1、 X-Change:Batteries The Battery Domino Effect December 2023 In partnership with:rmi.org/1 X-Change:Batteries The Battery Domino Effect Authors Daan Walter,Kingsmill Bond,Sam Butler-Sloss,Laurens Speelman,Yuki Numata,Will Atkinson Contacts Daan Walter daan.walterrmi.org Kingsmill Bond kbondrmi.org Sa

2、m Butler-Sloss sbutlerslossrmi.org Acknowledgments The X-Change:Batteries report was produced by RMI in partnership with the Bezos Earth Fund and is a contribution to Systems Change Lab.We would like to thank the following individuals for their input and expertise:Nigel Topping,Kelly Levin,Mark Dyso

3、n,Clay Stranger,Lena Hansen,Joel Jaeger,Tilmann Vahle,Zhe Wang,Carmelita Miller,Aparajit Pandey,James Newcomb,Achim Teuber,Leonardo Buizza,and Stephanie Schenk.About RMI RMI is an independent non-profit,founded in 1982 as Rocky Mountain Institute,that transforms global energy systems through market-

4、driven solutions to align with a 1.5C future and secure a clean,prosper-ous,zero-carbon future for all.We work in the worlds most critical geographies and engage businesses,policymakers,communities,and non-governmental organizations(NGOs)to identify and scale energy system interventions that will cu

5、t greenhouse gas emissions at least 50 percent by 2030.RMI has offices in:Basalt and Boulder,Colorado;New York City;Oakland,California;Washington,D.C.;and Beijing.rmi.org/2 X-Change:Batteries The Battery Domino Effect Table of Contents Summary.3 1 Exponential change so far.5 Energy density rose.5 Co

6、sts fell.6 Demand increased.7 Supply responded.8 Innovation accelerated.10 The battery domino effect.11 Experts keep underestimating batteries.16 2 The drivers of change will strengthen.18 Policy action will increase.18 Corporate investments will grow.19 Costs will continue to fall.19 Energy density

7、 will continue to rise.20 New dominoes will tip.21 Barriers are rapidly diminishing.23 3 Implications of continued growth.26 Battery demand to grow by an order of magnitude.26 The battery domino effect will capture all of road transport.27 Batteries put over half of fossil fuel demand at risk.28 Bat

8、teries put climate goals within reach.30 Change will not happen by itself.32 Appendix 1:Different ways to model battery sales.33 Sector-by-sector S-curve outlook.34 Aggregate S-curve outlook.40 Growth rate outlook.42 Announced capacity outlook.42 Expert outlooks.42 Appendix 2:Barriers to change.43 C

9、harging infrastructure.43 Mineral demand.44 What is the X-Change.47 References.48 rmi.org/3 X-Change:Batteries The Battery Domino Effect Summary Battery demand is growing exponentially,driven by a domino effect of adoption that cascades from country to country and from one sector to the next.This ba

10、ttery domino effect is set to enable the phaseout of half of global fossil fuel demand and be instrumental in abating transport and power emissions,propelling us over 60%of the way toward a zero-carbon energy system.Demand is growing on an S-curve.Battery sales have been doubling every two to three

11、years,and we are on track for a six to eight times increase by 2030,with sales of 5.5-8 TWh per year.A reinforcing feedback loop of scale,cost,and quality.As the battery market grows,unit cost keeps falling and quality keeps rising.Both battery cost and energy density are on learning curves:for ever

12、y doubling of battery production,costs fall by 19%-29%and the density of leading batteries rises by 7%-18%.At this rate,by 2030,battery cell costs will fall to$32-54 per kWh and top-tier batteries will have an energy density of 600-800 Wh/kg.The domino effect by sector.As one sector scales up demand

13、,the cost and quality feedback loops enable batteries to start uptake in the next.Batteries started in consumer electronics,moved to motorbikes and buses,and then into cars.Now they are moving into stationary electricity storage and trucking and will be ready to enter short-haul ships and planes by

14、2030.The domino effect by country.Once new battery technology is successful,it jumps geographies.The shift of batteries into the car market was started by early adopters;China is the largest domino to fall;and the transition is now shifting across the rest of the world,from Europe to the US,from Sou

15、theast Asia to India.The biggest capacity ramp-up since World War 2.The race to the top means we are building 400 gigafactories,with capacity to make 9 TWh of batteries a year by 2030 over 1 kWh for every person on the planet.The largest clean tech market.In 2022 we spent more on building battery fa

16、ctories($45 billion)than on solar and wind factories combined;and by the end of the decade,the battery market will be larger than both solar panels and wind turbines.Over half of fossil fuel demand is at risk.Batteries will enable renewable technologies to replace 175 EJ of fossil fuel demand in the

17、 electricity sector and 86 EJ of demand in the road transport sector,and put at risk the remaining 23 EJ of transport demand from shipping and aviation.Batteries put climate goals within reach.As batteries help phase out fossil fuels,they enable the reduction of global emissions by 22 GtCO2 per year

18、,which is over 60%of global energy-related emissions today.On the current S-curve trend,battery uptake is set to outpace net-zero scenarios.Incumbent modelers are behind the curve.Current models keep underestimating the speed of change in batteries.If we stay on the current S-curves,battery sales in

19、 2030 will be up to double consensus expectations of around 4 TWh a year.Barriers are soluble.Although supply chains are stressed,thanks to constant innovation and in-vestment,we have enough raw materials that can be sourced equitably and sustainably and can act fast enough to build the charging inf

20、rastructure required for the future battery-dominated energy system.Change does not happen by itself.Batteries got this far through the concerted efforts of compa-nies,governments,researchers,and climate advocates.Continued growth will require continued effort.The path ahead for batteries is clear,b

21、ut we do still need to walk it.rmi.org/4 X-Change:Batteries The Battery Domino Effect The battery domino effect in seven charts Figure 1:The Battery Domino Effect in Seven ChartsBattery energy density keeps risingBattery demand outlook,TWh/y0030FasterFastBattery cell cost outlo

22、ok,$/kWhTop-tier battery cell energy density outlook,Wh/kg030FastFaster2004005006007008002000 2005 2010 2015 2020 2025 2030FastFaster while battery cost keeps fallingdriving exponential growth of battery demandThe result is a domino effect of battery uptake across se

23、ctors and countrieswhich,in turn,further increases energy density and lowers cost through economies of scale and learning effects.The battery domino effect by sector80%40%100%0%20%60%Battery storageElectronicsCarsHeavy trucksLight trucksBusesTrains,ships,&airplanes(illustrative)2/3 wheelers80%0%20%4

24、0%60%100%JapanOther OECD AsiaChinaSoutheast AsiaAustraliaEuropeIndiaSouthKoreaUSRest of WorldShare of addressable market realized,2022The battery domino effect by country EV car exampleBattery technologyIncumbent technologyBattery technologyIncumbent technologywhich puts over half of todays fossil d

25、emand at riskand propels us 60%of the way to a zero-carbon energy system-24-22-20-18-16-14-12-10-8-6-4-20202520302035204020452050Abatement of CO2 enabled by batteries,GtCO2/yFossil demand put at risk by batteries,EJ8617523TotalOther transportElectricityRoad transport284(57%of total)Road transportEle

26、ctricityOther transport60%of global energy CO2 emissionsrmi.org/5 X-Change:Batteries The Battery Domino Effect 1 Exponential change so far Battery demand is growing exponentially.Rising energy density keeps unlocking new uses while declining costs enhance affordability and accelerate market uptake.T

27、his uptake,in turn,drives further cost reduc-tions and continuous innovation a cycle of self-perpetuating progress.The result is a domino effect,whereby batteries enter new markets,from country to country and from one sector to the next.Geopolitical tension has brought new players into the markets,s

28、peeding up the race to the top.The pace of change keeps confounding experts,who consistently underestimate the potential and expo-nential growth of batteries.Defying past predictions,batteries now play a key role in the energy transition and their continued rapid growth signals a seismic shift in th

29、e energy system to come.Energy density rose Batteries have been around for over 200 years.Yet they have only recently emerged as a key technology in the global energy system.For most of their history,batteries were simply too heavy to be practical in sectors that could otherwise have readily used th

30、em,such as transport.In the early days of the car,in 1900,over one-third of cars in the United States were battery powered,but their limited range due to battery weight could not compete against internal combustion engine(ICE)cars,so battery cars rapidly faded away.1 The development of battery energ

31、y density the amount of energy carried per unit of weight stagnated for over half a century until the 1970s and 1980s,when innovation in the United States and Japan picked up.2 In the 1990s,lithium-ion batteries entered the market.Rapid innovation in energy density and cost declines in the decade th

32、at followed enabled novel applications in electronics,such as battery use in cell phones and laptops.As energy density rose,battery applications in transport such as electric motorbikes,cars and light trucks became useful.Energy density has kept rising ever further,raising expectations that new appl

33、ications in heavy trucks and even aviation may become possible as well.This is illustrated below by the rise of top-tier battery density over time.Figure 1:Top-tier battery cell energy density by decade,Wh/kg Source:Zu and Li(2011),3 for 1900s-2000s,Bloomberg New Energy Finance(BNEF)Long-Term Electr

34、ic Vehicle Outlook(2023)4 for 2010s and 2020s Figure 1:Top-tier battery cell energy density by decade,Wh/kgMinimum viable energy density1,examplesSodium/sulfurLithium-ionNickel/zincSolid-state(multiple chemistries)Lead-acidShort-haul airplanesHeavy trucksLight trucksCarsElectronics0500300

35、3504004505005506006507007501920s 1930s 1940s 1950s 1960s1900s1980s 1990s 2000s 2010s1910s2020s1970s1.Minimum density at which first full battery-electric models are feasibleAchieved by 2023Further potential growthrmi.org/6 X-Change:Batteries The Battery Domino Effect If we look at the development si

36、nce 1993,top-tier energy density has increased by 7%for every doubling of battery deployment.Since 2012,top-tier energy density has been growing even faster,at 18%for every doubling of deployment.The innovation potential of batteries is in part driven by the wide range of elements that can make up a

37、 battery.Batteries can be made from many different chemical compositions,which means the chemical properties of lithium,sodium,nickel,cobalt,manganese,and many other minerals can be used to improve the technology.When contrasted with,for example,a combustion engine which effectively only relies on s

38、teel(for the engine itself)and hydrocarbons(as fuel)it becomes clear why battery innovation has so much potential:the solution space is vast,and we have only just started to explore it.Figure 2:Top-tier battery cell energy density vs.cumulative battery market size Source:Ziegler and Trancik(2021)5 b

39、efore 2018(end of data),BNEF Long-Term Electric Vehicle Outlook(2023)6 since 2018,RMI analysis Costs fell As innovation accelerated and battery uptake rose,so battery costs fell.Battery cost decline closely fol-lowed Wrights law over the past three decades as production grew,economies of scale and R

40、&D learning drove down cost.7 The learning rate for battery costs since the first introduction of the lithium-ion battery in 1991 has been 19%.That means that battery prices fell by 19%for every doubling of battery deployment.To arrive at this learning rate for battery cells,we use total battery dem

41、and in all applications,including consumer electron-ics,and look at the average cost decline for every doubling since 1991.i The battery cell learning rate has increased over time,and over the past two decades,the learning rate was 29%.i Learning rates for batteries can be calculated for cells or fo

42、r packs and over different time periods.One further complication is that some ana-lysts such as BNEF do not include consumer electronics in their calculations;the impact of this is to make battery demand lower in early years and so increase the growth rate and thus reduce the apparent learning rate.

