Electric light-duty vehicles have decarbonization potential but may not reduce other environmental problems

Electric light-duty vehicles have decarbonization potential but may not reduce other environmental problems | Communications Earth & Environment

Abstract

Electric vehicles are promoted as ‘clean’ technologies and offer promising reductions in transportation emissions. Nevertheless, their environmental benefits critically depend on the local electricity grid mix and the type of emission being considered. Here, we conduct a comparative life cycle assessment of the four dominant light-duty vehicle categories at both the global scale and in three representative countries: Norway, the US, and China. By analyzing different environmental indicators, particularly global warming potential and respiratory effects, and quantifying related parametric uncertainties, we reveal that the advantages of electric vehicles vary across these regions and across environmental impact types. While electric vehicles offer considerable decarbonization potential as the grid mix becomes cleaner, they might not mitigate other environmental impacts, such as increased respiratory effects on rural, low-income communities. Our results support stakeholders in identifying environmentally friendly vehicle and policy options while considering multiple factors, and emphasize the importance of tailored approaches over one-size-fits-all solutions in sustainable transportation.

Introduction

Transportation accounts for roughly 25% of energy-related greenhouse gas emissions 1. It also experienced the most substantial growth rate in greenhouse gas emissions among all end-use industries, with an average growth rate of 1.7% per year during the period 1990–2022 2. Without relevant interventions, transportation-related emissions are expected to increase by 16–50% (medium-confidence prediction) in 2050 compared to 2020 3. Light-duty vehicles, mainly used for commuting or daily activities, are the primary source (35.6%) of transport emissions worldwide 4. To mitigate the rising greenhouse gas emissions from the transportation sector, the Intergovernmental Panel on Climate Change (IPCC) promotes electric vehicles (EVs) 3, including battery electric vehicles (BEVs) and plug-in hybrid electric vehicles (PHEVs) 5.

The global EV market has increased substantially in recent years with growth projected to continue. Annual EV sales exceeded 10 million in 2022, an increase of 55% from 2021, accounting for 14% of new car sales that year 6. Among all EV sales, more than 70% were BEVs. By 2030, EVs are projected to account for 35% of new car sales 5.

One motivation behind the rapid growth of the EV market is BEVs’ potential to decarbonize the transportation sector, aligning with the IPCC strategy. This potential primarily derives from the lack of tailpipe emissions during battery-powered vehicles’ operation and the ability to charge the batteries via a low-carbon electricity grid 3. However, despite their reputation as ‘clean vehicles,’ 7 it is imperative to acknowledge the potential variability in their well-to-wheel emissions which are contingent upon the fuel mix of the local electricity grid 8. As we will demonstrate, in some cases, the lifecycle emissions of EVs can be greater than those of hybrid electric vehicles (HEVs) or even internal combustion engine vehicles (ICEVs).

Considerable attention in academic discourse has been devoted to analyzing the fuel mix of electricity grids. In countries like France and Brazil, where the electricity grid relies more on low-emission sources, BEVs are considered sustainable and reliable for countries’ carbon emission reduction targets 9. Conversely, in regions that depend heavily on fossil fuels, particularly coal, for electricity production, BEVs exhibit increased environmental emissions 10; for instance, studies have found that BEVs in Australia have higher carbon emissions than ICEVs. Reviews across ten countries and the European Union indicated that electricity grid mixes account for 60% of EVs’ lifetime carbon emissions variations 11, highlighting the dynamic nature of the evolving grid and its implications for the long-term sustainability of EVs.

Global warming potential (GWP) is a major focus within the EV literature. Except in countries heavily dependent on fossil fuels, EVs show promise in reducing GWP 12, 13, 14, 15, 16, 17. Studies have shown that despite high initial production emissions, BEVs achieve a net 36% reduction in lifecycle impact compared to ICEVs 12. Likewise, research focusing on the US and European Union concluded that BEVs offer a 50% reduction in greenhouse gas emissions compared to ICEVs 14, 15. In China, BEVs were reported to achieve a 32% decrease in CO2 emissions 16.

Alongside GWP, countries aim to mitigate the toxic impact of transportation emissions on human health, especially to minimize respiratory effects (RE). In the literature, RE associated with acute exposure to fine particulate matter (PM) 18 is increasingly recognized but remains relatively underexplored. Studies have demonstrated that a 27% fleet electrification in key Chinese regions by 2030 could prevent up to 17,456 premature deaths, highlighting a major health benefit due to reduced PM2.5 exposure. Similarly, one life cycle assessment (LCA) study on light-duty vehicles has found that EVs can substantially reduce carbon dioxide, volatile organic compounds, and nitrogen oxide emissions but can also lead to increased PM2.5 and sulfur dioxide emissions 19. By combining other impact categories, such as human toxicity, with GWP, researchers create a more holistic understanding of EV sustainability.

Environmental justice is another primary concern for EV implementation, as their emissions are unevenly distributed, affecting regions with lower average incomes differently than the areas that primarily use EVs. A comprehensive analysis of the EV deployment necessitates the comparison of uneven distribution of impact categories. Even though urban populations tend to benefit more from EVs’ zero tailpipe emissions, poor or rural populations experience more environmental burdens 20. Studies have estimated that urban areas have up to a 66% higher rate of avoided premature deaths per million people—a notable benefit—compared to rural areas; however, approximately 52% of emissions from urban EV usage, primarily from electricity production, are impacting non-urban populations, concentrated in rural areas where land is cheaper, and population density is lower 21. Research on the introduction of EVs into 34 major cities in China found EVs greatly exacerbated disparities in environmental justice 22. Further, this study has shown that the poorest regions experience a 40% greater burden of pollutant exposure from EVs compared to the wealthiest areas

In the current study, we conduct a comprehensive life cycle assessment, comparing the lifecycle environmental impacts of four dominant light-duty vehicle categories: ICEVs, HEVs, PHEVs, and BEVs. This analysis addresses critical gaps in existing literature, which predominantly focuses on GWP as the primary impact metric. By expanding to include other impact metrics (e.g., respiratory effects, acidification, and eutrophication), we aim to determine whether these metrics follow similar trends to GWP or whether they reveal notable deviations or trade-offs. Our study also provides a broader global perspective rather than a narrow focus on individual countries in existing research. Although other countries are also important in the global EV landscape, we select three prominent countries leading the global EV market—Norway, the US, and China. Moreover, owing to their distinct geographical locations, transportation profiles, and electricity grid mixes (see Fig. 1), these countries serve as representative case studies. For instance, Norway is an ideal example because it is overwhelmingly powered by hydroelectricity, resulting in a grid that is low in GWP and RE impact per kWh of electricity. Our approach enables a more nuanced examination of the benefits and challenges associated with EV adoption, providing valuable insights that can be applied broadly to other countries worldwide.

