Appendix F
Estimation of Lithium-Ion Battery Pack Costs

BACKGROUND

Battery pack costs will be the prime determinant of when PHEVs will become competitive, but projections of these costs are highly uncertain. This appendix provides additional detail on the committee’s methodology for developing the cost estimates presented in Chapter 2 of the report.1 The committee reviewed the literature and heard presentations given at NRC meetings by experts. It then discussed its preliminary conclusions about current and anticipated battery pack costs with industry and government experts. Its estimates were subject to the National Academies’ report review process during which additional experts commented on the committee’s assumptions and analysis. The committee’s estimates in its final report include a broad range of costs, reflecting the information the committee reviewed.

Several factors must be known before various battery pack cost estimates can be compared. They include whether the estimated cost per kWh applies to nameplate or usable energy, and, if the former, the assumed state of charge (SOC) range to know how much energy is available, and whether the cost is based on the beginning or end of life capacity (Li-ion batteries deteriorate at about 2 percent per year). To calculate the actual battery pack costs for a vehicle with a specified electric driving distance, the Wh/mile (propulsion energy) required to propel the vehicle must also be estimated, and the cooling (e.g., liquid or air) that is required must be specified. Due to the early state of PHEV development and lack of road experience, there is considerable variability in these cost estimates.

The costs reported in Chapter 2 of the report were based on usable energy, that is, the fraction of the total energy stored in the battery that can be withdrawn for propulsion without risking damage to the battery or causing safety issues. That is the most useful measure when determining the performance of the vehicle and is the measure used by the Department of Energy and the U.S. Advanced Battery Consortium in their published goals for battery storage. However, when comparing costs of different batteries, assuming equal propulsion energy (Wh/mile), nameplate (or nominal) capacity better describes the battery pack costs actually put into the PHEV. The difference between these two measures for a specific battery pack (essentially the SOC selected by the vehicle manufacturer) has caused considerable confusion. The discussion below of various sources of estimates compares the nameplate cost unless otherwise specified.

Battery lifetime and safety are also very significant PHEV development issues for the industry. The current life expectancy of Li-ion batteries for computers and power tools is around 3 to 4 years,2 but 10 years or more will be needed for PHEVs to be competitive. R&D programs are reducing battery costs and improving battery durability and safety, but these goals must be met simultaneously and may interfere with each other. The committee concluded that durability and safety goals are more likely to be met than cost goals. Therefore the focus of this discussion is on costs after durability and safety problems have been solved.

CURRENT COSTS

The committee reviewed a variety of sources to establish the most probable and optimistic costs for the current generation of battery packs. The review, discussed below, indicated a range of $500 to $1500/kWh nameplate. The range is so wide in part because different technologies at different stages of development are reported. Based on this range, the committee selected $875/kWh as the most probable value and $625/kWh as an optimistic value for batteries that have already been ordered to be used in the first generation of

1

This appendix was added to the report after the release of the prepublication version to clarify the basis for the committee’s cost estimates. The estimates themselves are unchanged.

2

The Hymotion kit made by A123 Systems to convert the Toyota Prius to a PHEV comes with a 3 ­year warranty against defects. The all ­electric range is not specifically guaranteed. See http://www.a123systems.com/hymotion/pop_ups/warranty.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 54
Appendix F Estimation of Lithium-Ion Battery Pack Costs BACKGROUND performance of the vehicle and is the measure used by the Department of Energy and the U.S. Advanced Battery Con- Battery pack costs will be the prime determinant of when sortium in their published goals for battery storage. However, PHEVs will become competitive, but projections of these when comparing costs of different batteries, assuming equal costs are highly uncertain. This appendix provides additional propulsion energy (Wh/mile), nameplate (or nominal) capac- detail on the committee’s methodology for developing the ity better describes the battery pack costs actually put into cost estimates presented in Chapter 2 of the report.1 The com- the PHEV. The difference between these two measures for mittee reviewed the literature and heard presentations given a specific battery pack (essentially the SOC selected by the at NRC meetings by experts. It then discussed its preliminary vehicle manufacturer) has caused considerable confusion. conclusions about current and anticipated battery pack costs The discussion below of various sources of estimates com- with industry and government experts. Its estimates were pares the nameplate cost unless otherwise specified. subject to the National Academies’ report review process Battery lifetime and safety are also very significant PHEV during which additional experts commented on the commit- development issues for the industry. The current life expec- tee’s assumptions and analysis. The committee’s estimates tancy of Li-ion batteries for computers and power tools is in its final report include a broad range of costs, reflecting around 3 to 4 years,2 but 10 years or more will be needed the information the committee reviewed. for PHEVs to be competitive. R&D programs are reducing Several factors must be known before various battery battery costs and improving battery durability and safety, but pack cost estimates can be compared. They include whether these goals must be met simultaneously and may interfere the estimated cost per kWh applies to nameplate or usable with each other. The committee concluded that durability and energy, and, if the former, the assumed state of charge (SOC) safety goals are more likely to be met than cost goals. There- range to know how much energy is available, and whether fore the focus of this discussion is on costs after durability the cost is based on the beginning or end of life capacity and safety problems have been solved. (Li-ion batteries deteriorate at about 2 percent per year). To calculate the actual battery pack costs for a vehicle with a CURRENT COSTS specified electric driving distance, the Wh/mile (propulsion energy) required to propel the vehicle must also be estimated, The committee reviewed a variety of sources to establish and the cooling (e.g., liquid or air) that is required must be the most probable and optimistic costs for the current genera- specified. Due to the early state of PHEV development and tion of battery packs. The review, discussed below, indicated lack of road experience, there is considerable variability in a range of $500 to $1500/kWh nameplate. The range is these cost estimates. so wide in part because different technologies at different The costs reported in Chapter 2 of the report were based stages of development are reported. Based on this range, the on usable energy, that is, the fraction of the total energy committee selected $875/kWh as the most probable value stored in the battery that can be withdrawn for propulsion and $625/kWh as an optimistic value for batteries that have without risking damage to the battery or causing safety already been ordered to be used in the first generation of issues. That is the most useful measure when determining the 2The Hymotion kit made by A123 Systems to convert the Toyota Prius 1This appendix was added to the report after the release of the pre­ to a PHEV comes with a 3­year warranty against defects. The all­electric publication version to clarify the basis for the committee’s cost estimates. range is not specifically guaranteed. See http://www.a123systems.com/ The estimates themselves are unchanged. hymotion/pop_ups/warranty. 