43、1001,00010,000.00100.0010.001,000.000.010.101.00202320121993Top energy density,Wh/kgBattery market size,GWh,cumulativeFigure 2:Energy density of top battery cell in production by decade18%7%Improvement rate,%rmi.org/7 X-Change:Batteries The Battery Domino Effect As batteries dropped in cost,they did

44、 not just improve the economics of existing battery technology they also enabled competitive market entry in new sectors that were unlocked by rising energy density.Figure 3:Lithium-ion battery prices,$/kWh(left),$/kWh log scale(right)Source:Ziegler and Trancik(2021)8 for 1991-2014,BNEF Lithium-Ion

45、Battery Price Survey(2023)9 for 2015-2023,RMI analysis Demand increased As batteries became more attractive through improvements in energy density and cost,market uptake took off.Battery demand has grown at an average annual rate of 33%for nearly three decades.Electronics drove early uptake with cam

46、corders,cell phones,laptops,and other niche applications(such as grid frequency control).As saturation was reached in these sectors and growth started to ease to 20%per year,batteries became a viable technology in the transport market in the 2010s.This led to a resurgence of market growth.Since 2014

47、,battery demand has been growing at an annual average of 41%,doubling every two years.As the global battery market scaled up,it reinforced the rise in energy density and fall in cost,which in turn accelerated uptake.The result was a rapid,exponential uptake of battery demand.1001,00010,0000.011.0010

48、0.0010,000.002003Figure 3:Historical lithium-ion battery cell price decline,USD/kWh01,0002,0003,0004,0005,0006,0007,0008,0009,0001990 1995 2000 2005 2010 2015 2020 2025-99%Battery cost declineBattery cost versus market size,log-log scale19%Learning rate 1991-2023:Battery market size,GWh,cumulative29

49、%Learning rate 2003-2023:rmi.org/8 X-Change:Batteries The Battery Domino Effect Figure 4:Battery demand by sector,GWh/y Source:Ziegler and Trancik(2021),10 Placke et al.(2017)11 for 1991-2014;BNEF Long-Term Electric Vehicle Outlook(2023)12 for 2015-2022 and the latest outlook for 2023(*)from the BNE

50、F Lithium-Ion Battery Price Survey(2023).13 Supply responded As battery demand rapidly grew,so supply had to respond.It did so by mounting a supply chain scale-up,at a pace not seen since World War 2(as pointed out by the International Energy Agency(IEA).14 Manufacturing capacity built out Battery d

51、emand grew fast over the past decade,but even so,manufacturing supply has been able to out-pace it:demand grew by a factor of 24,yet battery manufacturing capacity went up by a factor of 42.As battery manufacturers are expecting more growth to come,annual manufacturing capacity additions are still r

52、ising todays annual commissions of new factories are more than 18 times those of just five years ago.15 As the prospect of rising demand became apparent,investment markets paid attention and made ample capital available.16 According to Bloomberg New Energy Finance(BNEF),investment in battery factori

53、es to-day($45 billion in 2022)is considerably higher than investment in solar and wind factories combined($33 billion).17 As capital availability ceased to be the primary limiting factor,the challenge of meeting demand has pri-marily been an engineering one:factories need time to be built and start

54、operations.Companies are be-coming more proficient at building out factories the first battery gigafactory opened in 2016,and now dozens are under construction.According to Benchmark Minerals,there are now 240 operational gigafac-tories across the world,and this is set to increase to over 400 by 203

55、0.18 These factories can be deployed rapidly:in Western countries it usually takes one to two years to build a large battery factory,but in China Figure 4:Battery demand by sector,GWh/y005006007008009001,0001,1001994 1996 1998 2000 2002 2004 200619922010 2012 2014 2016 2018 2020 20222008C

56、ommercial EVsTwo/three-wheelersStationary storagePassenger EVsConsumer electronicsE-busesOtherAutomotive(all)*rmi.org/9 X-Change:Batteries The Battery Domino Effect this can now be done in as little as six months.ii With sufficient advance planning and reliable forecasts on expected demand,battery m

57、anufacturing has been able to scale up rapidly to meet demand.Figure 5:Installed battery manufacturing capacity,GWh/y capacity Source:BNEF,19 RMI analysis Mineral mining sectors expand The rise of batteries disrupted many mineral mining sectors that had long been niche.As battery demand grows at a b

58、reakneck pace,these sectors are racing to catch up.Mining of minerals such as lithium,cobalt,and nickel is undergoing exponential growth.For example,in lithium mining,over 60%of current mining production assets is less than five years old.20 Investments in lithium extraction is increasing by over 30

59、%annually,and new technologies like direct lithium extraction(DLE)are attracting significant funding.Major companies,including Exxon with its new DLE site in Arkansas,are entering the field,indicating further acceleration of growth.21 Batteries have rapidly taken over the role of main demand driver

60、across minerals.According to the IEA,in 2022,EV batteries accounted for 60%of lithium,30%of cobalt,and 10%of nickel demand.22 This scale-up has not been without challenges.Issues such as labor rights,environmental impacts,and the socioeconomic effects on local communities are increasingly coming to

61、the fore.For example,cobalt mining in places such as the Democratic Republic of Congo has raised concerns over ethical sourcing due to reports of hazardous working conditions and child labor.23 We discuss how these challenges are being tackled in section 2 and provide more detail in the appendix.ii

62、For example,consider Teslas Giga Nevada(2014-2016),Giga New York(2014-2017),and Giga Texas&Giga Berlin(2020-2021),but also Chinas Giga Shanghai(2019,168 working days).2004006008001,0001,2001,4001,6001,8002,0002,2002,4002,60003,2003,0002,8002000004Figure 6:Installed b

63、attery manufacturing capacity,GWh/y capacity ChinaEuropeOtherUSArmi.org/10 X-Change:Batteries The Battery Domino Effect Innovation accelerated As the battery market grew,so did the R&D budgets that companies and governments were willing to spend on it,driving energy density and cost improvements.Cel

64、l innovation boomed R&D spending has been essential in improving energy density and battery cost.Analysis by Ziegler,Song and Trancik shows that some 55%of cell cost decline came from public and private R&D and only 45%from other drivers,such as economies of scale.24 New battery models need extensiv

65、e lab research time and costly testing equipment.Just opening up a battery lab for research can easily come with a$5 million start-up cost.25 As fundamental battery research got funded,the battery research community grew quickly;academic publications on batteries increased fivefold over the past dec

66、ade.26 As more money came into R&D,the pace of technology breakthroughs accelerated.Annual patent filings on batteries have risen by 15%per year over the past decade.In 2018(the most recent year with full data),the number of patents filed during the year was more than twice the rate from just seven

67、years before.27 This suggests that innovation continues to accelerate,and we can expect more breakthroughs in years to come.Figure 6:Battery patents filed over time by region,thousands of filings per year Source:IRENA,via Our World In Data28 Innovation in design and business and delivery models Equa

68、lly important breakthroughs were made in the way batteries were built into technologies,and which business and delivery models were able to make for a better user experience.Tesla understood early on that batteries are the most important part of the EV.Collaborating with Pana-sonic,Tesla designed it

69、s cars around the battery to optimize EV cost and operational performance.To ad-dress range anxiety,Tesla also developed one of the largest EV fast-charging networks in the world.29 The market valuation of Tesla today eclipses that of all other traditional car manufacturers.30 55404550055

70、20022004200620082001620182000Figure 7:Battery patents filed over time by region,thousandsJapanChinaEuropeOtherUSArmi.org/11 X-Change:Batteries The Battery Domino Effect Companies such as Sun Mobility,Gogoro,and Battery Smart have successfully adopted a battery swapping model for smaller v

71、ehicles such as scooters and mopeds in India,with companies such as NIO working to bring battery swapping to passenger vehicles in China as well.31 This reduces charging waiting times,grid constraints and charging anxiety,and is a reimagination of what it means to fuel your vehicle.Stationary batter

72、y companies have found significant cost reductions not only in cell technology but also in construction processes.32 In the battery-electric aviation sector,companies like Eviation,Airflow,and Beta Tech are rethinking the aviation business model shifting from traditional flight patterns to shorter,b

73、at-tery-feasible routes.33 Investment in innovation increased Venture capital has been paying close attention.Investments in battery innovation surged over the past decade.Early and growth stage funding are both an order of magnitude larger than just five years ago.Figure 7:Venture capital investmen

74、ts in batteries Source:IEA Global Electric Vehicle Outlook(2023)34 The battery domino effect Batteries are adaptable,modular,and can be used across key energy sectors and geographies.Their ver-satility has enabled scale and innovation in one domain to catalyze initial uptake in others,creating a dom

75、-ino effect of adoption.The sectoral domino effect This domino effect started from electronics and then moved to two-and three-wheelers,buses and cars.Today,battery technology already makes up 50%of two-/three-wheelers and 40%of bus sales.The largest domino has also fallen with cars as discussed in

76、the previous X-Change report on cars,EV cars have hit a tipping point and are set to capture all of passenger cars.Today,most electric passenger cars already are at cost parity with ICEs on a total cost of ownership basis.35 Sticker price parity is expected to be achieved soon as well.36 Early 2023

77、figures show that almost 20%of global car sales are EV today.37 After cars,more sectors are ready to tip as well.We will discuss them in section 2.00500600700800900200018Battery recycling and reuseBattery management systemBattery componentBattery makerFigure 7:Ventur

78、e capital investments in batteries00500600700800200022Early stageMillion$Growth stageBillion$rmi.org/12 X-Change:Batteries The Battery Domino Effect Figure 8:The battery domino effect by sector Source:BNEF38,RMI analysis;Electronics share of addressable market percen

79、tage indicative,transport percentage based on 2022 EV sales share,stationary storage defined as sales volume today divided by peak sales in long term(2050).Trains,ships,and airplane total addressable market sizes illustrative.Figure 9:Stage of battery adoption by sector concept chart Source:RMI anal

80、ysis 100%80%60%40%20%0%CarsLight trucksStationarystorageHeavy trucksBuses2/3 wheelersTrains,ships,&airplanes(illustrative)ElectronicsBattery technologyIncumbent technologyShare of total battery addressable marketShare of addressable market realized,2022Figure 8:The battery domino effect by sectorSha

81、re of adoptionTrainsShipsAirplanesElectronicsEarly adoptionSystem integrationMarket expansion2/3 wheelersBusesCarsStationary storageHeavy trucksFigure 10:Stage of battery adoption by sectorLight trucksTime since initial deploymentrmi.org/13 X-Change:Batteries The Battery Domino Effect The country do

82、mino effect Battery technology growth did not happen uniformly across countries.China has been a clear frontrunner in the battery revolution.But Western countries are catching up.As the battery domino effect gained mo-mentum,governments started to realize batteries are an essential part of their ene

83、rgy future.The geopo-litical competition over batteries accelerated innovation and turbocharged the scale-up of batteries.Figure 10:The battery domino effect by country EV car example Source:BNEF39,RMI analysis;share of addressable market based on sales share in 2022 Figure 11:Stage of EV car batter

