ARTICLE pubs.acs.org/est
Life Cycle Environmental Assessment of Lithium-Ion and Nickel Metal Hydride Batteries for Plug-In Hybrid and Battery Electric Vehicles Guillaume Majeau-Bettez,*,† Troy R. Hawkins,† and Anders Hammer Strømman† †
Industrial Ecology Program, Norwegian University of Science and Technology (NTNU), Høgskoleringen 5, NO-7491 Trondheim, Norway
bS Supporting Information ABSTRACT: This study presents the life cycle assessment (LCA) of three batteries for plug-in hybrid and full performance battery electric vehicles. A transparent life cycle inventory (LCI) was compiled in a component-wise manner for nickel metal hydride (NiMH), nickel cobalt manganese lithium-ion (NCM), and iron phosphate lithium-ion (LFP) batteries. The battery systems were investigated with a functional unit based on energy storage, and environmental impacts were analyzed using midpoint indicators. On a per-storage basis, the NiMH technology was found to have the highest environmental impact, followed by NCM and then LFP, for all categories considered except ozone depletion potential. We found higher life cycle global warming emissions than have been previously reported. Detailed contribution and structural path analyses allowed for the identification of the different processes and value-chains most directly responsible for these emissions. This article contributes a public and detailed inventory, which can be easily be adapted to any powertrain, along with readily usable environmental performance assessments.
1. INTRODUCTION 1.1. Context and Purpose. As part of the efforts to control anthropogenic greenhouse gas emissions, the replacement of internal combustion engine vehicles by electric vehicles (EV) stirs much scientific, political, and public interest. Depending on the electricity mix used, these vehicles harbor the potential for substantial emission reductions.1,2 Relative to the number of publications on the use phase and fuel life cycles of EVs, few studies focus on the production of traction batteries.3 Most hybrid electric vehicles (HEV) presently rely on nickel metal hydride (NiMH) batteries.4 In the near future however, lithium-ion batteries (Li-ion) are expected to dominate the market, especially with the rise of plug-in hybrid (PHEV) and purely electrically driven battery electric vehicles (BEV).5,6 The present study investigates a NiMH and two Li-ion batteries and highlights their environmental tradeoffs. Our results can contribute to the formation of transportation policies based on comprehensive environmental understanding and empower car and battery companies to focus research on the most environmentally intensive processes and value chains. 1.2. Present State of Research. In the late 1990s, Singh et al.7 and Rantik8 provided early public inventories of NiMH batteries for EV, with both inventories built in a material-share manner. Since then, the traction battery industry has developed significantly. As part of a cost analysis, Gaines and Cuenca9 published a thorough component-wise inventory of Li-ion batteries, which has the potential to be actualized and complemented by environmental considerations. Schexnayder et al.10 also presented componentwise inventories and LCAs of high power-density traction batteries. r 2011 American Chemical Society
These HEV-type batteries have mass compositions differing significantly from that of PHEV or BEV devices. The SUBAT project also compiled a comparative LCA of both Li-ion and NiMH batteries.11 13 Results were presented with a highly aggregated single-score indicator, and as their LCI is not public, it limits their usability and physical significance. For the same reasons, we were not able to adapt the comparative LCA of portable batteries by Parsons.14 LCAs and material-share inventories for NiMH and Liion are included as part of the GREET model.15 The developers have qualified the battery inventories as incomplete and under development,1 and as such their LCA represents a kind of a minimum impact estimate.3 The metal requirements of EV bat teries were addressed by Rade and Andersson.16 Material flows and energy requirements associated with NiMH and Li-ion batteries were investigated by Rydh and co-workers.17 20 Samaras and Meisterling1 notably incorporate their findings as part of a broader analysis of greenhouse gas emissions by PHEV. Recently, a detailed LCA of a manganese oxide Li-ion battery was published by Notter et al.2 In addition to providing a transparent inventory, their study is characterized by a distance-based functional unit (FU), and a strong focus on global warming and fossil fuel depletion. Notter et al.2 present energy requirements significantly smaller than those published by Rydh and Sanden.17 In our work, we follow the lead of the study by Received: October 27, 2010 Accepted: April 11, 2011 Revised: February 16, 2011 Published: April 20, 2011 4548
dx.doi.org/10.1021/es103607c | Environ. Sci. Technol. 2011, 45, 4548–4554
Environmental Science & Technology Samaras and Meisterling1 and build upon the energy estimates of Rydh and Sanden.17 For better comparability and complementarity, we thus provide an inventory fully independent from that of Notter et al.2 With this work, we complement the scientific literature on traction batteries for PHEV and BEV by providing a public process-wise inventory and a transparent LCA for three relevant battery chemistries. Thirteen impact categories are covered and broken down in terms of the contributions of the different life cycle stages and components. The system is further explored with the Taylor series expansion and structural path analysis (SPA) techniques, as described by Waugh21 and Peters and Hertwich.22 We thus add a significant level of detail and specificity to the discussion on traction batteries, especially with regards to the diversity of impacts covered and the analysis of their underlying causes.