OCR for page 54
 APPENDIX F PHEV-40s, and $825/kWh and $625/kWh for PHEV-10s. in the media. The additional systems, materials, and labor to Literature results are as follows: assemble a battery pack are substantial. The committee concluded, based on research and discus- • The NAS-NAE-NRC report America’s Energy Future sions, that the cost of assembling the pack is approximately concludes that automotive-grade Li-ion battery pack costs the same as the cost of the cells, corresponding to the total today are between $500/kWh and $1000/kWh nameplate of $14,000 for the PHEV-40. The committee also estimated (NAS-NAE-NRC, 2009). a range of costs, recognizing the uncertainty involved, and • DOE estimates of current costs are $1,000+/kWh concluded that under more optimistic assumptions the cost usable energy (Howell, 2009).3 DOE goals are for perfor- could be $10,000. In comparison, DOE estimates that a mance at the end of life. Li-ion batteries deteriorate over PHEV-40 would require 11.6 kWh usable energy in a pack time, typically at about 2 percent per year. Assuming a DOE that would cost over $11,600, consistent with the estimating start of life SOC of 70 percent, the committee estimates accuracy of this report. DOE’s nameplate cost at start of life to be $560+/kWh. • A recent McKinsey report concludes that battery PROJECTED FUTURE COSTS pack costs range from $700/kWh to $1,500/kWh nameplate (Hensley et al., 2009). The committee estimated future costs of Li-ion batteries • A 2 009 paper (Shiau et al.) from researchers at based on the technology status and cost projections in the Carnegie Mellon University uses $1000/kWh nameplate. literature. Based on this analysis, the committee judged that • Pesaran et al. (2007) estimated the cost of advanced battery pack costs are likely to decline by about 35 percent Li-Ion battery costs as ranging from $800/kWh to $1,000/kWh by 2020 and 45 percent by 2030, as shown in Tables 2.2 and nameplate or higher. 2.3. This yields a nameplate 2030 PHEV 40 battery pack • The Zero Emissions Vehicle Report projected a “cur- cost of about $500/kWh ($1000/kWh usable) or, under more rent” cost of about $500/kWh nameplate in 2006 (Kalhammer optimistic assumptions, about $360/kWh. The committee did et al., 2007). not attempt to estimate the future costs if a major technology breakthrough occurs, such as the development of a durable, The following two reports were released after the com- safe Li- air battery. mittee completed its analysis, but they are included here for The literature contains a wide range of projected future completeness. Li-ion battery and battery pack costs (all costs are nameplate unless otherwise noted): • The Electrification Coalition’s 2009 report Electrifica- tion Roadmap (available at www.electrificationcoalition.org) • The DOE goal is for a very rapid cost reduction from the estimated $1,000+/kWh current cost to $500/kWh in states that the current cost is $600/kWh nameplate • The Boston Consulting Group’s 2009 report Batteries 2012 to $300/kWh (all costs based on available energy base) in 2014.4 Assuming 70 percent SOC and 20 percent for Electric Cars (available at www.bcg.com/documents/ file36615.pdf) says it is $1000-1200/kWh deterioration conversion factors, DOE’s goals correspond to $280/kWh in 2012 and $168/kWh in 2014 on a nameplate The committee expects that these early PHEVs will capacity basis. Meeting these goals would result in a $1700 employ a conservatively low SOC, about 50 percent, to cost for a 3.4 kWh battery pack in a PHEV-10, and $3,400 ensure battery durability and safety. With experience and for an 11.6 kWh pack in a PHEV-40. Note that these are improved battery and control technology the SOC may be goals, not projections. Meeting these goals could result in increased to 70 or even 80 percent, but that is speculation PHEVs being competitive in the marketplace much more until several years of real-life operating experience indicate rapidly compared with HEVs and conventional vehicles, as whether battery durability would be jeopardized. discussed in Appendix C. At 50 percent SOC, the current cost for usable (or avail- • The U.S. Advanced Battery Consortium (2009) has the able) energy for a PHEV-40 comes to $1750/kWh (prob- same goals as DOE. able) as shown in Chapter 2, and the nameplate cost is • Nelson et al. (2009) projected pack manufacturing cost $875/kWh. Based on the report’s assumed propulsion energy of about $350/kWh at 100,000 unit volume for a PHEV-10 of 200 Wh/mile, a 16 kWh battery pack (8 kWh usable) and $200/kWh for a PHEV-40. such as will be used in the Chevrolet Volt costs $14,000. • The McKinsey report projected that costs will decrease While neither GM nor LG Chem, the battery supplier, has at 6 percent to 8 percent per year, and, with aggressive announced the costs, $7000 for the cells has been reported cost reduction, could reach $420/kWh nameplate by 2015 (Hensley et al., 2009). 3See also S. Satyapal and P. Davis, presentation to the Committee on 4T.Q. Duong, Update on electrochemical energy storage R&D, presenta­ Review of FreedonCAR & Fuel Partnership, Phase 3, Washington, D.C., tion to the committee, Washington D.C., June 2009. 2009.