84、y adoption by country concept chart Source:RMI analysis 60%0%20%40%80%100%ChinaJapanSoutheast AsiaAustraliaEuropeIndiaSouthKoreaUSRest of WorldOther OECD AsiaShare of total addressable battery marketShare of addressable market realized,2022Figure 10:The battery domino effect by country EV car exampl

85、eBattery technologyIncumbent technologyChinaEuropeUnitedStatesRest of WorldFigure 12:Stage of battery adoption by region Cars exampleSouth KoreaEarly adoptionSystem integrationMarket expansionTime since initial deploymentShare of adoptionrmi.org/14 X-Change:Batteries The Battery Domino Effect China

86、leads the charge China is the frontrunner in battery uptake,leading the EU and United States in battery sales across all key sectors.This is in part because China was early to the game.Already in 2015,EV sales for two-wheelers in China were over 50%,versus 0-1%in the EU and US.Similarly,EV sales for

87、 buses were over 50%in China in 2020,compared to about 10%in the EU and US.40 Over recent years,one country after another started to catch up.New policies were introduced to promote battery uptake.In the EU,a suite of incentives including tax exemptions and rebates,alongside non-monetary benefits su

88、ch as access to bus lanes spurred the rapid adoption of EVs.41,42,43 In the US,the Inflation Reduction Act of 2022 also introduced a set of incentives,offering substantial tax credits for electric passenger vehicles and home battery storage solutions.44 Subsidies historically focused on lighter vehi

89、cles,but in the past year a similar policy push has taken shape for heavy trucks as well.For example,the US Inflation Reduction Act is set to provide incentives for heavy duty electric trucks of up to$40,000 per vehicle,making the cost per mile close to parity with traditional diesel trucks.45 Other

90、 countries are also trying to enter the race.India is actively promoting the uptake of battery technol-ogy by implementing incentives like the FAME-II scheme,the PLI scheme for battery storage,a battery swap-ping policy,reductions in customs duties for battery materials,and establishing special e-mo

91、bility zones.46 Meanwhile,China is not sitting still.It has introduced significant incentives such as financial subsidies and tax deductions for EVs.For example,China spent about$57 billion to support electric car purchases be-tween 2016 and 2022,over five times what the United States spent in the s

92、ame period.47 National subsidies for EVs concluded in 2022,marking a pivotal transition where the economics of EVs,having reached cost parity,began to drive the market independently of government intervention.48 China dominates battery supply China not only dominates demand,it also leads battery sup

93、ply.Some 77%of batteries are made in China today.49 The countrys leading position is not limited to sheer volume China has also made remarkable strides in innovation and lead battery patents today(see Figure 6).A prime example of Chinas pioneering efforts is the recent unveiling by CATL of its cutti

94、ng-edge 500 Wh/kg battery cell.50 Chinas dominance in battery technology is a relatively recent development.The origins of lithium-ion bat-teries trace back to innovation by US oil companies in the 1970s(notably,Exxon).51 Initial progress was abandoned in the 1980s as oil&gas companies doubled down

95、on fossil fuels.It was not until the 1990s that lithium-ion batteries found commercial success in electronics,thanks to Japanese firms such as Sony and demand from rich OECD countries.It was only in the late 2000s that China ascended to the role of the primary consumer,innovator,and producer of batt

96、eries.Today,Western nations are reentering the fray.Bringing battery manufacturing to home soil has become a focal point of industrial policy in both the United States and the EU.52 The additional support for R&D and factory development is driving further battery uptake.rmi.org/15 X-Change:Batteries

97、 The Battery Domino Effect Figure 12:Share of battery value chain by region,%of total Source:IEA Global Supply Chains of EV Batteries(2022)53 Energy security concerns boost domestic demand and supply creation efforts In response to a resurgence in energy security concerns triggered by Putins invasio

98、n of Ukraine,both the EU and the United States have taken legislative actions that promote domestic battery manufacturing and the development of clean energy technologies.In the EU,the RePowerEU legislation seeks to reduce dependency on imported fossil fuels,especially from Russia.It does so through

99、 targeted programs that include promoting EVs to curb oil demand and stationary batteries to replace gas peaker plants.54 In the US,the Inflation Reduction Act contains$500 billion in new spending and tax breaks aimed at boost-ing clean energy.The act provides significant incentives for domestic bat

100、tery production as part of its broader strategy to catalyze investments in domestic manufacturing capacity and promote the procure-ment of critical supplies,such as battery minerals,domestically.55 The effects of these policies are already apparent.Over the past year,battery investment significantly

101、 picked up in the US and EU to re-shore supply.In the United States,14 new battery gigafactories with a total capacity of over 400 GWh/y were announced since the IRA was passed(a jump of almost 60%versus the year before).56 The EU is expected to come up with a response to the IRA that will target bu

102、ilding an equal amount of capacity in Europe.57 Chinas plans for the development of gigafactories remained un-changed since the passage of the EU and US policies;the additional capacity is coming on top of not instead of Chinese manufacturing build-out.Figure 12:Share of battery value chain by regio

103、n,%of totalNickelCobaltGraphiteLithiumNickelLithiumGraphiteCathodeAnodeBatteriesEVsCobaltRoWJapanDRCAustraliaIndonesiaEuropeRussiaUnited StatesChinaKoreaMiningProcessingProductionrmi.org/16 X-Change:Batteries The Battery Domino Effect Experts keep underestimating batteries Despite the clear exponent

104、ial trend batteries have been on for decades,experts kept underestimating the pace of change.A systematic underestimation of the improvement potential of energy density and costs led to battery uptake outlooks that were far too pessimistic.This trend persists today.Underestimating energy density It

105、is notoriously difficult to forecast technological advancements,and most experts have been careful not to make explicit assumptions on future energy density.Nevertheless,implicit assumptions on energy den-sity permeate the outlooks of incumbent energy modelers.New battery applications have been dism

106、issed based on low technology readiness levels or the lack of models on the market all of which implicitly assumed energy densities at the time of analysis would not improve.They almost always did.This underestimation has followed a clear pattern.Initially,experts estimate that low energy density wi

107、ll confine batteries to a niche role in a sector.For in-stance,the first laptops were expected to have limited use except for those who could really benefit,such as the military and traveling salesmen.58 Similarly,not too long ago,EVs were seen as suitable only for small cars in dense urban areas.Th

108、en,as battery-powered technology continues its ascent,expert outlooks change,acknowledging a signif-icant role for batteries in the sector,except for a final niche that needs even higher energy density.The IEAs 2009 EV technology roadmap envisioned that half of car sales would not switch to EV,even

109、by 2050,due to their limited range.59 Even in 2017,McKinsey estimated that a third of the car fleet would need hydrogen technology to decarbonize.60 Later,the narrative shifts again,with experts predicting a complete takeover by battery-powered technol-ogy.The latest IEA EV outlook states that the e

110、ntire car segment can switch to EV.61 This trend has played out in electronics(laptops versus desktops),two-/three-wheelers(EV versus ICE scoot-ers)and is unfolding in real time in cars today.Even when batteries are initially seen as unsuitable for a certain sector,they tend to find their entry in n

111、iche or hybrid applications.Many of the early successful EV models,for example,were smaller city cars,or hy-brid battery-ICE cars.Once early models are successful and drive further scale,improved energy density and costs enable further market capture.Overestimating cost While energy density has been

112、 underestimated,costs have consistently been overestimated.Outlooks from as little as five years ago overestimated todays battery cost by a factor of 2.Notably,until 2022 no main-stream forecasts seem to have underestimated cost,showing a clear bias towards conservatism.Only with the global inflatio

113、nary pressures of 2022 did costs rise above forecast levels.But battery cell costs have started to fall again to the lowest-ever level of$107/kWh in 2023(using real 2023 dollars),according to BNEFs annual Lithium-Ion Battery Price Survey.62 Monthly prices even fell below$100/kWh in August this year,

114、according to Benchmark Minerals monthly battery price tracker.63 And that would put battery costs well below forecasts again.rmi.org/17 X-Change:Batteries The Battery Domino Effect Figure 13:Battery cell costs,expert forecasts vs.actuals,$/kWh Source:Mauler et al.(2021)64 for expert forecasts of 201

115、0-2018,BNEF Lithium-Ion Battery Price Survey(2023)65 for actuals and most recent fore-casts.Underestimating sales Underestimation of battery energy density and overestimation of cost led to the underestimation of battery demand growth.For example,BNEF battery demand outlooks from as recently as two

116、years ago were off by a factor of two compared to actuals today.66 A similar trend is unfolding for near-future outlooks.The IEA only started to publish battery outlooks in 2018,and its first four forecasts(2018 to 2021)all had 2025 battery demand at no more than 700 GWh/y67 but we are already excee

117、ding this level in 2023.68 Experts keep being surprised by the pace of change.Figure 14:Automotive lithium-ion battery demand,IEA forecast vs.actuals,GWh/y Source:IEA Global EV Outlook(2018-2023)69 current policy scenarios and actuals;BNEF Long-Term Electric Vehicle Outlook(2023)70 for 2023 estimate

118、.Historical data are similar across the two IEA and BNEF sources,so the 2023 BNEF estimate is shown for comparison.005006007008009001,0001,1001,2001,30020000022202320242025Matteson and Williams(b)Matteson and Williams(a)Cole et al.Catenacci et a

119、l.Sakti et al.Gerssen-Gondelach and FaaijSchmidt et al.Baker et al.Berckmans et al.Nykvist and NilssonFew et al.Forecast(BNEF)Edelenbosch et al.Actual(BNEF)Figure 13:Battery cell cost,expert forecast versus actuals,USD/kWhFigure 14:Battery uptake from EVs,expert forecast versus actuals,GWh/y02004006

120、008001,0001,2001,4001,6001,8002,0002,2002,4002,6002,8003,0003,2002002020224202520262027202820292030IEA 2018IEA 2019IEA 2020IEA 2021IEA 2022IEA 2023Actuals2023 estimatermi.org/18 X-Change:Batteries The Battery Domino Effect 2 The drivers of change will strengthen Drivers of rapi

121、d change will intensify over the coming decade.Policy action and private sector investments will increase,costs will go down,energy density will go up,new markets will form,supply chains will mature,and barriers will continue to be overcome.The stage is set for continued rapid expansion of batteries

122、.Therefore,in line with the methodology we have used across the X-Change report series,we set out in the rest of this note two key scenarios for the future,described as fast and faster.Both scenarios assume that battery sales continue to follow S-curves in line with the trend seen in other cheap mod

123、ular technologies.As explained in more detail in the appendix,the fast scenario assumes a more conservative S-curve and the lower bound of learning rates for energy density and cost,while the faster scenario uses a more ag-gressive S-curve and learning rates.Our analysis of the forecasting power of

124、S-curves suggests that actuals are likely to lie between these two curves.Policy action will increase As batteries continue to rise in significance,so governments will make strategic policies to secure,improve and grow supply.Already today,governments are making targeted policy to accelerate the bat

125、tery industry,as outlined in the previous section.The battery“arms race”has only just started one can expect more aggressive policies to accelerate the build-out for more and better batteries in the coming decade.71 Despite protective policies from the West,China is expected to continue to lead the

