Cradle-to-Gate Emissions from a Commercial ... - ACS Publications

Article pubs.acs.org/est

Cradle-to-Gate Emissions from a Commercial Electric Vehicle Li-Ion Battery: A Comparative Analysis Hyung Chul Kim,*,† Timothy J. Wallington,† Renata Arsenault,† Chulheung Bae,† Suckwon Ahn,‡ and Jaeran Lee‡ †

Research and Innovation Center, Ford Motor Company, Dearborn, Michigan 48121-2053, United States Corporate R&D, LG Chem Research Park, 188, Munji-ro, Yuseong-gu, Daejeon, Korea

‡

S Supporting Information *

ABSTRACT: We report the first cradle-to-gate emissions assessment for a massproduced battery in a commercial battery electric vehicle (BEV); the lithium-ion battery pack used in the Ford Focus BEV. The assessment was based on the bill of materials and primary data from the battery industry, that is, energy and materials input data from the battery cell and pack supplier. Cradle-to-gate greenhouse gas (GHG) emissions for the 24 kWh Ford Focus lithium-ion battery are 3.4 metric tonnes of CO2-eq (140 kg CO2-eq per kWh or 11 kg CO2-eq per kg of battery). Cell manufacturing is the key contributor accounting for 45% of the GHG emissions. We review published studies of GHG emissions associated with battery production to compare and contrast with our results. Extending the system boundary to include the entire vehicle we estimate a 39% increase in the cradle-to-gate GHG emissions of the Focus BEV compared to the Focus internal combustion engine vehicle (ICEV), which falls within the range of literature estimates of 27−63% increases for hypothetical nonproduction BEVs. Our results reduce the uncertainties associated with assessment of BEV battery production, serve to identify opportunities to reduce emissions, and confirm previous assessments that BEVs have great potential to reduce GHG emissions over the full life cycle and provide local emission free mobility.

1. INTRODUCTION Assessing the energy and emissions benefits of plug-in hybrid electric vehicles (PHEVs) and battery electric vehicles (BEVs) requires a life cycle assessment (LCA) perspective. A number of LCAs have been conducted to understand the benefits of electric mobility. Most studies have focused on the vehicle use phase, particularly on the grid mix that powers PHEVs and BEVs as it is a key determinant of emissions and energy use. However, it is important to also include the environmental impact of electrified vehicle (EV) powertrain production especially the traction battery which is energy intensive to produce. Energy use and greenhouse gas (GHG) emissions associated with EV battery production are not well established. Published LCAs for lithium manganese oxide (LMO = LiMn2O4) and lithium nickel−cobalt manganese oxide (NCM = LiNixCoyMnzO2) BEV batteries have reported cradle-to-gate (raw material extraction, materials production, cell and component manufacturing, and battery pack assembly) GHG emissions of 39−63 and 121−196 kg CO2-eq per kWh of battery energy capacity, respectively.1−5 Most published LCAs have used secondary data (estimates from literature values, R&D production data, facility design parameters, proxy data based on similar processes) for cradle-to-gate energy and materials input and do not consider commercial lithium-ion battery designs used in mass-produced EVs. We report the first cradle-to-gate emissions for a massproduced commercial BEV battery; the lithium-ion battery pack © XXXX American Chemical Society

used in the Ford Focus BEV. The assessment was based on primary data from bill of materials (BOM) of the battery and energy and materials input data from the battery cell and pack supplier. We start with a review of the existing literature focusing on uncertainties associated with data sources. Detailed descriptions of system boundary, process flow, and inventory analysis for the present study are provided. The cradle-to-gate criteria pollutant and GHG emissions are determined and the GHG emissions results are compared with previous estimates. Finally, we discuss the life cycle GHG emissions benefit of BEVs based on the literature and present estimates.

2. REVIEW OF BEV BATTERY CRADLE-TO-GATE GHG EMISSIONS As a first step toward understanding the emission implications of BEVs, we reviewed published studies of the cradle-to-gate GHG emissions for BEV traction batteries representing current technologies. Table 1 provides a summary of the methods and results from previous studies. HEVs and PHEVs use batteries with a smaller size and lower energy density than those in BEVs. A wide range of cradle-to-gate GHG estimates are found Received: February 17, 2016 Revised: June 13, 2016 Accepted: June 15, 2016

A

DOI: 10.1021/acs.est.6b00830 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

LMO - 63.4; NCM 121 172 (240 and 487 for high estimates) 107 (175 and 424 for high estimates)

LMO - 1.8; NCM - 0

LMO −61.5; NCM −86.7 65 a

LMO = LiMn2O4; NCM = LiNixCoyMnzO2.

