《麻省理工学院:2022能源储存的未来研究报告(英文版)(405页).pdf》由会员分享,可在线阅读,更多相关《麻省理工学院:2022能源储存的未来研究报告(英文版)(405页).pdf(404页珍藏版)》请在三个皮匠报告上搜索。
1、The Future of Energy Storage An Interdisciplinary MIT StudyAN INTERDISCIPLINARY MIT STUDYFuture ofTheEnergyStorageii MIT Study on the Future of Energy StorageCopyright 2022 Massachusetts Institute of Technology.All rights reserved.Incorporated in the cover art is a 3D concept illustration of battery
2、 cells,a form of electrochemical energy storage.Getty ImagesISBN(978-0-578-29263-2)Second version,published June 3,2022.Other reports in the MIT Future of series:The Future of Nuclear Power(2003)The Future of Geothermal Energy(2006)The Future of Coal(2007)Update to the Future of Nuclear Power(2009)T
3、he Future of Natural Gas(2011)The Future of the Nuclear Fuel Cycle(2011)The Future of the Electric Grid(2011)The Future of Solar Energy(2015)The Future of Nuclear Energy in a Carbon-Constrained World(2018)MIT Study on the Future of Energy Storage iiiStudy participantsStudy chairRobert ArmstrongChevr
4、on Professor,Department of Chemical Engineering,MITDirector,MIT Energy InitiativeStudy co-chairYet-Ming ChiangKyocera Professor,Department of Materials Science and Engineering,MITExecutive directorHoward GruenspechtSenior Energy Economist,MIT Energy InitiativeStudy groupFikile BrushettCecil and Ida
5、Green Associate Professor,Department of Chemical Engineering,MITJohn DeutchInstitute Professor,Department of Chemistry,MITSeiji EngelkemierPhD Student,Department of Mechanical Engineering,MITEmre GenerResearch Scientist,MIT Energy InitiativeRobert JaffeMorningstar Professor of Science,Department of
6、Physics,MITPaul JoskowElizabeth and James Killian Professor of Economics and Management,Department of Economics,MITDharik MallapragadaResearch Scientist,MIT Energy InitiativeElsa OlivettiEsther and Harold E.Edgerton Associate Professor,Department of Materials Science and Engineering,MITCo-Director,M
7、IT Climate and Sustainability ConsortiumRichard SchmalenseeProfessor of Economics,Emeritus,Department of Economics,MITDean and Howard W.Johnson Professor of Management,Emeritus,Sloan School of Management,MITRobert StonerDeputy Director for Science and Technology,MIT Energy InitiativeChi-Jen YangForm
8、er Visiting Researcher,MIT Energy InitiativeContributing authorsBjorn BrandtzaegFormer Visiting Fellow,Sloan School of Management,MITPatrick BrownFormer Research Scientist,MIT Energy InitiativeKevin HuangResearch Scientist,Department of Materials Science and Engineering,MITJohannes PfeifenbergerVisi
9、ting Scholar,MIT Center for Energy and Environmental Policy ResearchResearch advisorsFrancis OSullivanSenior Lecturer,Sloan School of Management,MITYang Shao-HornJR East Professor of Engineering,Department of Mechanical Engineering,MITiv MIT Study on the Future of Energy StorageStudents and research
10、 assistantsMeia AlsupMEng,Department of Electrical Engineering and Computer Science(20),MITAndres BadelSM,Department of Materials Science and Engineering(22),MITMarc BarbarPhD,Department of Electrical Engineering and Computer Science(22),MITWeiran GaoPhD Candidate,Department of Chemical Engineering,
11、MITDrake HernandezSM,Technology and Policy(21),MITCristian JungeMSc,Engineering and Management(22),MITThaneer Malai NarayananPhD,Department of Mechanical Engineering(21),MITKara RodbyPhD,Department of Chemical Engineering(22),MITCathy WangSM,Technology and Policy(21),MIT MIT Study on the Future of E
12、nergy Storage vAdvisory CommitteeLinda Stuntz ChairPartner,Stuntz,Davis&Staffier,P.C.Norman BayPartner,Willkie Farr&Gallagher LLPTerry BostonStrategic Partner,AcelerexMark BrownsteinSenior Vice President,Energy,Environmental Defense FundJudy ChangUndersecretary of Energy,Massachusetts Office of Ener
13、gy and Environmental AffairsManlio CovielloPresident,Terna PlusGeorge CrabtreeDirector,Joint Center for Energy Storage Research(JCESR),Argonne National LaboratoryPhilip DeutchFounder and CEO,NGP Energy Technology Partners IIIJulien Dumoulin-SmithManaging Director and Head of U.S.Power,Utilities,and
14、Alternative Energy Research,Bank of America SecuritiesElizabeth E.EndlerSenior Principal Science Expert(Electrification,Integration,and Storage)and Principal Technology Advisor Electric Power,Shell International Exploration&ProductionAndy KarsnerCo-Founder,Elemental LabsArun MajumdarJay Precourt Pro
15、vostial Chair Professor,Stanford UniversityLucio MonariFormer Director,Infrastructure,Europe and Central Asia,World BankPedro J.PizarroPresident and CEO,Edison InternationalJohn PodestaFounder and Chair,Board of Directors,Center for American ProgressPraveer SinhaCEO and Managing Director,Tata Power
16、Co.,Ltd.Fredrick StaInvestment Manager,Equinor Ventures,Equinor ASAEllen WilliamsDistinguished University Professor,University of MarylandWhile the members of the advisory committee provided invaluable perspective and advice to the study group,individual members may have different views on one or mo
17、re matters addressed in the report.They are not asked to individually or collectively endorse the report findings and recommendations.MIT Study on the Future of Energy Storage viiTable of contentsForeword and acknowledgments ixExecutive summary xiChapter 1 Introduction and overview 1Chapter 2 Electr
18、ochemical energy storage 15Chapter 3 Mechanical energy storage 67Chapter 4 Thermal energy storage 113Chapter 5 Chemical energy storage 147Chapter 6 Modeling storage in high VRE systems 171Chapter 7 Considerations for emerging markets 233 and developing economiesChapter 8 Governance of decarbonized p
19、ower systems 271 with storageChapter 9 Innovation and the future of energy storage 291AppendicesAppendix A Cost and performance calculations for 301 electrochemical energy storage technologiesAppendix B Cost and performance calculations for 319 thermal energy storage technologiesAppendix C Details o
20、f the modeling analysis for 327 high-VRE systems with energy storage in three U.S.regionsAppendix D Details of the modeling analysis for 349 developing country marketsviii MIT Study on the Future of Energy StorageAcronyms and abbreviations 367List of figures 369List of tables 373Glossary 376 MIT Stu
21、dy on the Future of Energy Storage ixForeword and acknowledgments The Future of Energy Storage study is the ninth in the MIT Energy Initiatives Future of series,which aims to shed light on a range of complex and vital issues involving energy and the envi-ronment.Previous studies have focused on the
22、role of technologies such as nuclear power,solar energy,natural gas,geothermal,and coal(with capture and sequestration of carbon dioxide emissions),as well as systems such as the U.S.electric power grid.Central to all these studies is understanding the role these particular technologies can play in
23、both decar-bonizing global energy systems and meeting future energy needs.Energy storage will play an important role in achieving both goals by complementing variable renewable energy(VRE)sources such as solar and wind,which are central in the decarbonization of the power sector.The study will prove
24、 beneficial for a wide array of global stakeholders in government,industry,and academia as they develop the emerging energy storage industry and consider changes in planning,oversight,and regulation of the electricity industry that will be needed to enable greatly increased reliance on VRE generatio
25、n together with storage.The report is the culmi-nation of more than three years of research into electricity energy storage technologiesincluding opportunities for the development of low-cost,long-duration storage;system modeling studies to assess the types and roles of storage in future,deeply-deca
26、rbonized,high-VRE grids in both U.S.regions and emerging market,developing economy countries;and implications for electricity system planning and regulation.The study was guided by a distinguished external Advisory Committee whose members dedicated a significant amount of their time to participate i
27、n multiple meetings;to comment on our preliminary analysis,findings,and recommendations;and to make available experts from their own organizations to answer questions and contribute to the content of the report.We would especially like to acknowledge the wise and able leadership of the Committees Ch
28、air,Linda Stuntz.The study is certainly better as a result of this thoughtful,expert input.However,the study is the responsibility of the MIT study group;the Advisory Committee members do not necessarily endorse all of its findings and recommendations,either individually or collectively.The Future o
29、f Energy Storage study gratefully acknowledges our sponsors:Core funding was provided by The Alfred P.Sloan Foundation and The Heising-Simons Foundation.Additional support was provided by MIT Energy Initiative members Shell and Equinor.As with the Advisory Committee,the sponsors are not responsible
30、for and do not necessarily endorse the findings and recommendations.That responsibility lies solely with the MIT study group.This study was initiated and performed within the MIT Energy Initiative.Alexandra Goodwin,Senior Administrative Assistant at MITEI,provided support to both the study team and
31、the Advisory Committee.Special thanks are due to the MITEI events team,specifically to Carolyn Sinnes,Administrative Assistant;Debi Kedian,Events Manager;and x MIT Study on the Future of Energy StorageKelly Hoarty,Events Planning Manager,for their skill and dedication.Thanks also to MITEI communicat
32、ions team members Jennifer Schlick,Digital Project Manager;Kelley Travers,Communications Specialist;Turner Jackson,Communications Assistant;and Tom Melville,Communications Director.Additional thanks to Martha Broad,MITEI Executive Director,for her vital role in bringing the study to fruition.Finally
33、,we thank Marika Tatsutani for editing the report with great skill and dedication.Executive summary xiExecutive summaryThis interdisciplinary MIT study examines the important role of energy storage in future decarbonized electricity systems that will be central to the fight against climate change.De
34、ep decarbonization of electricity generation together with electrification of many end-use activities is necessary to limit climate change and its damages.Wind and solar generationwhich have no operating carbon dioxide emissions,have experienced major cost reductions,and are being deployed at scale
35、globallyare likely to provide a large share of future total generation.Unlike traditional generators,the output from these variable renewable energy(VRE)resources depends on weather conditions,which sometimes change rapidly;thus,VRE generators cannot be dispatched to follow variations in electricity
36、 demand.Electricity storage,the focus of this report,can play a critical role in balancing e lectricity supply and demand and can provide other services needed to keep decarbonized electricity systems reliable and cost-effective.As we discuss in this report,energy storage encompasses a spectrum of t
37、echnologies that are differentiated in their material requirements and their value in low-carbon electricity systems.As electricity grids evolve to include large-scale deployment of storage technologies,policies must be adjusted to avoid excess and inequitable burdens on consumers,to encourage elect
38、rification for economy-wide decarbonization,and to enable robust economic growth,particularly in emerging market developing economy countries.Social justice and equity must be included in system design.The time horizon for this study is 2050,consistent with previous Future of studies in this series,
39、though we are also interested in technologies that can be deployed at scale in the nearer timeframe of 2030.Energy storage enables cost-effective deep decarbonization of electric power systems that rely heavily on wind and solar generation without sacrificing system reliability.Assuming favorable co
40、st reduction trends for VRE technologies continue,the modeling analysis conducted for this study identifies cost-effective pathways for decarbonizing electricity systemsreducing emissions by 97%99%relative to 2005 levels in the United States,for examplewhile maintaining grid reliability.Efficient de
41、carbonization will require substan-tial investments in multiple energy storage technologies,as well as in transmission,clean generation,and demand flexibility.If“negative emissions”technologiesthat is,technologies for removing carbon dioxide from the atmo-spherebecome available,they can provide emis
42、sions offsets that enable small amounts of natural gas generation to be part of a cost-effective net-zero electricity system.Energy storage basicsFour basic types of energy storage(electro-chemical,chemical,thermal,and mechanical)are currently available at various levels of technological readiness.A
43、ll perform the core function of making electric energy generated during times when VRE output is abundant and wholesale prices are relatively low available at times when VRE output is scarce and whole-sale prices are relatively high.This flexibility provides a range of benefits to power systems.An e
