Cleaner Cars from
Cradle to Grave
How Electric Cars Beat Gasoline Cars on Lifetime
Global Warming Emissions
Cleaner Cars from
Cradle to Grave
How Electric Cars Beat Gasoline Cars on Lifetime
Global Warming Emissions
Rachael Nealer
David Reichmuth
Don Anair
November 2015
ii
© 2015 Union of Concerned Scientists
All Rights Reserved
Rachael Nealer is a Kendall Science
Fellow in the UCS Clean Vehicles Program.
David Reichmuth is a senior engineer in
the program. Don Anair is deputy director
and research director in the program.
The Union of Concerned Scientists
puts rigorous, independent science to
work to solve our planet’s most pressing
problems. Joining with citizens across the
country, we combine technical analysis and
eective advocacy to create innovative,
practical solutions for a healthy, safe, and
sustainable future.
More information about UCS and the
Clean Vehicles Program is available on the
UCS website: www.ucsusa.org
This report is available online (in PDF
format) at www.ucsusa.org/EVlifecycle.
Layout:
Rob Catalano, Catalano Design
Cover photo: © iStockphoto.com/
m-imagephotography
Printed on recycled paper
iiiCleaner Cars from Cradle to Grave
v Figures, Tables, and Boxes
vii Acknowledgments
1
Global Warming Emissions from Driving Electric Vehicles
Methodology for Comparisons
Rating the Regions
Expanding EV Options and Improving Vehicle Eciency
An Improving Electricity Grid
Regional EV Emissions: Main Findings
Beating the Average with Cleaner Electricity
The Future of EV Emissions
2
Global Warming Emissions from Manufacturing Electric Vehicles
How the Manufacturing of BEVs Diers from That of Gasoline Cars
Choice of Vehicles for Modeling Manufacturing Emissions
Estimating Emissions from Vehicle Manufacturing
Manufacturing Emissions of Today’s BEVs: Main Findings
Future Potential for Reducing BEV Emissions from Manufacturing
3
How Federal Policies Could Increase the Benefits of Electric Vehicles
Limit Power Plant Emissions and Expand Renewable Electricity
Generation
Directly Invest in Battery Technology
Facilitate Electric Vehicle Accessibility
[
]
iv
References
Appendix A: Operation Emissions Modeling
Appendix B: Manufacturing Emissions Modeling
Appendix C: Disposal, Recycling, and Reuse
Appendix D: Average vs. Marginal Electricity Global Warming
Emissions
vCleaner Cars from Cradle to Grave
[
, ,
]
Figure ES-1. Electric Vehicle Global Warming Pollution Ratings and Gasoline
Vehicle Emissions Equivalents by Electricity Grid Region
Figure ES-2. Life Cycle Global Warming Emissions from the Manufacturing and
Operation of Gasoline and Battery-Electric Vehicles
Figure 1. Percent of Electric Vehicle Sales by Make and Model for Model
Year 2014
Figure 2. Percent U.S. Electricity Generation by Fuel Type for 2009 and 2012
Figure 3. Electric Vehicle Global Warming Pollution Ratings and Gasoline
Vehicle Emissions Equivalents by Electricity Grid Region for 2012
Figure 4. Electric Vehicle Global Warming Pollution Ratings and Gasoline
Vehicle Emissions Equivalents by Electricity Grid Region for 2009
Figure 5. Electric Vehicle Global Warming Emissions Ratings by Population
Figure 6. Drivetrain Components of Battery-Electric Vehicles
Figure 7. Life Cycle Global Warming Emissions from the Manufacturing and
Operation of Gasoline and Battery-Electric Vehicles
Figure 8. Life Cycle Global Warming Emissions for a Midsize BEV for Three
Dierent Electricity Grid Mixes
Figure 9. State Renewable Electricity Standards (Including the District
of Columbia)
Figure B-1. Life Cycle Global Warming Emissions from the Manufacturing and
Operation of Gasoline and Battery-Electric Vehicles with More Than One
Battery Replacement
Table 1. Well-to-Wheels BEV Miles-per-Gallon Equivalent (MPG
ghg
) by
Electricity Source
Table 2. Global Warming Emissions Rating Scale for Electric Vehicles
Table 3. Electric Vehicle Eciency Ratings
Table 4. Midsize Battery-Electric and Gasoline Vehicle Characteristics
Table 5. Full-size Battery-Electric and Gasoline Vehicle Characteristics
vi
Table A-1. Mix of Generation Sources for Each Grid Region in 2012
Table A-2. Emissions Intensity from Electricity Generation by Region in 2012
Table B-1. Midsize Gasoline Vehicles Comparable with the Nissan LEAF
Table B-2. Full-size Gasoline Vehicles Comparable with the Tesla Model S
Table B-3. Composition of Vehicles Modeled by Material Type
Table B-4. Lithium-ion Battery Specifications for Midsize and Full-size BEVs
Table B-5. Global Warming Emissions Changes Based on Battery Chemistry
for 28 kWh Battery
Table B-6. First 15 Years of Vehicle Lifetime in Annual Mileage
Table C-1. Recycling Rates of Metals for All Vehicles Modeled
Box 1. Disposal Considerations
Box 2. Manufacturers Are Making Greener Choices
viiCleaner Cars from Cradle to Grave
This report was made possible by the generous support of the estate of the late
Henry Kendall and the Kendall Science Fellowship program, the William and Flora
Hewlett Foundation, the 11th Hour Project (a program of The Schmidt Family
Foundation), and the Energy Foundation.
The authors thank the following people for their help in providing information
or for reviewing this report:
Marcus Alexander (Electric Power Research Institute)
Travis Johnson (U.S. Environmental Protection Agency)
Jarod Kelly (Argonne National Laboratory)
Nic Lutsey (International Council on Clean Transportation)
Jeremy Michalek (Carnegie Mellon University)
Nick Nigro (Atlas Public Policy)
Julia Sohnen (BMW of North America)
Ken Srebnik (Nissan North America)
Luke Tonachel (Natural Resources Defense Council)
Marzia Traverso (BMW of North America)
Thomas Turrentine (University of California–Davis)
Russell Vare (Nissan North America)
Jacob Ward (U.S. Department of Energy)
and other individuals who wish to remain anonymous
The authors also thank many of their colleagues at the Union of Concerned
Scientists, including Je Deyette, Jimmy Nelson, and members of the Clean
Vehicles Program for their thoughtful input and advice. Special thanks go to
Bryan Wadsworth, Cynthia DeRocco, Heather Tuttle, Rob Catalano, and Steven
Marcus for their roles in the report’s editing and production.
Organizational aliations are listed for identification purposes only. The
opinions expressed herein do not necessary reflect those of the organizations
that funded the work or the individuals who reviewed it. The Union of
Concerned Scientists bears sole responsibility for the report’s contents.
[
]
viii
1Cleaner Cars from Cradle to Grave
Together with other oil-saving approaches, such as more
ecient vehicles and advanced biofuels, EVs can help cut
projected U.S. oil use in half over the next 20 years. EVs
will also be essential to achieving the deep emissions reduc-
tions by mid-century needed to avoid the worst impacts of
climate change.
This report compares battery-electric vehicles (BEVs)
with similar gasoline vehicles by examining their global
warming emissions over their “life cycles”—from the raw ma-
terials to make the car through manufacturing, driving, and
disposal or recycling. Toward that end, we performed up-to-
date assessments of the carbon footprints of BEVs, taking into
account the latest information about electricity generation
and BEV models. The two BEVs we modeled, midsize and
full-size, are not specific to any particular manufacturer but
are based on the two most popular BEV models sold in the
United States today: the Nissan LEAF and the Tesla Model S.
Our analysis reflects the BEVs available to American consum-
ers and comparable gasoline vehicles.
Our analysis revealed that:
• From cradle to grave, BEVs are cleaner. On average,
BEVs representative of those sold today produce less
than half the global warming emissions of comparable
gasoline-powered vehicles, even when the higher emis-
sions associated with BEV manufacturing are taken into
consideration. Based on modeling of the two most popu-
lar BEVs available today and the regions where they are
currently being sold, excess manufacturing emissions are
oset within 6 to 16 months of average driving.
• EVs are now driving cleaner than ever before. Driving
an average EV results in lower global warming emissions
than driving a gasoline car that gets 50 miles per gallon
(MPG) in regions covering two-thirds of the U.S. popula-
tion, up from 45 percent in our 2012 report. Based on
where EVs are being sold in the United States today, the
average EV produces global warming emissions equal to a
gasoline vehicle with a 68 MPG fuel economy rating.
• EVs will become even cleaner as more electricity is
generated by renewable sources of energy. In a grid
composed of 80 percent renewable electricity, manufactur-
ing a BEV will result in an over 25 percent reduction in
emissions from manufacturing and an 84 percent reduc-
tion in emissions from driving—for an overall reduction of
more than 60 percent (compared with a BEV manufac-
tured and driven today).
Global Warming Emissions from Driving
Electric Vehicles
Although a BEV has no tailpipe emissions, the total global
warming emissions from operating it are not insignificant;
they depend on the sources of the electricity that charge the
vehicle’s batteries and on the eciency of the vehicle. We
estimated the global warming emissions from electricity
consumption in the 26 “grid regions” of the United States
(see Figure ES-1, p. 2)—representing the group of power
plants that together serve as each region’s primary source of
Electric vehicles (EVs) are a critical part of the
American transportation future given their potential
to dramatically cut global warming emissions—
especially when charged by a clean electricity grid.
[
]
© Rachael Nealer
2
electricity—and we rated each region based on how charging
and using an EV there compares with driving a gasoline ve-
hicle. We also estimated, based on recent sales data, the aver-
age eciency of new EVs (battery-electric and plug-in
electric vehicles combined) sold in the United States in 2015.
We found that: (1) driving the average electric vehicle
in any region of the country produces lower global warming
emissions than the average new gasoline car achieving 29 MPG;
(2) our ratings in 20 out of 26 regions have improved since our
2012 report; and (3) about 66 percent of Americans—up from
45 percent just three years ago—live in regions where power-
ing an EV on the regional electricity grid produces lower global
warming emissions than a 50 MPG gasoline car.