0.11 26.6 253 Ellingsen et al . (2014)2

na

LMO; NCM NCM

na

0.08−0.1

design−software tool (BatPac); cell mfg.- measured for R&D facility and extrapolated for commercial scale design-industry mix; cell mfg.- mix of industry data and literature values design-industry (Miljøbil Grenland); cell mfg.- industry data 0.13 210 LMO

28

54 143 design-literature; cell mfg.-literature (top-down) 0.112 na NCM

na

in the literature; 39−196 kg CO2-eq per kWh of battery capacity. As shown in Table 1, the GHG emissions range 0− 107 kg CO2-eq per kWh of battery in the cell manufacturing phase while those in the materials production and component manufacturing range 37−143 kg CO2-eq per kWh of battery. The inconsistencies across studies can be attributed to a variety of factors. When there are no available primary data (collected directly from industrial operations) for emission and energy consumption, LCA studies use secondary data such as literature values, databases, engineering modeling, or proxy data based on similar processes, and often extrapolate or adjust them to approximate the actual operational data. Notter et al. (2010)4 rely on expert estimates for the energy inputs in the cell and battery manufacturing phase while Majeau-Bettez et al. (2012)3 employ literature proxy values measured for stationary battery manufacturing. The latter study takes a top-down approach and may partially double count the material production impacts.1,3 To estimate cell manufacturing energy, Dunn et al. (2012) estimated energy consumption in a R&D facility based on a climate control design and extrapolated to a large-scale production, 6 million cells per year, in conjunction with a direct measurement in the cell formation stage.1,6 Ellingsen et al. (2014) used real-world commercial production data for cell manufacturing and battery design and estimated GHG emissions of 172 kg CO2-eq/kWh battery for the representative case.2 The authors found that GHG emissions depend on production volume with emissions as high as 487 kg CO2-eq/kWh for low volume production.2,7 In the reviewed studies, the cradle-to-gate GHG emissions of NCM batteries are much higher than those of LMO batteries, that is, 121−196 versus 39−63 kg CO2-eq per kWh of battery (Table 1), mainly because of the higher energy demand during the cell and pack manufacturing phase assessed for NCM than for LMO batteries. We found no inherent differences between NCM and LMO batteries in the cell and pack manufacturing processes that cover electrode coating, lamination, packaging, and pack assembly. The larger emissions reported for NCM batteries reflect differences in data sources and methodological approaches employed in the different studies not the fundamental characteristics of battery chemistry and manufacturing processes. The approaches taken by Majeau-Bettez et al. (2011)3 and Ellingsen et al. (2014)2 result in higher GHG emission estimates for cell and pack manufacturing than those found by Notter et al. (2010)4 and Dunn et al. (2012).1 The study by EPA (2013) evaluated both battery types and assigns higher GHG emissions for NCM than for LMO batteries.5 This is not surprising because the EPA study uses data from Notter et al. (2010) and Majeau-Bettez et al. (2011) for LMO and NCM batteries, respectively. The EPA study assigns zero GHG emissions for cell manufacturing of the NCM battery but assigns the greatest value for pack manufacturing among the reviewed studies, which seems to be a misinterpretation of industry data. With the exception of the study by Majeau-Bettez et al. (2011), the reviewed studies are in reasonable agreement in their estimation of GHG emissions for the materials and component production phases at 37−87 kg CO2-eq per kWh battery. Majeau-Bettez et al. (2011) estimated substantially higher emissions of 143 kg CO2-eq per kWh, reflecting their assumption of polytetrafluoroethylene (PTFE) binder material, whose production is deemed to generate a significant amount of halogenated organic emissions.2,3 Since real battery design data are difficult to obtain, studies often use hypothetical

LMO - 0.06; NCM - 34 1

39 na 2.1 37

196 na

53 pack mfg.

0.6

cell mfg. 0.9 51 design-industry (Kokam); cell mfg. − expert estimates 0.114 34.2 300 LMO

cathode reference

Notter et al . (2010)4 Majeau-Bettez et al. (2011)3 Dunn et al . (2012)1 EPA (2013)5

specific energy (kWh/kg battery) total energy (kWh) mass (kg)

Table 1. Review of Cradle-to-Gate GHG Emissions of BEV Batteriesa

data source

materials/part mfg.

