scale electro-mechanical option is compressed air storage, housed,for example, in large underground reservoirs. In one version of CAES, electricityfrom wind turbines is used to compress air, which is then stored in caverns for use tosupercharge the burning of gas in a conventional turbine (Gaelectric 2017). Inanother, compressed air produced mechanically using electricity from offshore windturbines is stored in large inflatable bags mounted underwater around the turbinebases, for subsequent use in a separate turbine to generate electricity (Garvey 2011).

Energy Storage Systems

9

MIT have developed an even more ambitious concept, using large hollow concretespheres mounted in deep water (200 meters or more), with water pumped out usingenergy from floating wind turbines and then let back in through turbines forgeneration when needed (MIT 2013). Norwegian researchers have come up with asimilar idea, claiming 80% round trip efficiency (SINTEF 2013). A variant is beingdeveloped in Germany (BINE 2017).

Compressed air storage systems can be large. In the USA, there is a veryambitious $8bn project proposed involving a 2.1 GW wind farm in Wyomingsending electricity by HVDC power grid 525 miles to an underground salt caverncompressed air storage facility in Utah and then, after conversion back to electricity,490 miles on to Los Angeles (PennEnergy 2014, Gruver and Brown 2014). However,compressed air storage does not have to be very large scale. There are also somesmaller scale systems on offer (Lightsail 2017).

3.3 Thermal systems

Heat is easier to store than electricity, and heat stores can have high-energy storagedensities. Hot water can hold about 3.5 times as much energy by volume as naturalgas at atmospheric pressure and temperature. There are numerous mechanisms forproducing heat, including by direct capture of solar energy, biomass combustion, thetrapping of geothermal energy, and, if all else fails, burning fossil fuels. It can bestored as hot water in insulated tanks or in larger pond or pit-type heat stores builtinto the ground with insulating walls and covers. Large heat stores are better thansmall stores, since the surface-area-to-volume ratio decreases with increasing sizeand it is that ratio that defines the heat losses. There are some very large partly solar-fed interseasonal heat stores in Denmark for storing summer solar heat for winteruse, fed into local district heating networks (Marstall 2017).

However, these are just for heating. Converting heat back to electricity (e.g. byraising steam to drive turbines) can be inefficient, so depending on location, demandand the grade of the heat, it is usually better to use the heat directly, unconverted.

Nevertheless, the US IT company Apple is reportedly developing a system thathas a tank of water with a mechanical churn driven directly by a wind turbine. Itschurning action heats up the water, which is then stored, ready for use in electricitygeneration when needed. A perhaps more sophisticated variant makes use of eddycurrent electromagnetic induction for heating the water (Okazaki et al 2015).

Another approach is to use excess electricity from wind to heat water in a store viaan immersion heater, for use when needed in a district heating network. There issuch a 200 MW system in operation in Denmark. In Scotland, the SHEAP districtheating project on the Shetland Isles supplies heat from a waste-to-energy inciner-ator to 1100 customers. Surplus heat during the night is fed to a 12 MWh thermalstore, with an expansion planned to link 6.9 MW of wind generation to a 135 MWhcapacity heat store with immersion heaters, capable of providing five days’ extraheat (Building4Change 2012).

Heat storage does not have to just be in water. UK company Isentropic havedeveloped a gravel filled heat store system linked to a heat pump that upgrades the

Energy Storage Systems

10

heat flow. They claim the round-trip efficiency is 72%–80% (Isentropic 2016).Molten salt can also be used to store heat, as is widely used with concentratingsolar power plants, and there has been interest in using sand as a heat store medium.Crushed rocks or bricks, with hot air blown through, have also been used for somesolar heat stores and Siemens has been reported to be looking at rock-based storagesystems for use with wind turbines (Smith 2014).