44、nergy storage facility can be characterized by its maximum instantaneous power,measured in megawatts(MW);its energy storage capacity,measured in megawatt-hours(MWh);and its round-trip efficiency(RTE),measured as the fraction of energy used for charging storage xii MIT Study on the Future of Energy S
45、toragethat is returned upon discharge.The ratio of energy storage capacity to maximum power yields a facilitys storage duration,measured in hoursthis is the length of time over which the facility can deliver maximum power when starting from a full charge.Most currently deployed battery storage facil
46、ities have storage durations of four hours or less;most existing pumped storage hydro(PSH)facilities have durations of eight to twelve hours or more.Storage technologies also differ in energy density,which is the maximum amount of energy that can be stored per unit volume.Battery technologies with h
47、igh energy density are particularly well-suited for use in electric vehicles(EVs)and mobile electronics;technol-ogies with lower energy density can nonetheless be used for storage in electricity system applications where the efficient use of space is generally less important.Energy storage technolog
48、ies also differ in other attributes,including the extent of facility-specific scale economies(geographical footprint,modularity)and the extent to which their performance degrades with use.The technologies considered in this report fall into three main groups based on their power and energy capacity
49、costs(Figure ES.1).Generally,technologies with low energy-capacity costs and high power-capacity costs(the blue area in the figure)are most suitable for longer duration storage applications(up to multiple days)and less frequent charge-discharge cycles;these include thermal,chemical,metal-air battery
50、,and pumped hydro storage options.Technologies in the brown area,including lithium-ion batteries,are better suited to shorter duration applications(a few hours)and more frequent cycling.Technologies with intermediate capabilities,including flow batteries,are in the green area.Figure ES.1:Three group
51、s of storage technologies based on power-and energy-capacity costsThe blue region,with high power and low energy capacity costs,includes thermal,chemical(e.g.,hydrogen),metal-air battery,and pumped hydro storage technologies.Lithium-ion batteries fall in the brown area,with low power,but high energy
52、-capacity costs;flow batteries fall in the intermediate,green region.In addition to the two parameters displayed in this figure,other cost and performance attributes,e.g.,charge and discharge efficiencies,are also important when comparing storage technologies within and across each class.The full se
53、t of characteristics used in system modeling are discussed in Chapter 6.02004006008001,0001,2001,4000204060800180200Power capacity cost($/kW)Energy capacity cost($/kWh)Executive summary xiiiElectricity system storage technologiesThe study examines electricity-to-electricity storage techno
54、logies in four categories:electro-chemical,thermal,chemical,and mechanical.We do not catalog,let alone evaluate,all options within each of these categories;rather,we focus on examples of storage technologies in each category and seek to highlight issues that apply across a broad set of technologies
55、within these categories.Some of the technologies we consider,such as lithium-ion batteries,pumped storage hydro,and some thermal storage options,are proven and available for commer-cial deployment.Others would require further research,development,and demonstration,and may not be commercially availab
56、le at scale until the 2030s or 2040s.Table ES.1 summarizes our assessment of the availability of various storage technologies and storage-supporting technolo-gies and practices in the near term(by 2030).All the technologies we consider in this report could be commercially available by 2050.Successfu
57、l innovation for energy and many other manufacturing-related technologies typically passes through five stages:idea creation R&D engineering at pilot scale technology demonstration deployment.Table ES.1 indicates the current stage of inno-vation for various storage technologies.The private sector ha
58、s provided significant venture capital for storage technologies generally,and for lithium-ion batteries used in vehicles in particular.As discussed in this study,EV battery development has significantly improved pros-pects for short-duration electricity system storage.So far,long-duration storage te
59、ch-nologies have not experienced similar help from other market drivers.While the value of long-duration storage(12 hours)is low when VRE penetration is low,long-duration storage technologies clearly become more valuable as decarbonization requirements become more stringent and reliance on VRE gener
60、ation grows.This is especially true if grid operators are precluded from using natural-gas-fueled generation,with or without carbon capture and storage,to provide balancing capacity during extended supply troughs for VRE generation or during unusually high levels of demand due to extended extreme we
61、ather events.The value that long-duration storage could provide in a highly decarbonized electricity system argues for increased federal support of various kinds of long-duration storage options,depending on the stage of innovation different technologies have reached.The current policy focus on rela
62、tively near-term decarbonization goals pushes both public and private attention toward downstream technology demonstration and deployment involving relatively mature technologies.The U.S.Department of Energy(DOE)can play a helpful role in this area,but its involvement should reflect two important le
63、ssons learned from past demonstration and deployment efforts.First,Congress should enable more joint technology demonstration projects with industry,unfettered by the Federal Acquisition Regulation and other rules that constrain tech-nology development and demonstration on commercial terms.The purpo
64、se of public investment in technology demonstration and early deployment activity is to disseminate knowledge,which is inconsistent with policies such as requiring cost sharing in exchange for intellectual property rights.Second,efforts to accelerate the deployment of any commercial technology shoul
65、d rely on incentives and mechanisms that reward success but do not interfere in project management.The Biden administration has proposed tax credits for a wide range of storage technologies,in addition to tax credits for transmission and various clean generation technologies,including wind and solar
66、.In contrast to electricity genera-tion technologies,where performance-based payments such as production tax credits can be directly linked to output measures,xiv MIT Study on the Future of Energy Storageperformance-based support for non-generation energy technologies such as storage must be based o
67、n preset development and operational testing measures.Electrochemical storageElectrochemical storage systems,which include well-known types of batteries as well as new battery variants discussed in this study,generally have higher energy density than mechanical and thermal storage systems,but lower
68、energy density than chemical systems.Round-trip efficiency for battery storage ranges widely,from as much as 95%for lithium-ion(Li-ion)chemistries to as little as 40%for metal-air chemistries.A compact footprint and indepen-dence from hydrological and geological resources make batteries a versatile
69、and highly scalable technology that can be sized for a range of applications,from power plants down to residential uses.Our study yields several key takeaways.Lithium-ion batteries possess high energy density,high power density,and high roundtrip efficiency,facilitating their near-ubiquitous use in
70、electric vehicles and their widespread use in short-duration(typically 4 hours or less)electricity system storage applications.The dominant role of Li-ion batteries in the rapidly growing EV market has attracted significant investment from the private sector and is supporting rapid expansion of batt
71、ery manufacturing capacity in the United States(currently most of this investment is coming from foreign firms).Cost and limits on the availability of key materials currently used in battery manufacture have set a floor on Table ES.1:Summary of findings on the current innovation status of selected e
72、nergy storage technologiesTechnologyCurrent innovation statusChapterElectrochemical storage2Li-ion batteries2452Flow batteries(aqueous inorganic)2452Flow batteries(aqueous organic)1232NaS batteries452Metal-air batteries232Critical materials supply(metals and rare earths)1232Battery re-cycling12342Ba
73、ttery second use122Advanced power electronics234Pumped hydro storage453Thermal storage2344Hydrogen5Production,transport,storage1245H2 generationphotoelectric,very high temperature gas reformation,advanced electrolysis251 Idea creation,study,and analysispublic and private sponsors2 R&Duniversity,nati
74、onal laboratory,and private sector performers3 Pilot scale engineering4 Demonstration&testing5 Deploymentdepends upon progress and market conditions.Further discussion is found in the chapters listed in the right column.Executive summary xvLi-ion battery costs and may constrain future deployment,ins
75、piring a shift toward chemistries that use more earth-abundant elements.Other advances being vigorously pursued for Li-ion battery components will also support cost and performance improvements.With these trends,Li-ion batteries will continue to be a leading technology for EVs and for short-duration
76、 storage,but their storage capacity costs are unlikely to fall low enough to enable widespread adoption for long-duration(12 hours)electricity system applications.To enable economical long-duration energy storage(12 hours),the DOE should support research,development,and demonstration to advance alte
77、rnative electrochemical storage technologies that rely on earth-abundant materials.Cost,lifetime,and manufacturing scale requirements for long-duration energy storage favor the exploration of novel electro-chemical technologies,such as redox-flow and metal-air batteries that use inexpensive charge-s
78、torage materials and battery designs that are better suited for long-duration appli-cations.While several novel electrochemical technologies have shown promise,remaining knowledge gaps with respect to key scientific,engineering,and manufacturing challenges suggest high value for concerted government
79、 support.Innovation in these technologies is being actively pursued in other countries,notably China.Thermal energy storageThermal energy storage(TES)has attributes suitable for long-duration storage including the ability to store heat effectively in low-cost materials.This report discusses several
80、generic TES strategies that reflect varying degrees of technology readiness.One possible near-term TES approach focuses on reducing the cost of converting heat to electric power,the main component of overall TES system cost,by reusing steam turbines at existing power plants and adding thermal storag
81、e and new steam generators in place of existing fossil-fuel boilers.This retrofit can be done today using commercially available technologies,and it may be attractive to plant owners and local communities as a way to use assets that would otherwise be abandoned as electricity systems decarbonize.Che
82、mical energy storage:HydrogenHydrogen is widely considered a leading chemical energy storage medium because it can be directly produced from electricity in a single step and consumed either as a fuel to produce power or as a feedstock or heat source for other industrial processes.We focus on hydroge
83、n in this chemical storage section.Hydrogens role as a form of energy storage for the electricity sector will likely depend on the extent to which hydrogen is used in the overall economy,which in turn will be driven by the future costs of hydrogen production,transportation,and storage,and by the pac
84、e of innovation in hydrogen end-use applications.Hydrogen is currently produced,transported,and sold as a feedstock for numerous industrial processes.Today,the dominant technology for hydrogen production relies on fossil fuels and produces carbon emissions.The ability to produce low-carbon hydrogen
85、by splitting water(also known as electrolysis)using low-carbon grid electricity can support decarbonization in end-use sectors such as industry and transportation,as well as in the power sector.Figure ES.2 shows how hydrogen produced via electrolysis can serve as a low-carbon fuel for industry as we
86、ll as for electricity generation during periods when VRE generation is low.Use of electrolyzers as a dispatchable load for the power system could also reduce the costs of power system decarbonization by increasing capacity utilization of VRE resources.xvi MIT Study on the Future of Energy StorageWe
87、support the effort that the DOE is leading to create a national strategy that addresses hydrogen production,transportation,and storage.In particular,the ability of existing natural gas transmission pipelines to carry hydrogen without suffering embrittlement,either at reduced pressures or if hydrogen