GHG emissions (kg CO2-eq/kWh battery)

total

Environmental Science & Technology

B


Article


for the cathode and anode, respectively, and calendared. The next step is slitting and notching process to prepare the electrodes for tabbing and further assembly. The positive and negative electrodes are assembled in stacks, separated by a ceramic coated polyolefin separator, and then they are tabbed. The electrode stacks are packaged into laminated aluminum pouches and injected with electrolyte which consists of lithium hexafluorophosphate (LiPF6) salt in a mixture of organic carbonate solvents. Most of the electrode production and cell assembly operations are housed in a dry room with stringent air filtration requirements. Cells are cycled and stored at different temperatures during the formation step and, after degassing and resealing processes, the formed cells are tested before being shipped. In the pack manufacturing step, cells and the balance of battery (electrical system including sensors, battery management system (BMS), thermal management system, and enclosure) are assembled into the final battery pack. The detailed bill of materials (BOM) was used to estimate emissions from materials production. Figure 2 gives the mass

designs derived using modeling tools, or by combining literature and/or industry information. Only Ellingsen et al. (2014) examined a commercial battery design used in a BEV.2

3. METHODS The Ford Focus Electric is a compact size BEV with a 24 kWh lithium-ion battery, a city-highway combined fuel economy of 105 miles per gallon gasoline equivalent (=19.9 kWh/100 km; based on a conversion factor of 33.7 kWh of electricity per gallon of gasoline8), and a range of 122 km.9 The battery consists of 430 cells with a nominal voltage of 3.7 V and has a specific energy of 0.08 kWh/kg. The battery cells are produced by LG Chem in South Korea, a world leading supplier of lithium ion batteries for automotive applications, while the battery pack is manufactured by Piston Group in Michigan. We conducted a cradle-to-gate inventory analysis based on industry primary data and upstream emissions data from LCA databases. A process-based and attributional approach was used to compile the inventory. The functional unit of this study is 1 kWh of battery energy capacity. We also use a functional unit of 1 kg battery to compare GHG emissions across studies. The scope of this study covers the cradle-to-gate of the battery, that is, materials production, cell and component manufacturing, and battery pack assembly, including transportation. Figure 1 gives a schematic high level overview of

Figure 2. Mass breakdown of Focus BEV battery pack materials.

breakdown of the Focus BEV battery which has a total mass of 303 kg. Cell materials account for 55% of the total mass. Electrodes and collectors represent 73% of the cell mass, while the balance including electrolyte, separator, and pouch constitutes 27%. Steel used mostly for structural integrity and battery enclosure comprises 30% of the total mass. Plastics and composites (8%) are used in the enclosures and module components while nonferrous metals such as copper and aluminum alloys (3%) are used for the electrical architecture (bus bars, wiring etc.) and thermal management system. The BMS which consists of printed circuit board, electronics, and wiring harness accounts for the remaining 4% of the battery mass. Figure S1 in the Supporting Information (SI) provides a further breakdown of the battery pack by component. The upstream emissions for the production of cell and battery enclosure materials were calculated using the Ecoinvent 3.1 database, and the upstream emissions for production of materials for the BMS were calculated using the GREET 2014 model.10,11 Emissions associated with production of LMO were

Figure 1. Cradle-to-Gate flow diagram for Focus BEV battery. The white and gray box represents part/material flow and manufacturing process, respectively. BMS = battery management system.

process flows in the cradle-to-gate phase. The cathode material in the Focus BEV battery is a mixture of LMO/NCM, which is combined with the solvent N-methyl-2-pyrrolidone (NMP), polymer binder polyvinylidene difluoride (PVDF) and conductive carbon to make a cathode slurry. The anode uses an aqueous slurry system, and combines the active material graphite with a mixture of styrene butadiene rubber (SBR) and carboxymethyl cellulose (CMC). The slurry mixtures are coated and dried onto aluminum and copper current collectors C


Article


Figure 3. Cradle-to-gate GHG and criteria pollutant emissions per kWh of Focus BEV battery.

content but do not account for recycling potential at the endof-life. For example, the wrought aluminum includes 10% recycled content and the emission credits are accounted based on the recycled portion.10 Loss factors are assigned to metal and plastics manufacturing steps based on the GREET 2014 model, for example, 1.38 and 1.34 for aluminum and steel parts, respectively.11 Emissions associated with the lost scraps were allocated to materials production. Environmental impacts other than climate impact were not assessed mainly because the system boundary spans multiple continents, and thus a high degree of uncertainty would be introduced in attempting to judiciously associate impact indicators to the appropriate region or country.