The stacked bars show grid combinations by source for Global, Norway, the US, and China. Values are shown in percentages to represent the corresponding share of each energy source (e.g., coal, hydro, and wind) in the total electricity generation for these regions. Only sources contributing more than 5% are labeled with their respective percentages in the chart. The table at the bottom presents the lifecycle global warming potential and respiratory effects emissions for 1 kWh of low-voltage electricity delivered to the end-user side. The values include both direct emissions from the power plant and upstream emissions from the power plant and fuel production. For other impact results, such as acidification and eutrophication, please refer to Supplementary Note 3. Grid mix generation ratios for the current scenarios are derived from IEA 2020 data 48, and the ratios for the 2030 STEPS and APS scenarios are based on the IEA World Energy Outlook 2023 28.

To ensure we accurately capture the environmental benefits of EVs, we have considered the ongoing shift of the grid towards cleaner energy in our model. By assuming a linear trend in grid mix evolution and considering that vehicles typically have a service life of around 12 years, we use the 2030 grid mix scenario as a representative “mid-point.” This mid-point falls between the vehicle’s manufacturing and start of service in 2024 and its end-of-life around 2036. Essentially, the 2030 grid mix scenario can be used to estimate the environmental impacts of electricity usage over the entire service life cycle of EVs. This approach allows us to predict the most likely future pathway for EVs from today, while also accounting for the gradual transition of the grid mix.

To achieve this, we leverage the Ecoinvent 3.7 database 23 to obtain electricity-related input. Given that LCA is typically retrospective 24, we first present the results of the “current scenario” based on actual data from Ecoinvent for global, Norway, the US, and China. This provides a reference point for the readers, allowing them to understand the current situation if the grid mix does not change notably in the future. It also helps avoid potential inaccuracies in future trajectory predictions that could mislead them 25, 26, 27. Then, we model the environmental impacts of electricity generation specifically for the year 2030 for the US, and China, based on the forecast trajectories under the Stated Policies Scenario (STEPS) and the Announced Pledges Scenario (APS) 28 in the International Energy Agency (IEA)’s World Energy Outlook 2023 (see Supplementary Note 3). We use the 2030 STEPS scenario as the baseline, which represents the most likely results under conservative assumptions, and complement those with the results calculated using the 2030 APS scenario, which assumes relatively more aggressive changes in the grid mix 29.

We break up the vehicle lifecycle into the vehicle cycle (vehicle manufacturing, maintenance, and disposal) and fuel cycle (fuel extraction, combustion, and electricity production) for analysis, detailed in the Methods Section. Supplementary Note 10 provides results for all impact metrics. In the main text, we conduct a nuanced analysis of two impacts: GWP and RE, emphasizing GWP for its global significance and RE for a more localized impact as a counterpoint. Our results show that EVs can reduce carbon dioxide emissions as the electricity grid mix becomes cleaner, but they may not mitigate particulate matter emissions due to electricity generation. This study provides insights that can guide policymakers in crafting vehicle-specific environmental strategies, assist consumers in selecting environmentally sustainable vehicles, and provide industry and academic stakeholders with a comprehensive understanding of light-duty vehicle lifecycle emissions.

Results

This section first analyzes the environmental impact of the vehicle cycle (i.e., global warming potential and respiratory effects). Subsequently, we present the vehicle lifecycle global warming potential results under the current, baseline STEPS, and APS scenarios. We then show the results for one localized impact—respiratory effects—under these scenarios. Finally, we conduct a detailed analysis of the respiratory effects, splitting them into tailpipe and non-tailpipe emissions.

Vehicle cycle impact

Figure 2 shows a breakdown of the impacts of the vehicle cycle: Fig. 2a illustrates the global warming potential (GWP), and Fig. 2b depicts the respiratory effects (RE). The vehicle cycle includes everything except the fuel cycle operation of the vehicle over its lifetime 30. Each light-duty vehicle has a different environmental impact. BEVs have the highest global warming potential and respiratory effects. Their values are noticeably higher than those of PHEVs (44% and 33% more for GWP and RE, respectively), which in turn are greater than HEVs (18% and 13% more for GWP and RE, respectively). HEVs and ICEVs have similar impacts (below 5% variance for both GWP and RE). BEVs’ environmental footprint is attributed mainly to the batteries, contributing 50% of the global warming potential and 45% of the respiratory effects. In contrast, the smaller-capacity battery pack of PHEVs contributes to 17% and 15% of global warming potential and respiratory effects, respectively. Apart from battery-related effects, the environmental impacts of all other parts remain relatively consistent across all light-duty vehicles, except BEVs, which have considerably lower maintenance requirements compared to their internal combustion engine counterparts, primarily due to the absence of engine oil replacements (see Supplementary Table S2). In summary, the vehicle cycle of PHEVs and BEVs entails a considerable environmental impact, primarily due to batteries having emissions-intensive processes involved in the extracting and processing of materials, manufacturing, and assembly 31. Our objective in the following sections is to explore under what circumstances BEVs offset the higher environmental impact associated with their vehicle cycle and potentially outperform ICEVs, HEVs, and PHEVs across their entire lifecycle. This includes an in-depth comparison of the lifetime environmental impact of these vehicles under different use cases.

a Global warming potential (kg CO2 eq). b Respiratory effects (kg PM2.5 eq). Environmental impacts for each vehicle type are presented in six stacked sections: components, assembly, batteries, fluids, maintenance, and disposal. The total impact of each vehicle type is displayed at the top of its respective column. Only the values for the sections that contribute more than 5% to the total impact are shown within these columns. Percentages in parentheses indicate the proportion of each section’s contribution to the vehicle’s total impact.