2. METHODOLOGY 2.1. Functional Unit. As pointed out by Matheys et al.,12
batteries have differing use phase efficiencies that cannot be captured by mass-based functional units. In this respect, batteries are analogous to leaking tanks; they lose part of their content in the process of storing and delivering it. We chose to express our results for a given amount of energy (50 MJ) accumulated by the battery and then delivered to the powertrain. This charge discharge approach has the advantage of being intuitive, representative of the purpose of the device, free of any assumption concerning the powertrain, and inclusive of the majority of battery characteristics, such as specific energy capacity, depthof-discharge, cycle-life expectancy, and charge discharge energy efficiency. We decided not to define our functional unit in terms of driving distance or driving range, as such a functional unit would have been dependent on powertrain and driving cycle assumptions. As a result, any electricity consumed by the vehicle powertrain, including the energy requirements induced on the powertrain to transport the mass of the battery,23 were deemed beyond our system definition. Thus, the generality of our inventory is preserved, allowing it to be adapted to specific systems of interest. For greater usability and smoother comparison with previous studies, we also present the cradle-to-gate portion of our LCA in terms of battery mass (kg) and nominal energy capacity (Wh, at 1 C-rate). Also, for the sake of giving an idea of the order of magnitude of our functional unit, we estimated the distance that could be driven with 50 MJ. Thus, for comparison with other publications reporting environmental impacts per km driven, our main functional unit would roughly correspond to driving 100 km with the PHEV efficiency set as target by the Electric Power Research Institute,4 0.53 MJ 3 km 1. 2.2. Battery Characterization. Three popular battery chemistries are covered in this study. For NIMH batteries, the AB5 rareearth metal-based chemistry dominates the market at present.24,25 Conversely, several different Li-ion chemistries are being developed in parallel.4,26 As they are among the most mature technologies and have contrasting chemical compositions,4,27 iron phosphate (LFP) and nickel cobalt manganese oxide (NCM) Li-ion batteries were selected. While HEV batteries are used mostly as assistance during accelerations and thus require greater power densities, PHEVs and BEVs use batteries as their primary energy source and require optimal energy densities. A fundamental trade-off between energy density and power density dominates battery engineering.27 A
ARTICLE
Figure 1. Flow diagram of the battery system, defined by the functional unit of 50 MJ stored and delivered to the powertrain. BMS: Battery Management System.
battery with maximum active material content will hold the most energy but will not be able to deliver it as fast as a battery with a higher proportion of “supporting material” (current-collector, separator, etc.). The batteries in this study correspond to PHEV or BEV batteries. In a combined top-down and bottom-up approach, we reconciled electrochemical studies, dismantling reports, and industry figures. Based on the state of the art material properties of the different electrodes, we combined and retrofitted the inventories of Singh et al.,7 Gaines and Cuenca,9 Schexnayder et al.,10 and Ying et al.25 so that they should yield credible PHEV and BEV battery characteristics, in agreement with performances reported by the industry.4,27 29 The simple model underlying our inventory ensures that the charge capacities of the batteries are consistent with the amounts and specific capacities of the active materials and furthermore that cathodes and anodes have equal reversible charge capacities. With this approach, we inventoried NiMH,4,27 NCM,4,27,29 and LFP28 battery packs in such a manner as to obtain credible total energy densities of 55, 112, and 88 Wh 3 kg 1, respectively. We assumed an energy efficiency of 80% for NiMH30 and 90% for Li-ion13,23,31 33 batteries. With 80% depth-of-discharge, NiMH4,27,34 and NCM 27,29 are expected to reach a lifetime of 3000 cycles, while 6000 cycles are predicted for LFP.27,34,35 By simple multiplication of the cyclelife expectancy, the depth of discharge, and the nominal energy capacity of one kilogram of each battery type, we obtain its total lifetime specific energy capacity: 478, 969, and 1520 MJ 3 kg 1 for NiMH, NCM, and LFP, respectively. It should be noted that these “expected” battery performance assumptions are both general and prospective; real-world PHEV and BEV performance evaluations are needed in order to refine our results. The Supporting Information (SI, section 2) further details the simple model underlying our inventory, the complete component-wise mass breakdown of the batteries, and their performance characteristics. 