assessed using the Ecoinvent database, whereas the GREET model11 (modified to account for differences in the ratio of nickel, cobalt, and manganese) was used to assess emissions for NCM precursor production. The Korean national LCI database developed by the Ministry of Trade, Industry & Energy (MOTIE) and Ministry of Environment (MOE) was used for aluminum and copper parts in the cell to represent local emissions and energy inventories.12 The Process Economics Program (PEP) yearbook13 was used for the upstream emissions of electrolyte production. The Ecoinvent 3.1 database was used to assess emissions for the component manufacturing phase.10 The manufacturing processes evaluated for steel and aluminum parts are rolling and stamping (metal working) while wire drawing is considered for copper parts. For the manufacturing stage of plastics and polymer composite parts, injection or blow molding was taken into account. The gray boxes in Figure 1 represent the cell and battery manufacturing. The energy used during cell manufacturing was measured in the LG Chem plant, Ochang, South Korea over 1 year of operation (January to December 2014) producing approximately 1 million cells for the Focus BEV. The energy used in battery pack manufacturing was based on normal operation of the Piston Group facility in Michigan where cells are received from South Korea and built into packs. The energy used for cell fabrication was translated into emissions based on the energy conversion conditions, that is, average electric grid mix (coal, 39%; natural gas, 22%; nuclear, 30%; oil, 5%; hydro, 1%; renewable and other, 3%),14 natural gas processing, and water supply in Korea.12 The balance of battery, that is, electrical system, enclosure, battery management system (BMS), and thermal management system were assumed to be produced in the U.S. Emissions associated with transporting cells between LG Chem’s plant in Ochang, Korea and the Piston Group facility in Michigan, were assessed using emission factors for road and sea freight transportation in the Ecoinvent 3.1 databases.10 In this study, we used the so-called Recycled Content approach to account for the recycling phase, where credits in the material production phase are based on secondary material

4. RESULTS Cradle-to-gate GHG (100-year) and criteria pollutant emissions for the Focus BEV battery were estimated using the methodologies and assumptions described above. We found the cradle-to-gate GHG emissions to be 3.4 t of CO2-eq or 140 kg CO2-eq per kWh of battery. GHG emissions associated with the use of utilities (electricity, natural gas, and water) in cell manufacturing covering the steps from “Mixing” to “Test” in Figure 1 account for 45% of the GHG emissions, i.e., 64 kg CO 2 -eq/kWh battery (see Figure 3). Producing and manufacturing the cell materials and components including cathode, anode, current collectors, electrolyte, separator, and pouch materials in Figure 1 account for 19% of the GHG emissions. Of the noncell contributors, the BMS accounts for 9%, whereas the balance (module and pack enclosures, electrical system wiring and components, and thermal management system) account for 22% of the cradle-to-gate GHG emissions. The GHG emissions from pack manufacturing and transportation are small; 1% and 3% of the total, respectively. Criteria pollutant emissions including volatile organic compounds (VOC), carbon monoxide (CO), nitrogen oxides (NOx), particulate matter (PM), and sulfur dioxide (SO2) were estimated and are shown in Figure 3. Similar to GHG emissions, cell manufacturing, cell components, and battery enclosure dominate criteria pollutant emissions, accounting for 82%−92% depending on pollutant. Each profile of criteria D


Article


Figure 4. Comparison of cradle-to-gate GHG emissions (a) per kWh and (b) per kg of battery. A specific energy of 0.1 kWh/kg is assumed for batteries in EPA (2013).

