Assessing Trade-Offs in Climate Impacts across Various Circular Economy Strategies for Japan’s Battery Industry

Abstract

Circular strategies, such as extending in-use lifetimes, reuse, and recycling, are recognized as efficient ways to mitigate the climate impacts associated with lithium-ion battery demand. However, these strategies come with trade-offs, as extending lifetime through continued use in vehicles or through reuse in battery energy storage systems (BESS) delays recycling and secondary material supply. While recycling regulations with mandatory warranty periods accelerate material recovery, they may sacrifice carbon benefits from prolonged use. Focusing on Japan, we develop a dynamic assessment framework that combines stock-driven material flow analysis with life-cycle-based greenhouse gas (GHG) accounting to evaluate these trade-offs under time-evolving vehicle battery and solar/wind BESS demand and time-varying GHG emission factors. Our analysis indicates that, under the current warranty (8-year) lifetime scenario, the generation of reusable electric vehicle batteries is projected to reach 225 GWh by 2050, sufficient to meet most energy storage demand. Despite the trade-offs, implementing both reuse and recycling reduces cumulative battery emissions by 20–30% compared with scenarios that use only one strategy. In the early stage (2022–2035), compared with reuse in BESS and recycling, extending battery lifetime in electric vehicles can yield greater emissions-reduction benefits. These findings highlight the need for adaptive policies to optimize circularity in battery systems.

Introduction

The global transition toward clean energy has intensified demand for batteries. In 2023, over 2400 GWh of battery capacity were used worldwide, four times the amount used in 2020. Batteries play a crucial role in transportation, where they power electric vehicles (EVs), and in power generation, where they stabilize renewable energy through battery energy storage systems (BESSs). , This increased reliance on batteries is not merely a trend but a critical step toward a sustainable future. However, the rapid growth of battery deployment also raises environmental concerns and resource challenges. Battery manufacturing alone accounts for approximately 20–30% of the total life-cycle GHG emissions of EVs. , Massive extraction and processing of critical battery materials such as lithium, nickel, and cobalt introduce additional environmental pressures and exacerbate supply chain vulnerabilities. − For instance, in Indonesia, villages near nickel mines experienced deforestation at nearly double the rate of nonmining areas from 2011 to 2018. In early 2025, the Democratic Republic of the Congo, the world’s leading cobalt supplier, temporarily banned cobalt exports to control global cobalt prices. These examples alert us to the growing tension between clean energy ambitions and the sustainability of critical material supply chains.

In response, circular economy strategies centered on extending battery lifetime and recycling are gaining increasing attention. , Extending battery lifetime can be achieved by prolonging battery use in EVs or by repurposing retired batteries in second-life applications, most notably in BESSs. , These approaches maximize the utilization of embedded energy and materials before the batteries enter the recycling stream. Recycling enables the recovery of critical materials, thereby reducing dependence on primary resource extraction, and can mitigate the environmental burdens associated with mining and refining. ,

However, given the benefits of these circular economy strategies, important trade-offs exist between them. Extending battery lifetime, whether through continued use in EVs or through reuse in BESSs, delays the point at which batteries reach their end-of-life (EoL). This postpones recycling and limits the short-term availability of secondary raw materials, which may slow progress toward material circularity as measured by the share of recycled materials in production. , On the other side, prioritizing early recycling may forego the additional carbon and economic benefits that could be achieved through prolonging battery life in EVs and second-life applications. ,

In Japan, battery demand for stationary renewable energy storage is growing rapidlyexpanding at double-digit annual rates as solar and wind installations accelerate, while the market penetration of pure EVs remains modest, at only about 2% in 2023. , This structural imbalance between stationary and mobile applications creates temporal and sectoral mismatches in battery supply and demand, presenting both constraints and opportunities for establishing reuse systems that cascade retired EV batteries into BESSs and final recycling. Currently, Japan is one of the world’s leading countries in exploring and demonstrating the reuse of retired vehicle batteries in wind and solar energy storage systems. In 2014, Nissan and Sumitomo Corporation pioneered the world’s first large-scale second-life battery storage system in Osaka with a capacity of 400 kWh. In 2023, the advanced 23 MWh Chitose Second-Life Battery Station was commissioned in Hokkaido. , Therefore, a Japan-focused analysis is timely and essential for understanding how circular battery strategies may unfold in a market with an accelerating need for grid storage but limited EoL EV battery supply in the near term. Besides, many other countries undertaking parallel decarbonization of the transport and power sectors are likely to face similar temporal and sectoral misalignments. Therefore, the insights gained here can inform the anticipation and management of practical challenges in implementing circular battery strategies globally.