Comparisons between electric vehicles and gasoline cars
look even more attractive when one considers that many EVs
are currently being sold and driven in areas where the electric-
ity grid is cleaner than the U.S. mean. As a result, based on
calculations that weighted where EVs were sold in 2014, driving
an EV in the United States produced global warming emissions
equal to a gasoline vehicle with 68 MPG during operation.
Emissions from operating electric vehicles are likely to
keep falling, as national data from 2013 to 2015 show a declining
percentage of electricity generated by coal power and an in-
crease in renewable resources such as wind and solar. Addition-
ally, the Clean Power Plan finalized by the U.S. Environmental
Protection Agency (EPA) in 2015 oers opportunities for even
greater progress, as states must collectively cut their 2005 pow-
er-sector carbon emissions 32 percent by 2030. Meanwhile,
many EV owners are pairing electric vehicle purchases with
home investments in solar energy. With increasing levels of re-
newable electricity coming onto the grid, with carbon stan-
dards for fossil-fuel power plants beginning to be implemented,
and with continued improvements in vehicle technologies, the
emissions-reduction benefits of EVs will continue to grow.
AKGD
45 MPG
AKMS
112 MPG
FRCC
51 MPG
HIMS
46 MPG
HIOA
36 MPG
MROE
40 MPG
NYCW
79 MPG
NYLI
47 MPG
RFCE
68 MPG
SRVC
63 MPG
NEWE
86 MPG
NYUP
135 MPG
RFCM
38 MPG
RFCW
44 MPG
MROW
43 MPG
RMPA
35 MPG
SPNO
35 MPG
SRMW
36 MPG
SRSO
51 MPG
SRTV
45 MPG
SPSO
39 MPG
SRMV
55 MPG
AZNM
53 MPG
ERCT
52 MPG
CAMX
87 MPG
NWPP
94 MPG
Good (31–40 MPG)
Better (41–50 MPG)
Best (51+ MPG)
U.S. average (EV sales-weighted): 68 MPG
FIGURE ES-1. Electric Vehicle Global Warming Pollution Ratings and Gasoline Vehicle Emissions Equivalents by
Electricity Grid Region
Note: The MPG (miles per gallon) value listed for each region is the combined city/highway fuel economy rating of a gasoline vehicle that would have global
warming emissions equivalent to driving an EV. Regional global warming emissions ratings are based on 2012 power plant data in the EPA’s eGRID 2015 database
(the most recent version). Comparisons include gasoline and electricity fuel production emissions. The 68 MPG U.S. average is a sales-weighted average based on
where EVs were sold in 2014.
SOURCE: EPA 2015C; IHS 2015.
3Cleaner Cars from Cradle to Grave
Global Warming Emissions from
Manufacturing Battery-Electric Vehicles
Global warming emissions occur when manufacturing any
vehicle, regardless of its power source, but BEV production
results in higher emissions than the making of gasoline cars—
mostly due to the materials and fabrication of the BEV lithi-
um-ion battery. Under the average U.S. electricity grid mix, we
found that producing a midsize, midrange (84 miles per
charge) BEV typically adds a little over 1 ton of emissions to
the total manufacturing emissions, resulting in 15 percent
greater emissions than in manufacturing a similar gasoline
vehicle. However, replacing gasoline use with electricity re-
duces overall emissions by 51 percent over the life of the car.
A full-size long-range (265 miles per charge) BEV, with
its larger battery, adds about six tons of emissions, which in-
creases manufacturing emissions by 68 percent over the gaso-
line version. But this electric vehicle results in 53 percent
lower overall emissions compared with a similar gasoline
vehicle (see Figure ES-2).
In other words, the extra emissions associated with elec-
tric vehicle production are rapidly negated by reduced emis-
sions from driving. Comparing an average midsize midrange
BEV with an average midsize gasoline-powered car, it takes
just 4,900 miles of driving to “pay back”—i.e., oset—the extra
global warming emissions from producing the BEV. Similarly,
it takes 19,000 miles with the full-size long-range BEV com-
pared with a similar gasoline car. Based on typical usages of
these vehicles, this amounts to about six months’ driving for
the midsize midrange BEV and 16 months for the full-size
long-range BEV.
Meanwhile, the global warming emissions of manufac-
turing BEVs are falling as automakers gain experience and
improve production eciency. With a focus on clean manu-
facturing, emissions could fall even more. There are many
ways in which the EV industry might reduce these manufac-
turing-related emissions, including:
• Advancesinmanufacturingeciencyandinthe
recycling or reuse of lithium-ion batteries;
• Theuseofalternativebatterychemistriesthatrequire
less energy-intensive materials; and
• Theuseofrenewableenergytopowermanufacturers’
and suppliers’ facilities.
Recommendations
To accelerate the U.S. transition to a low-carbon future, we
recommend the following:
• UndertheEPA’sCleanPowerPlan,statesshoulddevelop
and implement strong compliance plans that prioritize
renewable energy and energy eciency in meeting their
emissions-reduction targets.
• Policymakersatalllevelsofgovernmentshouldadopt
new or strengthened policies and programs for increasing
energy eciency and the deployment of renewable energy.
These options include renewable electricity standards,
energy-eciency resource standards, carbon-pricing
mechanisms, tax incentives and other financial incentives,
The extra emissions
associated with BEV
manufacturing can be
rapidly oset by reduced
emissions from driving.
FIGURE ES-2. Life Cycle Global Warming Emissions
from the Manufacturing and Operation of Gasoline and
Battery-Electric Vehicles
Midsize
Gasoline
Car
Midsize
84-mile
BEV
Full-size
Gasoline
Car
Full-size
265-mile
BEV
0
100
200
400
300
51%
reduction
600
500
OperationBattery Manufacturing
Vehicle Manufacturing
53%
reduction
Life Cycle Global Warming Emissions
(grams of CO
2
e per mile)
Note: We assume that the midsize vehicles are driven 135,000 miles over their
lifetimes and the full-size vehicles 179,000 miles. The dierence in the two
mileages derives from the dissimilar uses of 84-mile-range and 265-mile-range
battery-electric cars, as described in Chapter 2. We further assume that a
consumer buying a BEV would drive it the same total of miles as a
corresponding gasoline vehicle. We use U.S. average electricity grid emissions
to estimate manufacturing emissions, while the average electricity grid
emissions intensity during vehicle operation are based on a sales-weighted
average of where EVs are being sold today.
4
Electric vehicles provide benefits both in carbon emis-
sions and oil savings, with the greatest emissions benefits oc-
curring in regions with the lowest-carbon electricity sources.
To reach their full potential, EVs must account for a larger
share of vehicle sales while the electricity grid shifts from
coal to low-carbon renewable sources. Moving forward with
both of these transitions constitutes a critical strategy for cut-
ting projected oil use in half over the next 20 years and put-
ting the United States on a trajectory toward net-zero climate
emissions by mid-century.
and improvements in grid operation, transmission, and
resource planning.
• Governmentandtheprivatesectorshouldsupportmore
research aimed at decreasing the global warming emis-
sions associated with making electric vehicles’ batteries,
increasing the eciency of their operation, and improv-
ing the processes for battery recycling or reuse. By sup-
porting this emerging sector, we can help encourage
manufacturers not only to reduce manufacturing emis-
sions but also to lower the batteries’ costs.
• Toincreasethebenetsofelectricvehicles—especially
those in regions where global warming emissions from
electricity generation are higher than the U.S. average—
policies should support consumers who consider invest-
ing in cleaner sources of electricity, such as by installing
rooftop solar photovoltaic systems or purchasing renew-
able energy credits.
• Electricvehiclemakersandtheirsuppliersshouldraisethe
percentage of renewable electricity they use to build these
cars. The Union of Concerned Scientists estimates that,
with a future 80 percent renewable electricity grid, manu-
facturing emissions alone could decrease by more than
25 percent compared with manufacturing BEVs today.
To reach their full
potential, EVs must
account for a larger share
of vehicle sales while
the electricity grid shifts
from coal to low-carbon
renewable sources.
5Cleaner Cars from Cradle to Grave
[
]
Electric vehicles (EVs) are becoming a critical part of the
American transportation future because they can dramati-
cally cut global warming emissions, especially when charged
by a clean-electricity grid. Together with other oil-saving ap-
proaches—such as more ecient vehicles and advanced bio-
fuels— EVs can help achieve the goal, advocated by the Union
of Concerned Scientists (UCS) and others, that the United
States cut projected oil use in half over the next 20 years. EVs
will also be an essential part of any plan for proposed deep
emissions reductions for avoiding the worst impacts of cli-
mate change (UCS 2012).
But EVs are not global warming emissions–free. They
produce emissions, for example, during generation of the elec-
tricity required to charge the vehicles. The 2012 UCS report
State of Charge examined the global warming emissions from
operating an EV in dierent areas of the United States, and the
authors found that in every region, driving the average EV
produced lower global warming emissions than driving the
average new gasoline car (Anair and Mahmassani 2012).
Since State of Charge was published, those EV-related emis-
sions have become even lower in many parts of the country.
Electricity generation has been getting cleaner—coal-fired gen-
eration has declined while lower-carbon alternatives have in-
creased—and electric vehicles are becoming more ecient. For
example, the Nissan LEAF—the most popular battery-electric
vehicle (BEV),
1
powered completely by electricity—now has
greater eciency than before, and other even more ecient
BEV models, such as the BMW i3, have come to market.
However, global warming emissions from driving an EV
are not the full story. We must be attentive to the “bottom
line”—the vehicle’s overall global warming emissions during its
life cycle—which takes into account its operation, production,
and disposal. The manufacture of EV batteries, for example, is
of particular concern regarding global warming emissions.
This report compares battery-electric cars with similar
gasoline cars by examining their global warming emissions,
both from driving and manufacturing, over their lifetimes
(see Figure 7, p. 21). Toward that end, we performed up-to-
date assessments of the carbon footprints of BEVs, taking into
account the latest information about electricity generation
and BEV models.
We proceeded by addressing the following two key ques-
tions, whose successive answers gave rise to Chapters 1 and 2:
1) What are the global warming emissions from operat-
ing an electric vehicle today? We updated our previous
State of Charge emissions analysis from 2012, using the
latest available information on regional electricity grid
emissions and current EV models. We thus derived
new estimates of the miles-per-gallon rating a gasoline-
powered vehicle would need in order to equal the
emissions of a comparable EV charged on the regional
electricity grid.