and numeric values used in these previous studies are listed in Table S2 in the SI. For materials and parts manufacturing (blue bar), the GHG estimate per kWh in our study, 76 kg CO2-eq , largely agrees with results from previous studies, 51−87 kg CO2-eq , except for that of Majeau-Bettez et al. (2011), 143 kg CO2-eq , which assumes PTFE as a binder material which was not considered in the other studies.2,3 The battery specific energy values used for the literature studies shown in Figure 4a, 0.1−0.13 kWh/kg,1−5 are hypothetical and much higher than the value measured for Focus BEV battery, 0.08 kWh/kg.9 For example, the specific energy, 0.114 kWh/kg assumed by Notter et al.(2010) for a BEV similar to Volkswagen Golf,4 is actually much higher than the rated value for the e-Golf battery, 0.076 kWh/kg.9 Thus, literature GHG assessments if presented using 1 kWh as the functional unit may not reflect real-world impacts. Figure 4b shows GHG emissions based on a functional unit of 1 kg of battery, which decouples the estimate of battery specific energy from the LCA result. As seen in Figure 4b, using 1 kg of battery as the functional unit places the present GHG estimate, 11 kg CO2-eq, in the lower-mid range of literature estimates compared to the upper-mid range position in Figure 4a. The above discussion emphasizes the importance of comparing results from different assessments by using both capacity (kWh)- and mass (kg)- based metrics to highlight any potential discrepancies. The high specific energy values (kWh/kg) assumed in the literature partially reflect simplified assumptions for battery pack mass (kg) balance. The fraction of noncell battery materials in our analysis and that by Ellingsen et al. (2014),2 40−45%, is much higher than that in other studies which is typically 10−20% (see Table S2 in the SI). The difference is mostly related to larger amounts of battery enclosure materials in the former studies which use industry-representative battery designs versus the hypothetical designs based on literature3 and software tools1 in the other studies. We eliminated these “balance-of-cell” parts, that is, enclosure, electrical system, BMS, and thermal management, from the analysis scope. The resulting cradle-to-gate GHG emissions for cells only are plotted in Figure S2 in the SI. The GHG emissions for materials and parts manufacturing decreases most in our study

pollutant emissions reflects environmental aspects of specific materials and processes. For example, the large proportion of CO and PM emissions associated with the battery enclosure is related to high upstream emissions of these pollutants from production of steel used in the enclosure while the high SO2 emission from the production of cell components is associated with copper smelting.10,15,16 We note that emissions of criteria pollutants are highly dependent on emission control and recovery measures in facilities. GHG emissions largely reflect energy usage which is typically less variable between facilities.16 The complete numerical results are available in Table S1 of the SI.

5. COMPARATIVE ANALYSIS We compared our GHG emissions results with those from previous studies to understand the variability depending on data sources and assumptions. Figure 4a shows the cradle-togate GHG emissions of BEV batteries using 1 kWh as the functional unit. Our GHG estimate per kWh for the Focus BEV battery, 140 kg CO2-eq, is in the upper midrange of literature estimates for BEV batteries, 39−196 kg CO2-eq. As discussed earlier, the large range of the existing estimates is mainly attributed to inconsistent reporting of GHG emissions from cell and pack manufacturing across the literature, 1.5−108 kg CO2eq/kWh. Our GHG estimate per kWh battery for this stage, 65 kg CO2-eq, is much higher than the range of 1.5−1.9 kg CO2eq estimated by Notter et al. (2010),4 Dunn et al. (2012),1 and EPA (2013)5 which are included but are barely visible at the top of the three bars on the left in Figure 4a. In contrast our estimate of 65 kg CO2-eq is much lower than the representative estimate by Ellingsen et al. (2014)2 of 108 kg CO2-eq shown in the bar sixth from the left in Figure 4a. To provide further insight, we determined the combined primary energy demand from utility use (electricity, natural gas, and water) during the cell and pack manufacturing of the Focus BEV battery to be 120 MJ per kg of battery. The low GHG estimates of the first three red bars correspond to primary energy demand of 2−5 MJ/kg battery while those estimates by Ellingsen et al. (2014)2 including low efficiency production cases correspond to primary energies of 180−730 MJ/kg battery using a primary to electric energy conversion factor17 of 0.35. The assumptions E


Article


Figure 5. Cradle-to-gate GHG emissions for ICEVs and BEVs.