We identify two factors that introduce substantial variability in the results across these four regions: BEV electricity consumption and PHEV electric driving share (see Methods). This main text focuses on analyzing one factor: PHEV electric driving share. For a detailed analysis of the other factor, BEV electricity consumption, refer to Supplementary Note 8.

Lifecycle global warming potential—current scenarios

We combine the vehicle cycle impacts (Fig. 2) with the fuel cycle impacts to calculate the full lifecycle impacts of each light-duty vehicle. Then, we compare the full lifecycle environmental impacts of these vehicles using two important parameters: the first is the lifetime vehicle driving distance [km], and the second is the PHEV electric driving share [%], i.e., the percentage of driving done by PHEV that is powered by the battery versus the gasoline engine. In Fig. 3, we map these two parameters to create a landscape split into different colored areas to denote which vehicle (pink for ICEV, blue for HEV, yellow for PHEV, and green for BEV) has the lowest global warming potential for the combination of these two parameters. This landscape is repeated for different regions: global (Fig. 3a), Norway (Fig. 3b), the US (Fig. 3c), and China (Fig. 3d). The electricity grid mixes for these four regions are based on the current scenarios. For a description of how this plot was developed, see the Methods Section.

Each panel shows the current scenarios of the vehicle type with the lowest global warming potential based on vehicle lifetime driving distance (horizontal axis) and PHEV electric driving share (vertical axis) for a globe, b Norway, c the US, and d China. The colored areas in each panel denote which vehicle type has the lowest global warming potential for that combination of the two variables: pink (ICEV), blue (HEV), yellow (PHEV), and green (BEV). Horizontal dashed lines denote the average PHEV electric driving share for each country or region 32. The global value (39%) is calculated based on the weighted proportions of the individual regions. The vertical dashed line in the US represents the calculated average vehicle lifetime driving distance 36.

Firstly, it is worth highlighting that across all four regions, ICEVs capture a noticeably small fragment of the landscape, with their share varying by region but consistently below 1700—lifetime vehicle kilometers. For all other vehicles, substantial disparities exist among these four regions.

For the global region (Fig. 3a, average grid emissions of 0.72 kg CO2-eq kWh−1), BEVs have the lowest global warming potential for only a small portion of the landscape, above 270,000 km and below a PHEV electric driving share of 26–31%. The rest of the landscape is almost evenly split between PHEV and HEV, with HEV dominating for low lifetime vehicle driving distance and PHEV electric driving share (toward the bottom left of the landscape), and PHEV dominant elsewhere.

Moving to a relatively cleaner grid, as represented by the US (Fig. 3c, average emissions of 0.55 kg CO2-eq kWh−1), we observe that BEVs capture more of the landscape, winning out over HEVs above a lifetime driving distance of 82,000 km and over PHEVs when the PHEV driving share is between 27 and 70%, depending on the lifetime driving distance.

In Norway where the grid is the cleanest among all four regions (Fig. 3b, average emissions of 0.02 kg CO2-eq kWh−1), BEVs clearly dominate the landscape, beating out HEVs over 26,000 lifetime vehicle kilometers and PHEVs even up to a PHEV electric driving share of 94% (again depending on the lifetime vehicle kilometers). This trend suggests that as the proportion of low-carbon energy in the electricity grid increases, the environmental advantages of BEVs increase as evidenced by the expansion of BEVs’ colored area towards lower driving distance values in the plot. This shift indicates that BEVs, powered with a cleaner energy mix, offset battery production’s environmental impact at lower driving distance compared to using higher-emission electricity grids for charging, i.e., BEVs’ ‘clean tech’ status is a function of both vehicle and grid together.

In China’s dirty grid (Fig. 3d, average grid emissions of 1.04 kg CO2-eq kWh−1), HEVs strongly dominate. Both BEVs and PHEVs fail to offset the extra environmental burden of battery production. Contrary to expectations, driving battery-powered kilometers results in a higher global warming potential than driving gasoline-powered kilometers. This is due to the high greenhouse gas emissions associated with electricity production from the Chinese grid. This cumulative effect escalates with increasing driving distance, resulting in the total impact of BEVs and PHEVs consistently exceeding that of HEVs (the sum of the vehicle and fuel cycle emissions). This issue is further discussed in Supplementary Note 6 and Note 7.

Lifecycle global warming potential—STEPS and APS scenarios

In this section, we present the most likely future pathway results for the global warming potential of PHEVs and BEVs (i.e., baseline STEPS scenarios), analyzed together with ICEVs and HEVs, complemented by results from the APS scenarios.

Figure 4 shows the global warming potential results under the baseline STEPS scenarios in the US (Fig. 4a) and China (Fig. 4b), and under the APS scenarios in the US (Fig. 4c) and China (Fig. 4d). In the US, BEVs dominate the landscape, beating out HEVs over 36,200 lifetime vehicle kilometers and PHEVs up to a PHEV electric driving share of 89% under the baseline STEPS scenario (0.24 kg CO2-eq kWh−1). Similarly, BEVs beat out HEVs over 32,600-lifetime vehicle kilometers and PHEVs up to a PHEV electric driving share of 90% under the APS scenario (0.18 kg CO2-eq kWh−1).

a Baseline STEPS scenario of the US. b Baseline STEPS scenario of China. c APS scenario of the US. d APS scenario of China. Each figure shows the vehicle type with the lowest global warming potential based on vehicle lifetime driving distance (horizontal axis) and PHEV electric driving share (vertical axis). The colored areas in each panel denote which vehicle type has the lowest global warming potential for that combination of the two variables: pink (ICEV), blue (HEV), yellow (PHEV), and green (BEV). Horizontal dashed lines denote the average PHEV electric driving share for each country or region 32. The vertical dashed line in the US represents the calculated average vehicle lifetime driving distance 36.