Any large temperature difference can be used to run a heat engine and there aresome systems that use ice as an energy storage medium, although this methodusually is part of an air-conditioning/cooling package (BAC 2012, Ice Energy 2013).Building upon that is the idea of cryogenic liquid air storage. UK companyHighview Power Storage has demonstrated a 300 kW prototype that stores excessenergy at times of low demand by using it to cool air to around minus 190 °C viarefrigerators, with the resulting liquid air, or cryogen, then being stored in a tank atambient pressure (1 bar). When electricity is needed, the cryogen is subjected to apressure of 70 bars and warmed in a heat exchanger. This produces a high-pressuregas that drives a turbine to generate electricity. The still relatively cold air emergingfrom the turbine is captured and reused to make more cryogen. If waste heat from anearby industrial or power plant is used to re-heat the cryogen, it is claimed that theround-trip efficiency rises to around 70% (Highview 2017). Highview now has a5 MW plant near Manchester. They are looking to ‘10 MW and up’ next and saythey have designed a 200 MW/1200 MWh plant (Ali 2016)

While ‘cold’ is clearly an interesting storage option, heat and hot water are easierto work with, and as noted earlier there are many hot water solar heat stores linkedto district heating (DH) networks in Denmark. Some of them are fed from fossilgas, biogas or straw-fuelled combined heat and power (CHP) units. CHP plantscan vary the ratio of heat to power output, so that they can be used to balancevariable grid power, especially if they also have linked DH/heat stores. The heatoutput is raised and power output lowered when there is excess wind, and if notneeded, is stored for later use when heat demand is higher. When wind is low, moreCHP power and less heat are produced, and any needed heat is supplied from thestore (IEA 2011, JRC 2012). As the use of renewables expands, flexible balancingsystems such as this may play a helpful role, with Germany showing increasinginterest (Agora 2015). Certainly there are some larger heat store projects planned inGermany (Michel 2017).

3.4 Hydrogen options

Hydrogen can be produced by steam reformation of methane (natural gas) or, asnoted earlier, by electrolysis of water. It is flexible fuel, useful for power generation,heating and transport. There are many systems being developed for hydrogenstorage, as a gas under pressure or cryogenically as low-temperature liquid, orchemi-absorbed in metals, with a graphene-based system also being considered.

In Safe Energy’s system, hydrogen is absorbed as metal hydride in a molten mix,which can then be made to release its hydrogen, when required, through a reactionwith water, producing hydrogen and heat. That process converts the metal hydride

Energy Storage Systems

11

to a metal hydroxide, which can be recycled back to a metal hydride. Themagnesium hydride slurry can be stored safely in large quantities at ambientconditions (Safe Hydrogen 2013). Hydrogen can also be stored in undergroundcaverns for long periods, suitably sealed, much as is done with methane in saltcaverns (NATGAS 2013). Though clearly there are safety issues, the volumeavailable is very large (it could also include old oil and gas strata), making hydrogenthe largest potential storage option. Indeed, some see hydrogen, produced fromrenewable sources, as being a key new energy vector, with large stores being linkedto hydrogen pipelines, or tankers/trucks being used for distribution, depending onend-use requirements (Andrews and Shabni 2012).

At the local level, there are some small domestic-scale examples of hydrogenstorage for use with PV solar and fuel cells. An advanced example is the FroniusEnergy Cell system in which excess electricity from a PV array is used to decomposewater into oxygen and hydrogen by electrolysis. The hydrogen is then stored readyto be converted back into electricity in a fuel cell when it is needed (Fronius 2013).Some small-to-medium scale hydrogen storage systems have also been integratedwith electrolytic hydrogen production from wind-derived electricity and hydrogenfuel cells to make a complete green energy system (Hydrogen Office 2017).

Perhaps the cleverest hydrogen option is the so-called power to gas (P2G)approach, which is basically a utility-scale extension of the ideas above of usingwind or solar generated electricity to make hydrogen for later use to make electricityagain when it is needed. In this case, however, it is surplus renewable electricity thatis used—electricity that would otherwise be wasted, unless of course, it is stored issome other way. However, as we have seen, besides cavern-based CAES, few of theoptions available can offer long-term bulk storage on the scale likely for hydrogen—tens of GW equivalents in salt caverns and the like (Gammer 2015).