6. DISCUSSION To reduce the uncertainties in assessments of the environmental impacts from real-world battery production, we report the cradle-to-gate emissions of the Ford Focus BEV battery based on primary data for large scale production and battery design. Our GHG emissions estimate of 140 kg CO2-eq/kWh battery lies in the midrange of the literature values for BEV batteries, 39−196 kg CO2-eq/kWh battery. Approximately half of the GHG emissions in our analysis 65 kg CO2-eq/kWh battery are associated with utility usage (electricity, natural gas, and water) during cell manufacturing and pack assembly. Due to a dearth of primary data, estimating the GHG emissions from cell manufacturing is the most difficult aspect of battery production to analyze and has the greatest uncertainty. Even estimates based on primary data can have a large degree of variation depending upon operating conditions and assumptions. The representative GHG estimate by Ellingsen et al. (2014)2 of 172 kg CO2-eq/kWh battery or 18 kg CO2-eq/ kg battery is based on primary industry data, but is 20% and 60% higher than our estimate, per kWh battery and per kg battery, respectively. These differences cannot be easily explained as information on facility design, production volume and plant capacity is not public. Differences in production scales are a possible explanation for the different results. Energy requirement per functional unit produced decreases as production scale and associated efficiencies increase, and as manufacturing technology matures. For example, the climate controlled space of a factory producing EV cells is expected to house more lines with higher production speeds as production technology advances, with a lowered energy demand per kWh produced. A significant decrease in per unit energy use with increasing production scale has been observed in emerging technologies such as in the production of nanomaterials and solar photovoltaics21,22 and similar progress seems likely for BEV batteries. Some studies employed a “top-down” approach where the energy for production of batteries used in different applications was modified to estimate the energy demand for production of BEV batteries. Majeau-Bettez et al. (2011)3 adjusted the primary energy estimates by Rydh and Sandén (2005),17 96− 144 MJ/kg battery, for stationery batteries for photovoltaics. Zackrisson et al. (2010)23 reported a primary energy demand of

and in that by Ellingsen et al. (2014) where the noncell mass fraction is the highest. In this cell only analysis, switching the functional unit from 1 kWh to 1 kg of battery does not lead to the substantial decrease in our GHG estimate as in Figure 4, since noncell parts are excluded from the calculation. Figure 5 shows cradle-to-gate GHG emissions for ICEVs and BEVs from literature studies and the present work. The GHG emissions for lithium-ion batteries in Figure 5 were obtained by multiplying the total capacity of batteries by the GHG emissions factor (kg CO2-eq per kWh battery) reported in the different studies. The ICEVs and BEVs cradle-to-gate GHG emissions were taken directly from the lithium ion battery study, that is, Notter et al. (2010)4 or those studies that use the battery GHG emission estimates, that is, Hawkins et al. (2013)18,19 and the GREET model.11 Data for the Focus BEV were taken from Ford’s Product Sustainability Index (PSI) report.20 The study by EPA (2013)5 does not report total battery energy capacity and thus is not included. Since the estimates by Ellingsen et al. (2014)2 have not been combined with total BEVs so far, only the GHG estimate for the battery is shown here. It is well-known that more GHG emissions are generated in the vehicle cycle of BEVs than ICEVs as the electric powertrain system (motor, generator, and battery) requires more energy to produce than the internal combustion engine powertrain. In the electric powertrain, the traction battery accounts for the largest amount of energy demand and GHG emissions. As illustrated in Figure 5, a wide range of GHG emissions have been reported for producing batteries resulting in a large degree of uncertainty in assessments of the cradle-to-gate GHG emissions of BEVs. The increase of cradle-to-gate GHG emissions switching from ICEV to BEV, therefore, varies across studies. A 27% increase was reported by Notter et al. (2010)4 while a 63% increase was reported by Hawkins et al. (2013).18,19 Combining data from the 2015 Product Sustainability Index report for the Focus ICEV20 with results from the present study for the Focus BEV gives a 39% increase in cradle-to-gate GHG emissions switching from ICEV to BEV. Estimates of the mass and energy demand to produce powertrain components of the BEV excluding the battery, such as motor and generator, differ somewhat across studies but do not explain the differences between ICEVs and BEVs in the literature discussed above. F