In contrast, in China, BEVs outperform HEVs at longer lifetime vehicle kilometers and PHEVs up to a lower PHEV electric driving share. Specifically, BEVs outperform HEVs at 90,600 lifetime vehicle kilometers and PHEVs up to a PHEV electric driving share of 67% under the baseline STEPS scenario (0.58 kg CO2-eq kWh−1), and at 70,500-lifetime vehicle kilometers and a PHEV electric driving share of up to 74% under the APS scenario (0.52 kg CO2-eq kWh−1). In both countries, the greater landscape expansion of BEVs in the APS scenarios compared to the baseline STEPS scenarios is attributed to a further decline in the average carbon emissions from the grid mix. We also provide a detailed analysis of vehicles with a fixed lifetime distance of 320,000 km (Supplementary Fig. S4).

Also shown in Fig. 3 and Fig. 4 are the average values for the PHEV electric driving share in the different regions (dashed horizontal lines) and the average lifetime vehicle driving distance in the US (dashed vertical lines). Taking the US as an example (Fig. 4a and Fig. 4c), for the average lifetime driving distance of 241,400 km (vertical dash line) and the average PHEV electric driving share in the US of 54% 32, BEVs have the lowest global warming potential (see the intersection of vertical and horizontal dashed lines in the figure). For the ‘average’ driver, BEV stands out as the lowest-carbon emission option. Consumers should evaluate if their personal use aligns with these average conditions, such as access to workplace charging stations. Policymakers could explore options to change the average scenarios, like increasing public charging station accessibility. These topics are further elaborated in the Discussion Section.

Lifecycle respiratory effects—current scenarios

Turning to a localized impact, Fig. 5 shows which vehicle type has the lowest respiratory effects under current scenarios. In the four regions examined, ICEVs have the lowest respiratory effects up to a specific vehicle distance threshold (ranging from 9200 to 10,000 km). Beyond these thresholds, HEVs demonstrate the lowest respiratory effects in the globe, the US, and China. This is because electricity production in these regions has a higher respiratory effect than gasoline use (i.e., the well-to-tank process). The gasoline use emissions include gasoline combustion during vehicle operation and upstream emissions required for gasoline production 33. Therefore, despite the widespread promotion of BEVs, HEVs offer better environmental benefits in terms of respiratory effects. However, Norway is an exception, resembling the results in Fig. 3b due to the country’s distinctive electricity grid, which primarily relies on hydropower with minimal respiratory effects.

Each panel shows the current scenarios of the vehicle type with the lowest respiratory effects based on vehicle lifetime driving distance (horizontal axis) and PHEV electric driving share (vertical axis) for a globe, b Norway, c the US, and d China. The colored areas in each panel denote which vehicle type has the lowest respiratory effects for that combination of the two variables: pink (ICEV), blue (HEV), yellow (PHEV), and green (BEV). Horizontal dashed lines denote the average PHEV electric driving share for each country or region 32. The global value (39%) is calculated based on the weighted proportions of the individual regions. The vertical dashed line in the US represents the calculated average vehicle lifetime driving distance 36.

Lifecycle respiratory effects—STEPS and APS scenarios

Similar to the global warming potential analysis, we examine the respiratory effects of light-duty vehicles under the baseline STEPS and APS scenarios in the US and China. However, the pattern of results in both scenarios remains unchanged from the current scenarios (see Supplementary Fig. S11), with HEVs continuing to strongly dominate.

Tailpipe vs. non-tailpipe respiratory effects

Given the localized nature of respiratory effects, which differ from the global impact of global warming potential, we split respiratory effects between the tailpipe and non-tailpipe emissions, which occur in different locations. This approach allows us to better understand the environmental and health implications at a local level.

We fix the vehicle lifetime distance at 320,000 km and break down respiratory effects impact into the tailpipe and non-tailpipe emission categories. The non-tailpipe emissions include the vehicle cycle, the well-to-tank process (which includes crude oil extraction, refining, storage, transportation, etc.) 34, and electricity generation. We consider both the baseline STEPS and APS scenarios for the grid mix in the US and China.

As shown in Fig. 6, respiratory effects for BEVs and PHEVs under APS scenarios are lower than under baseline STEPS scenarios in both countries due to reduced respiratory effects per kWh of electricity. However, the total respiratory effects from BEVs and PHEVs remain higher than those of HEVs (10.3 kg PM2.5 eq in the US and 10.1 kg PM2.5 eq in China). Surprisingly, the tailpipe respiratory effects constitute a small portion (less than 4%) of the total. This pattern is also observed for other impact categories except for global warming potential (See Supplementary Note 4). Notably, the primary sources of total respiratory effects are non-tailpipe emissions, mainly from electricity generation (for electric driving) and well-to-tank processes (for gasoline driving). Moreover, we note that PHEVs, when relying more on electricity, can increase total respiratory effects compared to gasoline. For example, under the US baseline STEPS scenario, as the PHEV electric driving share increases from 0% to 100%, total respiratory effects also rise from 10.8 to 18.1 kg PM2.5 eq.

Each panel shows the respiratory effects of different light-duty vehicles under the grid mix for the baseline STEPS and the APS scenarios in the US (a) and China (b), with a lifetime vehicle driving distance of 320,000 km. Respiratory effects are categorized into tailpipe emissions (depicted in red) and three non-tailpipe emission types: vehicle cycle (shown in dark blue), well-to-tank (shown in blue), and electricity production (shown in light blue). Parentheses next to PHEVs denote their electric drive share. Horizontal dashed lines highlight the vehicle type with the lowest respiratory effects.