2) How much does the manufacturing of BEVs aect
their total global warming emissions benefits? Broad-
ening the scope from UCS’s previous analysis, we exam-
ined the impact of BEV manufacturing on the vehicles’
global warming emissions. We used the Argonne Nation-
al Laboratory’s vehicle manufacturing emissions model
to analyze two composite vehicles representing midsize
and full-size BEVs. These composites were similar, re-
spectively, to the two best-selling BEVs in the United
States: the Nissan LEAF (an 84-mile-range car) and the
Tesla Model S (a 265-mile-range car).
2
We then com-
pared our results with the global warming emissions
from the manufacturing of like gasoline cars.
This report also discusses the global warming emissions
consequences of what may be done with the BEV after it has
finished its useful life. The disassembly of the vehicle, the re-
cycling of its parts, and the disposal of its battery are briefly
addressed in Chapter 2. Despite our conservative estimates of
the global warming emissions from these end-of-life process-
es, they account for only a small fraction of the life cycle to-
tals for BEVs—and for gasoline cars.
1 In this report, EVs refer both to plug-in hybrid vehicles (such as the Chevrolet Volt) and battery-electric vehicles such as the Nissan LEAF, while BEVs refer specifically
to battery-electric vehicles.
2 The Tesla Model S has numerous configurations; for our analysis, we used a rear-wheel-drive model with a battery rated at 85 kilowatt-hours (kWh).
6
A vehicle’s largest contribution of global warming emissions
comes from its fuel consumption. In the case of gasoline ve-
hicles, these emissions are the result of burning gasoline in
the engine—and also of producing that fuel in the first place.
With electric vehicles, which have little to no global warming
emissions at the tailpipe during operation, these emissions
are produced indirectly—from generating the electricity used
to charge the vehicles’ batteries and from producing the fuels
to enable that electricity’s generation.
In comparing electric and gasoline vehicles in 2012, our
State of Charge report found that in every region of the Unit-
ed States, EVs produced lower global warming emissions than
the average compact gasoline vehicle. However, because the
sources of electricity varied across the country, some electric
vehicles were cleaner in some regions than in others (Anair
and Mahmassani 2012).
Over the three years since State of Charge was published,
two major changes have occurred:
1) The eciencies of EVs have improved, while the number
of available models has widened. For example, the most
popular BEV, the Nissan LEAF, and the most popular
plug-in hybrid electric vehicle, the Chevrolet Volt, have
undergone improvements to increase their eciencies;
and the most lightweight and ecient BEV so far, the
BMW i3, has come to market.
2) The way we produce electricity across the country has been
evolving. Coal use in power plants has declined, and lower-
carbon sources of electricity have been used more often.
These changes have lowered the global warming emis-
sions of operating an electric vehicle, making it, in more parts
of the country, the most eective vehicle for reducing global
warming emissions today.
In the next two sections of this chapter, we describe the
methodologies for determining EV global warming emissions
by region. In the three sections that follow, we deepen our
discussion of the two kinds of major changes cited above. We
then specify what individual EV owners might do to “beat the
average,” and in the final section we build on the chapter’s
overall content to provide a look into the future.
Methodology for Comparisons
In comparing EVs’ global warming emissions with gasoline
vehicles’ emissions, we take a “well-to-wheels” approach that
accounts for the full fuel cycle for both types of vehicles.
To assess the global warming emissions from charging
electric vehicles, we address all contributions from electricity
production. This includes: (1) emissions that result from ex-
tracting raw materials, such as coal mining or natural gas
Global Warming Emissions from
Driving Electric Vehicles
[
1
]
The eciencies of EVs
have improved, the number
of available models has
widened, and lower-carbon
electricity sources are now
used more often.
7Cleaner Cars from Cradle to Grave
drilling, and from delivering these resources to the power
plants; (2) emissions from burning the fuel in the power plant
to generate electricity; (3) electricity losses that occur during
distribution from the power plant to the point where the elec-
tric vehicle is plugged in; and (4) the eciency of the vehicle
in using electricity.
To assess the global warming emissions from a compa-
rable gasoline vehicle, we address emissions that result from:
(1) oil extraction at the well; (2) transporting the crude oil to a
refinery; (3) refining the oil into gasoline; (4) delivering the
fuel to the gas station; and (5) combusting the fuel in the en-
gine of the vehicle.
STANDARDIZING THE UNITS OF COMPARISON: MPG
GHG
Most drivers are familiar with the concept of miles per gallon
(MPG), the number of miles a car travels on a gallon of gaso-
line. The greater its MPG, the less fuel burned and the lower
the car’s global warming emissions. But how can such emis-
sions be figured for electric vehicles, which don’t use gaso-
line? One way is by determining how many miles per gallon a
gasoline-powered vehicle would need to achieve in order to
match the global warming emissions of driving an EV.
The first step in this process is to calculate the global
warming emissions that result from generating the electricity
needed to charge a vehicle. Then we convert this estimate into
a gasoline miles-per-gallon equivalent—designated MPG
ghg
,
where ghg stands for greenhouse gas emissions. If an electric
vehicle has an MPG
ghg
value equal to the MPG of a gasoline-
powered vehicle, both vehicles will produce the same amount
of global warming emissions for each mile traveled.
For example, if one were to charge a typical midsize BEV
using electricity generated by coal-fired power plants, that
BEV would have an MPG
ghg
of 29. In other words, the global
warming emissions from driving it would be equivalent to the
emissions from operating, and producing the fuel for, a gaso-
line vehicle with a 29 MPG fuel economy rating over the same
distance (see Table 1).
3
Under this equivalency, the cleaner an
electricity generation source, the higher the MPG
ghg
. When
charging a BEV from sources such as wind or solar, for ex-
ample, the MPG equivalent is in the hundreds (or thousands)
because these options produce no global warming emissions
when generating electricity.
Finally, when estimating emissions from charging an
electric vehicle, we use regional average emissions (averaged
over the full mix of a region’s electricity sources). An
alternative approach is to consider only marginal emissions:
emissions from the power plants that operate to meet new
electricity demand on the grid. While the use of the marginal
generation mix for electric vehicles is important for evaluating
EV global warming emissions at a particular time, average
generation provides more practical emissions information to
consumers regarding a vehicle they might purchase and oper-
ate today. Note that no region gets electricity solely from one
source, but rather from a mix of the electricity generation fu-
els listed in Table 1. Further discussion on this issue is pro-
vided in Appendix D.
Rating the Regions
To further help consumers evaluate the global warming ben-
efits of electric vehicles in comparison with gasoline vehicles,
we developed ratings in our State of Charge report—Good,
Better, and Best—to characterize the country’s dierent re-
gions (see Table 2, p. 8).
GOOD
In regions rated Good, EVs are similar to the best convention-
al gasoline vehicles and some hybrids (31 to 40 MPG
ghg
). That
3 Note that MPG values in this report refer to combined city/highway operation estimates and that U.S. EPA window-label values should be used as the basis of com-
parison between specific vehicle models.
TABLE 1. Well-to-Wheels BEV Miles-per-Gallon
Equivalent (MPG
ghg
) by Electricity Source
Electricity
Source
1
Gasoline Vehicle
Emissions
Equivalent
(MPGghg)
2,3
% Reduction
from Average
New 2014 Car
4
Oil
Coal
Naturalgas
Geothermal
Solar
Nuclear
Wind
Hydro
Notes:
(1) Represents electricity available at the wall outlet and includes emissions
from power plant feedstocks (e.g., coal mining) and power plant combustion.
Power plant construction emissions are also included; they are the only emis-
sions associated with solar, wind, geothermal, and hydro sources.
(2) Gasoline vehicle emissions equivalents account for oil extraction and refin-
ing of crude oil, but not refinery construction.
(3) Average new car (excluding truck) fuel economy for model year 2014 is
28.7 MPG. Sources: EPA 2014; ANL 2014A.
(4) To calculate the MPG
ghg
estimate, we use the 2014 average sales-weighted
eciency of 0.33 kWh/mile, regarding both plug-in hybrid and battery-electric
vehicles (see Table 3, p. 10).
8
BETTER
EVs rated Better correspond to the most ecient hybrids
(41 to 50 MPG
ghg
). The most ecient gasoline hybrids, such as
the Honda Insight and the Toyota Prius, are in this category,
though the model year 2016 Toyota Prius is expected to ex-
ceed 50 MPG (Voelcker 2015).
BEST
Driving a typical EV in these regions is equivalent to gasoline-
powered vehicles with a combined city/highway fuel econo-
my of more than 50 MPG
ghg
. While the most ecient gasoline
hybrids are approaching this level of eciency, electric ve-
hicles can go well above—even exceeding 100 MPG
ghg
. EVs are
the best choice in these regions for reducing global warming
emissions and cutting oil use.
Expanding EV Options and Improving
Vehicle Eciency
Since 2010, when the Chevrolet Volt (a plug-in hybrid
[PHEV] powered by batteries and a conventional gasoline
engine) and the Nissan LEAF (a BEV) came to market, auto-
makers—spurred by government policy and consumer de-
mand—have commercially introduced new and updated EV
models. At this writing (in November 2015), consumers may
choose from among 21 dierent kinds, oered mostly in Cali-
fornia (see Figure 1).
Electric vehicles vary in how far they can go on a kilo-
watt-hour (kWh) of electricity. For example, the 2014 Nissan
LEAF is estimated to consume 0.30 kWh of electricity per
is, driving a typical electric vehicle in these regions will result
in global warming emissions equivalent to gasoline vehicles
with a combined city/highway fuel economy rating of 31 to
40 MPG. This level is better than that of the average new
gasoline car (29 MPG) on the market today (EPA 2014).
TABLE 2. Global Warming Emissions Rating Scalen for Electric Vehicles
EVRegionalGlobal
WarmingEmissionsRating Good Better Best
EVGlobalWarming
EmissionsEquivalent
(MPG
ghg
)
– –
WhatDoesItMean
RegardingEVGlobal
WarmingEmissions?