We looked briefly at the P2G idea earlier and its overall round-trip efficiency.Though the efficiency was not high, in economic terms that may not matter toomuch since the (surplus) input energy is essentially free and a valuable output can beproduced. The initial output is hydrogen, produced by electrolysis, which is thenstored. While one option is to use it, when needed, to make electricity, another is tofeed it into the gas mains, something that is being done in Germany, although insome cases it is first converted into methane gas. This can be done chemically usingcarbon dioxide gas captured from fossil fuel power station exhausts or direct fromthe air. Doing this adds another stage to the process, but it also reduces atmosphericCO2 and creates a valuable new easily stored fuel. In some cases, it is used as avehicle fuel, replacing (fossil) petrol and diesel with green gas. It can also haveindustrial uses (ITM 2016).

The P2G concept opens up a range of options for producing storable fuels(methanol and ammonia have also been suggested) from variable renewables,turning their variability from a problem into a solution. As renewables expand itis likely that surpluses will also expand, since, to ensure that the average level ofdemand can mostly be met, despite the variations in output, extra capacity has to beinstalled, the output of which will not be needed when demand is low. But, with P2G

Energy Storage Systems

12

that does not matter; the surplus energy can be used later to meet demand whenthere are major wind or solar lulls or above average demand peaks.

4 Outlook: the issues aheadAs can be seen from the survey above, which is far from exhaustive, there are manypossible storage options. In terms of the development of individual storage devicesand systems, apart from some specific applications there are currently no clearwinners or losers, with many rival systems being investigated. There are still alsomany new avenues to explore (e.g. the use of graphene-based systems and chemi-absorption of hydrogen). But at the same time the pattern of development cannot beeasily separated from the wider system optimisation issue, which involves integratedgeneration and balancing systems, with multiple energy vectors (electric power, heatand gasses) for generation, storage, transmission and use. It’s a challenging andexpanding interdisciplinary field.

A key practical requirement for any storage facility is to match the system to thesupply characteristic and end-use needs. In some cases, as we have seen, what isneeded is short-term storage (a few seconds or minutes) to deal with brief variationsin supply or demand. In other cases, storage may have to respond to longer lulls inpower supplies, e.g. from wind or solar farms (for hours, days or more). In the firstsituation, batteries, super capacitors or a flywheel may be suitable. In the lattersituation, although pumped hydro, hydrogen or compressed air stored in cavernsmay be suitable for a while, the cheapest option, and one suited to longer term lulls,is to have backup supply plants using stored fuel: a ‘pre-generation’ rather than‘post-generation’ approach to storage. Of course, hydrogen generated in P2G modeusing surplus renewable electricity can be used as a fuel for these plants, so in thatcase we have a link between pre- and post-generation.

Post- and pre-generation storage are not the only options for grid balancing.Storage is about shifting energy availability in time from when it is produced to whenit is needed. It can also be shifted in space—from where it is produced to where it isneeded. Moreover, the need for energy can also be delayed—its use being shiftedaway from peak demand times until later, when more is available. The optimal mixof storage and these other types of grid balancing, including smart-grid demandmanagement and long distance supergrid imports and exports of surpluses, are thesubject of many current studies, and the outcomes of these studies will influence thedirection in which storage technology develops (Elliott 2016a).

5 Additional resourcesThe field is fast-moving and interdisciplinary, so there are few sources that cover thewhole area, but it is well-served by web sites, as referred to in the text. Helpful recentoverviews and reports include the short review by UK Parliamentary Office forScience and Technology (POST 2015) and the longer study by the InternationalRenewable Energy Agency (IRENA 2015). Trade organisations include the USElectricity Storage Association (ESA 2017) and the European Association for theStorage of Energy (EASE 2016). The US Institute for Electric Innovation is also a

Energy Storage Systems

13

good source (IEI 2016). The Energy Storage Forum is a useful global network(ESF 2017). My short overview published in Nature Energy may also be of interest(Elliott 2016b).