Discussion

Our analysis of light-duty vehicles at both the global scale and across three countries reveals that EVs are not inherently ‘clean’ technologies; their environmental performance depends on the composition of the electricity grid. EVs exhibit considerable decarbonization potential as the grid mix becomes cleaner.

Building on our understanding of how the grid mix composition affects EVs’ environmental performance, we now turn to look more closely at the specific sources of emissions related to different vehicle technologies. EVs are often promoted as preferable due to their zero tailpipe emissions, which substantially enhance air quality in urban and densely populated areas. In contrast, ICEVs exhibit tailpipe emissions that accumulate in one place over a short period, leading to direct impacts such as noticeable spikes in air pollutants during rush hour. However, this common perspective—focusing solely on vehicles’ tailpipe emissions—may overlook the full lifecycle emissions picture. As illustrated in Fig. 6, tailpipe respiratory effects account for only a tiny fraction of the total for ICEVs, with the majority coming from sites within the gasoline supply chain—the locations involved in the well-to-tank processes, including crude oil extraction, refining, storage, and transportation. It is important to note that respiratory effects associated with these processes occur in different locations and at various times, unlike global warming potential (GWP) with a cumulative effect on climate change. Conversely, electricity generation is a major source of respiratory effects for EVs. This pattern suggests that promoting electrified transportation might shift the emission sources from the location involved in well-to-tank processes to the vicinities of power plants. Notably, both locations are typically situated in lower-income or rural communities. Here, we discuss only the LCA impact results and recognize the differences between these results and actual exposure, as discussed in Supplementary Note 9.

Beyond shifting the location of respiratory effects, it potentially aggravates the emission levels, revealing that decarbonization efforts for EVs do not always lead to a “win–win” scenario, especially when considering other environmental impacts. Our analysis, considering a lifetime driving distance of 320,000 km in US and China baseline STEPS and APS scenarios (Fig. 6), reveals that the respiratory effects of EVs remain higher than those from HEVs. Norway is a notable exception due to its reliance on hydropower as the primary energy source, which results in minimal electricity-related emissions and renders gasoline combustion considerably less environmentally favorable in comparison (Fig. 3b and Supplementary Fig. S4). Therefore, the transition towards electrified transportation may inadvertently aggravate environmental injustice, disproportionately increasing local pollution emissions in vulnerable communities. Considering this, conducting risk assessments is imperative to avoid unforeseen consequences. Such measures are particularly critical to prevent a disproportionate impact on rural communities, mitigating potential issues such as respiratory illnesses, air quality degradation, and acidification, resulting from air-emissions-intensive power generation.

A predominant strategy for EVs favors installing larger-capacity battery packs to alleviate range anxiety and limited charging infrastructure. However, such augmentations introduce trade-offs, such as reduced fuel economy due to heavier batteries 35 and a higher environmental impact (e.g., GWP and RE) during both the vehicle and the fuel cycle. Notably, when PHEVs operate entirely on electricity, i.e., their electric drive share is 100%, their functionality is like BEVs but with smaller battery packs. This results in PHEVs being lighter overall compared to their BEV counterparts (See Supplementary Note 2). Our analysis indicates that under such conditions, PHEVs have lower GWP and RE impacts than BEVs, mainly due to BEVs’ larger battery packs, which lead to increased vehicle weight and electricity consumption, thereby increasing the environmental impact. This observation aligns with our study’s assumption that the electricity consumption of PHEVs in charge depletion mode is 21.25 kWh per 100 km, slightly lower than the 23.36 kWh per 100 km of BEVs.

The choice between PHEV and BEV requires a trade-off between emission impacts and driving range. BEVs have larger battery packs that provide longer driving ranges at a single charge (e.g., 483 km, as used in this study), but the lifecycle of BEVs entails increased battery manufacturing emission, vehicle weight, and electricity consumption, resulting in increasing their lifecycle environmental impact (both in the vehicle and fuel cycle). On the other hand, PHEVs, when running primarily on electricity (high electric driving share, e.g., daily commutes within their all-electric range of 48 km), under a clean grid mix like Norway, have the potential to reduce lifetime emissions but offer limited driving range. They switch to gasoline for unplanned or long-distance travel needs, especially when charging stations are not accessible, highlighting practical challenges that could undermine their environmental benefits.

Given these technical insights, consumers face a complex decision-making process when seeking environmentally friendly light-duty vehicles. They need to consider factors such as the expected lifetime driving distance of vehicles (e.g., in the US, the average light-duty vehicle lifetime driving distance is 241,400 km 36), daily commuting distance (e.g., 81.2% of passenger trips were shorter than 16 km in the US in 2021 37), and individual driving and charging patterns (e.g., the PHEV electric driving share). Moreover, regional differences in electricity grid mix and charging station availability further complicate these considerations. Consumers exert greater control over the daily lifetime driving distance of vehicles, commuting distance, and driving and charging patterns, whereas factors such as regional electricity grid mix and charging infrastructure availability are beyond their control.

Our vehicle choice landscape plots (Figs. 3– 5 and Supplementary Fig. S11) can provide guidance for both policymakers and individual consumers. For individual consumers, before purchasing a vehicle, they can estimate how many kilometers they are likely to drive it (to determine vehicle lifetime driving distance) and how often they would be able to charge it (to determine the PHEV electrical driving share) to see which vehicle would have the lowest environmental impact for their region. From the perspective of the policymakers, there are a number of outcomes that can be targeted. Improving access to public charging infrastructure to increase the PHEV electric driving share could also influence the size of battery packs by reducing range anxiety. Incentivizing the early retirement of high-emission vehicles, e.g., by ‘cash for clunkers’-type schemes in which rebates are predicated on the scrapping of an existing vehicle. Policies to increase the vehicle lifetime of BEVs and PHEVs, e.g., investing in research to extend the battery life cycles. Creating tailored policies in areas without clean grids and providing incentives for consumers to choose HEVs, rather than BEVs or PHEVs, in the short term and promoting the deployment of low-emission electricity generation such as solar and wind in the long term. We suggest policymakers also consider the impact of local emissions from fossil fuel-based power plants on air quality and the exposure of vulnerable communities nearby. Specifically, it is advisable to carefully study the extent of exposure of these communities, considering meteorological, geographical, and other relevant factors; to closely monitor their respiratory health, and to institute policies to curb emissions from fossil fuel-based power plants. Implementing these measures could help ensure that the transition to electrified transportation does not inadvertently aggravate existing environmental injustices.