126、battery revolution for the foreseeable future.72 Chinese companies are adapting their strategy to deal with protective trade poli-cies in the West,for example by expanding battery production to Western shores although Western countries are also making policy that makes that harder.73 According to BN

127、EF estimates,Chinas battery production capacity will remain an order of magnitude larger than that in the West for at least another five years.74 In the coming decade,Chinese companies are expected to continue to produce the most and the highest quality batteries.Nevertheless,markets in the United S

128、tates,Europe,OECD Asia and India will continue to try and catch up.This arms race will speed up the battery domino effect.rmi.org/19 X-Change:Batteries The Battery Domino Effect Corporate investments will grow Companies worldwide are competing in an ambition loop to produce the best and most cost-ef

129、fective bat-teries.75 Investments in battery factories and EVs are expected to continue their sharp rise across the world.76 And investments in new battery start-ups and scale-ups continue to surge.So far this year,several battery start-ups have raised over$1 billion in funding,including:Verkor($2.1

130、 billion)Redwood Materials($1 billion in Series D funding)and Northvolt:($1.2 billion).77 The total funding going into battery start-ups is expected to exceed$12 billion in 2023 more than six times as much as only five years ago.78 It is notable that investment in battery factories is now the larges

131、t part of the capital expenditure on tech-nology in the energy transition,according to BNEF.This trend is set to continue.Battery manufacturers have already committed to hundreds of new factories that will lead to a surge in global manufacturing capacity this decade.79 Figure 15:Investment in clean

132、tech factories,$bn/y Source:BNEF80 Costs will continue to fall As batteries continue to grow,so costs will fall.Continuing the 19%-29%learning rate identified in section 1 and assuming sales growth of 29%-35%,as we will derive in section 3,would imply that 2030 battery cell costs will end up at$32-5

133、4/kWh,depending on the speed of the transition the faster the uptake,the faster costs will fall due to learning effects and economies of scale.And,in turn,as costs fall,uptake accelerates.EV cars are already competitive in most regions in the world on a total cost of ownership basis.81 As cell costs

134、 keep falling and reach about$80/kWh,EVs will be at sticker price parity with ICEs.82 Battery cell prices fell to below$100 USD/kWh for the first time this summer.83 In both a fast and faster transition,EV car sticker price parity will be achieved by 2025.This will further increase uptake of EV cars

135、,leading to continued technology learning and economies of scale.Figure 16:Investments in clean tech factories000212022SolarWindBatteriesrmi.org/20 X-Change:Batteries The Battery Domino Effect Figure 16:Battery cell cost outlook,USD/kWh Source:BNEF84,RMI analysis based on and h

136、istorical learning rates and deployment outlook.Fast assumes demand of 5.5 TWh/y by 2030 and 19%learning rate.Faster assumes 8 TWh/y demand by 2030 and 29%learning rate.Note 2022 USD.Other cost forecasts Our projections suggest that 2030 battery cell costs will be only a little lower that prominent

137、forecasts.For example,BNEF predicts a cell price of$56/kWh,based on long-term trends since 1992.85 Goldman Sachs estimates a slightly higher cost of approximately$60/kWh in 2030,attributing much of this reduction to falling raw material costs.86 Other recent research estimates are even higher,based

138、on higher assumptions for future mineral costs.87 The difference between our forecast and those of establishment modelers comes from higher learning rates and higher growth rates in our forecasts.Energy density will continue to rise As the battery market and R&D grow,so energy density will keep risi

139、ng.To get a sense of where top-tier energy density may end up by the end of the decade,we can extend the 7%-18%improvement rate trend(as identified in section 1)to 2030,again using the sales growth rates of 29%-35%derived in section 3.Projecting energy density forward with this improvement rate lead

140、s to a 2030 leading battery cell energy density of 600 to 800 Wh/kg.We focus here on the battery energy density of the leaders,not the average,because the focus is on new applications and markets for batteries,rather than the average across existing applications.As top-tier energy density rises,new

141、battery applications will unlock.At around 400 Wh/kg,long haul truck-ing becomes economically attractive.88 At 500-650 Wh/kg,short haul electric aviation becomes feasible.89 Both are within striking distance after CATLs 500 Wh/kg battery cell announcement.As more 500+Wh/kg batteries enter the market

142、 over the coming decade,long haul trucks will decidedly tip,and electric aviation can start to take off.Besides unlocking new applications,higher energy density will also enhance the attractiveness of existing battery technologies.The new generation of(semi-)solid state battery cells of 500+Wh/kg ar

143、e set to double EV car ranges compared to today.90 That would put an end to any residual EV range anxiety.32540030FasterFastFigure 16:Battery cell cost outlook,USD/kWhActualsrmi.org/21 X-Change:Batteries The Battery Domino Effect Not only top tier,but also average energy densit

144、y will rise.This will be driven by new chemistries and ex-piring patents.For example,CATLs patents on lithium iron phosphate(LFP)batteries are expiring,making their high energy density batteries more widely available for production,and spurring CATL to further in-novate to find a new battery chemist

145、ry to patent and use to maintain and capture market share.Already in 2021,CATL introduced its first-generation sodium-ion battery and is continually advancing this technol-ogy.91 Many other companies are following suit.92 Figure 17:Top-tier battery cell energy density outlook,Wh/kg Source:Ziegler an

146、d Trancik(2021)93 for 2000-2016,BNEF Long-Term Electric Vehicle Outlook(2023)94 for 2017-2023,RMI analysis.Fast assumed 5.5 TWh/y by 2030 and 7%improvement rate.Faster assumed 8 TWh/y by 2030 and an improvement rate of 18%.Other energy density forecasts Other bottom-up expert assessments of future b

147、attery energy density appear to confirm the feasibility of our energy density projection.Already today,there are lab-scale battery cells with energy densities that exceed 700 Wh/kg.95 These technologies are on their way toward commercialization.Argonne National Labs even states a potential of 1,200

148、Wh/kg if some breakthroughs are made sooner than expected.96 These batteries still have a long road to the market,but lab results show that there is ample reason to believe top-tier cell density can rise well above 600 and towards 800 Wh/kg by 2030.New dominoes will tip After battery car uptake took

149、 off last year,the battery domino effect is turning to new sectors.We deal with each in turn.Stationary storage takes off Batteries are an ideal complement to variable renewable power generation,such as wind and solar.The inherent intermittency of renewables requires backup,such as batteries,to mana

150、ge variability and ensure a consistent energy supply.Analysis by the IEA shows that over a third of flexibility needed on the 2050 grid will come from batteries.97 Hence as wind and solar continue to grow exponentially,so will the stationary storage market.Annual stationary storage additions have gr

151、own by over 45%per year since 2018,reaching 35 GWh/y in 2022 and are expected to nearly triple in 2023 to 99 GWh/y(according to BNEF).98 This 050030035040045050055060065070075080020002005200252030FastFasterFigure 17:Top-tier battery energy density outlook,Wh/kgActualsrmi.org/22

152、 X-Change:Batteries The Battery Domino Effect growth is taking place at the same time as the continued,exponential growth of renewables expected over the coming decade,as discussed in a previous entry of the X-Change series on electricity.99 The core driver of stationary storage growth is economics.

153、As noted by many energy analysts,solar power coupled with battery storage is the cheapest source of firm electricity generation available today.100,101,102 In 2022,over 40%of planned solar projects globally came with on-site battery storage.103 In California,which is a frontrunner in solar power dep

154、loyment,it was over 95%.104 Stationary storage is not only growing on the utility side of the electricity market.An increasing number of electricity consumers are buying batteries for onsite backup power.Already in 2021,17%of US residential solar power installations were combined with battery storag

155、e.105 Market reports expect home storage sales to grow by 20-30%per year for the coming decade.106 The expected long-term market size for stationary battery storage is 1 TWh/y in both the economic tran-sition and net zero scenarios by BNEF,which is similar to the amount implied by the IEAs net zero

156、sce-nario.107 The actual long-term market size for stationary battery storage is highly dependent on a wide range of variables,and we expect to adjust forecasts as technologies evolve.They include the cost of other storage technologies and the ability to incorporate the batteries of the road transpo

157、rt fleet into electricity storage.For example,DNV forecasts that in 2050,total storage battery deployment will be 22 TWh,backed up by 14 TWh of transport storage from 10%of the transport fleet.108 The EV revolution reaches trucks The battery domino effect is about to tip the trucking sector.Rapid ch

158、ange is already underway.Companies already have over 300 lighter EV truck models to pick from today,and 6%-7%of lighter truck purchases(and 1%-2%of medium-to-heavy trucks)are electric today.109 To put that in context,EV cars passed the 5%threshold only in 2020 and are at nearly 20%today.110 Availabi

159、lity is improving as well:there are over 100 heavy truck models already on the market.That is double the level of just two years ago and an order of magnitude more than five years ago.111 Total EV truck sales in China have doubled annually from 2020 to 2022 and stand at over 6%of total truck sales i

160、n 2022.The Western world is catching up,with the EU at about 4%and the United States at about 1%today.Truck driving ranges are rapidly improving as energy density rises.It is just a matter of time before the new generation of high-density batteries reach the market and allow competitive driving rang

161、es likely not much later than mid-decade.112 Once they do,they will likely pick up fast:at todays battery price point,battery trucks are already compet-itive against ICE trucks on a total cost of ownership basis in many cases.113 Analysis by BNEF foresees EV cost parity in all trucking sectors well

162、before 2030,and already cost parity today for lighter trucking seg-ments.114 Uptake of e-trucks will further accelerate battery growth,create more scale and allow for even larger R&D budgets.This in turn will benefit all other battery technologies.The total long-term market size for batteries in lig

163、ht,medium,and heavy trucks is likely to be 5 TWh/y(see appendix for sizing details).Knocking on the door of“hard-to-abate”sectors Batteries momentum shows no signs of slowing.New solutions are on the horizon,and although uncertain,there are promising signs of the integration of batteries in trains,s

164、hips and planes.Rail already appears closest to a tipping point.New train orders across Europe signal a shift from diesel to battery-powered rail.115 That should come as no surprise,as analysis shows that battery trains can be competitive with diesel trains at a battery price of$100/kWh,which we are

165、 rapidly approaching.116 Uptake is still small,but the early signs of tipping are there.rmi.org/23 X-Change:Batteries The Battery Domino Effect Ships have seen first battery uptake in niche,shorter haul segments.117 At a battery pack cost of$100/kWh(cell price of$75/kWh),battery ships can become com

166、petitive with fuel oil ships on routes shorter than 1,500 km.On the current cost decline trend,this will happen(well)before 2030 in both the fast and faster transition.With another halving of battery price,up to 3,000 km can become economical.118 When such long ranges are achieved,analysis suggests

167、some 40%of con-tainer ship traffic could electrify.119 The first models are being tested out today.In July of this year,China-based Cosco Shipping launched its first battery electric container ship,using swappable bat-teries.120 Aviation applications start to become possible at 500-650 Wh/kg.121 Tod

168、ays top battery density already scrapes the bottom of that range.CATL plans to start using its new 500 Wh/kg battery for aviation,expecting to launch a first model in 2027.Trial flights are already ongoing.122 Initial e-plane models will be small,but as battery density rises,larger planes will becom