88、 is blended with natural gas or other compounds,remains an open question that deserves g overnment-supported study by the DOE and the U.S.Department of Transportation.An important step in this direction is the call in recent legislation for the creation of at least four hydrogen hubs.Mechanical stor
89、ageElectrical energy can be converted into various forms of mechanical energy such as gravitational potential energy and kinetic energy;electrical energy can also be used to compress a gas such as air.Some of these forms of mechanical energy are suitable for large-scale and long-duration energy stor
90、age.As a category,mechanical energy storage includes a wide variety of technologies.A common feature of all these technologies,however,is that their energy density is much lower than the energy density of chemical or electrochemical storage technologies.Consequently,mechanical energy storage systems
91、 tend to have large footprints and require geologically favorable locationsthus,they are not well suited for use in small-scale facilities.Pumped storage hydropower(PSH)stores energy in the potential energy of water pumped uphill.PSH is a mature,widely deployed tech-nology that accounts for well ove
92、r 90%of the functional grid-scale energy storage capacity that currently exists,both globally and in the United States.Yet,PSH deployment has signifi-cantly slowed in the United States and in many other countries since the 1990s(the notable exception is China).This trend reflects,among other factors
93、,the reduced value of intraday energy arbitrage as a result of the increased use of flexible gas-fired generation.In addition,PSH projects have high initial costs and inflex-ible sizing and siting requirements;historically,these projects have also experienced long construction periods and major cost
94、 overruns.Figure ES.2:Illustration of cross-sector(power-industry)coupling of hydrogenCoupling leads to cost reductions through increased utilization of variable renewable energy assets and operation of electrolyzers as dispatchable loads.ElectrolysisH2 demand(industry)H2 to powerH2 storageGrid elec
95、tricityGrid electricityH2H2H2H2 Executive summary xviiWhile not strictly an electricity-to-electricity storage technology,existing conventional hydropower systems with storage reservoirs could play a larger role in balancing supply and demand in electricity systems that rely heavily on VRE generatio
96、n.Where there is significant potential to play this role,system planners should consider options for increasing the amount of water that is held behind dams for use in balancing electricity systems.Compressed air energy storage(CAES)systems store pressurized air in underground cavities or above-grou
97、nd tanks;some CAES systems also store the heat that is generated when the air is compressed.This technology has been widely discussed as a potential grid-scale energy storage option,but it faces significant hurdles to deployment at scale.Although cost estimates for CAES are subject to multiple uncer
98、tainties,estimates of energy cost for this technology are generally higher than estimates for other energy storage technologies that are expected to be available in the future.Co-locating energy storage systems with existing power plants that are being retired could reduce storage costs by enabling
99、the reuse of existing grid interconnections and,in some cases,other power plant components.Using existing interconnections would save time as well as cost.In addition,as noted above,existing turbines can be reused in thermal storage systems that repower existing turbines using zero-emissions heat or
100、 fuel.The DOE should investigate the cost and system impacts of thermal storage technologies and other options that offer promise for reusing existing assets,as well as the social acceptance of such reuse strategies by neighboring communities,and should sponsor demonstration projects where appropria
101、te.Efficient high-VRE electricity systems with storage:Modeling results and implications for governance and policyThis section examines potential roles for storage in a developed country context and in an emerging market developing economy country context.These two country contexts are illustrated b
102、y results for three different regions in the U.S.and for India,respectively.Modeling results for a developed country:Three U.S.regionsOur modeling for the U.S.power sector focused on three regions:the Northeast(New York and New England),the Southeast,and Texas for largely“greenfield”systems in 2050.
103、These regions differ significantly in their electricity demand profiles,wind and solar resources,and availability of hydropower and existing nuclear resources.These differences affect both the least-cost generation mix in the absence of emissions constraints and the cost of achieving different degre
104、es of decarbonization.Figure ES.3 shows modeled projections for annual generation,deliverable energy capacity,and system cost of electricity for each region in 2050 under two policy scenarios:no carbon constraint and emissions constrained to 5 grams of carbon dioxide per kilowatt-hour(gCO2/kWh).If 2
105、050 electricity demand remains the same as the 2018 level,then reducing the average carbon intensity of the U.S.power sector to 5 gCO2/kWh would lower 2050 emissions by 99.2%relative to 2005.On the other hand,if electricity use grows such that demand in 2050 is greater than in 2018,as projected in t
106、he electricity demand scenario used to model energy storage impacts for this study(Mai et al.2018),a U.S.sector-wide average carbon intensity of 5 gCO2/kWh would deliver a 98.7%reduction in power sector emis-sions relative to 2005.The illustrative results in Figure ES.3 are from scenarios that assum
107、e only Li-ion battery and pumped hydro storage xviii MIT Study on the Future of Energy Storageare available;our modeling of U.S.regions(discussed in Chapter 6)examines a wide range of other storage technologies.The ability of storage technologies to substitute for,or complement,essentially all other
108、 elements of a power system(including generation,transmission,and demand response),coupled with uncertain climate change impacts on electricity demand and supply,means that more sophisticated analytical tools are needed to plan,operate,and regulate the power systems of the future and to ensure that
109、these systems are reliable and efficient.Important focus areas include system stability and dispatch(including enabling the participation and compensation of distrib-uted storage and generation(PV)assets in system dispatch and wholesale markets),resource adequacy,and retail rate design.The developme
110、nt of new analytical tools must be accompanied by additional support for complementary staffing and upskilling programs at regulatory agencies.This effort should be led by the DOE in cooperation with independent system operators and regional transmission organizations(ISOs/RTOs).The distribution of
111、hourly wholesale prices or marginal value of energy will change in deeply decarbonized bulk power systems,with many more hours of zero or very low prices and more hours of high prices compared to todays wholesale markets.Figure ES.3:Annual generation relative to demandAnnual generation relative to d
112、emand,deliverable energy capacity from storage(measured in hours of discharge at mean load),and system average cost of electricity(SCOE)in the Northeast(NE),Southeast(SE),and Texas in 2050.Modeling results are shown for a scenario with no limit on emissions(bottom half of each chart)and for a policy
113、 scenario with an emissions intensity limit of 5 gCO2/kWh(top half of each chart)(note that the policy scenario assumes decarbonization to a level that reduces U.S.power sector emissions by approximately 99%relative to 2005).SCOE includes total annualized investment;fixed O&M;operational costs of ge
114、neration,storage,and transmission;and any non-served energy penalty.Emissions intensity under the“No Limit”policy case for each region is as follows:NE:253 gCO2/kWh,SE:158 gCO2/kWh,Texas:92 gCO2/kWh.For the Northeast region,“Wind”represents the sum of onshore and offshore generation.In this illustra
115、tion,Li-ion batteries are the sole new technology deployed for energy storage purposes in the power sector.The full report discusses modeling results for a wide range of storage technologies,of which Li-ion batteries are only one example.PHS=Pumped Hydro Storage.VOM=Variable O&M cost.No limit Annual
116、 generation(relative to demand)Emission policies(gCO2/kWh)®ion Deliverable energy capacity(hours mean load)CCGTCCGT_CCSWindUtility PVLi-ionNuclear Canadian hydro Existing hydroPHSDistr PV Gen/storage inv+FOMVOM+fuelDemand responseStartupNetwork expansionSCOE($/MWh)Executive summary xixThis is bec
117、ause VRE-dominant bulk power systems with storage will have relatively high fixed(capital)costs and relatively low marginal operating costs compared to todays bulk power systems,which largely rely on thermal generators.Figure ES.4 compares the distribution of historical hourly wholesale electricity
118、prices for 2018 and 2019 in the ERCOT system,which covers nearly all of the state of Texas,with 2050 scenarios.Bars represent the distributions of prices for the no-limit and carbon-constrained Texas modeling cases.Increased reliance on VRE generation,with zero marginal cost,greatly increases the pe
119、rcentage of hours when prices,represented by marginal system costs in our modeling,are under$5 per MWh.This effect increases as the carbon constraint becomes more stringent(i.e.,allowable emissions are ratcheted down).During the highest-price hours,shown at the top of the bars and in the exploded se
120、ction of the figure,modeled prices are significantly above those in the present ERCOT market.The combination of relatively high capital costs and many more hours when prices are very low will create financing challenges for both VRE generation and storage,particularly since regulators will likely co
121、ntinue to cap(as they do at present)extremely high prices that could otherwise support cost recovery.Future patterns of wholesale electricity prices and the goal of decarbonizing other sectors through elec trification with decarbonized electricity also reinforces the benefit of adopting retail prici
122、ng and retail load management options that reward all consumers for shifting electricity uses away from times when high wholesale prices indicate scarcity to times when low wholesale prices signal abundance.Figure ES.4:Hourly marginal wholesale price of energy for TexasHourly marginal wholesale pric
123、e of energy for Texas under various emissions scenarios ranging from no limit(NL,3rd bar from left)to 1 gCO2/kWh(right-most bar).The price bands are based on the known marginal cost of various generation technologies;we zoom in on the top 4%of the price bands to show the price distributions at that
124、extreme.Historical price distributions in ERCOT are shown for reference.For the purposes of this figure,we assume Li-ion battery storage only.The effect of including other storage technologies on these results is discussed in Section 6.3.4.Price bandsPrice distribution by price bandsPrice bandsPrice
125、 bandsxx MIT Study on the Future of Energy StorageFigure ES.5:Impact of Li-ion storage cost projections on cost-optimal bulk power system evolution in IndiaCO2 emissionsmill.tonnesper yearInstalled capacity(1st row),annual energy generation(2nd row),storage energy capacity(3rd row),and annual CO2 em
126、issions(4th row).Results in the left column are for a reference case,which uses a mean estimate for future Li-ion battery capital costs.The right column assumes a low-cost trajectory for future Li-ion battery capital costs.Executive summary xxiTransmission expansion,which allows for increased VRE de
127、ployment in locations with higher-quality VRE resources and improves VRE integration by balancing resource inter-mittency across connected areas and smoothing the effects of geographical differences in VRE supply and demand,is also important for cost-effective decarbonization.The current likelihood
128、that cost-effective transmission projects to bring generation from areas with high-quality VRE resources to major load centers will face extended delays or possible rejection suggests the need for statutory and regulatory changes to reduce barriers to transmission expansion.A shortfall in new transm
129、ission capacity may lead to a larger role for storage as well as higher costs in future decarbonized electricity systems.Modeling results for an emerging market,developing economy country:IndiaCoal-dependent emerging market and devel-oping economy countries that lack access to abundant low-cost gas
130、or gas infrastructure,such as India,represent a very large and important future market for electricity-system applications of energy storage technologies.Modeling for this study suggests that energy storage will be deployed predomi-nantly at the transmission level,with important additional applicati
131、ons within urban distribu-tion networks.Overall economic growth and,notably,the rapid adoption of air conditioning will be the chief drivers of energy storage deployment.Assuming continued technology cost declines,we find that VRE generation and storage compete favorably with new coal from a cost st