The most appropriate vehicle types identified in our country-level analysis may not directly apply to more specific local contexts. Our analysis does not provide a detailed comparison of the impact of state-level or local electricity grid mixes (e.g., US states or Chinese provinces, each characterized by distinct energy mix profiles). When implementing vehicle-specific strategies or policies, adopting an appropriate geographical resolution is necessary.

We acknowledge the limitations of our study (see Supplementary Note 9), specifically pointing out the exclusion of the battery end-of-life phase due to data scarcity. Given its considerable impact on the EV’s overall environmental assessment, this area warrants further research. The evolution of battery end-of-life, including secondary reuses or mineral recovery 31, 38, 39, 40, 41, 42, along with the expected substantial growth for the EV recycled batteries market 39, provide tremendous opportunities for further exploration to optimize EV environmental benefits. Moreover, the constrained nature of available datasets limits our ability to analyze the field comprehensively, including its dynamic nature. Based on our methodology and findings, adopting more extensive and transparent datasets will enable future scholars to deliver more detailed and comprehensive outcomes.

Conclusion

Despite their zero-tailpipe emissions, EV technologies alone are not inherently ‘clean’ or complete solutions for reducing transportation emissions. The environmental benefits of EV use are closely tied to the cleanliness of the electricity grid they utilize for energy. While EVs offer considerable decarbonization potential as the grid mix becomes cleaner, their contribution does not ensure a win-win solution to mitigate other environmental impacts, such as increased respiratory effects. Furthermore, the push for electrified transportation not only shifts local environmental emissions from one area to another but also potentially aggravates the emission levels. Both areas are typically lower-income or rural communities, highlighting the potential deterioration of environmental injustices.

Future research should focus on a deeper analysis of EV lifecycle impacts, particularly the state-level or local analyses with expected shifts towards cleaner energy and the implications of alternative battery technologies. Developing a comprehensive and adaptable database for light-duty vehicles that covers all critical environmental indicators is crucial. Further investigation of regional variations in EV charging behaviors, especially fast charging options, and their environmental impact, is essential. A thorough understanding of these behaviors could reveal strategies to decrease grid emissions by aligning charging times with the availability of renewable energy sources. Scholars should also explore policies for cleaner and more reliable electricity grids, including grid-scale storage and vehicle-to-grid solutions, as well as the expansion of EV charging infrastructure. A thorough investigation of the economic impacts of the EV coupled with LCA is also important.

Our research contributes to the ongoing communication on the environmental implications of transportation transition. Although decarbonization remains a crucial objective, promoting sustainable transport requires a holistic perspective, including environmental, transportation, and energy systems, and environmental justice. While global electricity supply transitions towards cleaner generation sources (e.g., solar and wind), and as EV technologies continue to evolve, our findings suggest that PHEVs and HEVs can serve as transitional solutions in pursuing low-emission transportation, highlighting the importance of adopting multiple strategies to achieve sustainable transportation.

Methods

We follow the guidelines of ISO 14040 and 14044 to conduct a cradle-to-grave Life Cycle Assessment (LCA) for the comparisons among ICEVs, HEVs, PHEVs, and BEVs 43, 44. We encompass the vehicles’ construction, operation, maintenance, and end-of-life phases, as detailed in Fig. 7. We select a driving distance of 320,000 km for each vehicle. We employ the openLCA software version 2.1, integrating data from two primary sources: the Ecoinvent version 3.7 cut-off database 23 and the Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (GREET) model version 2022 30. The type of light-duty vehicles selected for the study is medium-sized. PHEVs have a rated all-electric range of 48.3 km (30 miles), and BEVs have a range of 482.8 km (300 miles).

Schematic illustration of the system boundary employed in this study.

Regarding the possible future evolution trajectories of the grid mix in the US and China, we calculate relevant data based on the electricity generation forecasts released under both the Stated Policies Scenario (STEPS) and the Announced Pledges Scenario (APS) in the IEA’s World Energy Outlook 2023 28 and modify the current Ecoinvent database to obtain the electricity generation emissions (Supplementary Note 3). The Life Cycle Impact Assessment (LCIA) is carried out by the Tool for Reduction and Assessment of Chemicals and Other Environmental Impacts version 2.1 (TRACI) 45. Comprehensive methodological details, including the goal and scope, specific assumptions, and justification underlying the analysis, are available in Supplementary Note 2. Life cycle inventory is available in the Supplementary Data (See Data availability).

The environmental impact of vehicles throughout their life cycle is divided into two main phases: the fuel cycle and the vehicle cycle 30, as depicted in Fig. 8. The fuel cycle, including fuel extraction, combustion, and electricity production, quantified as the \({{\boldsymbol{\alpha }}}\times {{{\boldsymbol{vehicle}}}\; {{\boldsymbol{lifetime}}}\; {{\boldsymbol{distance}}}}\), encompasses six variables: the regional electricity mix, the local vehicle emission standard, the vehicle fuel economy, the grade of gasoline, the PHEV electric driving share (electric driving share refers to the proportion of the total distance driven using electricity when the combustion engine is not in operation, also known as the Utility Factor (UF) 46), and vehicle lifetime driving distance. On the other hand, the vehicle cycle, represented by the coefficient \({{\boldsymbol{\beta }}}\), includes activities related to vehicle construction and maintenance as well as end-of-life disposal.