■

around 150 MJ per kg battery for cell module and battery assembly in their LCA study of plug-in hybrid vehicle batteries based on an industry report.24 Although these approaches arrive at energy estimates comparable to our estimate of 120 MJ primary energy demand per kg of battery, inconsistent system boundaries in the previous studies may have led to doublecounted or omitted energy usages.1,3 Dunn et al. (2012)1 used a “bottom-up” approach and extrapolated facility design and data from R&D battery manufacturing to a scenario of 6 million cell production per year6 to estimate the energy required for climate control in commercial scale cell assembly (manufacturing). However, their energy estimate, 4 MJ of primary energy per kg battery,11 may not reflect real-world operating conditions since their estimated process flow and cycle time may differ from commercial continuous production lines. The real world commercial battery designs studied here and by Ellingsen et al. (2014)2 have larger amounts of battery enclosure materials than in other engineering designs in the literature. Cells only account for 55−60% of the battery mass in the commercial BEV battery designs, whereas they account for more than 80% of battery mass in the other studies.1,3,4 We argue that more attention should be addressed to noncell contributions in LCAs of BEV batteries. Recycling or remanufacturing of enclosure materials may be able to significantly reduce the life cycle impact of traction batteries. Finally, we place the results from the present work into the broader perspective of the environmental benefits of BEVs. As discussed here, it is well established that the cradle-to-gate life cycle stage for BEVs is more energy intensive than for ICEVs mainly reflecting energy use and GHG emissions associated with battery production. Our estimate of a 39% increase in the cradle-to-gate GHG emissions of the Focus BEV compared to the Focus ICEV falls within the range of literature estimates of 27−63% increases based on studies of hypothetical nonproduction ICEVs and BEVs. Despite their higher cradle-togate GHG emissions, switching from ICEVs to BEVs potentially saves a large amount of GHG emissions during their life cycle. Published studies have estimated approximately 30−40% life cycle GHG emissions reduction for BEVs powered by the average U.S. or European electric grid mix.4,11,18,19,25,26 Using our GHG estimate for BEV battery production, 11 kg CO2-eq/kg battery, in place of those in the literature gives an estimate of 31−37% life cycle GHG benefits for BEVs over gasoline ICEVs. Our results confirm the potential for BEVs to curb GHG emissions from the transportation sector. Current trends of increasing vehicle energy efficiency, decreasing burdens associated with battery production, decreasing burdens for electricity production, and increasing burdens for oil production27 are expected to increase the GHG emission benefits of electrification technology. We highlight the importance of further LCA studies for BEVs using real world data to capture future improvements in vehicle performance and battery materials.

■

Article

AUTHOR INFORMATION

Corresponding Author

*Phone: 313-323-9745; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Ford Motor Company (Ford) does not expressly or impliedly warrant, nor assume any responsibility, for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, nor represent that its use would not infringe the rights of third parties. Reference to any commercial product or process does not constitute its endorsement. This article does not provide financial, safety, medical, consumer product, or public policy advice or recommendation. Readers should independently replicate all experiments, calculations, and results. The views and opinions expressed are of the authors and do not necessarily reflect those of Ford. This disclaimer may not be removed, altered, superseded or modified without prior Ford permission. We thank Hyun Wha Oh, Byungeun Lee, In Seob Yoon, Yong Eon Kim, Jeongil Lee and Mohamed Alamgir in LG Chem, Regina Newberry in Piston Group, and Wulf-Peter Schmidt at Ford for providing data and supporting this study.

■

REFERENCES

(1) Dunn, J. B.; Gaines, L.; Sullivan, J.; Wang, M. Q. Impact of Recycling on Cradle-to-Gate Energy Consumption and Greenhouse Gas Emissions of Automotive Lithium-Ion Batteries. Environ. Sci. Technol. 2012, 46, 12704−12710. (2) Ellingsen, L. A.-W.; Majeau-Bettez, G.; Singh, Bhawna; Srivastava, A. K.; Valøen, L. O.; Strømman, A. H. Life Cycle Assessment of a Lithium-Ion Battery Vehicle Pack. J. Ind. Ecol. 2014, 18 (1), 113−124. (3) Majeau-Bettez, G.; Hawkins, T. R.; Strømman, A. H. Life Cycle Environmental Assessment of Lithium-Ion and Nickel Metal Hydride Batteries for Plug-In Hybrid and Battery Electric Vehicles. Environ. Sci. Technol. 2011, 45, 4548−4554. (4) Notter, D. A.; Gauch, M.; Widmer, R.; Wagner, 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, 6550−6556. (5) U.S. EPA Application of Life-Cycle Assessment to Nanoscale Technology: Lithium-ion Batteries for Electric Vehicles; EPA 744-R-12001; 2013. (6) Argonne National Laboratory. Material and Energy Flows in the Materials Production, Assembly, and End-of-Life Stages of the Automotive Lithium-Ion Battery Life Cycle; ANL/ESD/12-3 Rev.; 2014. (7) Ellingsen, L. A.-W.; Majeau-Bettez, G.; Strømman, A. H. Comment on “The significance of Li-ion batteries in electric vehicle life-cycle energy and emissions and recycling’s role in its reduction” in Energy & Environmental Science. J. Ind. Ecol. 2015, 19 (3), 518−19. (8) New Fuel Economy and Environment Labels for a New Generation of Vehicles, EPA-420-F-11-017; United States Environmental Protection Agency, May, 2011. (9) Department of Energy Model Year 2015 Fuel Economy Guide. https://www.fueleconomy.gov/feg/download.shtml. (10) Weidema, B. P.; Bauer, C.; Hischier, R.; Mutel, C.; Nemecek, T.; Reinhard, J.; Vadenbo, C. O.; Wernet, G. The ecoinvent database: Overview and methodology, Data quality guideline for the ecoinvent database version 3: 2013. www.ecoinvent.org. (11) Argonne National Laboratory (ANL). GREET Model. Transportation Research and Analysis Computing Center; Argonne National Laboratory: Argonne, IL, 2015. (12) MOTIE/MOE. Korean National LCI Database; Minstry of Trade, Indsutry & Energy/ Minstry of Environment. http://www. kncpc.or.kr/resource/lci_pass_db.asp (accessed February 2015).