2.3. System Description. Figure 1 presents our system definition, built around the delivery of 50 MJ to the EV powertrain. During charge and discharge, the battery consumes a fraction of the input energy, emitting heat due to internal resistance. Each functional unit of energy delivered to the powertrain represents a fraction of the total lifetime output of the battery. This defines the fraction of the initial production that is assigned to each functional unit. 4549
dx.doi.org/10.1021/es103607c |Environ. Sci. Technol. 2011, 45, 4548–4554
Environmental Science & Technology
ARTICLE
Figure 2. Life cycle environmental impacts of storing 50 MJ of electrical energy in NiMH, NCM, and LFP traction batteries and delivering it to a PHEV or BEV powertrain. Total impacts are expressed quantitatively (left) and also normalized against the worst performance (graphically, right), with impacts broken down between the production of the battery and the electricity consumption during the use phase. Impact Categories: global warming (GWP), fossil depletion (FDP), freshwater ecotoxicity (FETP), freshwater eutrophication (FEP), human toxicity (HTP), marine ecotoxicity (METP), marine eutrophication (MEP), metal depletion (MDP), ozone depletion (ODP), particulate matter formation (PMFP), terrestrial acidification (TAP), terrestrial ecotoxicity (ETEP) potentials, with the suffixes “eq”, “inf”, and “100” referring to “equivalent”, infinity, and 100 years, respectively. Abbreviations: 1,4-DCB refers to 1,4-dichlorobenzene, CFC-11 to trichlorofluoromethane, PM10 to “particulate matter less than 10 μm in diameter”, NMVOC to “non methane volatile organic carbon”.
Our system definition covers the whole production chain and the use phase. As the recycling of EV batteries is not yet widely implemented, we decided to keep our inventory clean of any endof-life scenario. As pointed out by Notter et al.2 this corresponds to a worst case scenario, in which no beneficial secondary metal production is allocated to the battery life cycle. For both Li-ion batteries, the negative electrode is composed primarily of graphite. The active materials of the positive electrodes for NiMH, NCM, and LFP are Ni(OH)2, LiNi0.4Co0.2Mn0.4O2, and LiFePO4, respectively. We assumed basic aqueous precipitation for the first two chemicals, followed by calcination in the case of NCM. For LiFePO4, hydrothermal synthesis was assumed among the many different synthesis paths available. This constitutes an important modeling choice which may significantly impact energy requirements. All three materials are produced from metal salts, which are byproduct or intermediates of hydro-metallurgy.36 38 It is an important characteristic of our inventory that it avoids using metal proxies in lieu of metal salts; none of these ionic metal compounds ever undergo the energy intensive step of being reduced to their neutral form. This contrasts with the production of the NiMH negative electrode, which is based on an alloy of nickel and lanthanum (or lanthanum-rich mischmetal). The complete system description with the synthesis of the active materials is illustrated in the SI, section 1. 2.4. Inventory and Analysis. Our inventory is linked to Ecoinvent 2.2 as a background system.39 Average European conditions were generally assumed, due to greater data availability. Infrastructure and transport requirements, though often generic, were always included. The inventory of the components and materials may be found in the SI (sections 4 7). In order to provide an energy inventory as complete as possible, the total manufacture energy requirements reported by Rydh and Sanden17 were taken as a starting point. From this total, the process-specific manufacturing energy requirements were separated out (see SI,
section 8). Midpoint indicator characterization factors were taken from the ReCiPe method.40 Contribution analyses and Taylor series expansions allowed for a breakdown of the environmental impacts along the different processes and tiers of the value chain, respectively. SPA was performed to determine the emission of specific production chains.