Extensive research has investigated the recycled material supply and environmental implications of circular battery strategies through material flow analysis (MFA) and life-cycle assessment. ,− Most previous studies have examined these strategies separately without explicitly accounting for the potential trade-offs. A limited number of recent studies have explored the interactions between reuse and recycling, and among these, very few have explicitly incorporated the evolving and increasingly dynamic demand for batteries required in BESSs. Without explicitly accounting for this additional demand, the environmental benefits of reuse and recycling may be overstated or misrepresented, leading to incomplete or even misleading policy insights. Recently, López et al. explored the material implications of vehicle-to-grid and second-life applications, while Zhou et al. investigated the interactions among multiple circular economy strategies. Nevertheless, these studies did not quantify the associated GHG emission impacts. Wesselkämper et al. compared the GHG emission reduction potentials of reuse and recycling pathways, but did not account for the temporal evolution of GHG emission factors driven by grid decarbonization. Incorporating such dynamics is critical, as static emission factors could lead to underestimating or overestimating the climate benefits of circular strategies in rapidly decarbonizing energy systems. ,

Building upon these insights, the present study addresses these research gaps by explicitly coupling dynamic BESS demand with a stock-driven material flow and emission assessment framework. By systematically analyzing the effects of circular battery strategies across multiple battery chemistries and lifetime configurations, this study integrates dynamic MFA with a life-cycle-based GHG emission assessment to capture the coupled dynamics among material demand, GHG emissions, and circular strategies. Furthermore, time-dependent GHG emission factor trajectories consistent with long-term energy transition scenarios are incorporated. This dynamic representation ensures coherence between the underlying decarbonization assumptions and the environmental assessment framework, thereby providing a more realistic understanding of how the climate benefits of circular battery strategies evolve over time as the electricity mix progressively decarbonizes.

Methods

Model Framework

An integrated modeling and assessment approach is used in this study to quantify and compare the climate impacts of activities associated with batteries and the mitigation potential of circular battery strategies in Japan from 2022 to 2050 (Figure ). The framework links three modules: (1) a battery stock-dynamics module that projects additions and retirements of lithium-ion batteries (LIBs) across vehicles and BESS; (2) a material flow module that translates stock flows into total material demand, reuse flows, scrap, secondary supply from recycling, and net primary demand; and (3) a GHG analysis module that applies material- and process-specific emission factors (or energy intensities with emission factors) to the stock-dynamics module and material flow module outputs. The vehicle segment covers internal combustion engine vehicles (ICVs), hybrid electric vehicles (HEVs), plug-in hybrid electric vehicles (PHEVs), and battery electric vehicles (BEVs). Consistent with Japan’s market, PHEVs are represented by an 18 kWh pack size, while BEVs are parametrized with 20, 40, and 70 kWh pack sizes. − GHG emissions are quantified for primary material production, battery manufacturing, and recycling, yielding time-resolved emission trajectories and mitigation contributions attributable to different circular strategies.

1.

Modeling framework for evaluating material flows, reuse and recycling processes, and GHG emissions of LIB in EV and BESS sectors of Japan. EVB: EV batteries; NSB: new stationary batteries; RUB: reused batteries; Remanu.: remanufacturing. Numbers shown in the boxes and along the flows are used to label and distinguish different processes in the system. The waste flow on the right side indicates the fraction of materials that cannot be recovered through recycling and is therefore assumed to enter landfill disposal.

Battery Stock Dynamics

The demand for vehicles is estimated using a stock-driven dynamic model based on population projections and car ownership rates. The inflows and outflows of the vehicle sector are calculated based on the lifetime definitions and the conditions for reuse and replacement of EVs and batteries. Future new vehicle sales are allocated across powertrains based on the projected powertrain-specific market shares under the net-zero pathway derived from the Japan Automobile Manufacturers Association report. This scenario envisions a rapid increase in PHEV and BEV adoption in Japan, with BEVs reaching 100% of new car sales by 2050. To reflect potential divergence between vehicle and battery service lives, we adopt a double-lifetime formulation for EVs that jointly tracks vehicle and battery cohorts. In addition, to minimize the sensitivity of model results to uncertainties in future vehicle segmentation, the market shares of mini, small, and standard BEVs are assumed to remain the same as their current levels of approximately 0.36, 0.25, and 0.39, respectively. Similarly, PHEVs are assumed to continue to exist only in the standard vehicle segment.

In the BESS sector, two types of battery inflows are considered: newly manufactured batteries and reused EV batteries. The inflow of reused EV batteries, estimated as an output from the EV battery stock model, is treated as an exogenous input into the BESS model. Given the lower performance requirements of BESS applications and the lower economic attractiveness of recycling lithium–iron-phosphate (LFP) batteries, and consistent with assumptions commonly adopted in previous studies, , we assume that only LFP batteries are reused, while other chemistries are directed to recycling. First, the aggregate BESS capacity requirement is derived from projected renewable deployment and system design assumptions (Section 5 in Supporting Information SI1). Second, total BESS inflows needed to satisfy this requirement are computed using a stock-driven model. We then compare these required inflows with the available flow of reusable EV LIBs. When available reuse supply is insufficient, the net new BESS demand is defined as the residual requirement after subtracting the reuse inflow. Retirements from the BESS stock are determined by the lifetime distributions assigned to newly manufactured systems and to reused EV LIBs, respectively.