ReferencesAgora 2015 The Danish experience with integrating variable renewable energy (Berlin: Agora

Energiewende) www.agora-energiewende.de/fileadmin/Projekte/2015/integration-variabler-erneuerbarer-energien-daenemark/Agora_082_Deutsch-Daen_Dialog_final_WEB.pdf

Ali Y 2016 Cryogenic storage offers hope for renewable energy, BBC Science and EngineeringNews, 10 December www.bbc.co.uk/news/science-environment-37902773#share-tools

Andrews J and Shabini B 2012 ’Where does hydrogen fit in a sustainable energy economy? Proc.Eng 49 15–25

BAC 2012 Barcelona district cooling with BAC’s ice thermal storage, Baltimore Aircoil Companywww.baltimoreaircoil.com/english/14032/barcelona-district-cooling-with-bacs-ice-thermal-storage

Blue and Green 2016 Your home can be powered by solar around the clock, Blue and Green website, 24 August http://blueandgreentomorrow.com/features/home-can-powered-solar-around-clock

BINE 2017 Concrete sphere for the sea successfully stores electricity, BINE Information Service, 9January www.bine.info/nc/en/newsoverview/news/betonkugel-im-meer-speichert-erfolgreich-strom/

Building4Change 2012 Storing heat energy makes the most of fluctuating wind power, BRE Trust,27 March www.building4change.com/page.jsp?id=1219

Burrows L 2017 Long-lasting flow battery could run for more than a decade with minimumupkeep, Harvard University, Cambridge, press release, 9 Feb www.seas.harvard.edu/news/2017/02/long-lasting-flow-battery-could-run-for-more-than-decade-with-minimum-upkeep

Cassey T 2016 New ‘super battery’ energy storage breakthrough aims at $54 per kWhCeantechnica 26 February cleantechnica.com/2016/02/26/new-energy-storage-solution-could-hit-magic-54-mark/

Conca J 2016 Vanadium-flow batteries: the energy storage breakthrough we’ve needed Forbes13 December www.energystorageforum.com/europe/blog/vanadium-flow-batteries-energy-storage-breakthrough-needed

Coyne B 2016 Somerset solar site installs ‘grid scale’ Tesla battery storage unit The Energyst 23September http://theenergyst.com/who-needs-an-efr-contract-somerset-solar-site-installs-grid-scale-tesla-battery-storage-unit/

Edie 2015 Could ‘hydricity’ be the next big energy breakthrough? Edie.net news service 15December www.edie.net/news/6/Hydricity-definition-renewable-energy-generation-business-trends/

EASE 2016 European Association for the Storage of Energy www.ease-storage.eu/technologies.html

EIA 2011 Chart from US Energy Information Administration web site, based on Energy StorageAssociation data www.eia.gov/todayinenergy/detail.cfm?id=4310

Elliott D 2013 Renewable Energy: A Review of Sustainable Energy Supply Options (Bristol:Institute of Physics Publications)) http://iopscience.iop.org/book/978-0-750-31040-6

Elliott D 2016a Balancing Green Power (Bristol: Institute of Physics Publications)) http://iopscience.iop.org/book/978-0-7503-1230-1

Energy Storage Systems

14

Elliott D 2016b A balancing act for renewables Nat. Energy 1 1 www.nature.com/articles/nenergy20153

ESA 2017 US Energy Storage Association http://energystorage.org/energy-storage/energy-storage-technologies

ESF 2017 Energy Storage Forum, network, conferences and global market reports www.energystorageforum.com/Energy_Storage_World_Markets_Report_2014-2020.compressed.pdf

Flynn J 2017 Stanford engineers create a low-cost battery for storing renewable energy, StanfordUniversity news release, 7 February http://news.stanford.edu/2017/02/07/stanford-engineers-create-low-cost-battery-storing-renewable-energy/

Fraenkel P 2016 Storing electricity by raising massive weights, UK Energy Research Centre Blog,16 August www.ukerc.ac.uk/network/network-news/guest-blog-storing-electricity-by-raising-massive-weights.html

Fronius 2013 Fronius Energy Cell, company web site www.fronius.com/cps/rde/xchg/SID-316A5419-05794498/fronius_international/hs.xsl/83_19199_ENG_HTML.htm

Gaelectric 2017 CAES system, company website www.gaelectric.i.e./energy-storage-projects/project-caes-larne-ni/