The equation is represented as ‘Lifecycle environmental impact = fuel cycle (A) + vehicle cycle (B) = α × vehicle lifetime driving distance + β’. Seven parameters influencing two cycles are detailed in the figure.

As depicted in Fig. 9, the vehicle cycle impact represents the starting point on the y-axis, with vehicle lifetime driving distance being the x-axis variable. The factor α is defined as the slope of the line in this context. By incorporating these six variables into the calculation of factor α, we generate graphs illustrating the GWP and RE impacts for vehicles. Specifically, Fig. 6.

Solid straight lines represent the default value for each vehicle: ICEVs (pink line), HEVs (blue line), PHEVs (yellow line), and BEVs (green line). Shaded areas denote uncertainties associated with the four parameters: pink for the ICEV fuel economy, blue for the HEV fuel economy, yellow for the PHEV electric driving share, and green for the BEV electricity consumption.

Based on our observations, the default values for four parameters in GREET, namely, PHEV electric driving share (70%), ICEV fuel economy (8.99 L per 100 km), HEV fuel economy (6.45 L per 100 km), and BEV electricity consumption (23.36 kWh per 100 km), may not accurately reflect the actual conditions of medium-size vehicles. Therefore, for our uncertainty analysis, we have adjusted these variables based on data from all medium cars currently on sale, as documented in ref. 47. This adjustment involves selecting the minimum and maximum values for highway and city fuel economies to better represent the range of real-world conditions. This approach covers most real-life fuel economy ranges for ICEVs, HEVs, and BEVs (See Supplementary Note 2). Moreover, we adjust PHEV electric driving to encompass the full spectrum of individual usage behaviors for PHEVs. All these variations are depicted by the shaded area in Fig. 9.

For the same environmental metric, ICEVs and HEVs present almost the same size of shaded areas in each region, as detailed in Supplementary Figs. S6 and S7. In contrast, the shaded areas for PHEVs and BEVs show notable differences across regions. Therefore, these differences motivate us to explore how variations in PHEV electric driving share and BEV electricity consumption affect the results of our study. In the main text, we specifically focus on the uncertainty of the PHEV electric driving share, which ranges from 0% to 100% due to consumer usage behaviors. This range underscores the direct influence of consumer behavior on the results. We also conduct the uncertainty analysis for BEV electricity consumption, as detailed in Supplementary Note 8.

In Fig. 9, a series of lines graphically depict this variation. These lines, all originating from the same point on the y-axis—representing the same vehicle cycle impact—vary in slope and collectively create shaded areas. These areas intersect with solid straight lines representing different light-duty vehicles: ICEV (pink line), HEV (blue line), PHEV (yellow line), and BEV (green line). This intersection allows us to calculate specific break-even driving distances, the points at which emissions from different vehicle types are equivalent. By comparing these driving distances, we determine which vehicle type has the lowest environmental impact over specific driving ranges.

To investigate heterogeneity across diverse countries and regions, we undertake our analysis at both the global and country levels (Norway, the US, and China).

Data availability

The data used and generated in this study are available in the Supplementary Information and Supplementary Data (Figshare repository: https://doi.org/10.6084/m9.figshare.26044702.v1). Source data are provided within the Methods section.