132、andpoint in India over the medium and long term,but existing coal plants linger absent carbon pricing,as shown on the left panel of Figure ES.5.Modeling results for a scenario that assumes the availability of low-cost storage and VRE generation technology in India are shown in the right panel of Fig
133、ure ES.5.These results point to significant reductions in both system cost1 and modeled carbon dioxide emissions from Indias electricity system relative to baseline projections(captured in the left panel).Reductions in system cost and CO2 emissions occur whether or not there are caps or taxes on car
134、bon emissions.This result highlights the global environmental benefit of lower costs for electricity storage.Additional studySeveral storage-related topics beyond those addressed in this study deserve attention.These include:(1)manufacturing and supply chain trends,and their impacts in terms of the
135、availability and cost of energy storage tech-nologies and U.S.competitiveness;(2)the relationship between the stability of an economic and regulatory policy framework for economy-wide decarbonization and the time required to achieve a net-zero-carbon electricity sector;(3)the establishment of expect
136、ations for recycling and reuse for end of life batteries;(4)identification of environ-mental,health,and safety aspects of specific electricity storage systems;and(5)the practi-cally available scope for load flexibility and demand response to reduce grid storage needs and associated costs.ReferencesM
137、ai,T.,P.Jadun,J.Logan,C.McMillan,M.Muratori,D.Steinberg,L.Vimmerstedt,R.Jones,B.Haley,B.Nelson.(2018).Electrification Futures Study:Scenarios of Electric Technology Adoption and Power Consumption for the United States.Golden,CO:National Renewable Energy Laboratory.1 The resulting average system cost
138、s of electricity in 2040 and 2050 are reduced by 22%and 39%,respectively.Chapter 1 Introduction and overview 1Chapter 1 Introduction and overview1.1 Motivation and focusThis study is the latest in a series of Future of studies produced by the MIT Energy Initiative that aims to provide useful referen
139、ces for decision makers and balanced,fact-based recommendations to improve public policy,particularly in the United States.Earlier studies in this series have considered the futures of nuclear power,coal(with capture and seques-tration of carbon dioxide emissions),natural gas,the electric grid,and s
140、olar energyall major features in todays energy landscape.These studies have in common a focus on the role that a specific technology(or infrastructure,in the case of the electric grid)might play in an economically efficient,carbon-constrained world and a focus on what needs to be done to facilitate
141、these contributions.The time horizon for these studies has been 2050(and beyond);this study likewise uses a 2050 horizon,though we are also interested in technologies that can be deployed at scale in the nearer timeframe of 2030.1.1.1 MotivationThis study considers the future of energy storage withi
142、n the electric power system.Electric energy storage is certainly not new.The initial application of pumped hydroelectricity storage occurred in the late 19th century in Europe,and in 2020 pumped storage hydro still accounted for around 99%of total grid-scale electricity storage capacity in the Unite
143、d States and globally.1 Global recognition of the need to mitigate damages from climate change by dramatically reducing economy-wide emissions of greenhouse gases,most importantly carbon dioxide(CO2),is the root cause of increased interest and investment in energy storage.A variety of state,regional
144、,national,and international targets and timetables for climate-change mitigation call for achieving net-zero CO2 emissions or zero CO2 emissions;timelines range from 2035 to 2050 and beyond.Net-zero emissions allows for the possibility of offsetting small amounts of emissions in future energy system
145、s via negative emissions technologies(National Academies of Sciences,Engineering,and Medicine 2019),which remove CO2 from the atmosphere and store it,whereas zero emissions targets do not allow for such offsets.Achieving very low economy-wide CO2 emissions will require all sectors to achieve signifi
146、cant reductions.Most studies conclude that the path to very low economy-wide emissions involves decarbonizing the power sector and substituting decarbonized electricity for fossil fuels as much as possible in transpor-tation,industry,and buildings.The main focus of this study is the role that energy
147、 storage can play in decarbonizing the power sector in a cost-effective manner.The most significant current and foreseeable change in the electricity sector is the rapid substitution of variable renewable energy(VRE)i.e.,wind and solar energyfor fossil fuels in electricity generation.Figure 1.1 illu
148、strates historical growth rates for wind and solar generating capacity in the United States and globally.Rapid growth of these technolo-gies has been accompanied by cost reductions,as shown in Figure 1.2 for the United States.Similar cost trends have been observed globally.Both capacity expansion an
149、d cost reductions for VRE resources are generally expected to continue.1 Measured in terms of power capacity(megawatts)rather than energy capacity(megawatt-hours),pumped hydro accounted for roughly 87%and 90%of U.S.and global storage,respectively,in 2020.2 MIT Study on the Future of Energy StorageIn
150、creased penetration of VRE generation makes storage more attractive because VRE generation is intermittent:Its output is variable over time and imperfectly predictable.One approach to coping with intermittency is to use storage to perform energy arbitragethat is,to move electric energy availability
151、from times when it is abundant(lower price)to times when it is scarce(higher price).At the same time as the penetration of VRE generation has grown,the cost of lithium-ion(Li-ion)batteries has declined rapidly,as shown in Figure 1.3.This cost reduction is due primarily to increasing electrification
152、of light-duty vehicles in the transportation sector and the use of Li-ion batteries in mobile consumer electronics.Figure 1.1 U.S.and global installed capacity of(a)solar and(b)wind generation0050060070080000708020002005201020152020PVcapacity(GW)U.S.World(a)(b)005006
153、00700800020406080002005201020152020Windcapacity(GW)U.S.WorldSource:BP(2021).Chapter 1 Introduction and overview 3Demand for these batteries for storage in the electricity sector has been quite small relative to demand in these other two areas(Figure 1.4),but deployment of Li-ion batteries
154、 in the electricity sector has become more attractive and has increased rapidly in the last several years.Recent years have also seen advances in a range of storage technologies,including new chemis-tries for lithium-ion batteries that aim to improve performance and reduce dependence on elements wit
155、h constrained supply chains.In addition,new approaches to thermal storage for electricity and chemical storage(for exam-ple,via the production and storage of hydrogen that can be used to generate electricity),suggest that these may be useful additions to the suite of technologies for electricity sys
156、tem storage.This study finds that a wide range of storage technologies,at different stages of technological readiness,show promise as economical means of coping with VRE intermittency.As Chapter 6 discusses in detail,deployment of storage is only one tool to cope with VRE intermittency efficiently.F
157、or example,it may be optimal to build sufficient VRE capacity to meet more than 100%of demand on average.As a consequence of such“overbuilding,”VRE generation could meet most demand even when VRE resources are low(for example,during the wintertime for solar),but at other times VRE generation would s
158、ubstantially exceed demand and would need to be curtailed.Transmission expansion can also alleviate some variability in VRE resources by averaging VRE generation over larger geographical regions.Advances in clean,dispatchable generation,such as geothermal or biomass,may obviate the need for at least
159、 some storage.In addition,increasing the ability of businesses and Figure 1.2 Installed cost of solar and wind generation in the United States as a function of time0.02.04.06.08.010.012.014.00.00.51.01.52.02.53.020002005201020152020Installedcost(2020$/W),U.S.averageWindPVWindPVresidentialPVutilityWi
160、nd data:U.S.Department of Energy(2021);solar data:Feldman et al.(2012);Ramasamy et al.(2021).Utility PV data are for fixed tilt installations.The solid line for PV represents available yearly data;the discrete points capture data not available on a yearly basis.4 MIT Study on the Future of Energy St
161、oragehouseholds to shift electricity demand from times when energy is scarce(and prices are high)to other times,or to curtail electricity use entirely in periods of energy scarcity,will facilitate VRE integration.Finally,advances in negative emissions technologies may provide economical offsets to f
162、acilitate reaching net-zero emissions while allowing some fossil fuel generation to firm up VREs.Natural gas generation with less than 100%complete carbon capture and sequestration(CCS)is a prime example of a technology that would benefit from economical offsets.1.1.2 Focus of the studyIn this study
163、,we limit our focus to future opportunities for storage within the electricity sector.That is,we include only storage that takes in electrical energy,stores that energy in a variety of forms,and then returns the stored energy to the electricity system as electricity.Examples of technologies we discu
164、ss in detail are lithium-ion batteries,redox flow batteries,metal-air batteries,pumped hydroelectric storage,heat pumps,and hydrogen storage.As noted above,the cost declines that lithium-ion batteries have enjoyed are largely due to their development and use outside the electricity system,particular
165、ly in electric vehicles(EVs).This study looks at potential benefits from similar cross-sector couplings for two other storage technologies:thermal energy storage and hydrogen storage.Although we study the use of heat as a mecha-nism for storing electricity,heat is used widely beyond electricity gene
166、ration.In the industrial sector,high-temperature heat is important for Source:Bloomberg New Energy Finance(2021).Figure 1.3 Global Li-ion battery prices for 2022632204082675248363684606393303226200.250.50.7514201
167、52001920202021Li-ion batterypackprices(2021$/kWh)%of2010priceCell Non-cellChapter 1 Introduction and overview 5core processes such as iron and steel manufac-turing,cement making,and chemical and refining operations.In the buildings sector,heat is needed for space temperature conditioning
168、and hot water.Storing the heat that is produced for thermal power generationor by trans-forming electricity into heatcan be very cost-effective in cross-sectoral applications,allowing more efficient use of thermal generation resources such as nuclear,fuel,geothermal,and VRE(via resistance heating)by
169、 buffering the heat source from end use(in multiple sectors).This decoupling idea leads naturally to the thermal storage retrofit strategies for thermal power plants examined in Chapter 4.There is certainly a much broader scope for thermal storage beyond the boundaries of this study.Hydrogen product
170、ion and use also illustrates the potential importance of cross-sector interactions.A challenge for hydrogen in electricity applications is that,for the foresee-able future,the cost-efficient volume of hydro-gen storage for the electricity sector is too small to drive significant reductions in the co
171、st of producing hydrogen by using VRE-generated electricity to split water.In addition,hydrogen does not currently benefit from significant demand in other sectors in the same way that demand from transportation applications has benefitted Li-ion batteries.We therefore investigate whether the indust
172、rial sector,which might benefit from the availability of inexpen-sive hydrogen to replace natural gas as a high-temperature heat source,might play that role.Sources:U.S.Department of Energy(2020);Kane(2021b);Kane(2021a);Wehling and Abraham(2021).For 2021,EV data were not available for the full year;
173、the number shown is obtained by applying the percent increase that occurred during the first half of the year to the second half of the year.Figure 1.4 Global deployment of Li-ion batteries over the period 20112021 for electric vehicles and energy storage in the electricity sector05002011
174、2000021Li-ion batteryannualdeployments(GWh)(Somedataunavailablefor2020&2021)EVs Stationary Other6 MIT Study on the Future of Energy StorageThis study is aimed primarily at the U.S.electricity system.In our modeling work,we address geographic variations within the Uni
175、ted States that might affect storage deploy-ment in future power systems.This diversity derives from regional variations in VRE and other resources;from public attitudes towards different generation options,land use,and siting for transmission;from the nature of the demand for electricity;and from d