169、e viable.At an energy density of 1,000 Wh/kg,most regional(1,000 nautical miles)aviation can turn full electric.123 Based on the current top density improvement rate trends,we may only see announcements for such batteries after 2030 though.Batteries can also be a good complement to other low carbon

170、fuels in shipping and aviation.For example,Heart Aerospace claims that its hybrid airplanes can be up to 50%more efficient than conventional planes.124 Such efficiency gains will greatly help these hard-to-abate sectors to get the most benefits out of the limited and costly sustainable aviation fuel

171、 supply available to them.125 And as history has shown,once batteries get a foothold in a niche of a new sector,they tend to grow faster than expected.The total market size of trains will be small,but ships and planes could be up to a terawatt-hour or two per year,depending on how far energy density

172、 and cost improvements carry battery technology.iii Most of this uptake will only come after 2030.Barriers are rapidly diminishing There are several barriers to battery growth today:the need for minerals,the sustainability and equity of the supply chain,and the build-out of charging infrastructure.E

173、fforts are under way to actively address them all.Mineral supply constraints are disappearing Mineral supply constraints should be examined on two timescales:the short-term challenges of ramping up production,and the long-term consideration of reserve and resource limitations.According to BNEF data

174、in the figure below,battery mineral supply will be able to keep up with even very ambitious demand growth in the short term.This is due to ongoing acceleration of new investments in the battery mineral market.Battery mineral investments are growing by 30%per year,and as fast as 50%per year for lithi

175、um.126 The construction of new mines and improvements in battery technology and efficiency are key factors contributing to meeting the projected demand for minerals such as lithium,graphite,cobalt and nickel.The anticipation of a short-term shortage and subsequent price hikes spurred investments tha

176、t have helped alleviate market tightness,with critical mineral prices declining again over the past year.127 iii There are few good battery market sizing studies available.We can only make a high-level assumption.The Tesla master plan includes a 40 TWh shipping market.So assuming a 25-year ship life

177、time means ships would be a 1.6 TWh/y market.The typical cross-Atlantic flight takes about 5 million liters of kerosene,or 500 MWh.After accounting for improved efficiency when switching from ICE to an electric motor,battery size per plane could be 300 MWh.There are 20,000 active planes in the world

178、,so the total airplane market size would not exceed 6 TWh.Assuming a 20-year plane lifetime,this results in a 0.3 TWh/y market.Therefore,total market for ships and planes combined could reach up to 2 TWh/y in demand.rmi.org/24 X-Change:Batteries The Battery Domino Effect Figure 18:Battery mineral su

179、pply and demand outlook,kiloton mineral per year Source:BNEF Battery Metals Supply and Demand(as of June 1,2023)In the long term,analysis by the Energy Transitions Commission indicates there are more than sufficient battery mineral resources available for the entire energy transition.128 Reaching ne

180、t zero will only take one quarter of todays lithium,one-third of nickel and a quarter of known cobalt resources.Moreover,as mineral supply has consistently been under pressure to scale,battery companies have in-vested heavily in innovation to help alleviate pressure on the mining sector.Innovation i

181、n higher battery energy density led to fewer minerals being needed per battery.The growth of the battery-recycling industry will alleviate demand for new mining of minerals.129 The development of new battery chemistries such as sodium-ion,iron,lithium iron phosphate(LFP)cathodes,and silicon anodes c

182、an diversify the demand for minerals.130 ETC analysis shows that circularity can reduce mineral demand by up to 10%in the short term and possibly 50%in the long term,by extending battery life and recycling materials at end-of-life.131 Driven by policies such as the EU battery pass and economic facto

183、rs,battery recycling is rapidly evolving.PwC foresees a 50%cost reduction potential when pursuing recycling at scale,leading to economic viability of battery recycling in the EU by 2025.132 As recycling scales,economics will increasingly take over from policy as the driver of battery recycling uptak

184、e.The value chain is rapidly cleaning up Emissions Like all high-tech manufacturing,producing batteries comes with an emission penalty from the materials,electricity and heat used in the production process.This is currently estimated at up to 100 kg of CO2 per kWh(but already as low as 45 kgCO2/kWh

185、in Sweden).However,by 2030,the battery industry is on track to dramatically reduce this footprint,targeting a reduction of over 75%.McKinsey analysis suggests that up to 80%decarbonization can be achieved with minimal additional costs,largely due to the growing role of re-newable energy in electrici

186、ty generation(which is in fact enabled by stationary battery storage,as discussed above).133 Efforts by companies and policymakers are converging,with technological advancements and Figure 18:Battery mineral mining outlook,kiloton mineral per year522202420262028Demand(Economic

187、Transition Scenario)Demand(Net Zero Scenario)Supply(Announced)5001,0001,5002,0002,5003,0003,5004,0002022 2024 2026 2028050060070020222024202620284005006007008009001,0001,1001,2001,3001,4001,5002022 2024 2026 2028CobaltGraphiteLithiumNickelrmi.org/25 X-Change:Batteries The Battery Domino E

188、ffect regulatory changes such as the EUs Carbon Border Adjustment Mechanism and the US Inflation Reduction Act driving better emission accounting standards.As the cost-effective 80%of abatement is realized,2030 battery production emissions will be no more than 160 MtCO2e,even in our faster transitio

189、n scenario that foresees demand of 8 TWh/y by 2030.This is more than an order of magnitude less than the thousands of megatons CO2 abatement that those same batteries will enable(see Figure 24).Social inequity Addressing social inequity in the battery value chain,particularly in cobalt mines,has bec

190、ome a priority for the industry.Companies and stakeholders are actively implementing measures to ensure ethical sourcing and fair labor practices.Initiatives include stringent supplier audits to enforce labor standards,investment in technology to trace mineral origins,and partnerships with local com

191、munities to improve working condi-tions and livelihoods.Furthermore,significant efforts are being made to support educational and health programs in mining communities.These actions are complemented by a shift toward sourcing materials from regions with more robust regulatory frameworks,and by inves

192、ting in alternative battery technologies that reduce reliance on conflict-prone minerals such as cobalt.134 This multi-pronged approach aims to cre-ate a more equitable and sustainable battery value chain,mitigating the social impacts associated with mineral extraction.Charging infrastructure will c

193、ontinue to scale As more EVs enter the fleet,the need for charging infrastructure rises.Early uptake,particularly in EV cars,required less charging infrastructure per car:a 2018 study showed that only 5%of EV charging in the EU happened at public charging stations,as most EV owners were suburban and

194、 had space at home to charge.135 As the EV revolution moved to urban car owners,publicly accessible charging points started to grow exponentially.Last year alone,they grew by 55%globally.136 For comparison,EV car fleet growth has been around 62%-63%annually for the last five years.137 National gover

195、nments are playing a pivotal role in this expansion.China is already home to the most EV charging stations globally,with over 1.8 million in 2022 and its 14th Five Year Plan continues to promote charging station build-out,targeting 60%coverage nationally and 80%in densely populated areas by 2025.138

196、 The EU has approved new EV legislation that includes the construction of fast-recharging stations of at least 150kW for cars and vans every 60 km across the EU.139 The United States is also making significant strides.Through the National Electric Vehicle Infrastructure(NEVI)Formula Program,it has a

197、llocated$5 billon of funding to support the development of over half a million chargers for a coast-to-coast highway network.140 The coalescence of strong government planning and proactive private sector involvement should be ex-pected to continue the exponential trend of charging infrastructure rol

198、lout.rmi.org/26 X-Change:Batteries The Battery Domino Effect 3 Implications of continued growth We expect battery demand to grow by an order of magnitude and battery technology to capture all of road transport,enable renewables to clean up the power grid,and help decarbonize ships and airplanes.In d

199、oing so,batteries will put over half of global fossil fuel demand at risk and help to put the power and transport sectors on track to reach climate goals.To concentrate on realistic near-term trends,we focus our forecasts on the period up to 2030.Battery demand to grow by an order of magnitude Every

200、 battery technology we have discussed so far is set to undergo exponential growth over the coming decade.EV cars(as discussed in more detail in the previous X-Change entry on cars),light trucks and heavy trucks will be the main drivers of battery demand growth over the coming decade rising above 3,6

201、00,500 and 600 GWh/y of demand by 2030,respectively.As renewables continue to grow(as discussed in the first X-Change entry on electricity),so stationary storage will grow above 500 GWh/y by 2030.The remaining small-demand areas of electronics,buses and two-/three-wheelers will collectively grow to

202、at around 400 GWh/y.Battery demand from new segments in trains,ships and aviation will likely still be limited by 2030.In total,we expect annual battery demand will grow to 5,500-8,000 GWh/y(or 5.5-8.0 TWh/y)by 2030,de-pending on the speed of the transition.This means that battery demand in 2030 wil

203、l be 5.5 to 8 times larger than in 2023(1 TWh).After 2030,we expect demand will continue to grow to a terminal demand of around 13 TWh/y,or a factor of 13 larger than in 2023.The battery industry is not unfamiliar with such growth:over the past decade,demand grew by a factor of 20 and manufacturing

204、capacity by a factor of 30.Battery manufacturers are already building out capacity for over 9 TWh per year of production by 2030,according to Benchmark Minerals and BNEF.141 Figure 19:Battery demand outlook,TWh/y Source:BNEF,142 RMI analysis Figure 20:Battery demand outlook,GWhHistoricals01234567820

205、00222024202620282030FastFasterrmi.org/27 X-Change:Batteries The Battery Domino Effect Batteries to become the biggest clean tech market Such exponential growth will lead to a total global battery cell market of around$300 billion per year by 2030.To put this in perspective,the

206、wind turbine and solar panel markets will each be below$200 billion in sales.iv Batteries are well on track to become the largest primary clean technology market in the world.The battery domino effect will capture all of road transport The battery domino effect will soon tip all road transport secto

207、rs.After cars,light and heavy trucks will follow.In the fast scenario,we expect sales growth in trucks to follow a similar growth trajectory as that of cars,albeit delayed by a few years.Light commercial EVs have already passed price parity with ICE equivalents on a total cost of ownership basis and

208、 will rapidly grow to 10%-20%of sales by mid-decade.Heavy com-mercial vehicles will take another couple of years to reach price parity,but once this happens,they can be expected to reach 10%-20%of sales by the late 2020s.Under this scenario,we expect EV sales in 2030 to be approximately 60%in cars,4

209、0%in light trucks and 20%in heavy trucks.In the faster scenario,EV sales reach approximately 90%,60%and 50%of sales across cars,light and heavy trucks,respectively,by 2030.Notably,in such a faster scenario,truck EV sales may follow a faster growth trajectory than car sales.This is in part due to the

210、 S-curve method we use:the earlier on the S-curve we are today,the wider the S-curve range between fast and faster(more on this in the appendix).But there is also good reason to believe commercial vehicle sales may move faster than cars:commercial players are more cost-driven and plan more long term

211、.Once cost parity is reached,one may expect a faster shift in commer-cial segment sales than in cars.Figure 20:EV sales in cars and commercial vehicles in fast and faster scenarios,%of sales Source:BNEF,143 RMI analysis iv RMI calculation assuming 2030 volume and price estimates for wind turbines an

212、d solar panels from the previous X-Change:electricity report.Figure 21:EV sales in cars,light and heavy trucks,%of total salesCarsLight trucksMid/heavy trucksOutlookOutlookFastFaster0070809020252030203500708090202520302035rmi.org/28 X-Change:Batteries The