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.6b00830. Figures S1−2 and Tables S1−2 (PDF) G


Article

Environmental Science & Technology (13) SRI Consulting, PEP Yearbook. 2015. http://chemical.ihs.com/ PEP/yearbook.htm. (14) Trend of Korean Electricity Generation MIx. http://kosis.kr/ statHtml/statHtml.do?orgId=388&tblId=TX_38803_A016. (15) American Iron and Steel Institute Steel Works: the online resource for steel. https://www.steel.org/making-steel/how-its-made/ processes/how-a-blast-furnace-works.aspx (accessed October 2015). (16) Fthenakis, V.; Wang, W.; Kim, H. C. Life cycle inventory analysis of the production of metals used in photovoltaics. Renewable Sustainable Energy Rev. 2009, 13, 493−517. (17) Rydh, C. J.; Sandén, B. A. Energy analysis of batteries in photovoltaic systems. Part I: Performance and energy requirements. Energy Convers. Manage. 2005, 46, 1957−1979. (18) Hawkins, T. R.; Singh, B.; Majeau-Bettez, G.; Strømman, A. H.; Corrigendum to; Hawkins, T.; Singh, B.; Majeau-Bettez, G.; Strømman, A. H. 2012. Comparative Environmental Life Cycle Assessment of Conventional and Electric Vehicles. J. Ind. Ecol. 2013, 17 (1), 158−160. (19) Hawkins, T. R.; Singh, B.; Majeau-Bettez, G.; Strømman, A. H. Comparative Environmental Life Cycle Assessment of Conventional and Electric Vehicles. J. Ind. Ecol. 2013, 17, 53−64. (20) Ford Motor Company Ford Focus Product Sustainability Index (PSI). http://corporate.ford.com/microsites/sustainability-report2014-15/doc/sr14-psi-focus-fact-sheet-electric.pdf. (21) Kim, H. C.; Fthenakis, V. Life Cycle Energy and Climate Change Implications of Nanotechnologies. J. Ind. Ecol. 2013, 17 (4), 528−541. (22) Kim, H. C.; Fthenakis, V.; Choi, J.-K.; Turney, D. E. Life Cycle Greenhouse Gas Emissions of Thin-film Photovoltaic Electricity Generation: Systematic Review and Harmonization. J. Ind. Ecol. 2012, 16 (S1), S110−121. (23) Zackrisson, M.; Avellán, L.; Orlenius, J. Life cycle assessment of lithium-ion batteries for plug-in hybrid electric vehicles - Critical issues. J. Cleaner Prod. 2010, 18, 1519−1529. (24) Saft. Annual Report; Saft Batteries: 2008. (25) Dunn, J. B.; Gaines, L.; Kelly, J. C.; James, C.; Gallagherc, K. G. The significance of Li-ion batteries in electric vehicle life-cycle energy and emissions and recycling’s role in its reduction. Energy Environ. Sci. 2015, 8, 158−168. (26) U.S. Department of Energy Cradle to Grave Lifecycle Analysis of Vehicle and Fuel Pathways, Record #: 14006, 2014. (27) Wallington, T. J.; Anderson, J. E.; Kleine, R. D.; Kim, H. C.; Maas, H.; Brandt, A. R.; Keoleian, G. A. Life cycle assessments of alternative fuel-vehicle systems should consider trends in oil production. J. Ind. Ecol. 2016, DOI: 10.1111/jiec.12418.

H


Cradle-to-Gate Emissions from a Commercial ... - ACS Publications

Recommend Documents