3. RESULTS AND DISCUSSION 3.1. Overall Environmental Impacts. The total life cycle environmental impacts of the three batteries are reported in Figure 2. Except for ozone depletion potential, the NiMH battery performs significantly worse than the two Li-ion batteries for all impact categories. This difference may be rationalized by the greater use phase efficiency of Li-ion relative to NiMH, and the fact that each kilogram of Li-ion battery is expected to store between 2 to 3 times more energy in the course of its lifetime. Moreover, the NCM and LFP batteries contain at least an order of magnitude less nickel and virtually no rare earth metals. Among Li-ion batteries, our analysis points to overall environmental benefits of LFP relative to NCM, which can be explained by a greater lifetime expectancy and the use of less environmentally intensive materials. As an approximate indicator, if we assume a vehicle powertrain efficiency of 0.5 MJ 3 km 1, our results indicate an overall global warming impact of 35 gCO2-eq 3 km 1 for NiMH, 19 gCO2-eq 3 km 1 for NCM, and 14 gCO2-eq 3 km 1 for LFP. The environmental impacts attributable to the electricity consumed by the battery during the use phase (darker nonhatched fraction of the bars, Figure 2) represents more than 40% of global warming potential and fossil depletion impacts and between 27% 45% of the eutrophication impacts. As this is dependent on the use phase electricity mix, in this case average European electricity mix, this portion of our results should be 4550
dx.doi.org/10.1021/es103607c |Environ. Sci. Technol. 2011, 45, 4548–4554
Environmental Science & Technology
ARTICLE
Figure 3. Cradle-to-gate life cycle impacts of battery production, reported quantitatively for different functional units, and broken down among key processes. Impact categories and abbreviations as in Figure 2.
adapted to local conditions. As the iron phosphate battery is expected to enjoy a longer lifetime than NCM, the use phase electricity consumption has a greater relative impact for the former than for the latter. The environmental impacts due strictly to the battery production required to fulfill our functional unit (paler hatched fraction of the bars, Figure 2) are reported numerically in the first column of Figure 3. This figure also presents total impacts for alternative functional units, along with a component-wise allocation of the environmental impacts of battery production. 3.2. Breakdown and Analysis. The NiMH battery is characterized by its lower lifetime energy capacity relative to Li-ion, and hence the need for more battery mass per unit of energy storage, and also its high mass concentration of nickel. SPA indicates that
mining and metallurgy activities required for the production of the nickel in the electrodes and the current collectors are responsible for more than ca. 70% of the toxicity and ecotoxicity impacts and more than 80% of particulate matter formation, terrestrial acidification, and metal depletion potential impacts. A certain number of characteristics differentiate the environmental breakdown of Li-ion battery production from that of NiMH. First, Li-ion batteries require active control from a battery management system (BMS), the production of which causes ca. 10% 30% of the Li-ion production impacts for most impact categories. For all impact categories, the graphite-based negative electrode of Li-ion cells is significantly less environmentally intensive than the LaNi5-based equivalent in NiMH. The production of the positive electrode materials is responsible for more 4551
dx.doi.org/10.1021/es103607c |Environ. Sci. Technol. 2011, 45, 4548–4554
Environmental Science & Technology than 35% of the Li-ion GWP impacts, while the cathode only accounts for 12% of the NiMH battery production impacts. The electrode substrates of Li-ion, made of copper and aluminum foil, contribute proportionally less to the environmental burden of their batteries than the nickel foam substrates do for NiMH, though the production of copper has non-negligible impacts. In fact, we find by SPA that the copper used in the BMS and the electrode substrates is responsible for 30% 50% of life cycle toxicity and ecotoxicity impacts of LFP and NCM batteries. Figure 3 also illustrates differences between the two Li-ion chemistries. The production of the iron phosphate active material has a lower impact across all categories, especially for toxicity and abiotic depletion impacts, as explained by the abundance, the relatively low toxicity, and the generally nonsulfidic nature of iron ore and metallurgy.41 For all three batteries, the manufacture energy requirements are a major cause of GWP. SPA revealed the environmental significance of using polytetrafluoroethylene as dispersant/binder in the