213、Battery Domino Effect As batteries are scaled across the transport sector,trains,ships,and planes may follow.By 2030,the first percentage points of ship and airplane sales may well have shifted to battery technology as well.Figure 21:The battery domino effect going forward,market share sales%,fast s

214、cenario Source:RMI analysis.Note:vehicles are sale shares;stationary storage and electronics are shares of peak sales Batteries put over half of fossil fuel demand at risk Battery-powered technology is set to capture all of road transport,enable renewables to push fossil power generation out of elec

215、tricity,and either have a direct or enabling role to play in aviation and shipping de-carbonization.This means batteries put over half of global fossil fuel demand at risk:18%from road transport,35%from electricity and another 4%from other transport.The end of the ICE age:batteries push out road tra

216、nsport oil demand As the battery domino effect tips all of road transport,new ICE vehicle sales will be rare by 2040.In our previous X-Change entry on cars,we noted that the ICE age is rapidly coming to an end for cars.In this piece,we note that the battery domino effect is set to end the ICE age no

217、t just for cars,but for all of road transport.Once ICE sales dry up,it will take about the average ICE lifetime of 10-15 years for the global fleet to turn over,and hence for the entire ICE fleet to effectively disappear.As lifetimes of trucks are shorter than those of cars(10-12 years versus 15 yea

218、rs),the fleet transition to EV in trucks is set to complete at around the same time as cars,even though truck sales are lagging cars by about five years.The implication is that by 2050,virtually no ICE vehicles will be left on the road.Well before that,ICE owners will start to experience increasing

219、difficulty in using their vehicles as the in-vestment market dries up.Car parts will become rare,and refueling stations will switch to EV charging,moving range anxiety from EVs to ICEs.These negative feedback loops could further accelerate the rate at which people discard their ICE and turn to an EV

220、.10%20%30%40%50%60%70%80%2020202520302035Figure 21:The battery domino effect going forward,market share sales%,fast scenarioTrainsShipsPlanesCarsStationarystorageLighttrucksHeavytrucksElectronicsrmi.org/29 X-Change:Batteries The Battery Domino Effect Figure 22:ICE sales by sector,%of annual sales So

221、urce:BNEF,144 RMI analysis As ICEs are phased out,so will road transport oil demand.Road transport oil demand was 43 mbpd in 2022 according to BNEF,which was just under half of global oil demand and 18%of global fossil fuel demand.This demand will fall by 65%-85%by 2040 and almost fully vanish by 20

222、50.Fossil fuel savings from batteries It is possible to calculate the fuel savings from batteries from either a top-down or a bottom-up perspective.Care must be taken to distinguish between flows and stocks.The conclusion is about the same that each kWh of battery(weighing around 3-4 kg)over its lif

223、etime will save around 300 kg of oil.For example,an ICE car uses around 1 ton of oil a year.If it is replaced by an EV with a 50 kWh battery,the battery will save 20 kg of oil per kWh per annum.Over a 15-year life,that is a 15 ton saving of oil demand,or 300kg per kWh of battery.Figure 22:ICE sales

224、by sector,%of annual sales00708090202520302035CarsLight trucksMid/heavy trucks00708090202520302035FastFasterOutlookOutlookrmi.org/30 X-Change:Batteries The Battery Domino Effect Figure 23:Road oil demand,mbpd Source:BNEF,145 RMI analysis Batteries enable r

225、enewables to push out fossil power generation As batteries enable wind and solar to continue to grow,so renewables will push out 175 EJ of fossil fuel demand for power generation,amounting to 35%of global fossil fuel demand.Renewables are likely to reduce fossil power generation by over 20%by 2030 a

226、nd push out fossil fuels entirely by 2050.146 Clearly batteries are just one among several enabling factors,but their role is nevertheless key.As noted above,the IEA states that one-third of 2050 flexibility is likely to be provided by batteries.More dominoes will fall:trains,ships and airplanes As

227、the battery domino effect starts to tip battery uptake in trains,ships and airplanes,another 13 mbpd of oil demand is put at risk,or about 4%of global fossil fuel demand.Batteries will either serve as a sole replacement for fossil fuel technology or help ease the uptake of bio-and synthetic fuels by

228、 making fossil fuel planes and ships more efficient through hybrid battery-ICE models.Ultimately,batteries will have a role to play in the phase-out of fossil fuels in these sectors be it as a direct replacement or as a(hybrid)enabler of another green solution.Batteries put climate goals within reac

229、h Batteries will help reduce over half of global energy-related CO2 emissions As batteries enable the phase-out of half of fossil fuel demand,they will have an outsized impact on global emissions reductions.Batteries are an essential ingredient for the decarbonization of over 22 GtCO2/y:5.9 GtCO2 fr

230、om the direct replacement of fossil fuel technology in road transport,14.6 GtCO2/y from enabling renewables in the electricity sector,and 1.6 GtCO2/y either replaced or enabled in other transport.That is more than 60%of global CO2 emissions,and much of the pathway to a net-zero energy system.0510152

231、025303540452020202520302035204020452050Figure 24:Road oil demand outlook,mb/dFastFasterOutlookrmi.org/31 X-Change:Batteries The Battery Domino Effect Figure 24:CO2 emission abatement enabled by batteries,GtCO2/y Source:IEA NZE scenario(other transport);RMI analysis(power and road transport)Batteries

232、 are on track to meet this challenge.As pointed out by the IEA and detailed below,batteries are already overperforming compared to what is needed for a net-zero scenario today.147 As battery technology continues to rise,net zero is within reach across road transport and the power sector.Figure 25:Ba

233、ttery technology uptake vs.what is required in a net-zero scenario,2030 Source:RMI analysis,IEA Net Zero Energy(NZE),BNEF Net Zero Scenario(NZS)148 Figure 25:CO2 emissions abated with battery technology,fast scenario,GtCO2/y-20-100-22-18-16-14-12-8-6-4-2-24202520302035204020452050Battery technologyS

234、ectorCommentRoad transportEV bikesEV busesEV carsEV light trucksEV heavy trucksDirect abatement of oil demandEnabled abatement of power emissions by renewables Power generationStationary storagePotential abatement in other transport as domino effect continuesOther transportEV trainsEV/hybrid planesE

235、V/hybrid shipsFigure 25:Battery technology uptake versus what is required in a net-zero scenario,20300%5%10%15%20%25%30%35%40%45%50%55%60%65%70%75%80%85%90%IEA NZEBNEF NZS0%5%10%15%20%25%30%35%40%45%50%55%60%65%70%75%80%85%90%IEA NZEBNEF NZS0%5%10%15%20%25%30%35%40%45%50%55%60%65%70%75%80%85%90%IEA

236、NZEBNEF NZS0500300350400450500550600650700750IEA NZEBNEF NZSFasterFastStationary storage,GWh/y,2030Heavy truck sales,%of total,2030Light truck sales,%of total,2030Car sales,%of total,2030rmi.org/32 X-Change:Batteries The Battery Domino Effect Change will not happen by itself As we have ou

237、tlined,change is not an autonomous force,but the result of deliberate actions and concerted efforts.The exponential growth of batteries has been made possible by hundreds of thousands of people working billions of hours over recent decades to make batteries better,cheaper,and successful.The im-plici

238、t assumption behind scaling laws such as Wrights and Moores and the theory of S-curve uptake is that human ingenuity and effort will continue to be poured into technological advancement.But this is not a given.Not all technologies make it up the S-curve.Concerted lobbying against technologies,compan

239、ies not setting up sustainable value chains and underinvesting in enablers can stop technology development dead in its tracks.Governments,companies,researchers,and climate advocates will have to unite to drive sustainable and equitable growth in the battery industry.This includes a focus on innovati

240、ve R&D for sustainable technology and adaptable solutions,supported by government funding and corporate investment.Collaboration is key,with an emphasis on establishing fair supply chains and enhancing transparency,bolstered by stand-ardized regulations to ensure safety,quality,and environmental int

241、egrity.Simultaneously,adopting circular economy principles will be crucial.This involves designing batteries for easier recycling and longer life,promoting the reuse of materials,and supporting infrastructure for efficient recycling processes.Researchers and companies can play a pivotal role in adva

242、ncing these technologies,while governments can aid by implementing supportive policies and regulations,ensuring a cohesive and sustainable approach across the industry.To ensure fairness in the battery industrys growth,policies must equitably distribute costs and benefits,especially in mining region

243、s.Significant investment is needed in local communities and sustainable job creation,such as emphasized by the White Houses commitment to a domestic supply chain for critical minerals.149 Moreover,a broader understanding of environmental justice impacts is crucial,beyond tradi-tional greenhouse gas

244、and cost analyses,to assess local socioeconomic and environmental effects fully.150 As batteries scale,they will continue to require lifetimes of work and generous support to develop.With ever-increasing interest in batteries from researchers,companies and policymakers,we expect this to hold true.Th

245、e road ahead for batteries is clear,but we do still need to walk it.rmi.org/33 X-Change:Batteries The Battery Domino Effect Appendix 1:Different ways to model battery sales There are many ways to project battery sales.In this appendix,we look at five ways to forecast battery demand and summarize our

246、 conclusions below for likely battery sales in 2030.Sector-by-sector S-curve outlook.A range of 5.5-8.0 TWh/y based on sector-by-sector S-curves to forecast demand.Aggregate S-curve outlook.A range of 5.0-8.4 TWh/y using a terminal demand of 13 TWh/y and a single S-curve for total battery demand.Gro

247、wth rate outlook.A range of 4-8 TWh/y based on a compound annual growth rate(CAGR)of 25%-35%.Announced capacity outlook.From the IEAs estimated announcements of 7.5 TWh/y to Benchmark Minerals capacity of 9 TWh/y and BNEFs of 12.5 TWh/y.Expert forecasts.From the E Source outlook at 2.7 TWh/y to McKi

248、nsey and Rethink Energys 4.5-5 TWh/y.All outlooks lead to very considerable levels of growth by 2030 versus this years 1 TWh/y market.Expert forecasts are the least optimistic,but conservatism has characterized expert outlooks on the battery indus-try for years.Figure 26:Comparison of battery demand

249、 outlooks,GWh/y,2030 Source:RMI analysis from charts below Figure 26:Comparison of battery demand outlooks,GWh/y,203001,0002,0003,0004,0005,0006,0007,0008,0009,00010,00011,00012,00013,000Sector-by-sectorAggregateCAGRAnnounced capacityExpert forecastsFasterFastrmi.org/34 X-Change:Batteries The Batter

250、y Domino Effect Sector-by-sector S-curve outlook The main battery demand outlook used in this note uses a sector-by-sector S-curve approach.This ap-proach runs as follows:We model each sector of battery demand independently,fitting an S-curve through the battery uptake per sector.We fit the S-curves

251、 based on historical data,and fit both fast S-curves(Gompertz curve)and faster S-curves(logistic curve)to the data,assuming 100%uptake in the long term.As set out in more detail in other reports in the X-change series,we have found that a Gompertz S-curve tends to have good predictive power for tech

252、nologies at an early stage,while a logistics S-curve has better pre-dictive power for later stage technologies.Because battery technologies are still in the early stages of their S-curve,it is hard to confidently fit S-curves to them.To remedy this,we add two years of the short-term forecasts from B