The lifetime of a product determines the frequency of replacements and the volume of spent products available for reuse and recycling. We assume that the lifetimes of all vehicles follow a normal distribution with an average of 14 years and a standard deviation equal to 50% of the mean. This assumption is consistent with the current lifetime distribution of passenger vehicles reported by the Automobile Inspection and Registration Information Association of Japan. Under exploratory settings, we include a replacement-only case in which the battery reuse rate is set to zero and an in-vehicle battery replacement rate of 80% is applied.

The lifetime distribution of batteries is assumed to follow a normal distribution, with different average lifetimes (Section 7 in SI1). Particularly, we distinguish two lifetimes for EV batteries to decouple “when” batteries exit vehicles from “how much capacity” they retain upon exit. The functional lifetime is defined here as the time required for the state of health (SOH) to decline to 80%, a technology-limited boundary governed by chemistry, design, and operating conditions. In this study, we assume the functional lifetime to be 12 years, representing the rounded mean of the 10–15 year range reported in previous studies. The in-use lifetime is the realized service time in a vehicle from installation to replacement or disposal, and is further influenced by warranties, user preferences, and economics. In the stock model, the in-use lifetime determines replacement outflows, while the functional lifetime maps each exiting cohort to a remaining-capacity distribution using scenario-specific degradation curves. To span plausible practice and technology ranges, we specify four in-use lifetime scenarios of 8, 10, 12, and 14 years. For reused EV batteries, previous studies commonly assume lifetimes of 6–10 years. , We therefore adopt an average value of 8 years in this study. For newly manufactured BESS units, we assume a corresponding lifetime of 20 years. For simplicity, all battery chemistries are assumed to have identical lifetimes in this study.

We adopt an 80% SOH threshold for reuse, as previous studies have shown that the profitability of second-life applications declines sharply with decreasing SOH. , This suggests that batteries with lower SOH levels are less economically feasible for reuse, and that a threshold near 80% represents an appropriate balance between technical performance and economic return. Operationally, in the 12-year functional lifetime case, age cohorts exceeding 12 years are considered ineligible for reuse and are therefore routed directly to recycling (eqs S34–S36 in SI1). Furthermore, among batteries retiring prematurely, such as those replaced within eight years due to warranty claims, we assume that 30% are nonreusable because of defects or safety concerns. This assumption is based on interviews with a battery reuse facility in Japan.

Material Flow Analysis

The total material demand is computed by multiplying battery demand from the stock-driven model by material-intensity coefficients and applying a manufacturing scrap rate of 10%, which is routed directly to recycling in the model. Material intensity data for cathode active materials, cells, modules, and battery packs are compiled from peer-reviewed publications. , For each year and sector, we estimate material demand, reused flows, and waste by multiplying the masses of new, reused, and spent batteries by the corresponding material-intensity coefficients (Supporting Information SI2). Battery chemistry strongly influences material composition. Accordingly, we consider two technology pathways for all new batteries: lithium–nickel-cobalt–aluminum/lithium–nickel-manganese–cobalt oxide (NCX) and LFP (Section 6 in SI1). We adopt global chemistry scenarios from the literature due to the lack of a publicly available, Japan-specific chemistry roadmap. We acknowledge that battery chemistry adoption differs across regions and is shaped by both global supply chain dynamics and region-specific market and policy conditions.

Recycling rates determine the quantity of secondary material obtained from scrap flows. A review of Japan’s battery-recycling industry indicates that recovery rates for lithium, nickel, and cobalt are expected to improve and move toward the targets in the EU Battery Regulation. We therefore model recovery rates that gradually converge to those targets over time. Net primary demand is calculated as total demand minus recovered secondary supply, and recovered materials are returned to the supply chain to offset future extraction. Detailed assumptions for recovery efficiencies are reported in Section 10, SI1. Equations for total material demand, reused flows, and waste are provided in Section 1, SI1.