electrode paste. Its production is responsible for more than 97% of the ozone depletion potential of all three batteries, along with 14 15% of the GWP of the two Li-ion batteries, mostly due to the halogenated methane emissions of this value chain. The final shipping and the productions of the cell containers, module packaging, separator material, and electrolyte contribute relatively little to the environmental damage, with collectively less than 10% of any impact category. It should be noted that the ReCiPe method has no abiotic depletion characterization factor for lithium or rare earth metals, including lanthanum. This environmental impact is thus expected to be significantly underestimated in our results for all three batteries. Our results demonstrate the importance of the choice of functional unit for batteries. While production of NiMH causes the least GWP impact per kilogram, its lower energy density makes it score worst both relative to its nominal energy capacity and our storage-based functional unit. Similarly, the GWP impacts of LFP and NCM production are roughly equal for a given mass or nominal energy capacity, but the greater life expectancy of LFP confers a net environmental advantage to this chemistry for a per-energy-delivered functional unit. 3.3. Result Robustness and Benchmarking. As previous studies either do not single out the use phase electricity losses of the battery relative to that of the powertrain or report their results on a single-score indicator basis, comparing our main LCA results (Figure 2) with previous studies would have required extensive system boundary adjustments. The environmental impacts of the cradle-to-gate section of our LCA (Figure 3) are more readily comparable. On a per-mass basis, the GWP impacts of our two Li-ion batteries are roughly twice that of Samaras and Meisterling1 (9.6 kgCO2-eq 3 kg 1). Though both studies base their manufacturing energy requirements on Rydh and Sanden,17 we find more emissions upstream in material processing. As our estimates for manufacturing energy requirements differ significantly, it is not surprising that our Li-ion battery production GWP impacts should be 3.6 times greater than that of Notter et al.2 (6 kgCO2-eq 3 kg 1). They report electricity and heat requirements summing up to 1.4 MJ 3 kg 1 for battery and subcomponent manufacture. To put this number in perspective, this is four times less than molding of plastic (6 10 MJ 3 kg 1)42 and 40 times smaller than the estimate by Rydh and Sanden (ca. 60 MJ 3 kg 1)17 for battery and component manufacture. Interestingly, due to differences in use phase system definitions, our three studies converge if battery production impacts
ARTICLE
are expressed per distance driven. Assuming, for the sake of comparison, a powertrain efficiency4 of 0.5 MJ 3 km 1, our Li-ion production impacts convert to an estimated 10 gCO2-eq 3 km 1 and 7 gCO2-eq 3 km 1 for NCM and LFP, respectively, which is comparable to the 12 gCO2-eq 3 km 1 of Notter et al.2 and the 7 10 gCO2-eq 3 km 1 of Samaras and Meisterling.1 Both Notter et al. and Samaras and Meisterling base their estimates on a certain battery mass for a desired driving range and then a total vehicle life expectancy, discarding any battery cycle-life consideration. The vehicle studied by Notter et al. has a driving range of 200 km, carries a 300 kg Li-ion battery, and has a lifetime of 150 000 km. For their 60 km and 90 km range PHEVs, Samaras and Meisterling chose instead a 240 000 km life expectancy and batteries of 168 and 255 kg, respectively. We purposefully refrain from vehicle design assumptions and base our results on intrinsic battery characteristics and expected lifetime energy outputs. Thus, Notter et al. and Samaras and Meisterling estimate lower production impacts than we do, but the cycle-lives implicit in their lifetime estimations are shorter than ours. For NiMH, our estimates of the GWP impacts are significantly higher than the estimates by Rantik8 and Burnham et al.15 Prospective LCAs of new technologies are subject to important assumptions and data uncertainties, and as such our results should be used with circumspection. However, our inventory indicates that environmental impacts of traction batteries may be higher than previously expected. Differences in battery designs can lead to important variations in environmental impacts. For example, alternative materials have been used by the battery industry in lieu of polytetrafluoroethylene as binder9,43,44 and nickel foam as NiMH current collectors,45 both of which are identified in our results as especially environmentally consequential. A sensitivity analysis was performed to help define the variability of the most critical system parameters. Each parameter was increased and