Gammer D 2015 The role of hydrogen storage in a clean responsive power system, UK EnergyTechnologies Institute www.eti.co.uk/insights/carbon-capture-and-storage-the-role-of-hydrogen-storage-in-a-clean-responsive-power-system/

Garvey S 2011 The dynamics of integrated compressed air renewable energy systems Renew.Energy 39 271–92

Gravity Power 2014 Gravity Power, company web site www.gravitypower.net/technology-gravity-power-energy-storage/

Gruver M and Brown M 2014 Renewable energy plan hinges on huge Utah caverns, AssociatedPress, 24 September http://bigstory.ap.org/article/3084cb4c459f4ffd9b666f5d5d2e44e3/wind-energy-proposal-would-light-los-angeles-homes

GPI 2013 Green Power Island, web site www.greenpowerisland.dkHarvard 2014 Organic mega flow battery promises breakthrough for renewable energy (Cambridge

MA: Harvard University) press release, 8 January http://www.seas.harvard.edu/news/2014/01/organic-mega-flow-battery-promises-breakthrough-for-renewable-energy

Highview 2017 Highview Cryogenic energy storage, company web site www.highview-power.com/

Hoster H 2015 Lithium-air: a battery breakthrough explained The Conversation 2 Novemberhttps://theconversation.com/lithium-air-a-battery-breakthrough-explained-50027

Hsu C 2016 Glow-in-the-dark dye could fuel liquid-based batteries, University at Buffalo, newrelease, 1 November www.buffalo.edu/ubnow/stories/2016/11/bodipy-batteries

Hydrogen Office 2017 Hydrogen projects, Scotland www.hydrogenoffice.com/Ice Energy 2013 Ice Energy, company web site www.ice-energy.com/IDTechEx 2014 Supercapacitors can destroy the lithium-ion battery market, IDTechEx

Research Article www.idtechex.com/research/articles/supercapacitors-can-destroy-the-lithium-ion-battery-market-00006649.asp

IEA 2005 Variability of wind power and other renewables: management options and strategies,International Energy Agency, Paris www.uwig.org/iea_report_on_variability.pdf

IEA 2011 Harnessing Variable Renewables, International Energy Agency, Paris www.iea.org/publications/freepublications/publication/harnessing-variable-renewables.html

Energy Storage Systems

15

IEI 2016 Key trends driving change in the electric power industry (Washington DC: Institutefor Electric Innovation) http://s2.q4cdn.com/825052743/files/doc_news/AES%20In%20the%20news/2016/Key_Trends_Driving_Change_Volume_II_FINAL_Web.pdf

ITM 2016 100 MW electrolyser plant designs to be launched at Hannover, ITM Power newsitem, 12 December www.itm-power.com/news-item/100mw-electrolyser-plant-designs-to-be-launched-at-hannover

ITM 2017 ITM Power, company web site http://www.itm-power.comIRENA 2015 Renewables and electricity storage (Abu Dhabi: International Renewable Energy

Agency) www.irena.org/DocumentDownloads/Publications/IRENA_REmap_Electricity_Storage_2015.pdf

Isentropic 2016 Isentropic gravel heat storage system, company web site www.isentropic.co.uk/JRC 2012 District Heating and Cooling (Petten: European Commission Joint Research Centre)

http://setis.ec.europa.eu/system/files/JRCDistrictheatingandcooling.pdfJRC 2013 Assessment of the European potential for PHS, European Commission Joint Research

Centre http://setis.ec.europa.eu/newsroom-items-folder/jrc-report-european-potential-pumped-hydropower-energy-storage

LaMonica M 2013 Ambri’s better grid battery, MIT Technology Review, 18 February http://www.technologyreview.com/featuredstory/511081/ambris-better-grid-battery/

Lightsail 2017 Lightsail compressed air system promotion www.lightsail.comLin et al 2015 An ultrafast rechargeable aluminium-ion battery Nature 520 324–8Lombardo T 2016 Massive 800 megawatt-hour battery to be deployed in China, Engineering.com,