EVshaveemissionscomparable
withthemostecientnon-
hybridgasolinemodelsavailable
EVshaveemissionscomparable
withthemostecient
gasolinehybridmodelsavailable
EVsoutperformthemost
ecientgasolinehybrid
modelsavailabletoday
ExamplesofModelYear
GasolineandHybrid
Vehicles
HondaFit(MPG)
FordFocusSFE(MPG)
ChevroletCruzeEco(MPG)
ToyotaPrius(MPG)
HondaAccordHybrid(MPG)
VolkswagenJettaHybrid
(MPG)
Nogasolineordiesel
comparisonsinmodel
year
Notes:
(1) Assumes 10,881 grams of global warming pollution per gallon of gasoline and average EV eciency of 0.333 kWh/mile.
(2) Model year 2015 combined city/highway fuel economy window-label value. Data from the 2015 Fuel Economy Guide are available at EPA 2015a. All vehicles
given as examples are classified by the EPA as midsize or smaller.
EVs are appealing not only to individual consumers but also to state and local
governments. By generating lower global warming emissions, these vehicles
(such as the plug-in hybrid truck above, tested by Massachusetts’ public transit
agency) are an important tool in city and state eorts to meet emissions-
reduction targets.
Massachusetts Oce of Energy and Environmental Aairs
9Cleaner Cars from Cradle to Grave
Nissan LEAF
Chevrolet Volt
Tesla Model S
Toyota Prius PHEV
Ford Fusion Energi
Ford C-Max Energi
BMW i3
Other
EV Make and Model
5.0%
24.5%
9.4%
6.9%
10.8%
14.1%
15.3%
14.1%
FIGURE 1. Percent of Electric Vehicle Sales by Make and
Model for Model Year 2014
Note: “Other” includes EVs produced by BMW, Cadillac, Chevrolet, Fiat, Ford,
Honda, Kia, Mercedes, Mitsubishi, Porsche, Toyota, Smart, and Volkswagen.
SOURCE: INSIDEEVS 2015.
mile traveled, while the BMW i3 consumes 0.27 kWh/mile
and the Tesla Model S consumes 0.38 kWh/mile (DOE
2015a).
4
Each vehicle’s design and capabilities determine its
energy eciency, as is the case for gasoline vehicles. For ex-
ample, the Tesla is a full-size BEV with substantial battery
capacity, which allows to it go farther on a single charge. Such
features also add to its weight, causing the Tesla to use more
electricity than a smaller, lighter-weight, shorter-range elec-
tric vehicle in going the same distance. Even for vehicles of
similar size (or footprint), changes in design can aect vehicle
eciency. The BMW i3, for example, is similar in size to the
Nissan LEAF but has greater eciency, due in part to a car-
bon-fiber body that helps reduce its overall weight.
Since our State of Charge analysis in 2012, eciency has
also improved in existing models such the Chevrolet Volt and
Nissan LEAF. The 2014 LEAF, for example, represented a
12 percent improvement in eciency compared with the
2011 LEAF, a result of vehicle upgrades that included better
regenerative braking, improved aerodynamics, and a more
ecient heater. The Chevrolet Volt also improved its electric-
drive eciency between model years 2011 and 2014, and the
new redesigned 2016 model boasts even better performance.
See Table 3 (p. 10) for a list of the top-selling EVs and their
energy eciencies.
Also since our 2012 report, the combination of recent
model introductions and upgrades to existing models has af-
fected the overall eciency of the full electric vehicle fleet.
Given the few models available at the time, the eciency of
the 2011 Nissan LEAF—0.34 kWh/mile—was chosen as repre-
sentative of EV eciency. However, there are now sucient
sales of EVs to use an average EV eciency; based on sales
data for the 2014 calendar year, that average eciency is now
0.33 kWh/mile—a 3 percent improvement from the eciency
of the 2011 LEAF. The improvement in overall EV eciency
means that electric vehicles are on average going farther on
a kilowatt-hour of electricity, resulting in lower emissions
per mile.
An Improving Electricity Grid
The sources of energy used to generate electricity in the
United States have been changing, with a consequent reduc-
tion in “emissions intensity”—the average global warming
emissions emitted from producing a given amount of electricity.
Sustained lower natural gas prices have led to a declining
share of coal-fired power (one of the highest-polluting sourc-
es of electricity) and a rising share of electricity generated
from natural gas. The deployment of renewable energy sourc-
es, such as solar and wind, have also contributed to the elec-
tricity grid’s improvements.
Figure 2 (p. 11) shows how the primary sources of U.S.
electricity generation have changed between 2009 (the year
on which our previous estimates of EV global warming emis-
sions were based) and 2012 (the most recent year for which
data are available, at this writing, from the EPA’s eGRID data-
base
5
). Coal power remains the predominant source of elec-
tricity in the United States. However, the last several years
4 The Nissan LEAF has a 24 kWh battery, the BMW i3 has a 22 kWh battery, and the Tesla Model S has a 85 kWh battery. Note the Tesla Model S is available in
more battery sizes.
5 The EPA periodically publishes the Emissions & Generation Resource Integrated Database (eGRID) of power plant electricity generation and emissions for
26 electricity grid regions around the country (EPA 2015C). These data are the basis for our regional emissions analysis of electric vehicles for 2009 and 2012 in
Figure 3 (p. 12) and Figure 4 (p. 13).
10
generation for that region, aecting in turn the estimates of
global warming emissions from electric vehicles being operated
in that region. These changes are discussed next.
Regional EV Emissions: Main Findings
Using the most recently available regional emissions data, we
find that electric vehicles are not only getting cleaner but also
becoming the best choice for more Americans aiming to help
cut global warming emissions and slash oil consumption.
Figure 3 (p. 12) shows, for each U.S. electricity grid region,
the fuel economy rating that a gasoline vehicle would need in
have seen a substantial shift toward other sources of electric-
ity, including natural gas and renewables. The share of coal in
powering the nation’s electricity grid dropped from 44 percent
in 2009 to 37 percent in 2012. Natural gas increased by a simi-
lar amount, producing 30 percent of the nation’s electricity in
2012, up from 23 percent in 2009. Non-hydro renewable elec-
tricity (biomass, wind, and large-scale solar), while still a
small portion of the national grid mix, increased significantly,
surpassing 5 percent of utility-level electricity generation.
These figures represent the average change across the
26 electricity grid regions of the United States. Changes in the
grid mix of each region aect the emissions from electricity
TABLE 3. Electric Vehicle Eciency Ratings
2014 Models
Tesla Model
S (265-mile
range)
Tesla Model
S (208-mile
range)
Chevrolet
Volt
Nissan
LEAF (84-
mile range)
Plug-in
Prius
2014 Sales-
Weighted Average
of All EVs
EVType BEV BEV PHEV BEV PHEV BEVsandPHEVs
ElectricEciency
(kWhmile)
Notes: The plug-in Prius uses a small amount of gasoline when operating from battery power (charge depleting mode). The eciency numbers here represent
only the electricity consumed per mile.
SOURCE: DOE 2015A; IHS 2015.
On average, battery-electric vehicles produce less than half the global warming emissions of comparable gasoline-powered vehicles; these emissions savings increase
in regions of the country that get more of their electricity from renewable resources.
© iStockphoto.com/joel-t
11Cleaner Cars from Cradle to Grave
the cutting of global warming emissions by 70 percent or more,
compared with the average new gasoline car in these areas.
Up from just 45 percent three years ago, about 66 per-
cent of Americans now live in regions where powering
an EV on the regional electricity grid produces lower
global warming emissions than a 50 MPG gasoline car
(Figure 5, p. 14). And down from 17 percent three years ago,
only 12 percent of Americans now live in the lowest-rated re-
gions, where powering an EV on the regional electricity grid
produces global warming emissions similar to that of the best
non-hybrid gasoline cars. Electricity grid improvements and
improved EV eciency mean several new regions of the
country are now “Best” regions, including Florida and Texas.
In these locales, EVs powered by the regional electricity grid
do better than a 50 MPG gasoline vehicle on global warming
emissions.
order to achieve the same global warming emissions as the
average EV. The numbers of Figure 3 are based on 2012 power
plant emissions data. For comparison, Figure 4 (p. 13) shows
the updated figures using power plant emissions from 2009.
Given these updated circumstances, our main findings
are as follows:
As a result of a cleaner electricity grid and improved
eciency of electric vehicles, emissions from charging an
electric vehicle on the grid have improved in 76 percent
of the regions over the past three years. Twenty out of 26
regions saw reduced emissions from EVs because of regional
changes in the mix of electricity sources. The biggest im-
provements occurred in the Pacific Northwest, where coal
and natural gas generation dropped nearly 10 percent—with
replacements largely by hydropower and wind—resulting in a
greater than 20 percent decrease in the emissions intensity of
electricity generation.
Driving an EV in any region of the country produces
less global warming pollution than the average new gaso-
line car. Even on the dirtiest U.S. regional electricity grid, EVs
produce the global warming emissions equivalent of a 35 MPG
gasoline vehicle—a 21 percent improvement over the new gaso-
line car’s average fuel economy of 29 MPG.
6
On the cleanest
electricity grids, including those covering parts of California
and New York, EVs emit lower global warming emissions than
a gasoline vehicle rated at more than 85 MPG. This results in
FIGURE 2. Percent U.S. Electricity Generation by Fuel Type for 2009 and 2012
Note: Non-hydro renewables include wind, solar, and geothermal electricity generation.
SOURCES: EPA 2015C; EPA 2012A.
Other
Petroleum
Non-hydro
Renewables
Hydro
Nuclear
Natural Gas
Coal
2012
2009
0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50%
Emissions from charging
an electric vehicle on the
grid have improved in 76
percent of the regions over
the past three years.
6 For the model year 2014, the EPA Trends report estimates average new car fuel economy, excluding trucks and SUVs, to be 28.7 MPG (EPA 2014).
12
EVs perform on average in each region. However, individual
EV owners may be able to achieve lower-than-average emis-
sions. Some utilities within the region may rely on cleaner
sources of electricity than the regional average, many EV
consumers may pair their EV purchase with rooftop solar,
and green power programs may provide additional opportu-
nities to heighten the global warming benefits of today’s EVs.
PAIRING EVS WITH ROOFTOP SOLAR
A 2013 survey of new EV owners in California, which repre-
sents more than 40 percent of the market for EVs, found that
32 percent of respondents had solar photovoltaic (PV) sys-
tems in their homes. An additional 16 percent indicated they
planned to install a PV system in the future (CCSE 2013).