176、ifferent regional policies and approaches to decarbon-ization.Lessons learned from regional studies in the U.S.context can be useful in informing possible roles for electricity storage in future energy systems in other parts of the world.We look specifically at opportunities in the Indian context,wh
177、ich differs from the U.S.situation in a number of important ways,particularly in the lack of a significant domestic supply of natural gas and in very rapid growth in electricity demand.A number of storage-related topics beyond those addressed in this study deserve attention in other work:(1)manufact
178、uring and supply chain trends,and their impacts on the availabil-ity and cost of storage and U.S.competitiveness;(2)the relationship between the stability of an economic and regulatory policy framework for economy-wide decarbonization and the time and cost required to achieve a net-zero-carbon elect
179、ricity sector;(3)the need to establish expectations for end-of-life recycling and/or reuse for batteries;(4)identification of envi-ronmental,health,and safety aspects of specific electricity storage systems;and(5)the practi-cally available scope for load flexibility and demand response to reduce gri
180、d storage needs and associated costs.1.2 Roles for storage in electricity systemsEnergy storage services can broadly be classified in four categories:energy arbitrage,ancillary services,transmission and distribution infra-structure services,and customer energy management services.This section provid
181、es definitions and examples of services that can be provided by storage(definitions are adapted from the International Energy Agencys Technology Roadmap:Energy Storage(2014).In practical usage,a single energy storage technology or several storage technologies may support multiple services.Energy arb
182、itragedefined as moving electrical energy from low-value to high-value periodsis the principal role for energy storage in the electricity system today and is likely to be its principal role going forward.Integration of intermittent VRE generation drives this oppor-tunity,which will grow as VRE penet
183、ration increases.Figure 1.5 shows daily and seasonal variations in solar and wind resources and load for all of 2021 in ERCOT,the electricity system that covers most of Texas.Sometimes wind availability complements the solar resource from day to night(and vice versa),but there are extended periods,s
184、ay the latter part of June into the beginning of July 2021,with minimal wind generation.This results in a significant potential gain from intraday shifting,today most often over a time span of 4 hours or less,as well as shifting across multiple days,over timespans of,say,greater than 12 hours.We exa
185、mine multiple storage technologies to understand which ones will be suited for these short-and long-duration roles in future energy systems.Energy storage assets that provide ancillary services to the bulk power system deliver power for short durations but require faster response times(from less tha
186、n a second to minutes).These ancillary services include the following:Frequency regulation is the use of storage to dampen fluctuations caused by momentary differences between power generation and load demand.This is often performed automatically on a minute-to-minute,or shorter,basis.In VRE-dominan
187、t systems,this function Chapter 1 Introduction and overview 7replaces the inertia provided by spinning turbines in thermal generators.Load following,similar to frequency regulation,is a continuous electricity balancing mechanism that manages system fluctuations.However,in this case,the time frame of
188、 the intervention is longer,ranging from 15 minutes to 24 hours.Voltage support refers to the maintenance of voltage levels in the transmission and distribu-tion system.Black start capability refers to a power stations ability to restart without relying on the trans-mission network in the event of a
189、 wide-area power system collapse.Supplemental reserves can supply extra power to the grid(historically from extra generating capacity)with a response time of less than 10 minutes(and sometimes meeting other requirements).These reserves can be used to maintain system frequency stability during unfore
190、seen load swings or emergency conditions(U.S.Department of Energy 2011).Transmission and distribution(T&D)infrastructure services help defer the need for capital-intensive T&D upgrades or investments to relieve temporary congestion or potential substation overloads in the T&D network.These services,
191、which work by injecting energy into the grid between the bottleneck and load during peak periods,turn out to be particularly important in the Indian context discussed in Chapter 7.Customer energy management services,including enhanced reliability and reduction of peak loads,may be provided by relati
192、vely small storage systems located on customer premises.When managed by aggregators,these systems can also provide energy services at the bulk power level.Figure 1.5 Daily variability of wind and solar resources in TexasDaily variability of wind(blue)and solar(red)resources in Texas relative to load
193、(black line)in 2021.Days of the month are in the columns and months of the year are in rows.CF=Capacity factor.Source:Electric Reliability Council of Texas(2021).JanFebMarAprMayJunJulAugSepOctNovDecERCOT 2021Load(fraction of peak)Wind CFSolar CF8 MIT Study on the Future of Energy Storage1.3 Key attr
194、ibutes,cost dimensions,technology classes,and environmental,health,and safety considerations for storage technologiesElectrical storage technologies considered in this study have a range of characteristics with respect to technological readiness,cost and performance,modularity,abundance and cost of
195、component materials,energy storage density per unit of volume or weight,and environmen-tal,health,and safety(EHS)impacts.Among these attributes(and others)we focus on cost characteristics,because cost will be a dominant factor in deployment at scale in the electricity system.Factors like energy dens
196、ity or efficiency at small scale matter much less in this sector than in EV or consumer electronics applications,for example.We expect that all the storage technologies included in our modeling could be ready for commercial deployment by 2050.1.3.1 Cost dimensions and technology classesIt is useful
197、to review the standard cost analysis of electric generation facilities as background for the more complex analysis of energy storage systems.The capacity of generating facilities is typically described by their maximum instanta-neous power capacity,which is measured in megawatts(MW).To a first appro
198、ximation,the cost of a generation plant has two components:the annualized capital cost per MW of capacity(including any annual costs that do not vary with generation)and the operating cost per megawatt-hour(MWh)of electric energy produced(including fuel,if any,plus any other annual costs that vary w
199、ith generation).Some analysts summarize technology-specific generation costs by using a quantity called the levelized cost of energy or LCOE.2 Computing the LCOE for a particular thermal generation technology requires making assumptions about fuel cost and annual output profile for a typical facilit
200、y.The LCOE can then be calculated as the average revenue per MWh produced that just covers a typical facilitys capital and operating cost.It should be apparent that the LCOE cannot sensibly be used to compare the attrac-tiveness of investments in thermal plants with very different output profiles:a
201、nuclear plant that is run all the time to provide baseload power and a gas-fired plant that is used only to meet peak demand,for instance.For the same reason and because the value of VRE output depends on when the output occurs,using LCOE to compare VRE generators and thermal generators makes even l
202、ess sense(Joskow 2011).We do not use LCOE in this study.Describing storage facility costs is considerably more complex than describing generating facility costs.(Table 6.3 summarizes the storage cost assumptions used in modeling for this study.)Storage capacity has at least two and possibly three di
203、mensions of capacity that can generally be independently varied.Like gener-ating plants,one element of a storage facilitys capacity is the maximum instantaneous power,measured again in megawatts(MW),that it can supply to the gridin other words,its discharge power capacity.For some technologies,the m
204、aximum instantaneous power that a facility can take from the gridalso called its charge power capacitycan be different from its discharge power capacity.In addition,every storage facility can be characterized by its 2 See U.S.Energy Information Administration(2021)for a general discussion of LCOE an
205、d LCOS.We share the EIAs conclusion that“LCOE and LCOS do not capture all of the factors that contribute to actual investment decisions,making direct comparisons of LCOE and LCOS across technologies problematic and misleading as a method to assess the economic competitiveness of various generation a
206、lternatives.”Chapter 1 Introduction and overview 9energy storage capacity,measured in megawatt-hours(MWh).The ratio of a facilitys energy storage capacity to its maximum discharge power capacity is its duration,measured in hours:This is the length of time the facility can provide maximum power start
207、ing from a full charge.Most existing battery storage facilities currently have durations of four hours or less;most existing pumped storage hydro(PSH)facilities have durations of twelve hours or more.3A facilitys round-trip efficiency(RTE),defined as the fraction of energy used for charging storage
208、that is available for discharge,is gener-ally determined by the technology employed and does not vary with power or energy capacity.(It is sometimes useful to distinguish between a facilitys charge(or up)power efficiency,the fraction of energy taken from the grid that ends up as useful charge,and it
209、s discharge(or down)power efficiency,the fraction of useful charge that can be discharged to the grid.The product of these two parameters is round-trip efficiency.)Finally,energy stored using some technologies is gradually lost over time.The rate at which this occurs is called the self-discharge rat
210、e.Some analysts have summarized the costs of various storage technologies by computing a quantity called the levelized cost of storage(LCOS).To compute this quantity for a par-ticular technology requires specifying the storage duration for a typical facility and making assumptions about charge/disch
211、arge cycles over time and the cost of the power used to charge the facility.4 To compute annual facility cost,operation and maintenance costs must be added to the annualized cost of capacity;these O&M costs may depend on assumed usage and(for batteries)the cost of degradation implied by assumed usag
212、e.As in the case of LCOE,LCOS can then be computed as the average revenue per MWh discharged that just covers the facilitys assumed capital and operating cost.The estimated LCOS for any particular technology will thus depend on how the typical facility is assumed to be used,so that storage technolog
213、ies likely to play different roles,like generation technologies that play different roles,cannot be usefully compared.Moreover,LCOS computations rest heavily on assumptions about the cost of charging that are essentially arbitrary.We do not use LCOS in this study.Instead,in the modeling exercises de
214、scribed in Chapter 6,we use mathematical programming to determine optimal capacities and operations of available generation and storage technologies given assumptions about costs,regional wind and solar resources and other regional resources,the demand for electricity over time,and constraints on CO
215、2 emissions.The technologies considered in this report fall into three main groups based on their per-unit discharge power and energy capacity costs(Figure 1.6).Generally,technologies with low energy-capacity costs and high power-capacity costs(the blue area in the figure)are most suitable for longe
216、r-duration storage applications(up to multiple days)and less frequent charge/discharge cycles;these include thermal and chemical forms of storage,metal-air batteries,and pumped storage hydro options.Tech-nologies in the brown area,including lithium-ion batteries,are better suited to shorter-duration
217、 applications(a few hours)and more frequent cycling.Technologies with intermediate capabilities,including flow batteries,are in the green area.As we discuss in Chapter 6,however,3 The unusually long duration of PSH facilities explains the fact that,although PSH accounted for around 99%of energy stor
218、age capacity(MWh)in the United States and globally in 2020,it accounted for only 87%of power capacity(MW)in the United States and 90%globally.4 Lazard(2021)provides a particularly clear example of LCOS calculations.10 MIT Study on the Future of Energy Storagethe efficient use of any particular stora