GHG Emission Analysis

We quantify GHG emissions over the full life cycle of lithium ion batteries and resolve four source categories: (1) material specific emissions from the production of primary materials; (2) direct process emissions from battery manufacturing, including remanufacturing; (3) indirect emissions associated with energy supply for battery manufacturing and remanufacturing, such as grid electricity and on-site fuels; and (4) recycling process emissions from collection, pretreatment, and material recovery. The GHG emissions associated with the landfill stage are expected to be minor relative to the main recycling-process burdens and were not explicitly included in our accounting. For each year, sector, and scenario, process-level activity data (material masses, manufacturing and remanufacturing throughputs, and recycling throughputs) are taken from the stock-dynamics and material flow modules and multiplied by the corresponding emission factors or energy intensities. The functional unit is defined as the production, remanufacturing, and recycling of 1 kg of battery over its life cycle. Emission factors and energy intensities for primary material production, battery manufacturing and remanufacturing, as well as battery recycling are compiled from peer-reviewed LCA studies (Sections 8–12 in SI1, SI2). ,, These studies are based on the ecoinvent database and predominantly reflect global average production conditions. For grid electricity, we apply emission factors consistent with the Japanese electricity mix. All emission factors are characterized using the 100-year global warming potential (GWP100) metric. Although all referenced studies employ GWP100, minor discrepancies may arise due to differences in the IPCC assessment report versions underlying the characterization factors. A detailed description of the GHG emissions calculation methodology is provided in Section 3, SI1.

To reflect future grid decarbonization, we allow temporal changes in emission intensities for energy supply, battery recycling, and primary material production. Electricity-related GHG emission intensity declines over time under the renewable deployment settings of Japan (Section 12 in SI1), decreasing from 131 g CO₂-eq MJ^–1 to 20 g CO₂-eq MJ^–1 by 2050. We then use the assessment tool for the recycling of lithium-ion traction batteries provided by ref to quantify the associated decline in recycling-stage GHG emission intensity driven by grid decarbonization (Section 10 in SI1). The results indicate that recycling-stage GHG emission intensity decreases from 1.96 kg CO₂-eq per kg battery at present to 1.57 kg CO₂-eq per kg battery by 2050. For primary material production, we do not explicitly model process redesign; instead, we approximate reductions in material-specific emission factors using ranges reported in prior studies for key battery materials and map these reductions to the analysis horizon (Section 11 in SI1). , Other carbon emission factors, including direct emissions from battery manufacturing processes and the emission factor of natural gas, are assumed to remain constant over time. In addition, we conduct a sensitivity analysis on the underlying intensity assumptions and report the impact on total emissions and mitigation results. Total life cycle emissions are obtained by summing stage-level results across materials and processes.

Mitigation attributable to circular battery strategies is quantified by scenario contrast. The benefit of extending in-use battery lifetime in EVs is calculated as the difference between scenarios with different lifetime assumptions. The benefit of reuse is calculated as the difference between scenarios with and without reuse while holding other parameters constant. The benefit of recycling is calculated as the difference between scenarios with and without recycling under the same conditions.

Sensitivity Analysis

A baseline projection was established as the reference case for the sensitivity analysis. This scenario reflects a high-ambition pathway characterized by rapid EV deployment and strong growth in BESS capacity. It incorporates moderate technological improvements in battery manufacturing and recycling efficiency. The average in-use lifetime of EV batteries is assumed to be 8 years. Current market trends favoring LFP-based chemistries are maintained, with only LFP batteries assumed to be reused, while other chemistries are routed to recycling to recover valuable metals such as nickel and cobalt. This configuration represents the prevailing market conditions observed today.

Subsequent variations in individual parameters were introduced to examine how changes in specific assumptions influence overall GHG emissions and mitigation outcomes. The parameters and their corresponding scenario descriptions used in the sensitivity analysis are summarized in Table .

1. Summary of Parameter Variations and Corresponding Scenario Descriptions Used in the Sensitivity Analysis.

Change in parameter	Description
Shift to all reuse	Represents a case where all battery chemistries are eligible for reuse, reflecting no chemistry-based preference in second-life applications.
Shift to post LIB	Captures a gradual but significant transition from conventional NMC chemistries to high-Ni and postlithium-ion chemistries, where sodium-ion (SIB) and solid-state batteries (SSB) account for 55% of sales by 2050, NMC955 and NMC811 for 40%, and the remainder from other NMC and LFP (Section 6 in SI1).
Slower EV penetration	Reflects a slower increase in the EV fleet, consistent with current policy commitments without additional decarbonization efforts (business-as-usual scenario), whereby ICVs continue to dominate Japan’s market through 2050, accounting for approximately 40% of the vehicle stock (Section 5 in SI1).
Lower BESS demand	Represents short-duration flexibility primarily for intraday balancing, ramp control, and ancillary services, corresponding to a low energy-to-power configuration (Section 5 in SI1).
Lower failure rate	Reflects improved battery reliability and technology, resulting in a reduced failure ratio (10%) during the warranty period.
Lower collection rate	Models an imperfect collection system with a 90% recovery rate of spent batteries.
Longer functional lifetime	Reflects a high-technology scenario in which EV batteries degrade more slowly, requiring 15 years to reach the 80% SOH threshold.
Longer BESS lifetime	Represents enhanced BESS technology, extending the average lifetime to 10 years for reused LIBs and 25 years for new BESS installations.
Longer vehicle lifetime	Represents a shift toward greener consumer behavior, where vehicle replacement is delayed, extending the average vehicle lifetime to 16 years.
Moderate recycling efficiency	Describes a gradual improvement in recycling performance, where the recovery rates of key battery materials increase slowly and do not reach the ambitious targets set by the EU regulation by 2031, but only achieve them by 2050 (Section 10 in SI1).
Static GHG emission factor	Assumes no temporal reduction in GHG emission factors, effectively representing a static emission factor baseline.