GREEN POWER PROGRAMS
Not everyone has the option of installing solar panels to
Based on 2014 EV sales data, the average EV in the
United States produces global warming emissions while
driving similar to a 68 MPG gasoline vehicle, thereby cut-
ting emissions 60 percent compared with the average
new gasoline car. In 2014, approximately half of all EVs sold
were in California, where plugging in produces emissions
similar to that of an 87 MPG gasoline vehicle. Using data on
new EV registrations across the country in 2014, we estimate
that the sales-weighted average MPG
ghg
for EVs is 68 MPG,
which means they produce less than half the emissions of the
average new car while driving (IHS 2015).
Beating the Average with Cleaner Electricity
Our analysis of global warming emissions from charging an
electric vehicle is based on the average mix of electricity
sources in a given region. This provides an estimate of how
AKGD
45 MPG
AKMS
112 MPG
FRCC
51 MPG
HIMS
46 MPG
HIOA
36 MPG
MROE
40 MPG
NYCW
79 MPG
NYLI
47 MPG
RFCE
68 MPG
SRVC
63 MPG
NEWE
86 MPG
NYUP
135 MPG
RFCM
38 MPG
RFCW
44 MPG
MROW
43 MPG
RMPA
35 MPG
SPNO
35 MPG
SRMW
36 MPG
SRSO
51 MPG
SRTV
45 MPG
SPSO
39 MPG
SRMV
55 MPG
AZNM
53 MPG
ERCT
52 MPG
CAMX
87 MPG
NWPP
94 MPG
Good (31-40 MPG)
Better (41-50 MPG)
Best (51+ MPG)
U.S. average (EV sales-weighted): 68 MPG
FIGURE 3. Electric Vehicle Global Warming Pollution Ratings and Gasoline Vehicle Emissions Equivalents by
Electricity Grid Region for 2012
Note: The MPG (miles per gallon) value listed for each region is the combined city/highway fuel economy rating of a gasoline vehicle that would have global
warming emissions equivalent to driving an EV. Regional global warming emissions ratings are based on 2012 power plant data in the EPA’s eGRID 2015 database
(the most recent version). Comparisons include gasoline and electricity fuel production emissions. The 68 MPG U.S. average is a sales-weighted average based on
where EVs were sold in 2014.
SOURCE: EPA 2015C; IHS 2015.
13Cleaner Cars from Cradle to Grave
green pricing program (DOE 2015b; EPA 2015a). These initia-
tives allow consumers, by paying a small premium for renew-
able electricity, to support their utility’s greater investment in
renewables (Swezey and Bird 2001). The types of renewables
and program details vary by utility (EPA 2015a). In some de-
regulated utility markets, consumers have the ability to
choose their power provider. In those locales, choosing a pro-
vider that supplies electricity from renewable sources or that
maintains a green pricing program may be distinct options.
States oering this type of choice for at least some consumers
include California, Connecticut, Illinois, Maine, Maryland,
Massachusetts, New Jersey, New York, Pennsylvania, Rhode
Island, Texas, and Virginia. The District of Columbia oers
such a choice as well (DOE 2015b).
Purchasing renewable energy certificates (RECs), which
are available nationwide, is another option. RECs are directly
tied to electricity generated by renewable sources and are
power their EV. But participating in a green power program
oered by your utility, or independently purchasing renew-
able energy certificates, are two additional ways of supporting
renewable energy.
According to the U.S. Department of Energy (DOE), near-
ly 850 utilities across the nation are oering some type of
FIGURE 4. Electric Vehicle Global Warming Pollution Ratings and Gasoline Vehicle Emissions Equivalents by
Electricity Grid Region for 2009
AKGD
44 MPG
AKMS
110 MPG
FRCC
48 MPG
HIMS
42 MPG
HIOA
36 MPG
MROE
40 MPG
NYCW
41 MPG
NYLI
84 MPG
RFCE
64 MPG
SRVC
60 MPG
NEWE
75 MPG
NYUP
115 MPG
RFCM
38 MPG
RFCW
42 MPG
MROW
39 MPG
RMPA
34 MPG
SPNO
35 MPG
SRMW
37 MPG
SRSO
46 MPG
SRTV
46 MPG
SPSO
38 MPG
SRMV
57 MPG
AZNM
49 MPG
ERCT
48 MPG
CAMX
78 MPG
NWPP
73 MPG
Good (31-40 MPG)
Better (41-50 MPG)
Best (51+ MPG)
Note: The MPG value listed for each region is the combined city/highway fuel economy rating of a gasoline vehicle that would have global warming emissions
equivalent to an EV. Regional global warming emissions ratings are based on 2009 power plant data in the EPA’s eGRID 2012 database. Comparisons include
gasoline and electricity fuel production emissions.
SOURCE: EPA 2012A.
By providing additional
revenue for renewable
energy projects, the
purchase of RECs can help
to increase the supply of
renewable electricity.
14
more than 150 MPG
ghg
, compared with the regional average of
94 MPG
ghg
(UCS 2015).
States and utilities can help consumers determine how
clean their electricity is by disclosing data on emissions and
on electricity produced and consumed. Eorts have been
made in this regard, and some utilities also report to their
customers the specific mix of energy sources that generate
their electricity (CEC 2015).
7
However, such disclosures are
not consistent across all utilities, and they often do not con-
tain an estimate of the actual emissions intensity of the deliv-
ered electricity.
Table A-1 (p. 34) provides the grid mix for each region. If
a utility does provide a breakdown of its sources of electricity,
consumers can compare that utility’s mix with the regional
mix. As a general rule of thumb, its percentage of coal indi-
cates whether the utility is providing electricity that has high-
er or lower global warming emissions intensity than the
regional average.
The Future of EV Emissions
While electric vehicles have begun to reduce global warming
emissions today, if EVs are to deliver on their full potential—if
they are to help us avoid the worst consequences of climate
change—we must move systematically in the coming years to a
clean-power grid. That is, a “decarbonizing” of the transporta-
tion and electricity sectors over the next several decades, in
sold in a voluntary market (CRS 2015). By providing addition-
al revenue for renewable energy projects, the purchase of
RECs can help to increase the supply of renewable electricity
(Heeter, Belyeu, and Kuskova-Burns 2014).
When reviewing options for buying green power, con-
sumers should look for the Green-e certification label, which
indicates that the products have been independently verified
(see www.green-e.org).
ESTIMATES BY STATE OR INDIVIDUAL UTILITY
Finally, particular states or utilities within a region may have
lower or higher emissions than the regional average, based on
the types of power purchases they make. Utilities procure elec-
tricity from (1) power plants that they own, (2) direct contracts
with owners of other power plants, (3) short-term purchases
through the regional power grid, and (4) trades across regions.
Emissions data at the individual utility level are not consistent-
ly or readily available across the country, which is in part why
this report uses grid-level information for its primary estimates.
Similarly, individual states within a region may have
cleaner or dirtier electricity than the regional average, based
on the sources of energy used by the utilities in those states.
In Washington, for example, carbon dioxide emissions associ-
ated with the power delivered to customers in the state are
reported by utilities, which supply a great deal of renewable
electricity (WA DOC 2015). Using data specific to Washing-
ton, an EV charged with electricity in that state averages
FIGURE 5. Electric Vehicle Global Warming Emissions Ratings by Population
Note: Global warming emissions ratings are based on 2012 and 2009 power plant data in the EPA’s eGRID 2015 and 2012 databases, respectively.
Good
Better
Best
22%
66%
12%
38%
45%
17%
2009 2012
7 For example, California requires utilities to provide a Power Content Label specifying the mix of generation sources that are supplying their customers, but no
emissions intensity values need to be reported. See CEC 2015.
Good (31-40 MPG)
Better (41-50 MPG)
Best (51+ MPG)
15Cleaner Cars from Cradle to Grave
ergy, energy eciency, natural gas, and nuclear power; and
shifting from coal-fired power. States have until September
2016 to submit their final compliance plans, and emission re-
ductions must begin in 2022 (EPA 2015b).
By 2050, much greater deployment of renewable energy
sources could lower global warming emissions even further.
UCS has modeled a scenario in which, by mid-century, 80 per-
cent of U.S. electricity is produced from renewable sources
(Cleetus et al. 2014; Mai et al. 2012). Under this scenario,
electricity emissions intensity is reduced to below 100 grams
of carbon dioxide–equivalent emissions (CO
2
e) per kWh of
electricity generated (g/kWh), resulting in global warming
emissions from electric vehicles comparable with 300 MPG
ghg
gasoline vehicles.
the United States as well as in other countries, is needed.
Standards to reduce global warming emissions from elec-
tricity generation are invaluable for moving the United States
away from high-pollution sources such as coal. Several indi-
vidual states—including California and the nine Northeast
states participating in the Regional Greenhouse Gas Initia-
tive—have adopted such standards and are successfully im-
plementing them.
In August 2015, the EPA finalized the first-ever national
standards for cutting power sector carbon emissions. Under
the Clean Power Plan, states are collectively required to re-
duce power plant carbon emissions by 32 percent below 2005
levels by 2030. The plan provides for a number of options to
cut carbon emissions, including investing in renewable en-
As the United States develops more renewable energy resources, electric vehicles will become even cleaner.
© iStockphoto.com/123ArtistImages
If EVs are to deliver on
their full potential we
must move systematically
in the coming years to a
clean-power grid.
16
The use of dierent parts, materials, and processes to build
components unique to electric vehicles means that global
warming emissions from building a battery-electric vehicle
dier from those of building comparable gasoline vehicles.
The relative significance of those two sets of emissions dif-
fers as well. With gasoline cars, we have found that vehicle
operation accounts for almost 90 percent of the lifetime
global warming emissions, making the manufacturing emis-
sions a smaller portion of the life cycle burden. By contrast,
BEVs produce lower emissions during operation, with emis-
sions from manufacturing being a more significant contribu-
tor to the total life cycle emissions.
To determine the impact of manufacturing emissions on
the overall global warming emissions benefits of EVs, we
modeled two common configurations of battery-electric ve-
hicles available today: an 84-mile-range midsize BEV and a
265-mile-range full-size BEV.