decreased by a fraction that we estimated representative of its variability. When altering the ratio of the components to optimize for either more energy or power density, a change of 25% in energy density changed the GWP impact of the life cycle of the battery by 3 10%. Battery efficiency and life expectancy assumptions are associated with important uncertainties, as they are a function of usage conditions, charge and discharge rates, depth-of-discharge, overcharge, and temperature control. In contrast with HEV battery data, publicly available real-world PHEV and BEV battery performance results are scarce. The efficiency of the battery proves to be a crucial parameter, with an alteration of 5 percentage points (80 ( 5% for NiMH and 90 ( 5% for Li-ion) leading to 8 23% changes life cycle GWP due to changes in use phase electricity waste. A reduction of lifetime estimations by one-third increases all categories of impacts by 30 45%. As this study is general and not based on a specific battery, modeling choices had to be made concerning synthesis routes of the different active materials. Nevertheless, material requirements of most production processes rely upon reasonably robust dismantling inventories and mass balance guidelines. In contrast, estimations of process-wise energy requirements and direct emissions are generally accompanied by larger uncertainties. If we consider a relative uncertainty of 33% on the energy requirements of the “Battery and subcomponent manufacture”, this would translate into a relative uncertainty of at most 8.5% for any given life cycle impact. If average Chinese electricity mix is used for all inventoried production processes instead of average European electricity mix, the life cycle impacts of the batteries 4552
dx.doi.org/10.1021/es103607c |Environ. Sci. Technol. 2011, 45, 4548–4554
Environmental Science & Technology increase by 10 16% for GWP and by 10 29% for particulate matter and photochemical oxidant formation. Conversely, it is expected that introducing secondary metals and end-of-life recycling in our system would bring down the environmental impacts. By Taylor series expansion, we found that more than 70% of GWP emissions occurred in processes more than 6 tiers upstream of the use phase in the value chain. This type of situation is inherently challenging in terms of system completion and boundary issues, and a conventional process-based LCA approach is likely to suffer from severe truncation bias.46 To alleviate these, we see great potential for a hybrid input-outputLCA study of traction batteries.47,48 Batteries for PHEV and BEV are emerging technologies and as such constitute an opportunity for the LCA field to prevent problem shifting issues. Yet, simultaneously, the novelty and proprietary nature of these technologies complicates data accessibility. In this paper, we have extensively reviewed publicly available data and produced transparent LCIs for three traction batteries for PHEV and BEV. The AB5 NiMH, the NCM Li-ion, and the LFP Li-ion devices were compared with a midpoint indicator approach. The two Li-ion batteries generally presented better environmental performances than NiMH, with LFP presenting environmental advantages relative to NCM. A shift from NiMH to Li-ion may thus be viewed positively. Though associated with important uncertainties, our results point to a higher than expected level of environmental impacts for the production and use of traction batteries. This inventory and LCA provide a basis for further benchmarking and focused development policies for the industry.
’ ASSOCIATED CONTENT
bS
Supporting Information. Further methodological details and the complete numerical results (SI sections 9 11). This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT The authors wish to thank the Norwegian Research Council for supporting this work under the E-Car Project, grant number 190940. The views expressed are those of the authors. For their precious help, our gratitude goes to Ann Mari Svensson, Bjarte Øye, and Yasushi Kondo. ’ REFERENCES (1) Samaras, C.; Meisterling, K. Life Cycle Assessment of Greenhouse Gas Emissions from Plug-in Hybrid Vehicles: Implications for Policy. Environ. Sci. Technol. 2008, 42, 3170–3176. (2) Notter, D. A.; Gauch, M.; Widmer, R.; W€ager, P.; Stamp, A.; Zah, R.; Althaus, H.-J. Contribution of Li-ion batteries to the environmental impact of electric vehicles. Environ. Sci. Technol. 2010, 44, 6550–6. (3) Hawkins, T. R.; Gausen, O. M.; Strømman, A. H. Environmental Impacts of Hybrid and Electric Vehicles: A Critical Review. Int. J. Life Cycle Assess. 2011, In Review, . (4) Axsen, J.; Burke, A.; Kurani, K. Batteries for plug-in hybrid electric vehicles (PHEVs): goals and the state of technology circa 2008; Davis, CA, 2008.