Results

Projecting Future Battery Demand

Figure illustrates the results of the battery dynamics module, showing the projected evolution of EV stocks, LIB demand for EVs and BESS, the generation of spent EV batteries, and the availability of reusable EV LIBs in Japan from 2022 to 2050. According to the net-zero pathway, BEVs are expected to account for 100% of new vehicle sales by 2050. This rapid transition to electrified transport leads to a sharp increase in the number of BEVs, which is projected to grow by approximately 142 times and reach 58 million units by 2050 in Japan. In addition, this rapid electrification substantially increases battery demand in the transport sector. As shown in Figure b, the LIB demand for EVs is projected to reach 268–389 GWh by 2050. Correspondingly, the volume of spent EV batteries increases to 201–334 GWh by 2050. The wide range reflects uncertainty associated with different battery in-use lifetime assumptions, which strongly influence the replacement cycle, new battery demand, and EoL output. A longer in-use lifetime reduces total demand and EoL output by lowering the replacement frequency, whereas a shorter lifespan results in higher demand and EoL output due to earlier retirements.

2.

Projected evolution of EV and BESS battery dynamics in Japan. (a) EV stocks by powertrain type; (b) EV LIB inflows (annual demand) and outflows under different in-use lifetime assumptions; (c) Total BESS capacity under low, medium, and high demand; (d) Annual BESS demand and potential supply of reusable EV batteries.

For the stationary energy storage sector, the total installed capacity of BESS is projected to rise to 1359, 2700, and 4050 GWh under the low, medium, and high demand scenarios, respectively (Figure c). The corresponding annual addition of new batteries reaches approximately 149, 204, and 258 GWh by 2050, as shown in Figure d. This figure also presents the potential supply of reusable EV batteries, estimated based on the number of spent EV batteries and the degradation curves of their remaining capacities over time. Batteries whose in-use life exceeds the defined functional lifetime are excluded, as their capacities fall below 80% of the initial value and are therefore considered unsuitable for reuse from an economic perspective. Under the 8-year in-use lifetime assumption, the available reusable EV batteries reach approximately 225 GWh and could fully meet the total annual BESS demand by 2050 in the low and medium demand scenarios and approximately 87% of the demand in the high scenario. Under the 10-year in-use lifetime assumption, reusable EV batteries reach approximately 167 GWh and could cover the entire low-demand BESS requirement and 65% of the high-demand case. With longer in-use lifetimes of 12 and 14 years, reusable EV batteries reach approximately 118 GWh and 100 GWh, which could satisfy approximately 46 and 39% of the high BESS demand case, respectively. These findings highlight the substantial potential of second-life EV batteries to alleviate the need for new BESS production.

Reuse Impacts on Material Inflow and Outflow

Figure presents the impacts of battery reuse on material demand and waste generation under NCX and LFP chemistry scenarios, assuming an 8-year in-use lifetime of EV batteries and high BESS demand. Under the LFP scenario (Figure a and b), implementing reuse substantially reduces both total material demand and waste generation. Specifically, the total material demand decreases from 47.7 to 43.8 Mt, representing an 8.2% reduction, while the total waste generation for recycling declines from 21.1 to 16.3 Mt, corresponding to a 22.7% reduction. In contrast, under the NCX chemistry scenario (Figure c and d), both the total material demand (35.9 Mt) and waste generation (16.5 Mt) are inherently lower than those of the LFP scenario, reflecting the strong influence of battery chemistry on the overall material balance. However, because the reuse scenario assumes that only LFP batteries are eligible for second use, the reuse effect in the NCX case remains minimal, resulting in only a 0.2% reduction in material demand and a 1.0% reduction in waste generation.

3.

Cumulative material flows of batteries in Japan from 2022 to 2050 under four scenarios: (a) LFP chemistry without reuse; (b) LFP chemistry with reuse; (c) NCX chemistry without reuse; (d) NCX chemistry with reuse. The Sankey diagrams present total material demand, stock accumulation in EV and BESS applications, and eventual waste to recycling across major battery components, including cathode materials, carbon black (CB), graphite, electrolyte, and other structural materials. All scenarios assume an 8-year in-use lifetime of EV batteries and high BESS demand.

It is also noteworthy that both the inflow and outflow of BESS are higher under the reuse scenario than in the baseline case. This is mainly because reused batteries have a lower residual capacity, typically around 80% of their original capacity, so a larger quantity of batteries is required to meet the same storage demand. In addition, reused batteries generally exhibit shorter lifetimes compared to newly manufactured BESS batteries, leading to more frequent replacements and consequently higher overall inflow and outflow volumes throughout the study period.