How the Manufacturing of BEVs Diers
From That of Gasoline Cars
Unlike gasoline vehicles, BEVs have no fuel tank or internal-
combustion engine but instead have a battery pack, electric-
drive motor, power-control electronics, and regenerative
braking systems (see Figure 6). The greatest dierence in the
manufacture of BEVs compared with gasoline vehicles is in
the type and size of batteries required. Gasoline vehicles only
have a small lead-acid battery for starting the engine and
powering accessories while the engine is o, whereas BEVs
rely on much larger lithium-ion batteries to power the
vehicle itself.
The energy-storage capacity of a BEV battery, measured
in kWh, varies in dierent electric vehicle models and is a key
factor in determining how far a vehicle can travel on a single
charge.
8
The larger the battery, the greater the vehicle range;
but also the greater the weight added to the vehicle and the
greater the emissions from manufacturing the battery. For
example, the 24 kWh battery for the Nissan LEAF, which al-
lows a driving range of 84 miles and weighs about 650 pounds
(Nissan 2015). The much larger 85 kWh battery in the Tesla
Model S weighs about 1,200 pounds and carries the vehicle
265 miles (Tesla Motors 2015).
In addition to a large battery pack, BEVs also have dier-
ing components, such as the power train, transmission, and
traction motor. These components are not a large portion of
the vehicle’s materials and weight (less than 15 percent); and
Global Warming Emissions from
Manufacturing Electric Vehicles
[
2
]
8 Additionally, various battery chemistries are used for the batteries in dierent BEVs.
The greatest dierence
in the manufacture of
battery-electric vehicles
compared with gasoline
vehicles is in the type and
size of batteries required.
17Cleaner Cars from Cradle to Grave
Choice of Vehicles for Modeling
Manufacturing Emissions
Even within the electric vehicle community, not all BEVs are
created equal. While all BEVs do use similar components,
their designs and capabilities vary. The top-selling BEV mod-
els in 2014, the Nissan LEAF and Tesla Model S, reflect this
diversity. The LEAF is a midsize five-seat car that travels
84 miles when fully charged, and the Model S is a full-size
five- to seven-seat car with a range of 265 miles (DOE
2015a).
9, 10
These two electric vehicles provided the basis for
our estimates of global warming emissions from BEV manu-
facturing—in which we took into account both a midsize mid-
range car and a full-size long-range car. But we note that the
they replace the functionality of the engine, which is a
large part of the gasoline vehicle’s weight (about 30 percent)
(ANL 2014b).
Similarly, battery-electric and gasoline vehicles tend to
have dierent ratios of the materials—such as copper, alumi-
num, and steel—used to make the components. Weight and
material composition of the vehicle determine the majority of
global warming emissions from manufacturing it. More mate-
rial means more global warming emissions; and some materi-
als—depending on the processes involved from resource
extraction all the way up to ultimate disposal—generate more
emissions than others. For more information about the
weight and composition dierences of gasoline and electric
vehicles, see Appendix B.
9 We retrieved the data on vehicle model specifications from the Department of Energy (2015a).
10 The Model S is available with dierent battery sizes. We chose the 2015 Model S that is equipped with an 85 kWh battery pack (rated at a 265-mile range).
FIGURE 6. Drivetrain Components of Battery-Electric Vehicles
Power-control electronics. Directs
electricity from the batteries to the
motor when driving or from brakes to
batteries during regenerative braking.
Directs electricity from the outlet to the
batteries during recharging.
Electric drive motor. Converts
electrical energy into mechanical
energy to propel the car. Can
deliver eciency of more than
90 percent during driving and
oers quick acceleration.
Regenerative braking. Conventional cars
rely entirely on friction brakes to slow down,
dissipating the vehicle’s kinetic energy as
heat. Regenerative braking allows some of
that energy to be captured, turned into
electricity, and stored in the batteries.
Battery pack. The battery cells,
assembled into a battery pack, are
recharged by plugging into an outlet
or vehicle charger. When driving, the
batteries provide power to the
electric motor to turn the wheels as
well as power to run accessories
such as heating and cooling systems.
There are a number of dierences between the components of battery-electric vehicles and gasoline vehicles (the most significant are shown above). The eciency
and range of electric vehicles on the market today vary based on the size, composition, and configuration of these components.
18
vehicles (FHWA 2009). For consistency, we made our com-
parisons under the assumption of a 135,000-mile,
15-year lifetime for both midsize electric and gasoline cars.
We also assumed that this lifetime mileage requires only one
battery, which is similar to most other peer-reviewed life cy-
cle assessments (Hawkins et al. 2013).
For a comparable gasoline vehicle, we modeled a midsize
car with a fuel economy of 29 MPG and a vehicle weight of
3,000 lbs. These specifications were chosen based on the av-
erage fuel eciency and vehicle weight of several models
similar in size to the Nissan LEAF, including the Ford Focus,
Mazda 3, and VW Golf.
FULL-SIZE LONG-RANGE BEV
The full-size BEV we modeled was similar to a Tesla Model S,
reflecting the larger vehicle size in terms of battery capacity
and vehicle range, weight, and structure (see Table 5). Several
battery sizes are available with the Model S; we chose to mod-
el the largest, 85 kWh. The full-size gasoline vehicle we mod-
eled for comparison resembled the full-size BEV in terms of
size (or footprint), with an average fuel economy of 21 MPG
and average weight of 4,300 lbs. These characteristics were
based on the average of several similar gasoline vehicles,
including the Audi A8, Hyundai Equus, and Porsche Pana-
mera (see Appendix B).
underlying data used for modeling in this analysis were not
specific to any manufacturer. Given the diculty in obtaining
specific information about a company’s supply chain, energy
use, and specific component manufacturing processes, we
made assumptions applicable to the industry as a whole and
included vehicle specific attributes when possible.
Similarly, for the midsize and full-size gasoline vehicles
used for comparison, we examined the fuel economy of several
available gasoline models, similar in size to the corresponding
BEVs, to arrive at an average MPG rating for comparison.
MIDSIZE MIDRANGE BEV
The midsize BEV we modeled was similar to a Nissan LEAF,
reflecting the range, eciency, battery size, battery chemistry,
and vehicle weight (see Table 4). The LEAF’s 24 kWh battery
pack and eciency rating of 0.3 kWh per mile allows it to
travel an EPA-rated 84 miles on a single charge (DOE 2015a).
Data collected over the last several years on the Nissan
LEAF show that the car’s drivers put on about 9,000 miles per
year, less than the 12,000 miles per year seen with typical gas-
oline-powered cars (Carlson et al. 2014). Total BEV lifetime
mileage data are not available, given the short time these
vehicles have been on the road, but extrapolating from early
data we projected full lifetime to be about 135,000 miles,
compared with the 179,000-mile average lifetime of gasoline
TABLE 4. Midsize Battery-Electric and Gasoline Vehicle
Characteristics
Midsize
BEV
Comparable
Gasoline
Vehicle
Similarto
VehicleModel
NissanLEAF Average
EnergyEciency kWhmile MPG
VehicleCurbWeight(lbs)
VehicleFootprint(ft
)
RangeonFullCharge
(miles)
-
BatteryCapacity(kWh) -
NumberofLithium-ion
BatteriesoverLifetime
-
Note: Our results, based broadly on these specifications and the manufactur-
ing process emissions typical of the auto industry, are not specific assessments
of any manufacturers. The average gasoline vehicle comparable to the midsize
BEV is a composite of five gasoline vehicles available today (see Appendix B).
SOURCES: DOE 2015A; NISSAN 2015.
TABLE 5. Full-size Battery-Electric and Gasoline Vehicle
Characteristics
Full-size
BEV
Comparable
Gasoline
Vehicle
Similarto
VehicleModel
TeslaModelS Average
EnergyEciency kWhmile MPG
VehicleCurbWeight(lbs)
VehicleFootprint(ft
)
RangeonFullCharge
(miles)
-
BatteryCapacity(kWh) -
NumberofLithium-ion
BatteriesoverLifetime
-
Note: Our results, based broadly on these specifications and the manufacturing
process emissions typical of the auto industry, are not specific assessments of
any manufacturers. The average gasoline vehicle comparable to the full-size
BEV was a composite of five gasoline vehicles available today (see Appendix B).
SOURCES: DOE 2015A; TESLA MOTORS 2015.
19Cleaner Cars from Cradle to Grave
Vehicle recycling. In this study we took into consider-
ation the recycling rates of various materials and the energy
and global warming emissions associated with recycling and
disposing of vehicles at the end of their lives. Our assessment
included recycling of vehicle components at levels that are
typical today, but it did not include any recycling of the lithi-
um-ion battery (due to limited data). As the scale of BEV use
expands, the economic pressure to find second uses or recy-
cling options for batteries will expand as well. To the extent
this occurs, the emissions attributed to lithium-ion batteries
in our analysis could decrease. For the other vehicle compo-
nents, we used the recycling rates of similar materials—such
as iron and aluminum—that the Argonne National Laboratory
has collected from industry and that has been optimized in
gasoline vehicles over time (Burnham 2012). Without recy-
cling of the vehicle materials (excluding the batteries), the
BEV’s and gasoline car’s global warming emissions from man-
ufacturing each increase by about 15 to 20 percent. See Ap-
pendix C for these recycling rates and other related
information.
Vehicle disposal. Vehicle disposal refers to vehicle com-
ponents that are sent to a landfill. The literature suggests that
the global warming emissions from disposing either of gaso-
line vehicles and BEVs are small, and similar for both vehicle
types—except for the BEV battery (Nealer and Hendrickson
2015; Hawkins, Gausen, and Strømman 2012). We calculated
ADDITIONAL ASSUMPTIONS
Battery life. For this analysis we expected the midsize and
full-size BEV to need only one lithium-ion battery pack over
its lifetime. As most BEVs on the road today were sold less
than five years ago, there is little experience with actual long-
term battery durability. However, many manufacturers oer
warranties of 100,000 miles or more, indicating an expecta-
tion that the batteries will have acceptable performance for at
least that distance. Early data on battery replacements are
consistent with this assumption (DeMorro 2015), though we
note that factors such as regional temperature and weather
conditions may aect the longevity of the battery.