219、ge technology generally involves a mix of storage cycles of different durations.1.3.2 Environmental,health,and safety considerationsU.S.firms generally pay attention to environ-mental,health,and safety(EHS)concerns in investment and operations decisions and in innovation efforts.An unanticipated EHS
220、 issue can create lengthy project delays,significant cost increases,and loss of public support.There are several notable examples of such issues in the history of the energy industry in the United States:the Three Mile Island nuclear accident,particulate emissions from coal-fired electric power plan
221、ts and diesel trucks,and air and water quality impacts from hydraulic fracturing techniques employed in unconventional oil and natural gas production.EHS concerns figured prominently in the 2003 Future of Nuclear Power study and the 2007 Future of Coal study.The MIT Energy Initiative strives to incl
222、ude EHS aspects on a par with technical and economic aspects in its work on energy and climate subjects and related research projects.Several EHS issues arose in the course of this Future of Energy Storage study.The diversity of the storage technologies we considered and the disparate roles those te
223、chnologies might play in a decarbonized electricity system precluded the construction of a single hand-book of EHS procedures and best practices.Similarly,this diversity makes the creation of an industry-wide electricity storage institute similar to the Institute of Nuclear Operations(INPO)problemat
224、ic.EHS management for an energy storage facility must be tailored to the specific technology,scale,application,and physical location of the facility,resulting in Figure 1.6 Three groups of storage technologies based on power-and energy-capacity costsThe blue region,with high power and low energy cap
225、acity costs,includes thermal,chemical(e.g.,hydrogen),metal-air battery,and pumped storage hydro technologies.Lithium-ion batteries fall in the brown area,with low power,but high energy-capacity costs;flow batteries fall in the intermediate,green region.02004006008001,0001,2001,4000204060801001201401
226、60180200Power capacity cost($/kW)Energy capacity cost($/kWh)Chapter 1 Introduction and overview 11distinct approaches on a largely case-by-case basis.However,it can be expected that different technology classes will have characteristic EHS concerns that may have an impact on siting challenges,inform
227、 operating constraints,and require the introduction of auxiliary systems to safely manage storage facilities.Rather than list these concerns here,we integrate this information in the technology chapters.Several examples of storage-related EHS issues deserve mention here,however.Perhaps the most sign
228、ificant and far-reaching involve the mining of metal ores such as cobalt,nickel,and vanadium,which must be expanded substantially to enable electrochemical storage technologies to play a meaningful role in future energy systems.Currently,the extraction and beneficiation of these ores is concentrated
229、 in developing nations in Africa and Latin America that may lack adequate regulatory resources and enforcement mechanisms to effectively manage EHS concerns.The lack of regulatory capacity in these countries heightens EHS concerns and deserves greater attention.Other technologies raise other potenti
230、ally important EHS concerns:Examples include flammability hazards associated with lithium-ion batteries,risks from hydrogen-induced embrittlement of conventional pipelines and the impacts of hydrogen leakage,as well as the ecological and geological impacts of pumped hydroelectric storage.Although th
231、is study does not address these or other storage-technology-related EHS issues in depth,it is important to make readers aware of the importance of these EHS topics.It should also be mentioned that EHS concerns are increasingly being raised in the context of a broader focus on environmental,social,an
232、d governance(ESG)matters that encompasses the community and income distribution impacts of climate change itself,and of climate-related policies.At present,ESG impacts from energy storage innovation,manufacturing,and deployment are nuanced and depend on project,technology,and location specifics that
233、 challenge a broad assessment.Thus,the recom-mendation in Chapter 6 of this study to invest public resources in improving modeling and simulation tools includes tools that would enable a better understanding of environmental and economic outcomes from alternative approaches to decarbonized electrici
234、ty systems with energy storage.1.4 Report structureThe remainder of this report is comprised of three major sections:storage technologies,system modeling,and implications for policy and innovation.The storage technologies section follows this chapter and is divided into four chapters,each focused on
235、 one of the four technology areas of importance to the electricity sector:electrochemical storage,mechanical storage,thermal storage,and chemical storage.Each chapter develops high-,medium-,and low-cost estimates for promising technologies in these four categories,which are used for analyzing future
236、 electricity systems(described below).Chapter 2,which deals with electrochemical energy storage,focuses on three types of battery storage technologies involving different chemistries and formats:lithium-ion batteries,redox flow batteries,and metal-air batteries.Brief consideration is also given to o
237、ther battery technologies that have been deployed for stationary energy storage in the past.A major section of this chapter deals with materials costs and availability,since these considerations may have a significant bearing on the future scalability and adoption of electrochemical technologies.Mat
238、erials criticality,supply chains,diversity of sourcing,and considerations for recycling are all part of this discussion.12 MIT Study on the Future of Energy StorageChapter 3 examines mechanical energy storage technologies and consists of two distinct sections.The first deals with pumped hydro-power
239、storage,which still dominates energy storage globally(International Energy Agency 2019).The second section of the chapter deals with compressed air energy storage(CAES).Since compressing air generates a great deal of heat,CAES systems can be distinguished by whether this heat is discarded or saved f
240、or re-expanding the compressed air.Only the latter is a true energy storage technology and is the main focus of the latter half of Chapter 3.Chapter 4 discusses a variety of thermal energy storage options.A general challenge for these technologies is the low efficiency of converting heat back to ele
241、ctricity.One option for mitigating this low conversion efficiency involves taking advantage of opportunities for retrofitting existing thermal power plants.The chapter also examines options like heat pumps that may be available in a mid-term(2030)timeframe.Finally,very high temperature options,which
242、 promise much more efficient discharge from heat to electricity,are examined for a 2050 horizon.Chapter 4 also looks at the potential for using very low-cost materials for thermal storage,which can lead to attractive costs for long-duration storage.Finally,Chapter 5 looks at options for chemical ene
243、rgy storage.Because of the large amount of energy stored in chemical bonds,chemical energy storage has advantageous energy density,which can be useful when space is limited and/or for very-long-duration storage applications.In addition,the chemical stability of these bonds provides very low self-dis
244、charge rates,again making chemicals attractive for long-duration storage.Hydrogen is used as an example for chemical energy storage in this chapter,because hydrogen can be directly produced from electricity in a single step(by using electric current to split water molecules)and consumed either as a
245、fuel to produce power or as a feedstock or heat source for other sectors.The second section of the report uses cost estimates and information about other technol-ogy attributes from the four technology chapters in combination with capacity expansion models to analyze future power systems in two dist
246、inct contexts:developed countries(Chapter 6)and emerging market and develop-ing economy(EMDE)countries(Chapter 7).To explore the characteristics of future decar-bonized systems in a developed country context we study three U.S.regions:the Northeast,Southeast,and Texas;in addition,we present results
247、at a national level,although with less detail.India is used as an example of power system evolution in an emerging market,developing economy country.In these country-and region-specific studies,we consider a number of different constraints on power sector carbon emissions,ranging from no limit at al
248、l down to 5 grams of carbon dioxide per kilowatt-hour(gCO2/kWh).To provide a sense of the stringency of this lower limit on emissions,holding the U.S.electricity system to an average emissions intensity of 5 gCO2/kWh in 2018 would have reduced power-sector emissions(relative to actual 2005 emissions
249、)by 99.2%.The India case study is interesting not only because the policies India is likely to enact will be different,but because electricity demand in India and other EMDE countries will likely grow much faster than in developed nations.At the same time,India does not have a significant source of
250、domestic natural gas.This means that new energy storage options in India will not have to compete against natural gas as vigorously as they would in the U.S.context.This study concludes with a section that examines the implications of our technology and modeling findings for regulatory and policy de
251、cisions.Chapter 8 looks at the governance of decarbonized power systems with storage and considers how alternative organizational,regulatory,and policy arrangements can enable storage to play different roles in these systems at the lowest possible total cost,with appropri-ate attention to equity con
252、siderations.Chapter 1 Introduction and overview 13The chapter also discusses arrangements that are important to avoid,because they undermine the efficient deployment and utilization of storage on the grid and because they would make it more difficult to replace fossil fuels by decarbonized electrici
253、ty in other sectors.We focus primarily on the United States,though the general issues we discuss are relevant in other developed regions.The final chapter,Chapter 9,examines the role of technology innovation in ensuring that energy storage can play a significant role in future electric power systems
254、.It discusses the nature and pace of innovation needed in order for energy storage to have a positive impact on future electricity systems with large amounts of variable renewable energy resources.In addition,Chapter 9 points to specific U.S.programs for advancing innovation,discusses who should lea
255、d them,and considers how these programs can be conducted so as to avoid previous mistakes.Finally,this report includes several appendices that provide additional detailed material,reference data,and calculation results that can be helpful to some readers.14 MIT Study on the Future of Energy StorageR
256、eferencesBloomberg New Energy Finance.2021.“Battery Pack Prices Fall to an Average of$132/kWh,But Rising Commodity Prices Start to Bite.”Bloomberg New Energy Finance.November 30.https:/ Review of World Energy.”BP.https:/ Reliability Council of Texas.n.d.“Generation.”ERCOT.https:/ Barbose,Robert Marg
257、olis,Ryan Wiser,Nam Darghouth,and Alan Goodrich.2012.“Photovoltaic(PV)Pricing Trends:Historical,Recent,and Near-Term Projections.”U.S.Department of Energy SunShot Program.November.https:/www.nrel.gov/docs/fy13osti/56776.pdf.International Energy Agency.2014.“Technology Roadmap-Energy Storage.”Interna
258、tional Energy Agency.https:/www.iea.org/reports/technology-roadmap-energy-storage.2019.“Will pumped storage hydropower expand more quickly than stationary battery storage?Analysis from Renewables 2018.”International Energy Agency.March 4.https:/www.iea.org/articles/will-pumped-storage-hydropower-exp
259、and-more-quickly-than-stationary-battery-storage.Joskow,Paul L.2011.“Comparing the Costs of Intermittent and Dispatchable Generation Technologies.”American Economic Review 101(3):238-241.10.1257/aer.101.3.238.Kane,Mark.2021a.“Global Passenger xEV Battery Market Stats:H1 2021.”Inside EVs.September 26
260、.https:/ Research:Global xEV Battery Market-142.8 GWh In 2020.”Inside EVs.February 18.https:/ Levelized Cost of Storage,Version 7.0.”Lazard.https:/ Academies of Sciences,Engineering,and Medicine.2019.Negative Emissions Technologies and Reliable Sequestration:A Research Agenda.Washington,DC:The Natio
261、nal Academies Press.https:/doi.org/10.17226/25259.Ramasamy,Vingesh,David Feldman,Jal Desai,and Robert Margolis.2021.“U.S.Solar Photovoltaic System and Energy Storage Cost Benchmarks:Q1 2021.”National Renewable Energy Laboratory.November.https:/www.nrel.gov/docs/fy22osti/80694.pdf.U.S.Department of E
262、nergy.2011.“Electric Market and Utility Operation Terminology.”U.S.Department of Energy Solar Energy Technologies Program.https:/www.nrel.gov/docs/fy11osti/50169.pdf.2020.“Energy Storage Grand Challenge:Energy Storage Market Report.”U.S.Department of Energy Energy Storage Grand Challenge.https:/www.
