While conventional vehicle (CV) ownership and electricity consumption in China are both increasing rapidly—annual growth rates during the past decade were ∼25% and ∼10%, respectively—e-bike ownership is skyrocketing: 86% annual growth during the past decade (doubling time: ∼13 months). Ten years ago, e-bikes were nearly unheard of, with vehicle ownership rates 26× lower for e-bikes than for CVs. Today, e-bikes outnumber CVs 2:1.

For EVs, combustion emissions occur where electricity is generated rather than where the vehicle is used. In China, 85% of electricity production is from fossil fuels, of which ∼90% is from coal. Most electricity generating units (EGUs) in China lack advanced pollution controls. Compared to typical vehicle emissions, EGUs are often located further from population centers; therefore, the exposure and health impacts per mass emitted tend to be lower for EGUs than for CVs. The net result for China is that it is unclear a priori whether EVs are an environmental health benefit or disbenefit relative to CVs.

—Ji et al.

The team evaluated five vehicle types (gasoline and diesel cars, diesel buses, e-bikes, e-cars) and considered how environmental impacts varied depending on the emission location. The study used an intake-based, rather than concentration-based, risk assessment for primary PM2.5.

The team found that using concentration rather than intake is suboptimal for health comparisons. Because electricity generation typically occurs farther from people than do tailpipe emissions, intake factors (iF) values are often lower for EVs than for CVs. For example, they noted, comparing PM_2.5 averages per passenger-km, emissions are 5× higher for an e-car than for a bus, but health impacts from primary PM_2.5 are about equal between the two modes. Comparing averages for e-bikes and buses, based on PM_2.5 emissions the two modes are similar (30% higher for buses) but based on PM_2.5 mortality rates, impacts are 7× greater for buses as for e-bikes.

Emission factors for EVs and CVs (g person-km−1), for four pollutants. Large circle icons indicate CVs. Small non-circle icons indicate EVs (e-car: triangle-icon; e-bike: plus-icon), with emission factors that vary among the 15 electricity grids. Large non-circle icons indicate the arithmetic mean of the 15 values per EV. Lines from icons indicate magnitude of well-to-station emissions; diamond end points of lines indicate well-to-wheel emission factors. (Missing lines indicate indistinguishable impacts.) Assumed average passenger load factors are car: 1.5, bus: 50, motorcycle: 1, e-bike: 1. Credit: ACS, Ji et al. Click to enlarge.

Among their findings were:

• The order-of-magnitude variability in EGU emission factors by region yields the same degree of variability in EV emission factors and with the same spatial pattern (highest in the Northeast because of heavy reliance on coal). EV emission factors vary by the city they are in; they estimated that an e-car (180 Wh/km in Beijing emits 220 gCO₂/km, equivalent to a gasoline car with a fuel economy of 9 L/100 km (26 mpg US), whereas in Chengdu the same e-car would emit only 135 gCO₂ km⁻¹, equivalent to a gasoline car with a fuel economy of 5.6 L/100km (or 42 mpg US).

• PM_2.5 emission factors generally are lower for CVs (gasoline or diesel) than comparable EVs. However, intake fraction is often greater for CVs than for EVs because combustion emissions are generally closer to population centers for CVs (tailpipe emissions) than for EVs (power plant emissions).

  For most cities, the net result is that primary PM_2.5 environmental health impacts per passenger-km are greater for e-cars than for gasoline cars (3.6× on average), lower than for diesel cars (2.5× on average), and equal to diesel buses. In contrast, e-bikes yield lower environmental health impacts per passenger-km than the three CVs investigated: gasoline cars (2×), diesel cars (10×), and diesel buses (5×).

• Compared to a new (Euro IV) gasoline car, average e-car emission factors are about the same for CO₂ and 19× greater for PM_2.5. E-bikes outperform cars, motorcycles, and buses on most emission metrics.

• Well-to-station emissions represent a larger proportion of total emissions for CVs relative to EVs for many pollutants.

• In terms of health impacts, e-cars typically perform better than diesel cars, worse than gasoline cars, and comparably to diesel buses; e-bikes perform much better than diesel cars and buses but are comparable to or slightly better than gasoline cars.

• CO₂ emissions (g/km) vary and are an order of magnitude greater for e-cars (135–274) and CVs (150–180) than for e-bikes (14–27).

China provides a useful case study because of the large number of EVs (in 2009, 100 million EVs) and because of government policies aimed at increasing the number of EVs. Unique aspects of China include the large population and coal-heavy electricity system. Our findings show that replacing gasoline cars with e-cars will result in increased CO₂ from combustion emissions and all-cause mortality risk from primary PM_2.5 in most cities. Health risks attributable to other pollutants, including secondary PM_2.5, are uncertain. Lightweight EVs such as e-bikes can have environmental and health benefits because of their energy efficiency. Chinese policy makers should carefully proceed with deployment of plug-in vehicles and consider aggressive improvements in the power sector to realize anticipated gains in emissions and health.

Future research could explore whether results presented here hold for other countries and could model impacts of secondary PM2.5. We highlight one distributional aspect of CV versus EV emissions (urban-rural exposure differences), leaving for future research a more significant exploration of environmental justice.

—Ji et al.

Resources

• Shuguang Ji, Christopher R. Cherry, Matthew J. Bechle, Ye Wu, and Julian D. Marshall (2012) Electric Vehicles in China: Emissions and Health Impacts. Environmental Science & Technology doi: 10.1021/es202347q