Regardless of climate, over time the eectiveness and
range of the battery decreases during use. Such reductions
would result in a loss of range in original battery capacity, but
the battery could still meet the needs of most EV drivers
(Saxena et al. 2015). We expect this reduction to be more
important for a vehicle with shorter original range (under
100 miles) and less significant for a vehicle with longer
range (greater than 200 miles). Whether BEVs typically
get battery replacements, or find drivers who have suitable
range requirements, or boast larger batteries (installed by
manufacturers to maintain the car’s residual value) will be-
come clear over the next few years as the BEV market grows
and matures.
© Creative Commons/chris connors (Flickr)
Technicians test the performance of EV battery packs during vehicle manufacturing and assembly. Batteries increase EVs’ manufacturing-related emissions
compared with gasoline vehicles, but the excess manufacturing emissions are oset in 6 to 16 months of average driving.
20
Estimating Emissions from Vehicle
Manufacturing
Having identified the many materials and components, across
numerous supply chains, that are assembled to produce the
final vehicle, we summed the global warming emissions at-
tributable to each item. Specifically, as noted earlier, we did
this for two types of BEVs—a midsize midrange battery-elec-
tric car and a full-size long-range battery-electric car—along
with their comparable gasoline models, using a life cycle
that the disposal process accounts for less than 5 percent of
the global warming emissions attributable to production of
the vehicles. Nevertheless, the data available for BEV end-of-
life procedures are limited because the majority of BEVs are
still on the road—i.e., they have not yet been retired. In the
absence of pertinent data, we made conservative assumptions
by allocating no emissions savings to the reuse or recycling of
the lithium-ion battery and by applying industry averages to
other recycled materials. (See Box 1 and Appendix C.)
Our sensitivity analysis of battery disposal options shows that
possible dierences in end-of-life scenarios do not alter the
conclusions of our life cycle assessment (see Appendix C).
When gasoline cars reach the end of their service lives, they
are disassembled for parts and materials that are either reused
or recycled. Only a small remnant of the vehicle is sent to a
landfill (American Recycler 2009). Assuming that most parts
of BEVs are recycled similarly to gasoline cars, the only major
dierence will involve the BEVs’ lithium-ion batteries. There
are three possibilities for these batteries: reuse, recycle, or
landfill.
Reuse. The lithium-ion battery at the end of the vehicle
application is assumed to have 75 percent of its original
capacity to store energy. The battery could then be used in
other applications—for example, storage for intermittent
renewable energy sources such as solar and wind. Such a
second life for BEV batteries on the grid, after they are no
longer being used by the car, could oset fossil fuel–related
global warming emissions by displacing coal- or natural gas–
based electricity generation.
Recycle. Parts or materials of the BEV lithium-ion battery
can be recycled for use in new batteries—a service currently
performed by two major companies: Umicore of Belgium and
Retriev Technologies of the United States and Canada (Retriev
Technologies 2015; Umicore 2015). The ability to recycle the
battery materials or parts depends significantly on the design
of the battery and the economics of recycling the materials.
Recycling requires energy and produces global warming emis-
sions, but recycling can reduce emissions compared with using
new materials (see Appendix C) (Hendrickson et al. 2015;
Dunn et al. 2012).
Landfill. The battery can go directly to a landfill, where it
is neither reused nor recycled. This scenario is the least
expected by experts; they cite concerns about localized pollu-
tion beyond global warming emissions), resource scarcity, and
BOX 1.
Disposal Considerations
the market for batteries even in second use (Dunn et al. 2014).
In other words, if the value of the lithium-ion battery is high
enough, there will be an incentive to reuse or recycle it. Also, if
federal, state, or local governments regulated the recycling of
lithium-ion batteries, as they do for other materials such as
tires and lead-acid batteries, demand might be more stable and
predictable.
Resource scarcity. Production of most lithium-ion BEV
batteries requires not only lithium but also cobalt, nickel, and
other metals, most of which are mined outside the United
States (NMA 2015). Early in the development of BEVs, there
were concerns that the demand for lithium for battery produc-
tion would be greater than its global supply. However, more
recent studies have quantified that supply and concluded there
is enough lithium for large increases in BEV manufacturing
(Gruber et al. 2011). Cobalt and nickel are today’s biggest
economic drivers for recycling because the market prices of
these metals are relatively high; recycling them not only reduces
cost but decreases the amounts of virgin materials extracted and
lowers the risk of resource scarcity (Dunn et al. 2014).
Disposal toxicity. Beyond global warming emissions,
cars—electric and gasoline vehicles alike—produce air pollut-
ants (such as nitrogen oxides, sulfur oxides, and particulate
matter) and contribute to other environmental impacts (for
example, water pollution such as eutrophication and acidifica-
tion) that can be toxic to humans and other species. With
BEVs, these degradations can occur throughout the car’s life
cycle—directly during its manufacture and disposal (in partic-
ular, disposal of the battery) and indirectly through electricity
generation to charge the vehicle during its service life. In the
United States, virtually all of the associated production,
disposal, and generation processes are subject to air and water
quality regulations. Quantifying these impacts is outside the
scope of this report, which focuses on global warming emis-
sions.
21Cleaner Cars from Cradle to Grave
manufacturing. For a midsize 84-mile-range BEV, manufac-
turing emissions are approximately 15 percent, or 1 ton of
CO
2
e higher than those of a comparable conventional gasoline
vehicle. However, total global warming emissions of the
midsize BEV, when powered by the electricity grid mix repre-
sentative of where BEVs are sold today, are 51 percent lower
than the comparable midsize gasoline car, thereby saving
29 tons of CO
2
e. The global warming emissions from manufac-
turing a midsize BEV are about 30 percent of its lifetime global
warming emissions; the remaining 70 percent come from
driving it.
For a full-size 265-mile-range BEV, manufacturing emis-
sions are approximately 68 percent, or 6 tons of CO
2
e higher
than a comparable conventional gasoline vehicle. Total global
warming emissions of the full-size BEV, when powered by
the electricity grid mix representative of where BEVs are sold
today, are 53 percent lower than the comparable full-size
gasoline car, thereby saving 54 tons of CO
2
e. The global
warming emissions from manufacturing a full-size BEV are
model developed by Argonne National Laboratory (ANL
2014a; ANL 2014b). Most of the components, made largely
of steel, aluminum, and plastic, of the battery-electric and
gaso line cars were similar; the vehicle body and chassis made
up about 60 percent of the total vehicle weight for both gaso-
line and electric vehicles (Burnham 2012).
The largest manufacturing dierence between gasoline
and electric cars, of course, is the production of the lithium-
ion battery. Emissions from producing the battery come from
extracting raw materials such as lithium, cobalt, copper, and
iron ores, processing these materials into finished metals, and
then fabricating them into the parts of the battery. Finally,
when the battery is assembled and installed in the car, there
are global warming emissions from the assembly.
Excluded from the life cycle assessments are the global
warming emissions from building the infrastructure (such as
factories and industrial equipment) required to do all of the
processing and assembling, and the emissions from transpor-
tation of raw materials for manufacturing. We not only expect
these emissions to be small on a per-vehicle basis but also that
they are likely to be about the same in gasoline and electric cars.
The gasoline and battery-electric cars we modeled were
“representations” of what is available on the market today.
They were representations—similar, say, to the Nissan LEAF
or the Tesla Model S, but not those actual vehicles—because
instead of using data unique to specific models or manufac-
turers to estimate the global warming emissions of producing
cars, we used industry averages and data from the published
literature (Burnham 2012). Given that our analyses involved
representative cars and not particular models—due to the
lack of specificity in the data, we cannot model the actual
cars—we have not captured areas in which individual auto-
makers are better or worse than the average, and the results
are not detailed assessments of these companies’ manufactur-
ing processes.
For more detail on the methodology used in this report
and on the representative components of the vehicles mod-
eled, see Appendix B.
Manufacturing Emissions of Today’s BEVs:
Main Findings
The lifetime global warming emissions from vehicle manu-
facture and operation, both for the midsize midrange BEV
and the full-size long-range BEV, are shown in Figure 7, along
with the totals for their comparable gasoline vehicles.
Our main findings are as follows:
On average, battery-electric vehicles have much
lower global warming emissions than comparable gaso-
line vehicles, despite higher emissions from vehicle
FIGURE 7. Life Cycle Global Warming Emissions from
the Manufacturing and Operation of Gasoline and
Battery-Electric Vehicles
Midsize
Gasoline
Car
Midsize
84-mile
BEV
Full-size
Gasoline
Car
Full-size
265-mile
BEV
0
100
200
400
300
51%
reduction
600
500
OperationBattery Manufacturing
Vehicle Manufacturing
53%
reduction
Life Cycle Global Warming Emissions
(grams of CO
2
e per mile)
Notes: We assume that the midsize vehicles are driven 135,000 miles over their
lifetimes and the full-size vehicles 179,000 miles. The dierence in the two
mileages derives from the dissimilar uses of 84-mile-range and 265-mile-range
battery-electric cars, as described in Chapter 2. We further assume that a
consumer buying a BEV would drive it the same total of miles as a corre-
sponding gasoline vehicle. We use U.S. average electricity grid emissions to
estimate manufacturing emissions, while the average electricity grid emissions
intensity during vehicle operation are based on a sales-weighted average of
where EVs are being sold today.
22
For a midsize 84-mile-range BEV, on average, the extra
manufacturing emissions are oset within 4,900 miles, or in
less than six months, based on the sales-weighted electricity
emissions of where EVs are sold today (assuming an 84-mile-
range BEV travels 75 percent of the typical first-year mileage
of a new gasoline car). When driving a BEV charged from the
cleanest regional grids in the United States, these extra manu-
facturing emissions are oset even quicker, within the first
3,700 miles of driving. On the dirtiest grid they are oset
within 13,000 miles, slightly more than a year for the typical
vehicle owner (though this time can be reduced if the BEV
driver finds a cleaner source of power).
For a full-size 265-mile-range BEV, on average, the extra
manufacturing emissions are oset within 19,000 miles, or in
about 16 months of driving, based on the sales-weighted elec-
tricity emissions of where EVs are sold today (assuming a
265-mile-range BEV travels the same first-year mileage as the
typical new gasoline car). When driving a BEV recharged
from the cleanest regional grids in the United States, these
extra manufacturing emissions are oset within the first
15,000 miles of driving, or in just under one year for the aver-
age driver. On the dirtiest grid they are oset within 39,000
miles, or in less than three years for the typical vehicle owner.