Impacts of Recycling on Primary Material Demand

We further investigated the effects of recycling after EV batteries, reused batteries, and newly manufactured energy storage batteries reach their EoL. Figure illustrates the resulting changes in material demand under the LFP and NCX chemistry scenarios, assuming an 8-year in-use lifetime of EV batteries and high BESS demand. Under the LFP chemistry scenario, the total demand for battery production materials is estimated to be 55.9 Mt during the period from 2022 to 2050. Primary materials remain the dominant source throughout the period, although the share of secondary (recycled) materials gradually increases, contributing an annual average of 23.4% between 2022 and 2050. Under the NCX chemistry scenario, the total demand for battery production materials is estimated to be 45.3 Mt during the same period, with secondary materials contributing a similar annual average share of 40.2%. The different trends observed between the two chemistry scenarios can be attributed to the higher recycling rates of nickel and cobalt in NCX batteries, while recycling in the LFP case is limited primarily to lithium recovery. Extending the in-use lifetime of EV batteries can further reduce the total demand for both production and primary materials; however, the effects of extended in-use lifetimes, reuse, and recycling interact in ways that create substantial trade-offs rather than cumulative benefits (Figures S9–S10 in SI1).

4.

Projected total material demand and secondary material supply for battery production in Japan from 2022 to 2050. (a) and (b) are under LFP chemistry scenarios; (c) and (d) are under NCX chemistry scenarios, assuming an 8-year in-use lifetime of EV batteries and high BESS demand. Lithium salts: Li₂CO₃ and LiOH.

Figure b and d provide a breakdown of material-specific contributions, focusing on lithium salts, nickel sulfate, and cobalt sulfate. Under the LFP chemistry scenario, the proportion of secondary lithium salts, nickel sulfate, and cobalt sulfate is projected to reach 47, 70, and 74%, respectively, by 2050. In contrast, under the NCX chemistry scenario, these ratios are expected to reach 40, 66, and 69%, respectively. Overall, the results indicate that while recycling can substantially offset primary material demand, most battery production materials will continue to depend heavily on virgin resources by 2050.

GHG Emissions and Mitigation Benefits

Figure illustrates the system-wide GHG emissions under the LFP and NCX chemistry scenarios, highlighting the effects of extending the in-use lifetime of EV batteries, reuse, and recycling on emission reduction. In the absence of reuse and recycling, the differences in GHG emissions across various battery lifetime scenarios are pronounced. Under the LFP scenario, system GHG emissions in 2050 range from 12.3 to 15.2 Mt, depending on the assumed lifetime. In the NCX scenario, the corresponding range is from 13.9 to 17.2 Mt by 2050. When both battery reuse and recycling are implemented, total GHG emissions decrease substantially, and the variations among different lifetimes become much smaller, converging to 7.4–8.4 Mt by 2050 under the LFP scenario and 7.8–8.9 Mt by 2050 under the NCX scenario. Considering cumulative emissions over the entire period from 2022 to 2050, the LFP scenario results in 310–378 Mt of GHG emissions, while the NCX scenario produces 324–399 Mt. With the adoption of both reuse and recycling, the cumulative emissions can be further reduced to 243–270 Mt under the LFP scenario and 246–279 Mt under the NCX scenario.

5.

Projected annual and cumulative system-wide GHG emissions in Japan from 2022 to 2050. (a) and (b) are under LFP chemistry scenarios; (c) and (d) are under NCX chemistry scenarios, assuming an 8-year in-use lifetime of EV batteries and high BESS demand. Panels a and c show the annual GHG emissions across different in-use lifetime scenarios of EV batteries. Solid lines represent the case without reuse or recycling (baseline), while dotted lines correspond to both reuse and recycling (RU+RC, dotted lines). Panels b and d present cumulative GHG emissions from 2022 to 2050 for in-use lifetime scenarios of EV batteries, highlighting the reduction potential from reuse (dark gray bars), recycling (light gray bars), and their combined effect (hatched bars). Percentages above the bars indicate the relative reductions compared to the baseline.

A further breakdown of mitigation contributions indicates that recycling provides the dominant emission reduction, while reuse plays a supporting role by extending the useful life of batteries and delaying the demand for new production. Specifically, under the LFP scenario, reuse contributes approximately 4–7% of the total mitigation, whereas recycling accounts for 18–21%. In the NCX scenario, reuse contributes less than 1%, while recycling delivers a substantially larger share of the reductions, ranging from 24 to 30%. Overall, achieving the lowest possible GHG emissions requires the combined implementation of reuse, recycling, and longer in-use lifetimes of EV batteries. In Figures S12–S14 (SI1), we further present the contributions of different life cycle stages to total GHG emissions. The results show that with longer in-use lifetimes, emissions from all stages decline, while reuse and recycling mainly reduce emissions associated with the material production stage.