263、energy.gov/sites/prod/files/2020/12/f81/Energy%20Storage%20Market%20Report%202020_0.pdf.2021.“Land-Based Wind Market Report:2021 Edition.”U.S.Department of Energy Office of Energy Efficiency and Renewable Energy.https:/emp.lbl.gov/sites/default/files/land-based_wind_market_report_2021_edition_final.
264、pdf.U.S.Energy Information Administration.2021.“Levelized Costs of New Generation Resources in the Annual Energy Outlook 2021.”U.S.EIA.February.https:/www.eia.gov/outlooks/archive/aeo21/pdf/electricity_generation.pdf.Wehling,Mark,and Allen Tom Abraham.2021.“Global Energy Storage Outlook.”Bloomberg N
265、ew Energy Finance.August 11.https:/ 2 Electrochemical energy storage 15Chapter 2 Electrochemical energy storage2.1 IntroductionElectrochemical energy storage devices,com-monly called batteries,interconvert chemical and electrical energy through reduction and oxidation(redox)reactions.These reactions
266、 occur on two different electrodes(positive and negative)that are electrically connected through an external circuit and physically separated by an ionically conducting medium(i.e.,an electrolyte).While so-called“primary”batteries,which are discharged once and then disposed of or recycled,are of int
267、erest for many applications,the battery technologies that are relevant for grid storage and emphasized in this study are,almost without exception,recharge-able or“secondary”batteries in which the redox processes are electrically reversible.Since Alessandro Volta first conceived the idea of a battery
268、 in 1800,electrochemical energy storage has enjoyed a rich history of research,development,demonstration,and commercial-ization resulting in a number of battery tech-nologies that now play important roles in modern society.A notable example is the lead-acid battery,which is widely used for engine st
269、art and onboard power in internal combustion automobiles and for back-up power in homes and industry.Another is the lithium-ion battery,which underpinned the portable electronics revolution and for which the 2019 Nobel Prize in Chemistry was awarded to John B.Goodenough,M.Stanley Whittingham,and Aki
270、ra Yoshino.While electrochemistry can play a role in numerous technologies across the broader energy system,here we limit our focus to electrochemical technologies that complete the loop of receiving,storing,and delivering electricity.We do not attempt to be encyclopedic in our coverage of batteries
271、,rather,we limit our analysis and modeling to battery chemistries that have either reached a technology readiness level(TRL)of 6 or higher1 in grid-scale electricity storage,or demonstrated,at an early stage,particularly promising attributes for future deployment.These batteries primarily fall into
272、three catego-ries:lithium-ion(Li-ion)batteries,redox flow batteries(RFBs),and metal-air batteries,as illustrated in Figure 2.1.The chapter is organized as follows:Section 2.2 discusses,in general terms,the technical and cost performance metrics used to compare different battery technologies for grid
273、-scale energy storage.Subsequent sections detail the current status and outlook for individual battery technologies.Specifically,Sections 2.32.6 cover Li-ion batteries,RFBs,metal-air batteries,and other closed-system batteries(i.e.,lead-acid and high-temperature batteries).For Li-ion,redox flow,and
274、metal-air batteries,we summarize technical performance parameters and estimate likely cost ranges in 2050these estimates are used as inputs to the modeling analyses described in Chapters 6 and 7.The following part of the chapter,Section 2.7,discusses the materials availability issues that may arise
275、at projected levels of electrochemical energy storage deployment over the next 10 to 30 years(i.e.,by 2030 and 2050),with an emphasis on the materials or elements that are central to state-of-the-art Li-ion and RFB chemistries.The last section,Section 2.8,summarizes key takeaways from the chapter.1
276、Technology Readiness Levels(TRL)are a type of measurement system used to assess the maturity level of a particular technology(TRL 1 is the lowest and TRL 9 is the highest).TRL 6 is defined by the U.S.Department of Energy as one in which“engineering-scale models or prototypes are tested in a relevant
277、 environment.”Source:Technology Readiness Assessment Guide(U.S.Department of Energy 2011).16 MIT Study on the Future of Energy Storage2.2 Electrochemical systems for grid-scale energy storageElectrochemical energy storage systems generally have higher energy densities than mechanical(Chapter 3)or th
278、ermal(Chapter 4)storage systems and can achieve a wide range of roundtrip energy efficiencies(the ratio of energy output to energy input),from as high as 95%for certain Li-ion battery chemistries to as low as 40%for certain metal-air battery chemistries.Because of their compact footprint and indepen
279、dence from geographical and geological resources,batteries are a versatile technology that can be readily deployed at a variety of scales,from centralized large-scale facilities down to the level of distributed residential users,and face fewer siting con-straints.Whereas energy-dense batteries are f
280、avored for mobile applications,which empha-size compact cell formats and high-capacity chemistries,battery cost and service lifetime are of greater importance in stationary applica-tions.This means a broader range of battery chemistries and system configurations can be considered for grid-scale ener
281、gy storage.The elements and compounds that comprise the positive and negative electrodes in which chemical energy is stored or extracted are a large component of the batterys cost.On this basis,the chemical cost of stored energy,in dollars per kilowatt-hour($/kWh),varies by more than two orders of m
282、agnitude among known battery types,from less than$1/kWh for the most earth-abundant elements,to more than$30/kWh for some high-specific-energy Li-ion chemistries,to more than$100/kWh for vanadium RFBs(Figure 2.2).It should be noted,however,that these costs are not purely a function of the materials
283、cost,but also the batterys energy density and efficiency.Further,the installed cost of a battery system comprises more than its chemical costs because it includes Figure 2.1 Categories of electrochemical storage technologiesCategories of electrochemical storage technologies,lithium-ion(Li-ion)batter
284、ies,redox flow batteries,and metal-air batteries,and their role in an integrated electricity system.Dashed and solid arrows show the directions of electron and cation transport during charging and discharging,respectively.Depending on the application,additional components not shown in the figure(e.g
285、.,thermal and battery management systems,etc.)may also be needed to ensure the proper and safe operation of the storage system.Li-ionbatteryMetal-airbatteryRedox flowbatteryBattery storage systemSystem couplingGrid integrationElectron TransportCation TransportDCACAirElectron TransportAnion Transport
286、+-Electron TransportLi+Transport+-Chapter 2 Electrochemical energy storage 17costs for the supporting materials needed to produce a battery cell,pack,or electrochemical stack(depending on the system architecture);the cost of mechanical and power electronics components;the cost of manufacturing;and t
287、he cost of installation and interconnection,among other costs.Thus,the chemical cost of stored energy represents a floor on the cost of any battery.These costs can dominate overall cost,as is the case for todays high-specific-energy Li-ion batteries.The recognition that chemical costs may fall more
288、slowly than other costs as battery technologies mature warrants a two-stage learning curve model for battery cost projections(Hsieh et al.2019).However,recent developments in grid-scale storage favor low-cost,widely available chemical components(e.g.,iron,manganese,zinc,and sulfur)that shift more of
289、 the cost burden to other system components.For the lowest cost chemistries,the cost structure for battery storage systems is not unlike that for pumped hydroelectric or compressed air storage systems,in which the cost of the energy storage medium(water or air)is a small to negligible percentage of
290、total system cost.Battery chemistries that utilize earth-abundant elements can also benefit from a diverse and secure supply chain that may enable rapid scaling of systems for grid applications.The chemical cost of stored energy for battery electrochemistries,shown against the year in which the syst
291、em first appeared in the public domain.Technologies pre-1900 are shown against the left axis.Chemical cost is calculated as the cost of the negative electrode material,positive electrode material,and electrolyte,divided by the stored energy.Abbreviations:LMO=lithium manganese oxide,LCO=lithium cobal
292、t oxide,LFP=lithium iron phosphate,LNMO=lithium nickel manganese oxide,LTO=lithium titanium oxide,NMC=lithium nickel manganese cobalt oxide,NCA=lithium nickel cobalt aluminum oxide,P2-MN=P2-type sodium manganese nickel oxide,NTP=sodium titanium phosphate,NMO=nickel manganese oxide,NiCd=nickelcadmium
293、,NiMH=nickel metal hydride,AQDS=9,10-anthraquinone-2,7-disulfonic acid(adapted from(Li et al.2017);see Appendix A for details).Figure 2.2 The chemical cost of stored energy for battery electrochemistriesNa ion(aqueous)Na ion(non-aqueous)Redox flowMetal airLi-ion(C6 anode)Li-ion(Si anode)High tempera
294、tureLithium MetalLi-ion(SiO/C anode)Other100.0Chemical cost($/kWh)Year2020SiO-C/NMC(8:1:1)Li/Pb2SbLi/LCONiMHLi/TiS2Li/SLi/MoS2Ca/SbSi/NMC(8:1:1)Si/NMC(6:2:2)C6/NMC(6:2:2)C6/NMC(8:1:1)C6/NMC(1:1:1)C6/LCONa/SZn/Br2Zn/MnO2AQDS/Br2Cu/CuFe/CrS/BrV/VFe/FeZn/FeNa/NiCl2Na2Sx/air(OH-)Na/P2-MNONa2Sx/air(H+)Al
295、/airZn/air,1878Zn/NiOOH,1899NiCd,1899Pb acid,1859Fe/airLi/airC6/LNMOC6/LMOC6/LFPC6/NCALTO/LMONTP/NMO2099001.010.0Pre-1900 chemistries18 MIT Study on the Future of Energy StorageEvery battery chemistry comprises a trade-off in attributes such as energy density,safety,durability,
296、and cost,and system architectures are generally optimized for a specific chemistry-application fit.Historically,rechargeable battery development has been driven by mobile applications,which favor high-capacity chemis-tries and compact system designs.In emerging stationary applications,by contrast,sy
297、stem cost and lifetime are prioritizedin some cases,at the expense of energy density or roundtrip efficiency.Notably,Li-ion batteries are con-strained to specific architectures that tend to yield lower power costs but higher energy costs(Table 2.1).This has made Li-ion batteries most competitive,on
298、a capital cost basis,for short-duration storage applications(less than four hours of storage).Projected cost declines may make Li-ion batteries competitive for storage durations up to about eight hours.Emerging alternatives,such as RFBs and metal-air batter-ies,allow a broader range of chemistries t