253、NEF until 2025.This enables us to fit S-curves with a greater degree of certainty.The range between the two curves therefore provides a reasonable framing for future sales,as discussed in the previous X-Change note on cars.151 For road transport,we take 100%uptake to be 100%of sales,which we get fro

254、m BNEF.For stationary storage,we assume total long-term demand of 1 TWh/year,in line with the peak demand levels in BNEF.We then fit an S-curve to that total,based on historic data.For electronics,we take the outlook from BNEF via Avicenne.152 This is a very minor part of end demand and is not expec

255、ted to grow as fast as other sectors.Below we provide some details on the outlooks per sector,going deeper on four key sectors of battery demand:cars,light trucks,heavy trucks and stationary storage.Figure 27:Battery uptake by sector,TWh/y Source:RMI analysis 02468025203020352040611Heavy

256、trucksLight trucksCarsStationary storage2/3 WheelersElectronicsBuses02468025203020352040812FastFasterFigure X:Battery demand outlook,TWh/yrmi.org/35 X-Change:Batteries The Battery Domino Effect Cars The passenger car sector is the largest piece of the battery industry.We start by fitting

257、two S-curves(Gompertz and logistic)on historic sales share data from 2015 to 2023,building to an end point of 100%sales share.This gives us a projection of EV market share in sales over time.Then,for each year,we multiply the projected share by BNEFs outlooks for total car sales to get the number of

258、 EV sales.We calculate battery demand assuming a constant battery size of 61 kWh per car,which is the average battery size in BNEFs long-term outlook.Using the methodology set out in X-Change Cars,the results of our analysis suggest that the EV sales share for cars could reach anywhere between 62%(f

259、ast;Gompertz)to 86%(faster;logistic)by 2030.This translates to battery demand of 3.5-5 TWh from the car sector in 2030.Figure 28:Battery-powered car uptake and battery demand Source:BNEF historical data,153 RMI analysis Light trucks In the light commercial vehicles sector,we assume a similar methodo

260、logy as in cars,taking the average battery size of 74 kWh per vehicle from BNEF.We take the BNEF short-term outlook for the period up to 2025 and then model an S-curve after that.The BNEF short-term outlook is mostly based on actual orders,and not projections.While such data is less firm than the hi

261、storical actuals,it provides a robust glimpse into the future that we use to fit more precise S-curves(without the data up to 2025,the range between Gom-pertz and logistic would become almost unusably large).The results of our analysis suggest that the EV sales share for light commercial vehicles co

262、uld reach between 42%(fast;Gompertz)to 62%(faster;logistic)by 2030.This translates to annual battery demand of 0.5-0.7 TWh from the light commercial vehicles sector in 2030.Rapid uptake in this segment is already underway with electric delivery vans and other light duty vehicles already commercially

263、 available on the market.For example,in October 2023,Amazon announced that it had 10,000 electric vans fulfilling deliveries across Europe and the United States,with plans to expand this fleet tenfold to 100,000 by 2030.154 UPS has announced plans to purchase 10,000 electric vehicles,with a possibil

264、ity of increasing this number to at least 66,000 by 2028.155 Other delivery companies are following closely behind.156 0070809020222024202620202030FasterFast202805001,0001,5002,0002,5003,0003,5004,0004,5005,00020222024202620202030FasterFast2028CarsSales share,%Annual new battery demand,GW

265、h/yFigure 28:Battery-powered car uptake and battery demandrmi.org/36 X-Change:Batteries The Battery Domino Effect Figure 29:Battery-powered light truck uptake and battery demand Source:BNEF historical data,157 RMI analysis Heavy trucks As for cars and light commercial vehicles,we assume a similar me

266、thodology for medium and heavy com-mercial vehicles.We use historic market share data and the BNEF short-term outlook up to 2025.We take an average battery size of 431 kWh per truck from BNEFs long-term outlook.The results of our analysis suggest that the EV sales share for medium and heavy commerci

267、al vehicles could reach between 24%(fast;Gompertz)to 46%(faster;logistic)by 2030.This translates to battery demand of 0.6-1.1 TWh from the medium and heavy commercial vehicles sector in 2030.We expect that heavy truck sales will pick up in the late 2020s and will transition from 10%to 80%of sales wi

268、thin 7-15 years.Figure 30:Battery-powered heavy truck uptake and battery demand Source:BNEF historical data,158 RMI analysis 007020222024202620202030FasterFast20280050060070080020222024202620202030FasterFast2028Light trucksSales share,%Annual new battery demand,GWh/yFigure 29:B

269、attery-powered light trucks uptake and battery demandActualsShort-term outlook ActualsShort-term outlook 00222024202620202030FasterFast202802004006008001,0001,20020222024202620202030FasterFast2028Heavy trucksSales share,%Annual new battery demand,GWh/yFigure 30:Battery-powered heavy truck

270、s uptake and battery demandActualsShort-term outlook ActualsShort-term outlook rmi.org/37 X-Change:Batteries The Battery Domino Effect Comparison with expert forecasts We compare our outlook with other forecasts for truck sales.Our analyses suggest that for the light com-mercial vehicles segment,EV

271、sales will make up 42%-62%of new vehicle sales in 2030.This is a bit more optimistic than the BNEF ETS scenario,which estimates light truck EV sales to be approximately 35%of new light trucks sales in 2030.Similarly,for the medium and heavy commercial vehicles segment,we expect that EV sales will ma

272、ke up 24%-46%of new vehicle sales in 2030.This is somewhat higher than the global number in BNEF(18%),and in line with external organizations that focus on US,Europe,and China alone(where the proportion is likely to be higher).Figure 31:Other forecasts for 2030 truck sales,%of sales Source:Interact

273、Analysis(2022),159 BNEF Electric Vehicle Outlook(2023),160 DNV Energy Transition Outlook(2023),161 OPEC World Oil Outlook(2023),162 EIA Annual Energy Outlook(2023),163 NREL(2023),164 McKinsey(2022),165 PwC(2022),166 BCG(2022),167 IEA Global Electric Vehicle Outlook(2023),168 BP Energy Outlook(2023)1

274、69 Other road segments:buses and two-/three-wheelers There are two smaller road segments that we include in our analysis as well:both buses and two-/three-wheelers are undergoing rapid EV uptake.We assume the same approach as for the other road segments to model buses and two-/three wheelers,taking

275、an average battery demand per vehicle of 179 and 3 kWh,respectively,from BNEF.One innovation that has dramatically accelerated the uptake of two-/three-wheelers is battery-swapping stations.In Taiwan,for example,there are more battery-swapping stations than gas stations across major cities,with more

276、 than 1 million swappable batteries in service.170 In Kenya,there are plans to deploy 1 million electric motorcycles supported by 3,000 battery-swapping stations by 2030.171 Similar battery-swap-ping networks are quickly expanding across other countries in Asia and Africa.Electric buses are already

277、in commercial use in many markets and growing rapidly.For example,the Euro-pean e-bus market is growing at 45%per year.172 Ambitious policies are paving the way for fast growth in other markets as well:In New York,all 47,000 school buses in the state will be electric by 2035,173 and Bermuda aims to

278、have an entirely electric fleet of public buses by 2030.174 Figure 31:Other forecasts for 2030 truck sales,%of sales0%10%20%30%40%50%Interact AnalysisNREL(for US)McK-insey*PwC*BCG*BNEF ETSIEA STEPSBP New Mom-entumEIA(for US)BNEF ETSDNVOPECBNEF ETSEIA(for US)External viewEstablishment viewOil viewAll

279、 trucksLight trucksMedium and heavy trucks*US,Europe,China onlyrmi.org/38 X-Change:Batteries The Battery Domino Effect Despite the very large numbers of two-/three-wheelers(their current oil demand reduction benefits that are more than from EV cars for now)175,they are a relatively small piece of 20

280、30 battery demand,at 120-270 GWh.Likewise,we expect battery demand from buses in 2030 to be very small compared to other sectors at around 34-38 GWh.Stationary storage We model the size of the stationary storage market in a different way.The amount of stationary storage needed is a(complex)function

281、of the amount of variable renewable power that is deployed,which requires careful,hourly capacity expansion modelling.To account for that,we take the S-curve renewable outlook from the previous X-Change note on electricity and match it to comparable BNEF and IEA scenarios to find the required batter

282、y demand.The BNEF NZS and IEA NZE scenarios show very comparable renewables uptake to the X-Change electricity outlook and are therefore used to analyze battery demand.We use BNEF for consistency with our historical data,taking the net zero scenario 2050 value(of 954 GWh of sales)as the endpoint and

283、 fitting the two S-curves to the historical data,plus BNEFs 2023-2025 outlook,which is based on near-term grid announce-ments and plans.In both scenarios,batteries are only part of the total flexibility need.For example,other solutions such as demand side flexibility,dispatchable firm capacity,inter

284、connectors,and long duration storage solution fill in two thirds of the flexibility need in the IEA scenario.176 As noted above,there is much uncertainly in this terminal number of 954 GWh of sales,and there are good reasons to believe that it could be considerably higher or lower than this.If we ar

285、e successful at incorpo-rating a large share of the transport fleet to provide system flexibility,then the need for stationary storage would be lower.If batteries get cheap enough,they may supplant other flexibility solutions and demand for storage would be higher.For the purposes of this analysis,t

286、he number is not especially material be-cause demand for batteries from the transport sector is forecast to be nearly 10 times larger than the battery storage market.As with so much in the energy transition,we will build large amounts of batteries and then optimize our use of them using the superior

287、 technologies of the future.Figure 32:Renewables uptake and associated required battery demand Source:BNEF historical data,177 RMI analysis The next dominos:trains,ships,and airplanes The battery domino effect is heading toward three additional sectors:rail,aviation and shipping.Battery-powered tech

288、nology has not yet made any significant entrance in these sectors,hence there is no data to 01,5003,0004,5006,0007,5009,00010,50012,00013,50015,000202020222024202620282030FasterFast00500600700800202020222024202620282030FasterFastStationary StorageSolar and wind total generation,TWhAnnual

289、new battery demand,GWh/yFigure 32:Renewables uptake and associated required battery demandrmi.org/39 X-Change:Batteries The Battery Domino Effect fit an S-curve.Nevertheless,there are hopeful signs that battery technology may start picking up before the end of this decade.Trains Although rail especi

290、ally passenger rail is already widely electrified,batteries can still play a crucial role by replacing diesel to operate trains on segments of rail that do not have overhead catenary lines.Studies show that the range of an average freight train in the United States(150 miles,or 241km)can be achieved

291、 with a 9-14 MWh battery.The same studies also find that battery-electric trains can achieve price parity with diesel-electric trains at current battery prices of$100/kWh.178 Because most diesel trains operate a diesel-electric drive,hybrid and/or battery retrofits are possible.This means that batte

292、ry uptake may grow faster than in other sectors as there is no need to wait for fleet turn-over.The first battery trains are already being rolled out.In May 2023,The Pacific Harbor Line in California un-veiled its new battery electric locomotive,the EMD Joule.179 In November 2023,Wabtec Corp.launche

293、d a 7 MWh battery electric locomotive for Australias Roy Hill,an iron-ore mining company.180 In Europe,Hitachi has announced plans to release in the next two years a full battery powered train that can travel up to 62 miles.181 Ships As a general rule of thumb,the lighter the ship and the shorter th