Sensitivity Analysis of GHG Emissions and Mitigation Benefits

Figure presents the sensitivity of GHG emissions and circular mitigation outcomes to changes in key model parameters. In the upper heatmap, both baseline and residual GHG emissions respond strongly to variations in several key drivers, most notably the shift toward PLB, slower EV market penetration, and lower BESS demand. These changes lead to total system-level variations of about 30% relative to the reference projection, underscoring the sensitivity of overall emissions to assumptions governing future technology composition and market dynamics. Parameters related to the failure rate, collection rate, BESS lifetime, and vehicle lifetime exhibit relatively minor effects (<10%), suggesting that their influence on system-wide emission outcomes is limited.

6.

Sensitivity of accumulated (2022–2050) GHG emissions to alternative assumptions. Heatmaps report percentage changes relative to the baseline reference case. (a) Changes in accumulated GHG emissions, shown separately for Baseline (total emissions without implementing reuse and recycling) and Residue (the remaining emissions after implementing reuse and recycling). Negative values indicate lower total emissions than the baseline reference case, while positive values indicate higher emissions. (b) Changes in accumulated GHG mitigation benefits, disaggregated into Reuse (mitigation attributable to the reuse of batteries in BESS) and Recycling (mitigation attributable to battery recycling). Negative values indicate reduced mitigation benefits relative to the baseline reference case, while positive values indicate increased mitigation benefits. In both panels, darker colors indicate better outcomes, and lighter colors indicate worse outcomes.

The lower heatmap further differentiates the mitigation contributions from reuse and recycling, revealing distinct yet interlinked sensitivities. The reuse-related climate benefit is highly responsive to assumptions regarding chemistry eligibility, market evolution, and lifetime behavior. Allowing all chemistries to be reused, shifting toward SIB and SSB, or accelerating EV market penetration markedly amplifies reuse benefits, with variations exceeding ± 70%. These results demonstrate that reuse performance is closely tied to the diversity of chemistries and the pace of electrification. Lower failure rates and longer BESS lifetimes enhance the reuse benefit but simultaneously reduce the subsequent recycling benefit, illustrating a temporal trade-off between extending in-use lifetimes and advancing the recovery of critical materials. The recycling benefit increases primarily with shifts toward SIB and SSB chemistries and with faster EV fleet expansion, both of which enlarge the future pool of recyclable materials. In contrast, applying a static emission factor, which neglects future grid decarbonization, inflates both reuse and recycling benefits.

Discussion

With the rapid pace of vehicle electrification and renewable energy deployment, battery demand is expected to continue increasing. This expansion will inevitably intensify the need for critical materials and increase GHG emissions associated with battery production. Therefore, circular battery strategies are of great value for sustainable energy transition. Our findings indicate that even under a relatively stringent reuse threshold, defined here as spent EV batteries with more than 80% remaining capacity, retired EV batteries in Japan still have considerable potential to contribute to grid-scale energy storage. However, there might be a significant potential timing mismatch between EV battery retirement and BESS deployment. When BESS demand grows faster than the availability of retired EV batteries, reuse potential will be constrained by supply limitations, especially when in-use battery lifetimes in EVs are extended. Conversely, if EV battery retirements accelerate ahead of storage deployment, a temporary surplus of reusable batteries may emerge. This temporal mismatch is not unique to Japan; it is evident in other countries as well and is reflected in regional differences. ,, Such temporal misalignment complicates the coordination between the transport and power sectors and underscores the need for integrated planning that synchronizes EV fleet turnover, battery collection, and storage expansion.

Our dynamic MFA model results reveal that extending EV battery lifetimes during the early stages of circular economy system development will achieve more benefits of GHG emission mitigation. These findings highlight that, when collection, testing, remanufacturing, and recycling infrastructure are still underdeveloped, extending battery lifespan may be the most practical option for reducing environmental impacts. Moreover, a longer lifespan delays the accumulation of battery waste and reduces the pressure on underdeveloped EoL systems, which also allows time for governments and industries to build the necessary technical, regulatory, and logistical foundations for managing spent batteries effectively. As reuse and recycling systems mature and efficiencies improve, however, the performance gap between different lifetime assumptions narrows. In a fully developed circular system, earlier battery retirement could become advantageous if advances in cell chemistry and manufacturing substantially lower the environmental impacts of new battery production, allowing recovered materials to re-enter supply chains more rapidly. However, these potential advantages are highly context-dependent, and future research should examine when accelerated retirement may outperform life-extension strategies as circular infrastructures continue to develop.