299、o be used,including high-abundance and low-cost active species(e.g.,iron,zinc,oxygen)that make capital costs for these batteries attractive for longer storage durations(over eight hours).Furthermore,flow and metal-air battery architectures can be designed to allow for periodic maintenance of some sy
300、stem compo-nents,which reduces the levelized cost of storage over long service life.However,the relative immaturity of these battery technolo-gies,including limited engineering experience,uncertainty in manufacturing costs and methods,and the lack of a fully developed supply chain,presents barriers
301、to deployment and increases perceived risks from practical,operational,and financial points of view.Conversely,demand for longer duration applications is still emerging,challenging the development of technologies that are primarily competitive in these spaces.As part of this study,we estimate figure
302、s of merit for performance and cost for Li-ion batteries,RFBs,and metal-air batteries for the present day(2020)and the future(2050).Our estimates of future cost include a low,medium,and high value.These cost estimates are then used in the grid modeling analyses discussed in Chapter 6.Numerous studie
303、s have examined historic,current,and projected future costs for Li-ion batteries.We use numbers from the National Renewable Energy Laboratory(NREL)Annual Technology Baseline(ATB)of 2020,a widely cited source that is in good agreement with many other published reports.For current(2020)costs,the NREL
304、ATB uses a bottom-up cost model that contains detailed cost information for components of the battery storage system(Feldman et al.2021),including Li-ion battery pack,inverter,and the balance of system needed for installation.For future(2050)Li-ion battery costs,the NREL ATB makes projections based
305、on a literature review of 19 sources published in 2018 or 2019(Cole and Frazier 2020).We use the lower-bound,median,and higher-bound projections from this literature review as the low-,mid-,and high-cost assumptions in our modeling analy-sis,respectively.A techno-economic assessment for the less-dev
306、eloped technologies we consider(i.e.,RFBs and metal-air batteries)is challenged by limited sources of information for compo-nent and system costs as well as a lack of established field-wide standards in engineering design and manufacturing.To mitigate these information gaps,we surveyed the published
307、 literature and engaged industry experts.Our estimated values are in general agreement with those in other published reports around the time of writing(February 2022)and rely on inputs,assumptions,and calculations that are based on peer-reviewed work.Nevertheless,these values should be considered ea
308、rly-stage estimates that may be refined in the future as commercialization expands and specific chemistries develop.Details on our cost calculations for RFBs,metal-air batteries,and balance-of-plant subsystems can be found in Appendix A.Chapter 2 Electrochemical energy storage 192.3 Lithium-ion batt
309、eries 2.3.1 Technology overviewLithium-ion(Li-ion)batteries are a family of rechargeable batteries that utilize solid com-pounds at both the negative and positive electrodes as hosts for reversible lithium-ion storage.During discharge,lithium ions migrate internally from the negative electrode to in
310、tercalate into the positive electrode through a liquid electrolyte,while electrons simultane-ously move in the same direction through an external circuit,powering the device to which the battery is connected.During charge,the process is reversed,with lithium ions migrating from the positive to the n
311、egative electrode,and electrons flowing through the external circuit,under voltage supplied by an external power source.The Li-ion battery is a relatively mature technology that has benefited from more than three decades of commercial development.Thanks to several factorsincluding the low atomic mas
312、s of lithium;the development of positive and negative electrodes that are capable of reversibly storing lithium ions at high mass and volume concentrations and with large differences in electrode potential(cell voltage);and the development of high conductivity electrolytes,supporting components,cell
313、 designs,and manufacturing methodsLi-ion batteries today offer energy and power densities that are superior to most other battery types.State-of-the-art Li-ion battery cells have a nominal voltage of 3.64.0 volts(V),a specific energy(or gravimetric energy density)between 100 and 250 watt-hours per k
314、ilogram(Wh/kg),and an energy density between 300 and 650 TechDischarging capital cost ($/kW)Storage capital cost ($/kWh)FOM ($/kW-year)FOM ($/kWh-year)Efficiency-charge (%)Efficiency-discharge (%)Self-discharge rate (%/month)Li-ion20202572771.46.892%92%1.5Li-ion2050 Low3270.90.31.492%92%1.5Li-ion205
315、0 Mid110125.80.82.292%92%1.5Li-ion2050 High154177.01.43.292%92%1.5RFB20205836501714.10.092%88%0.0RFB2050 Low29715.54.10.092%88%0.0RFB2050 Mid39648.04.10.092%88%0.0RFB2050 High530102.24.10.092%88%0.0Metal-air20201,0681,1353.726.728.40.172%60%7.3Metal-air2050 Low5950.114.90.070%59%1.5Metal-air2050 Mid
316、6432.416.10.173%63%1.5Metal-air2050 High9503.623.70.172%60%1.5Estimated and projected capital costs,operating costs,efficiencies,and self-discharge rates for lithium-ion(Li-ion)batteries,redox flow batteries(RFBs),and metal-air batteries in 2020 and 2050,respectively.For each technology,a range of p
317、ossible 2050 costs(low,mid,and high)is given.All operations and maintenance(O&M)costs are treated as fixed operations and maintenance(FOM)costs and reflect routine component servicing and replacement due to degradation.For Li-ion and metal-air batteries,FOM costs are assumed to equal 2.5%of capital
318、cost per year,while for RFBs the FOM cost assumes replacement of membranes and electrodes every decade.Table 2.1 Estimated and projected capital costs,operating costs,efficiencies,and self-discharge rates20 MIT Study on the Future of Energy Storagewatt-hours per liter(Wh/L)(Li et al.2018;Zubi et al.
319、2018).They have high roundtrip energy efficiency(85%95%,depending on the rate of charge and discharge),low maintenance requirements,adequate cycle life for many applications(up to several thousand full charge/discharge cycles),and a low self-discharge rate.These merits have made Li-ion batteries the
320、 incumbent technology for a wide range of applications,from portable electronics and power tools to electric vehicles(EVs)and stationary energy storage systems.A Li-ion cell contains several key components within its external housing(Figure 2.3):a positive electrode,a negative electrode,alumi-num an
321、d copper foil current collectors to which the positive and negative electrodes are respec-tively adhered,a liquid electrolyte,and a porous separator to electrically isolate the two electrodes from one another.The positive electrode(commonly referred to as the cathode,although this terminology is tec
322、hnically correct only during the discharge step)is typically a lithium transition-metal oxide such as lithium cobalt oxide(LCO),lithium manganese oxide(LMO),lithium nickel-manganese-cobalt oxide(NCM or NMC),or lithium iron phosphate(LFP).The electrode also contains electrochemically inactive materia
323、ls that improve electrical and structural characteristics,typically conductive carbon powders and a polymeric binder.The mixture of active and inactive materials is coated on an aluminum foil current collector,which,in turn,is connected to the external electrical terminals of the cell.The negative e
324、lectrode(or anode during discharge)is typically a graphite-based material,with Figure 2.3 Schematic of a Li-ion battery cellElectron transportCu currentcollectorGraphenestructureLiMO2 layerstructureLi+SolventmoleculePositive electrodeNegative electrodeAl currentcollectorLi+transportDashed and solid
325、arrows show the directions of electron and lithium-ion transport during charging and discharging,respectively.Chapter 2 Electrochemical energy storage 21higher-specific-energy Li-ion battery cells now incorporating silicon in varying amounts.These active materials are also mixed with conductivity en
326、hancers and binders and subsequently coated on a copper current collector.To a large extent,the choice of compounds for positive and negative electrodes defines battery performance and favored applications of different types of Li-ion battery cells.The liquid electrolyte enables the move-ment of lit
327、hium ions between the two electrodes during charge and discharge;it consists of a lithium salt(e.g.,lithium hexafluorophosphate,LiPF6)dissolved in an organic solvent,which is most commonly composed of a blend of alkyl and cyclic carbonates(e.g.,ethylene carbonate,propylene carbonate,ethyl methyl car
328、bonate,etc.).Various chemical additives in the electro-lyte are used to improve the performance,lifetime,and safety of the cell.The liquid electrolyte can also be infused into a polymer,forming a gel electrolyte;Li-ion batteries that use this type of electrolyte are typically called lithium-polymer(
329、or Li-poly)batteries(Dahn and Ehrlich 2010).In addition to gel electro-lytes,fully solid polymer electrolytes have been usedalbeit in batteries produced at relatively low volumes;today,inorganic compounds(ceramics)are being widely studied as possible successors to liquid electrolytes(Doughty 2010).S
330、olid-state batteries using either organic or inorganic electrolytes have potential advan-tages in safety and energy density compared to liquid electrolyte systems,but lag in commer-cial maturity.Li-ion cells are manufactured in a wide range of sizes and in two basic forms:cylinders and rectangular p
331、risms.Cylindrical cells are typi-cally contained in a metal can,while prismatic cells may be contained in a metal can or in a sealed bag made from a multilayered polymer sheet,forming a so-called pouch cell.Individual Li-ion cells,for which the nominal cell voltage is determined by the specific comb
332、ination of positive and negative electrode materials,can be directly used in small-scale applications such as cell phones.To deliver the increased capacity and operating voltage required for larger-scale applications,multiple cells are interconnected in various series and parallel configurations to
333、form battery modules and packs.Applications that require a large number of interconnected Li-ion cells,such as EVs and grid-scale energy storage systems,also require several additional subsystems to ensure proper and safe operation.These subsystems include thermal management systems that help maintain a proper cell temperature range and battery management systems(BMSs)that electronically monitor a