ARTICLE
(5) Armand, M.; Tarascon, J.-M. Building better batteries. Nature 2008, 451, 652–7. (6) Scrosati, B.; Garche, J. Lithium batteries: Status, prospects and future. J. Power Sources 2010, 195, 2419–2430. (7) Singh, M.; Cuenca, R.; Formento, J.; Gaines, L.; Man, B.; Santini, D.; Wang, M.; Adelman, S.; Kline, D.; Mark, J.; Ohi, J.; Rau, N.; Freeman, S.; Humphreys, K.; Placet, M. Total energy cycle assessment of electric and conventional vehicles: An energy and environmental analysis; Argonne National Laboratory, Argonne, Ill: Argonne, IL, 1998; Vol. ANL/ES/ RP-, p 233. (8) Rantik, M. Life Cycle Assessment of Five Batteries for Electric Vehicles under Different Charging Regimes; Chalmers University of Technology: Stockholm, 1999; Vol. 28. (9) Gaines, L.; Cuenca, R. Costs of lithium-ion batteries for vehicles; Argonne National Laboratory, Center for Transportation Research: Argonne, IL, 2000. (10) Schexnayder, S. M.; Das, S.; Dhingra, R.; Overly, J. G.; Tonn, B. E.; Peretz, J. H.; Waidley, G.; Davis, G. A. Environmental Evaluation of New Generation Vehicles and Vehicle Components; Oak Ridge National Laboratory, US Department of Energy: Washington, DC, 2001; p 133. (11) Matheys, J.; Timmermans, J.-M.; Mierlo, J. V.; Meyer, S.; Van den Bossche, P. Comparison of the environmental impact of five electric vehicle battery technologies using LCA. Int. J. Sustain. Manuf. 2009, 1, 318–329. (12) Matheys, J.; Autenboer, W.; Timmermans, J.-M.; Mierlo, J.; Bossche, P.; Maggetto, G. Influence of functional unit on the life cycle assessment of traction batteries. Int. J. Life Cycle Assess. 2007, 12, 191–196. (13) Van Den Bossche, P.; Vergels, F.; Van Mierlo, J.; Matheys, J.; Van Autenboer, W. SUBAT: An assessment of sustainable battery technology. J. Power Sources 2006, 162, 913–919. (14) Parsons, D. The environmental impact of disposable versus re-chargeable batteries for consumer use. Int. J. Life Cycle Assess. 2006, 12, 197–203. (15) Burnham, A.; Wang, M.; Wu, Y. Development and Applications of GREET 2.7 — The Transportation Vehicle-Cycle Model; Argonne National Laboratory: Chicago, 2006; Vol. ANL/ESD/06. (16) Rade, I.; Andersson, B. A. Requirement for metals of electric vehicle batteries. J. Power Sources 2001, 93, 55–71. (17) Rydh, C. J.; Sanden, B. A. Energy analysis of batteries in photovoltaic systems. Part I: Performance and energy requirements. Energy Convers. Manage. 2005, 46, 1957–1979. (18) Rydh, C. J.; Sanden, B. A. Energy analysis of batteries in photovoltaic systems. Part II: Energy return factors and overall battery efficiencies. Energy Convers. Manage. 2005, 46, 1980–2000. (19) Rydh, C. J.; Sv€ard, B. Impact on global metal flows arising from the use of portable rechargeable batteries. Sci. Total Environ. 2003, 302, 167–184. (20) Rydh, C. J. Environmental Assessment of Battery Systems: Critical Issues for Established and Emerging Technologies, Ph.D. Thesis, Chalmers University of Technology: G€oteborg, 2003. (21) Waugh, F. V. Inversion of the Leontief Matrix by Power Series. Econometrica 1950, 18, 142–154. (22) Peters, G. P.; Hertwich, E. G. Structural analysis of international trade: Environmental impacts of Norway. Econ. Systems Res. 2006, 18, 155–181. (23) Shiau, C.-S. N.; Samaras, C.; Hauffe, R.; Michalek, J. J. Impact of battery weight and charging patterns on the economic and environmental benefits of plug-in hybrid vehicles. Energy Policy 2009, 37, 2653–2663. (24) Zhao, X.; Ma, L. Recent progress in hydrogen storage alloys for nickel/metal hydride secondary batteries. Int. J. Hydrogen Energy 2009, 34, 4788–4796. (25) Ying, T.; Gao, X.; Hu, W.; Wu, F.; Noreus, D. Studies on rechargeable NiMH batteries. Int. J. Hydrogen Energy 2006, 31, 525–530. (26) Whittingham, M. S. Lithium Batteries and Cathode Materials. Chem. Rev. 2004, 104, 4271–4302. 4553
dx.doi.org/10.1021/es103607c |Environ. Sci. Technol. 2011, 45, 4548–4554
Environmental Science & Technology
ARTICLE