1.
Transport | UNEP—UN Environment Programme.
https://www.unep.org/explore-topics/energy/what-we-do/transport.
2.
Transport—Energy System—IEA.
https://www.iea.org/energy-system/transport.
3.
Kahn Ribeiro, S. et al. SPM Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [P Transport Coordinating Lead Authors: Lead Authors: Contributing Authors: Review Editors: Chapter Scientist. 2022: Transport. In IPCC 2022 (2022).
4.
Global CO2 emissions from transport by subsector, 2000-2030—Charts—Data & Statistics—IEA.
https://prod.iea.org/data-and-statistics/charts/global-co2-emissions-fr….
5.
International Energy Agency, I. Global EV Outlook 2023: Catching up with climate ambitions. (2023).
6.
Executive summary—Global EV Outlook 2023—Analysis—IEA.
https://www.iea.org/reports/global-ev-outlook-2023/executive-summary.
7.
New and Used Clean Vehicle Tax Credits | Department of Energy.
https://www.energy.gov/energysaver/new-and-used-clean-vehicle-tax-credi….
8.
Alternative Fuels Data Center: Emissions from Electric Vehicles.
https://afdc.energy.gov/vehicles/electric_emissions.html.
9.
Xia, X. & Li, P. A review of the life cycle assessment of electric vehicles: considering the influence of batteries. Sci. Tot. Env. 814, 152870 (2022).
10.
Marmiroli, B., Messagie, M., Dotelli, G. & Van Mierlo, J. Electricity generation in LCA of electric vehicles: a review. Appl. Sci.
https://doi.org/10.3390/app8081384(2018).
11.
Hoekstra, A. The underestimated potential of battery electric vehicles to reduce emissions. Joule 3, 1412–1414 (2019).
12.
Pero, F. Del, Delogu, M. & Pierini, M. Life cycle assessment in the automotive sector: a comparative case study of Internal Combustion Engine (ICE) and electric car. in Procedia Structural Integrity vol. 12 521–537 (Elsevier B.V., 2018).
13.
Temporelli, A., Carvalho, M. L. & Girardi, P. Life cycle assessment of electric vehicle batteries: an overview of recent literature. Energies 13, 2864 (2020).
14.
Burnham, A., Lu, Z., Wang, M. & Elgowainy, A. Regional emissions analysis of light-duty battery electric vehicles. Atmosphere 12, 1482 (2021).
15.
Cucinotta, F., Raffaele, M. & Salmeri, F. A well-to-wheel comparative life cycle assessment between full electric and traditional petrol engines in the European Context. Lect. Notes Mech. Eng.
https://doi.org/10.1007/978-3-030-70566-4_30/FIGURES/2(2021).
16.
Xiong, S., Wang, Y., Bai, B. & Ma, X. A hybrid life cycle assessment of the large-scale application of electric vehicles. Energy 216, 119314 (2021).
17.
Shafique, M., Azam, A., Rafiq, M. & Luo, X. Life cycle assessment of electric vehicles and internal combustion engine vehicles: a case study of Hong Kong. Res. Transp. Econ. 91, 101112 (2022).
18.
Particle Pollution and Respiratory Effects | US EPA.
https://www.epa.gov/pmcourse/particle-pollution-and-respiratory-effects.
19.
Yang, L. et al. Life cycle environmental assessment of electric and internal combustion engine vehicles in China. J. Clean Prod. 285, 124899 (2021).
20.
Sharma, A., Shiwang, J., Lee, A. & Peng, W. Equity implications of electric vehicles: a systematic review on the spatial distribution of emissions, air pollution and health impacts. Env. Res. Lett. 18, 053001 (2023).
21.
Liang, X. et al. Air quality and health benefits from fleet electrification in China. Nat. Sustain. 2, 962–971 (2019).
22.
Ji, S. et al. Environmental justice aspects of exposure to PM2.5 emissions from electric vehicle use in China. Env. Sci. Technol. 49, 13912–13920 (2015).
23.
Database—ecoinvent.
https://ecoinvent.org/database/.
24.
Hawkins, T. R., Singh, B., Majeau-Bettez, G. & Strømman, A. H. Comparative environmental life cycle assessment of conventional and electric vehicles. J. Ind. Ecol. 17, 53–64 (2013).
25.
Irsyad, M. I., Halog, A. & Nepal, R. Renewable energy projections for climate change mitigation: an analysis of uncertainty and errors. Renew. Energy 130, 536–546 (2019).
26.
Energy Information Administration, U. Annual Energy Outlook 2022 Retrospective: Evaluation of Previous Reference Case Projections (2022).
27.
Shen, H., Wen, X. & Trutnevyte, E. Accuracy assessment of energy projections for China by Energy Information Administration and International Energy Agency. Energy Climate Change 4, 100111 (2023).
28.
World Energy Outlook 2023—Analysis—IEA.
https://www.iea.org/reports/world-energy-outlook-2023.
29.
Stated Policies Scenario (STEPS)—Global Energy and Climate Model—Analysis—IEA.
https://www.iea.org/reports/global-energy-and-climate-model/stated-poli….
30.
GREET | Argonne National Laboratory.
https://www.anl.gov/topic/greet.
31.
Porzio, J. & Scown, C. D. Life-cycle assessment considerations for batteries and battery materials. Adv. Energy Mater. 11, 2100771 (2021).
32.
Real-World Usage Of Plug-In Hybrid Electric Vehicles Background.
33.
GREET—Free and maintained.
https://greet.anl.gov/greet/gettingstarted/wtw.html.
34.
Tracing Gasoline From the Well to Your Tank | Department of Energy.
https://www.energy.gov/energysaver/tracing-gasoline-well-your-tank.
35.
Milovanoff, A., Posen, I. D. & MacLean, H. L. Electrification of light-duty vehicle fleet alone will not meet mitigation targets. Nat. Clim. Change 10, 1102–1107 (2020).
36.
Average Age of Vehicles in the US Increases to 12.2 years, according to S&P Global Mobility | S&P Global.
https://www.spglobal.com/mobility/en/research-analysis/average-age-of-v….
37.
NHTS NextGen OD Data.
https://nhts.ornl.gov/od/downloads.
38.
Koroma, M. S. et al. Life cycle assessment of battery electric vehicles: Implications of future electricity mix and different battery end-of-life management. Sci. Tot. Env. 831, 154859 (2022).
39.
Battery recycling takes the driver’s seat | McKinsey.
https://www.mckinsey.com/industries/automotive-and-assembly/our-insight….
40.
Casals, L. C., Amante García, B. & Canal, C. Second life batteries lifespan: rest of useful life and environmental analysis. J Env. Manag. 232, 354–363 (2019).
41.
Hossain, E. et al. A comprehensive review on second-life batteries: current state, manufacturing considerations, applications, impacts, barriers potential solutions, business strategies, and policies. IEEE Access 7, 73215–73252 (2019).
42.
Kim, S. et al. A comprehensive review on the pretreatment process in lithium-ion battery recycling. J. Clean. Prod. 294, 126329 (2021).
43.
ISO 14044:2006—Environmental management—Life cycle assessment—Requirements and guidelines.
https://www.iso.org/standard/38498.html.
44.
ISO 14040:2006—Environmental management—Life cycle assessment—Principles and framework.
https://www.iso.org/standard/37456.html.
45.
Bare, J. TRACI 2.0: the tool for the reduction and assessment of chemical and other environmental impacts 2.0. Clean Technol. Env. Policy 13, 687–696 (2011).
46.
Plötz, P., Funke, S. Á. & Jochem, P. The impact of daily and annual driving on fuel economy and CO2 emissions of plug-in hybrid electric vehicles. Transp. Res. Part A Policy Pract. 118, 331–340 (2018).
47.
Fuel Economy Guide 2024.
https://www.fueleconomy.gov/feg/pdfs/guides/feg2024.pdf.
48.
Electricity Information—Data product—IEA.
https://www.iea.org/data-and-statistics/data-product/electricity-inform….

Acknowledgements

The authors acknowledge that there are no additional contributors to this paper beyond those listed as co-authors. No external funding was received for this research.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Communications Earth & Environment thanks Xiaoning Xia and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Pallav Purohit, Martina Grecequet. A peer review file is available

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

About this article

Cite this article

Chen, H., Can Sener, S.E., Van Emburg, C. et al. Electric light-duty vehicles have decarbonization potential but may not reduce other environmental problems. Commun Earth Environ 5, 476 (2024). https://doi.org/10.1038/s43247-024-01608-z

-
Received:
-
Accepted:
-
Published:
-
DOI: https://doi.org/10.1038/s43247-024-01608-z

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.