Our findings further highlight that, while reuse temporarily limits the flow of EoL batteries into recycling, the integration of reuse and recycling ultimately achieves higher GHG reduction compared with a recycling-only pathway. In addition, by delaying large-scale recycling activities, reuse can also provide valuable time for the industry to enhance recycling efficiency and expand processing capacity. However, several technical and economic barriers persist. The heterogeneous SOH of retired batteries complicates reuse, safety certification systems are underdeveloped, and the rapid decline in new battery costs erodes the financial competitiveness of reuse. ,, Moreover, when EoL batteries are severely degraded, recycling may become both economically and environmentally preferable, particularly when high market prices for cobalt or nickel increase the value of recovered materials. Regulatory frameworks such as the EU Battery Regulation, which mandates minimum collection and recycling targets, may further accelerate recycling investments while diverting attention from reuse. To reconcile these dynamics, transparent battery data systems and harmonized safety standards are crucial to ensure operational reliability and strengthen investor confidence in second-life applications.

Battery chemistry plays a key role in shaping both GHG emissions and the mitigation potential of circular battery strategies. Nickel-based batteries offer greater recycling value, yet their limited chances for reuse may result in missed opportunities for greater emissions reductions. In contrast, LFP batteries have advantages in the reuse market; however, the longer service life delays the timing of recycling, which may create short-term profitability challenges for recycling companies. Looking ahead, next-generation battery chemistries such as SSBs may offer improved durability during use, but their EoL pathways remain uncertain. The solid electrolyte complicates material separation, and the overall reuse potential of SSBs has not yet been clearly established. These differences may affect both GHG emissions and the overall effectiveness of circular battery strategies. Therefore, policymakers need greater flexibility to accommodate the system-level changes introduced by different battery chemistry pathways.

Assessing how GHG emission factors evolve over time is essential for understanding the impacts of circular pathways on environmental sustainability. As renewable energy deployment expands, the emission intensity of primary production is expected to decline, narrowing the emissions differential that underpins the climate “credit” attributed to reuse and recycling. However, depletion of high-grade deposits is also driving a shift toward lower-grade ores, often in remote or geologically complex settings, which can reduce energy efficiency and increase the GHG emissions of extraction. , This trend toward lower-grade and more challenging resources increases the marginal environmental burden of primary supply, potentially increasing the mitigation potential associated with circular pathways. To enable robust evaluation, future research should adopt an integrated, dynamic perspective that explicitly incorporates key determinants of primary and secondary production emission factors, such as ore grade, processing technologies, and energy mix.

Our study is based on several assumptions regarding key parameters that influence both the demand for battery materials and environmental performance. First, we assume consistent performance degradation behavior across all battery chemistries, although in practice, LFP batteries generally exhibit longer in-use lifetimes than NMC batteries. In addition, due to the lack of publicly available Japan-specific battery chemistry scenarios, we adopt global chemistry pathways as a proxy, although actual chemistry transitions may vary across countries under different local policy and market conditions. Consequently, the timing and condition of battery retirements may deviate from the modeled patterns, particularly for LFP batteries, which could influence the volume and the remaining capacity of retired packs available for second-life energy storage. Second, our analysis considers only one form of battery reuse and does not account for alternative second-life pathways or the possibility of a third-life stage, such as redeploying batteries to additional applications after their use in BESS. This limitation may lead to an underestimation of the total reuse potential and the associated emission-reduction benefits, as additional cascaded uses could further extend battery in-use life. Third, improvements in cell design and manufacturing efficiency can reduce the material intensity of battery production and enhance energy efficiency without compromising performance. These advancements could influence the relative mitigation potential of reuse and recycling, as reduced material inputs diminish the marginal benefits achievable through secondary use and material recovery. Additionally, while we assume high recovery and reuse rates to test their potential, technological uncertainties and regulatory constraints, such as the mechanical preprocessing of batteries into black mass for export, could result in different real-world performance, introducing complexities in value capture, environmental burden distribution, and material circularity. These factors collectively highlight the need for further research to address the evolving technological and regulatory landscape.

Using Japan as a case study, this research illustrates how the interplay among chemistry diversification, lifetime behaviors, energy system evolution, and EoL pathways collectively shapes the sustainability and resilience of future battery supply chains. The analytical framework developed here provides a transferable approach for evaluating dynamic interactions between material circularity and GHG mitigation under diverse regional contexts. As global electrification accelerates, integrated and adaptive strategies, which link lifetime management, chemistry transition, and EoL pathways, will be essential for balancing material security, technological feasibility, and climate goals within the rapidly evolving circular systems.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.est.5c16978.

• Model details, electric vehicle demand and battery energy storage system deployment scenarios, battery chemistry scenarios, model input data, and additional figures (SI1) (PDF)

• Life cycle inventory of cell production and material and energy inputs for battery pack manufacturing (SI2) (XLSX)

Conceptualization: Z.H., G.X., K.N.; methodology: Z.H., G.X., X.S.; project administration: K.N., G.X.; original draft: Z.H., G.X.; review and editing: K.N., S.N., M.F., X.S.

The authors declare no competing financial interest.

Published as part of Environmental Science & Technology special issue “Advancing a Circular Economy”.