(27) Kalhammer, F. R.; Kopf, B. M.; Swan, D. H.; Roan, V. P.; Walsh, M. P. Status and Prospects for Zero Emissions Vehicle Technology; State of California Air Resources Board: Sacramento, CA, 2007. (28) Deveney, B.; Nechev, K.; Guseynov, T.; Jow, R.; Xu, K. Large Size Lithium Ion Cells Based on LiFePO 4 Cathode Material; In Tenth Electrochemical Power Source Research and Development Symposium; 2008. (29) Kokam Superior Lithium Polymer Battery (SLPB) Cell Specification Data; Kokam Co. Ltd.: Korea, 2006. (30) Ikeya, T. Multi-step constant-current charging method for an electric vehicle nickel/metal hydride battery with high-energy efficiency and long cycle life. J. Power Sources 2002, 105, 6–12. (31) Karden, E.; Ploumen, S.; Fricke, B.; Miller, T.; Snyder, K. Energy storage devices for future hybrid electric vehicles. J. Power Sources 2007, 168, 2–11. (32) Tanaka, T.; Ohta, K.; Arai, N. Year 2000 R\&D status of largescale lithium ion secondary batteries in the national project of Japan. J. Power Sources 2001, 97, 2–6. (33) Matheys, J.; Mierlo, J. V.; Timmermans, J.-M.; Van Den Bossche, P. Life-cycle assessment of batteries in the context of the EU Directive on end-of-life vehicles. Int. J. Vehicle. Des. 2008, 46, 189–203. (34) Shukla, A. K.; Kumar, T. P. Materials for next-generation lithium batteries. Curr. Sci. 2008, 93, 314–331. (35) Takahashi, M.; Ohtsuka, H.; Akuto, K.; Sakurai, Y. Confirmation of Long-Term Cyclability and High Thermal Stability of LiFePO[sub 4] in Prismatic Lithium-Ion Cells. J. Electrochem. Soc. 2005, 152, A899. (36) Lascelles, K.; Morgan, L. G.; Nicholls, D.; Beyersmann, D. Nickel Compounds. In Ullmann’s Encyclopedia of Industrial Chemistry; WileyVCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2005; p 16. (37) Donaldson, J. D.; Beyersmann, D. Cobalt and Cobalt Compounds. In Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2005; p 35. (38) Wellbeloved, D. B.; Craven, P. M.; Waudby, J. W. Manganese and Manganese Alloys. In Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2000. (39) Hischier, R.; Weidema, B.; Althaus, H.-J.; Bauer, C.; Doka, G.; Dones, R.; Frischknecht, R.; Hellweg, S.; Humbert, S.; Jungbluth, N.; K€ollner, T.; Loerincik, Y.; Margni, M.; Nemecek, T. Ecoinvent report v2.2; St. Gallen, 2010. (40) Goedkoop, M.; Heijungs, R.; Huijbregts, M.; Schryver, A. D.; Struijs, J.; Zelm, R. V. ReCiPe 2008; Netherlands, 2009; pp 1 126. (41) Classen, M.; Althaus, H.-J.; Blaser, S.; Scharnhorst, W.; Tuchschmid, M.; Jungbluth, N.; Emmenegger, M. F. Life Cycle Inventories of Metals. Ecoinvent v2.0 Reports; D€ubendorf, 2007; Vol. 10. (42) Hischier, R. Life Cycle Inventories of Packaging and Graphical Papers. Ecoinvent report No. 11; D€ubendorf, 2007; pp 1 242. (43) Yang, D.-C.; Jang, I.-S.; Jang, M.-H.; Park, C.-N.; Park, C.-J.; Choi, J. Optimization of additive compositions for anode in Ni-MH secondary battery using the response surface method. Met. Mater. Int. 2009, 15, 421–425. (44) Zhang, S.; Jow, T. R. Study of poly(acrylonitrile-methyl methacrylate) as binder for graphite anode and LiMn2O4 cathode of Li-ion batteries. J. Power Sources 2002, 109, 422–426. (45) Soria, M. L.; Chacon, J.; Hernandez, J. C. Metal hydride electrodes and Ni/MH batteries for automotive high power applications. J. Power Sources 2001, 102, 97–104. (46) Majeau-Bettez, G. Evaluation of Process- and Input-Output-Based Inventories for Environmental Assessment of Production and Consumption Activities, MSc Thesis in Industrial Ecology, Norwegian University of Science and Technology (NTNU): Trondheim, Norway, 2010; pp 1-69. (47) Strømman, A. H. Applying Leontief’s Price Model to Estimate Missing Elements in Hybrid Life Cycle Inventories. J. Ind. Ecol. 2008, 12, 26–33. (48) Hawkins, T. R.; Hendrickson, C.; Higgins, C.; Matthews, H. S.; Suh, S. A Mixed-Unit Input-Output Model for Environmental LifeCycle Assessment and Material Flow Analysis. Environ. Sci. Technol. 2007, 41, 1024–1031. 4554
dx.doi.org/10.1021/es103607c |Environ. Sci. Technol. 2011, 45, 4548–4554