5 June www.engineering.com/ElectronicsDesign/ElectronicsDesignArticles/ArticleID/12312/Massive-800-MegaWatt-hour-Battery-to-Be-Deployed-in-China.aspx

Ludington 2013 Pumped storage plant on lake Michigan www.consumersenergy.com/content.aspx?id=1830

Maloney P 2016 With big gains in the Q3, energy storage market set to finish strong in 2016,Utility Drive, 13 December www.utilitydive.com/news/with-big-gains-in-the-q3-energy-stor-age-market-set-to-finish-strong-in-201/432211/

Marstal 2017 Marstal Sunstore project details, PlanEnergi, Denmark: www.solarmarstal.dk/media/2854117/summary-technical-description-marstal.pdf

Michel J 2017 Hamburg considers innovative heat storage scheme, Enerfgy Post, 8 Februaryhttp://energypost.eu/hamburg-considers-innovative-heat-storage-scheme/

Mis M 2017 British village gets solar battery boost in green energy trial Reuters 20 January www.reuters.com/article/us-britain-renewables-solar-idUSKBN154296

MIT 2013 Wind power even without the wind, MIT News release, 25 April http://web.mit.edu/newsoffice/2013/wind-power-even-without-the-wind-0425.html

Moixa 2017 Moixa battery, company web site http://www.moixa.comNaan R 2015 Why energy storage is about to get big—and cheap, Ramez Naan Energy Blog, 14

April http://rameznaam.com/2015/04/14/energy-storage-about-to-get-big-and-cheap/NATGAS 2013 Natural gas Storage in the USA, Natgas web site http://naturalgas.org/natural-

gas/storage/Nykvist B and Nilsson M 2015 Rapidly falling costs of battery packs for electric vehicles Nat.

Clim. Change 5 329–32Okazaki T, Shira Y and Nakamura T 2015 Concept study of wind power utilizing direct thermal

energy conversion and thermal energy storage Renew. Energy 83 332–8

Energy Storage Systems

16

PennEnergy 2014 $8B renewable energy initiative proposed for Los Angeles, PennEnergy, 23September www.pennenergy.com/articles/pennenergy/2014/09/8b-renewable-energy-initiative-proposed-for-los-angeles.html

PD Energy 2013 Vanadium flow cells, company web site www.pdenergy.comPOST 2015 Energy Storage, UK Parliamentary Office of Science and Technology, POSTNote

492: www.parliament.uk/briefing-papers/POST-PN-492/energy-storageRadford T 2015 Safer battery could spark investment in renewables Climate Network 30

September http://climatenewsnetwork.net/safer-battery-could-spark-investment-in-renewables/Reid G 2016 The coming of the hybrid ultracapacitor and lithium-ion system and what it means

for EVs, power generation and stationary storage Linkedin Pulse 16 December www.linkedin.com/pulse/coming-hybrid-ultracapacitor-lithium-ion-system-what-means-reid

Robertson J 2016 Half the price in half the time solar storage innovation harnesses new energyfrontier The Guardian 23 February www.theguardian.com/sustainable-business/2016/feb/23/half-the-price-in-half-the-time-solar-storage-innovation-harnesses-new-energy-frontier

Safe Hydrogen 2013 www.safehydrogen.comSandia 2012 Sandia National Laboratories researchers find energy storage ‘solutions’ in MetILs,

press release https://share.sandia.gov/news/resources/news_releases/metils/SINTEF 2013 Storage power plant on the seabed Science Daily 15 May www.sciencedaily.com/

releases/2013/05/130515085343.htmSmith P 2014 Siemens developing thermal energy storage system Windpower Monthly 2

November www.windpowermonthly.com/article/1320266/siemens-developing-thermal-energy-storage-system

Spector J 2016 Storage costs come down across technologies and applications according to LazardReport GreenTech Media 19 December www.greentechmedia.com/articles/read/energy-storage-costs-lcos-lazard-lithium-ion-flow-batteries

Stornetic 2017 Stornetic Company promotion: http://stornetic.com/assets/downloads/stornetic_general_presentation.pdf

Tesla 2015 Tesla Powerwall promotional web site www.teslamotors.com/powerwall

Energy Storage Systems

17