OCR for page 54
 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—PHEVS FIGURE F.1 Historical cost reduction experience for NiMH battery packs and for Li-ion battery packs. Recent experience does not sug - gest rapid further cost reductions. SOURCE: T.Q. Duong, Update on electrochemical energy storage R&D, presentation to the committee, Washington D.C., June 2009. Figure F-2 R01653 uneditable bitmapped image • Anderman predicts that the cost of Li-ion batteries will one-column otherwisew very different in either chemistry or manufac- size belo not remain at around $600/kWh even with increased production turing processes. Thus the potential for large cost reductions (Anderman, 2007). from technology improvements is limited. Furthermore, • Kalhammer et al. (2007) project costs from $350/kWh materials represent more than half the cell cost (Nelson et al., to $400/kWh (nameplate) for PHEV-40 battery packs at 2009), and these costs are unlikely to drop dramatically. volume production (100,000 to 200,000 units per year). Economies of scale are often cited as a factor that can Costs for PHEV-10 battery packs are projected to be $560 drive down costs, but hundreds of millions to billions of Li- to 860/kWh for production at 100,000 to 625,000 units per ion cells already are being produced in optimized factories. year. Building more factories is unlikely to have a great impact on costs. The cost of the battery pack enclosure that holds the The future cost estimates in this report are higher than cells, the electronics required to monitor and control each most, but not all, other projections, especially the DOE cell to prevent over-charging and run-away, and the tempera- goals. The committee concluded that reductions greater than ture control system to manage battery pack temperature are 50 percent in battery costs are unlikely over the next two a major portion of the total battery pack cost. These com- decades without a major technology breakthrough, because ponents are unlikely to undergo larger cost reductions than meeting battery durability and safety goals could slow cost the cells, and so the committee maintained the same ratio of reductions. For example, raising the SOC range would be a twice the cell cost for the pack. significant cost saver but could compromise durability if that Li-ion batteries have undergone large cost reductions put too much stress on the cells. over the last 10 years, but the costs seem to be leveling out. Cost reductions are likely to come mainly from improve- Costs of NiMH battery pack for HEVs have declined only ments in technology, with lesser contributions from manu- modestly in recent years, as shown in Figure F.1, suggesting facturing improvements, improved yield, and manufacturing that further major cost reductions are not very likely without scale.5 Technology will continue to improve, but it is already technology breakthroughs, which this study did not try to well developed for current Li-ion cells. Cells for automotive project. applications will be bigger than current Li-ion cells but are The committee considered all these projections and other information to come up with its estimates for 2030 future 5According to one estimate, cell costs could drop by more than 50 percent costs of about $500/kWh or, under more optimistic assump- by 2015, with almost all of that decrease coming from technology and pro - tions, of about $360/kWh. cess improvements (D. Vieau, A123 Systems, presentation to the committee, Washington, D.C., May 2009).

OCR for page 54
 APPENDIX F VALIDATION OF COSTS Based on the above and other related information, dis- cussions with industry and government experts, and its own judgments, the committee developed Tables 2.2 through 2.6 with what it considered the most probable set of numbers for the key battery pack performance parameters and bat- tery pack cost. These were discussed with the companies acknowledged in the front matter of this report to ensure that battery costs and performance were realistic.6 Some adjustments were made based on these discussions, and the estimates presented above were finalized. The committee’s report was subjected to the National Academies’ report review process in which another set of experts critiqued the report and the committee’s assumptions and analysis. 6Toyota Motor Corporation, General Motors, Ford Motor Corporation, A123 Systems, Compact Power Inc. (LG Chem), Delphi Corporation, DENSO International America, Inc., U.S. Department of Energy.

OCR for page 54