294、e distance travelled,the easier it will be to electrify.Passenger ferries are a prime candidate for electrification,and there are already several models of electric ferries in operation in Scandinavia,with battery capacities ranging between 1 and 5 MWh.182 Electric ferries also operate in places suc

295、h as New Zealand,California,and New York.183 In passenger ship-ping,electrification is now expanding to cruise ships,which are much bigger and longer distance:in 2023.Norwegian cruise line company Hurtigruten announced plans to deliver a battery-electric cruise ship in 2030.The ship will be powered

296、by a 60 MWh battery and will be 443 meters long.184 Container shipping is also not far behind,with one study finding that over 40%of global containership traffic could be electrified cost-effectively with current technology.185 The first battery electric container ships have already been unveiled:In

297、 2021,Yara debuted a fully electric autonomous container ship pow-ered by eight batteries totaling 6.8 MWh.186 In 2023,China launched its first battery electric container ship with swappable batteries.Each swappable battery unit is housed in 20-foot containers with a capacity of 50 MWh.187 These swa

298、ppable batteries may be a game-changer for battery uptake in the shipping sector,in much the same way they were for the two-/three-wheeler sector.Finally,it is also important to note that with some changes to current shipping practices such as relying more on smaller ships with frequent stops instea

299、d of larger ships with one long route batteries could power a larger portion of shipping even faster than expected.As Steve Henderson,co-founder and CEO of FleetZero(a battery cargo ship startup),points out,the current shipping system is optimized for fossil fuels and can be re-optimized for batteri

300、es.188 According to Rystad,45%of global ton mileage is to transport fossil fuels.189 Lower fossil fuel demand will therefore tend to drive down demand for long haul shipping services in any event.Airplanes Though battery powered planes are unlikely to happen at scale in the next few years,there is p

301、lenty of evidence to suggest it is possible,and many actors are already working to make it a reality.Studies have found that a typical twin-engine narrowbody aircraft with a range of 600 miles would require 800Wh/kg of specific energy from the battery pack.190 Larger models could require around 1,00

302、0 Wh/kg.191 Like vehicles,near-term electric planes are expected to be 60%-70%cheaper to operate than current options,such as helicopters and other short-haul aircraft.192 Beta Technologies recently completed a 16-stop,1,730-mile journey across the US East Coast,using a plane that has already flown

303、386 miles on a single charge.193 Meanwhile,Eviation Aircraft an electric aircraft manufacturer has received nearly 300 orders for the nine-seater plane“Alice”,with goals for commer-cialization by 2027.194 rmi.org/40 X-Change:Batteries The Battery Domino Effect At least 60 companies are currently wor

304、king to research and develop fixed-wing electric aircraft,including major manufacturers such as Boeing and Airbus as well as airlines such as United Airlines and Virgin Atlan-tic.195 Well over 100 electric aviation programs are in development around the world,in a market that has already reached$9 b

305、illion.196 Aggregate S-curve outlook A much simpler approach to sizing battery demand is an aggregate outlook i.e.where we fit an S-curve through total battery demand.An aggregate outlook results in a very similar result to the detailed,sector-by-sector approach.Sizing the total battery market To fi

306、t an aggregate S-curve to battery uptake,one must first define the total addressable market size.If we take data from BNEF on long-term sales and average battery sizes,we get an implied addressable market of nearly 13 TWh/y in 2050.Figure 33:Aggregate peak battery market size Source:BNEF,197 RMI ana

307、lysis We take historical data and near-term expectations to 2025 for total battery demand and combine it with the total potential demand(calculated above)as the end point to fit an S-curve through total battery de-mand.We follow the same approach as with our individual sector outlooks,fitting both a

308、 fast(Gompertz)and faster(logistics)S-curve to the data.Matching sector-by-sector with the aggregate S-curve Comparing the results of the aggregate and sector-by-sector approach reveals a striking alignment in out-come.The aggregate result matches the sector-by-sector outlook to within less than 1 T

309、Wh/y.This finding hints at a deeper insight into S-curve growth patterns.Every global S-curve is made up of more granular S-curves by(sub-)sector or region,each with its own distinctive uptake trajectories.Like a fractal,Figure 33:Top-down battery market sizing1.06.12.32.90.10.30.2Total12.9Total mar

310、ket size,TWh/yRationale0.2 TWh/y as forecasted by BNEF via AvicenneAnnual sales in long termVehicle battery size,kWh/vehicle7 million vehicles p.a.143123 million vehicles p.a.174100 million vehicles p.a.1610.3 million vehicles p.a.1179115 million vehicles p.a.11 TWh/y as forecasted by BNEF NZS31.As

311、forecasted by BNEF 2023Heavy trucksLight trucksCarsBusesStationarystorage2/3 wheelersElectronicsrmi.org/41 X-Change:Batteries The Battery Domino Effect these uptake trajectories are self-similar in shape to the whole:they also follow an S-curve.Such uniformity of uptake for cars across countries was

312、 discussed in X-Change Cars.The consistency of the S-curves shape at various scales provides a valuable insight top-down S-curve analysis can yield results that can be as accurate as those from more granular analysis.This reinforces one of the leading thoughts behind this X-Change series:high-level

313、S-curve trend assessments tend to yield equally valuable insights as more complex,granular models.Figure 34:Sector-by-sector vs.aggregate battery market S-curve fits,TWh/y Source:BNEF,198 RMI analysis FastFaster02468025203020352040Heavy trucksLight trucksCarsBusesStationary storage2/3 Whe

314、elersElectronics02468025203020352040Figure 34:Bottom-up versus top-down battery market S-curve fits,TWh/yAggregate fitAggregatefitSector-by-sector fitsSector-by-sector fitsrmi.org/42 X-Change:Batteries The Battery Domino Effect Growth rate outlook We can also use a simple continued growth

315、 rate perspective to project demand.Over the past 30 years,annual sales growth rates have ranged from 20%to 60%;they have been 33%on average and over 40%in recent years.If we continue at growth rates of 25%-35%pa,total battery demand by 2030 would lie between 4 and 8 TWh.Figure 35:Future battery sal

316、es GWh at different CAGRs,TWh/y Source:BNEF Long-Term Vehicle Outlook 2023(actuals),199 RMI analysis Announced capacity outlook Another approach to sense-check the forecasts for battery demand growth is to look at the amount of battery capacity currently being built.According to Benchmark Minerals,r

317、oughly 9 TWh/y of battery capacity is already planned for 2030,200 while the IEA has an estimate of 7.5 TWh/y.201 As of late November,BNEFs battery manufacturing announcement pipeline already includes announcements for 12.5 TWh/y of battery manufacturing capacity by 2030 including more than 8.5 TWh/

318、y in China alone.202 Given that we have seven years until 2030 and it takes under a year to construct a battery factory,it is fair to say that capacity plans are no constraint to the amount of demand that we foresee.Rather,it is a signal that battery production capacity will be a driver of further a

319、doption.As production capacity comes online,battery manufacturers will create their own demand and accelerate the battery domino effect.Expert outlooks A final approach to forecasting battery demand is to look at expert outlooks.Expert forecasts range be-tween 3 and 5 TWh/y of battery sales in 2030,

320、or up to 7 TWh/y when including scenarios that do not fore-cast but goal-seek to reach net zero.Those at the bottom of the range assume that change will significantly 05001,0001,5002,0002,5003,0003,5004,0004,5005,0005,5006,0006,5007,0007,5008,000202020224202520262027202820292030Faster(35%

321、CAGR)Fast(25%CAGR)Figure 35:Future battery sales GWh at different CAGRs,%Actualsrmi.org/43 X-Change:Batteries The Battery Domino Effect decelerate from the 40%annual growth we have seen in recent years.The lowest of the range assumes a 20%CAGR of sales to 2030,and the top end of the range implies a

322、33%CAGR.As discussed in section 1,expert forecasts have historically underestimated battery uptake.We foresee a similar trend in these forecasts:the sector-by-sector battery demand forecast we created for 2030 is 5.5-8 TWh/y,which is 40%-100%more than the expert average of about 4 TWh/y.Figure 36:Ex

323、pert forecasts of 2030 battery sales,TWh/y Source:Benchmark Minerals(2023),203 IEA Global Electric Vehicle Outlook(2023),204 BNEF Long-Term Electric Vehicle Outlook(2023),205 Rethink Energy(2023),206 McKinsey Battery 2030(2023),207 Goldman Sachs(2023),208 Rystad Battery Market Outlook(2023),209 BP E

324、nergy Outlook(2023),210 E Source(2023)211 Appendix 2:Barriers to change In this section we consider two key barriers to change charging infrastructure and mineral demand in more detail to see if they can slow the battery domino effect.Charging infrastructure As more EVs enter traffic,the need for ch

325、arging points rises.As laid out in the previous entry of this series,chargers are rising rapidly,and EV range anxiety due to a lack of charging stations will soon disappear in most of the world.The expansion of charging stations is notable in several key areas:Rapid expansion of public charging infr

326、astructure:The global increase in publicly accessible charging points is exponential.Last year alone,the number grew by about 55%,a testament to the accelerated de-ployment led by regions like China and Europe.Fast chargers grew by the same rate in the EU.212 Government initiatives and funding:Natio

327、nal governments are playing a pivotal role in this expansion,as noted in section 2 of this note.According to the IEA,there are over 140 charging infrastructure-promot-ing policies across 35 countries,from North America to Europe,Asia and Oceania.213 Figure 22:Expert forecasts for 2030 total lithium-

328、ion battery sales,TWh/y0.00.51.01.52.02.53.03.54.04.55.05.56.06.57.0IEA NZEBNEF NZSRethink EnergyMcKinsey IEA APS Goldman SachsBNEF ETSIEA STEPSBenchmark present pathwayRystad base case demandBP AcceleratedE SourceBenchmark net zero pathwayExternal viewEstablishment viewNet-zero goal-seek viewOil vi

329、ewrmi.org/44 X-Change:Batteries The Battery Domino Effect Addressing the needs of heavy-duty vehicles:Charging a 500+kWh battery can take a long time and requires a large grid connection.Fast charging networks are being rolled out across geographies to meet demand.Analysis by T&E on the EU market ha

330、s shown that there is ample room to accommodate rapid growth of heavy trucks.214 At the same time,innovations such as battery swapping,and electric road charg-ing systems are emerging to cater to the unique charging needs of heavy-duty vehicles.China is leading in this area,particularly in the devel

331、opment and implementation of battery swapping technologies for trucks.215 Tackling power grid expansion:The expansion and upgrade of the power grid to connect charging points efficiently is being actively tackled.In the EU,initiatives like the Green Deal and the Alternative Fuels Infra-structure Reg

332、ulation focus on grid upgrades and the integration of renewable energy sources to support EV charging infrastructure.In the United States and China,substantial government investments are being made under the Infrastructure Investment and Jobs Act and national environmental strategies,respectively,to

333、 modernize the electric grid and integrate renewable energy,ensuring it can support the growing demand from EV charging stations.Private sector involvement:Companies are increasingly leaning in as well.Smart charging is already a$35 billion industry today and is expected to grow by 30%in the coming decade.216 Car OEMs are also getting involved:for example,Teslas announcement to open a portion of i