Overall, oset occurs as fast as six months or at most
within three years, which means that everywhere in the
about 33 percent of its lifetime global warming emissions; the
remaining 67 percent come from driving it.
Battery manufacturing accounts for the most signifi-
cant dierence between the manufacturing emissions of
BEVs and gasoline vehicles, but it represents only a small
percentage of the cars’ respective total emissions. For a
midsize 84-mile-range BEV, the battery-manufacturing global
warming emissions account for 24 percent of the total manu-
facturing emissions of the midsize BEV, but less than 8 per-
cent of the lifetime emissions.
For a full-size 265-mile-range BEV, the battery-manufac-
turing global warming emissions are larger than those of the
midsize BEV because of the increased battery size and its
longer range. The battery-manufacturing emissions for
this vehicle represent 36 percent of its total manufacturing
emissions, but still less than 12 percent of the lifetime
emissions.
The higher manufacturing emissions of a BEV are
quickly oset by emissions savings from driving the ve-
hicle, but how long it takes to realize this benefit depends
on where the owner plugs in. The regions that use the most
fossil fuel energy to generate electricity, especially coal, have
longer “break-even mileages”—the number of miles it takes a
BEV to oset the global warming emissions of producing the
lithium-ion battery.
© Creative Commons/Maurizio Pesce (Wikimedia)
With the exception of EVs’ lithium-ion battery, most of the materials and components used in EV and gasoline vehicle manufacturing are similar. The additional
emissions associated with EV battery manufacturing are more than oset by the emissions generated from driving a gasoline vehicle.
23Cleaner Cars from Cradle to Grave
generation by using renewable electricity sources. In the fu-
ture, there will be more possibilities to reduce emissions as
we add more renewable sources of electricity to the U.S. grid
(see Box 2). By producing a BEV using 80 percent renewable
energy, the global warming emissions from manufacturing
can be reduced by about 30 percent compared with the aver-
age U.S. electricity grid emissions today.
Advances in batteries may reduce global warming
emissions of battery manufacturing. Because the automo-
tive battery industry is growing and BEV battery chemistry is
still being researched, we examined a range of lithium-ion bat-
tery chemistries (see Appendix B). Our sensitivity analysis on
battery chemistry shows that global warming emissions from
battery production could range from a 45 percent increase to a
43 percent decrease, depending on the vehicle battery chemis-
tries we modeled. Data for these new technologies are often
only available at the laboratory scale, so the emissions as a
function of battery chemistry could diminish due to econo-
mies of scale. Also, we expect the emissions per battery to de-
crease as industry scales up the production of BEV batteries.
Making batteries with recycling options can reduce
global warming emissions of batteries. Many companies
recycle lithium-ion batteries for small electronics, but the
batteries needed for vehicles are much larger. There are cur-
rently two major companies that are capable of recycling lith-
ium-ion batteries at vehicle sizes. Being able to recycle
batteries may prove to be an advantage in the long term. For
our estimates, we attribute all the global warming emissions
of battery manufacturing to the first use of the battery, but
reusing or recycling it could reduce the battery emissions
over the vehicle’s lifetime. See Box 1 for more information.
United States BEVs will produce net emissions savings well
before the end of the vehicle life.
Future Potential for Reducing BEV
Emissions from Manufacturing
The global warming emissions from manufacturing BEVs can
be reduced (1) through the use of renewable energy to power
the production facilities, (2) through advances in lithium-ion
battery fabrication and recycling, and (3) through the applica-
tion of alternative materials and designs. Also, the BEV mar-
ket is still relatively small; the economies of scale realized by
manufacturing greater numbers of cars for a much larger
market will possibly reduce per-vehicle costs and improve
eciencies (Dunn et al. 2014).
Global warming emissions from producing a vehicle
can vary by as much as 30 percent, depending on the
source of electricity. If a vehicle is manufactured using re-
gional electricity with relatively high global warming emis-
sions, as is presently the case in the U.S. Midwest and in
China, this will result in higher manufacturing emissions
than in areas, such as California, that have significantly
reduced their global warming emissions from electricity
Automakers are already showing interest in using renewable
energy to produce BEVs and their parts. A few examples:
BMW. The i3 BEV uses carbon fiber–reinforced plastics
made at a supplier’s facility in Moses Lake, Washington, that is
powered exclusively by hydro-based electricity generated
nearby (SGL 2015). Although the process to make carbon
fiber–reinforced plastics can result in higher manufacturing
global warming emissions than from making steel, at least a
portion of these emissions are oset by producing the material
with renewable energy; further, emissions are reduced during
vehicle operations as a result of a lighter and more energy-
ecient vehicle.
BOX 2.
Manufacturers Are Making Greener Choices
Nissan. The LEAF vehicle and its batteries are assembled
in Smyrna, Tennessee, where the electricity grid produces
lower global warming emissions than in other international
regions where batteries are often made (Nissan 2015).
Tesla. The company plans to build a large-scale battery-
manufacturing facility in Nevada, with wind and solar elec-
tricity generation located nearby. Tesla is already using the
California grid, one of the cleanest grid regions in the United
States, to power its Model S production and assembly facility
in Fremont (Tesla Motors 2014).
Across the country, BEVs
will produce net emissions
savings well before the
end of the vehicle life.
24
The use of alternative materials and redesign for
weight reduction and vehicle eciency can reduce BEV
manufacturing’s share of global warming emissions. Be-
cause bigger batteries add to vehicles’ production costs and
total vehicle weight, manufacturers have a great incentive to
use lighter materials and designs. Lighter vehicles will also be
more ecient in their fuel use, with consequent decreases in
emissions while driving. Strategies to reduce vehicle weight
can involve replacing heavier materials (such as steel) with
lighter materials (aluminum or carbon fiber–reinforced plas-
tics). Changes in vehicle design—e.g., using welds instead of
bolts—can also result in weight reductions.
Because bigger batteries
add to vehicles’ production
costs and total vehicle
weight, manufacturers
have a great incentive
to use lighter materials
and designs.
25Cleaner Cars from Cradle to Grave
The research described in this report shows that even with
the greater global warming emissions from manufacturing
(largely because of lithium-ion battery manufacturing), a
battery-electric vehicle still results in significantly
lower global warming emissions over its lifetime than its
gasoline counterpart. Other studies on this topic have come
to similar conclusions (Nealer and Hendrickson 2015;
Hawkins, Gausen, and Strømman 2012).
Thus given the potentially major role of BEVs—if they
are widely deployed—in reducing global warming emissions
from the transportation sector, we recommend the adoption
of innovative policies in the following areas: (1) increased re-
newable electricity generation; (2) advanced battery technol-
ogy; and (3) facilitation of electric vehicle accessibility.
Limit Power Plant Emissions and Expand
Renewable Electricity Generation
How electricity is generated greatly aects the global warm-
ing emissions of electric vehicles, both in their manufacture
and operation. As such, renewable electricity will be the main
mechanism for reducing global warming emissions from EVs.
Consider one achievable future: if the nation’s grid relied on
renewable energy for 80 percent of its power supply by 2050,
the emissions from BEVs (including manufacturing emis-
sions) would be 60 percent lower than those under the aver-
age U.S. mix today. That level is roughly equivalent to a
gasoline car that achieves more than 300 MPG
ghg
. Even with
today’s grid mixes, as represented by the range of Figure 8
(p. 26), we can see about a 39 percent reduction if the EV is
manufactured and operated in a region similar to the national
average, and another over 20 percent reduction from the na-
tional average in a high renewable grid region.
The EPA recently finalized its Clean Power Plan, which
aims to reduce global warming emissions from power plants
by 32 percent below 2005 levels by 2030 (EPA 2015b). This is
the first-ever national policy designed to lower the carbon
emissions from electricity generation. To ensure its success,
states should now develop and implement strong compliance
plans that prioritize the use of renewable energy and energy
eciency to meet their emissions-reduction targets.
Although the Clean Power Plan is an important step for-
ward in limiting carbon pollution from the electricity sector,
it is not sucient to meet long-term U.S. climate goals. Our
nation’s response to climate change should also include a fed-
eral policy that puts a price on polluters’ carbon emissions
and sets mandates that by 2050 we reduce power-sector
emissions by at least 90 percent below 1990 levels and reduce
economy-wide emissions by at least 80 percent.
Congress should enact a federal Renewable Electricity
Standard (RES), and encourage the strengthening of state
How Federal Policies Could Increase
the Benefits of Electric Vehicles
[
3
]
Renewable electricity
will be the primary
mechanism for reducing
global warming emissions
from electric vehicles.
26
power-sector global warming emissions 11 percent by 2030
(Bailie, Clemmer, and Deyette 2015).
In addition to establishing renewable electricity stan-
dards, policy makers at all levels of government should en-
courage greater deployment of renewable energy through
carbon-pricing mechanisms, tax incentives and other finan-
cial incentives, and improvements in grid operation, trans-
mission, and resource planning.
Consumers and organizations should invest directly in
renewable energy technologies. Homeowners, businesses,
and diverse institutions can also accelerate the transition to
greater renewable energy use through on-site generation,
green power purchasing, and REC purchases. Net metering
allows consumers who generate their own electricity from
renewable technologies—such as a rooftop solar panel or a
small wind turbine—to feed excess power back into the elec-
tricity system and thereby spin their meter backward. Forty-
four states and the District of Columbia now have net
metering requirements, with only four states that do not have
such policies, and two states that have net metering policies
that are not uniform across the state (DSIRE 2015).
In some deregulated utility markets, consumers have the
ability to select their power provider. In those locales, choos-
ing a provider that supplies electricity from renewable sourc-
es or that maintains a green pricing program may be distinct
options. States oering this type of choice for at least some
consumers include California, Connecticut, Illinois, Maine,
Maryland, Massachusetts, New Jersey, New York, Pennsylva-
nia, Rhode Island, Texas, and Virginia. The District of Colum-
bia oers such a choice as well.
11
Purchasing RECs, which are available nationwide, is an-
other option. RECs are directly tied to electricity generated
by renewable sources and are sold in a voluntary market.
12
By
providing additional revenue for renewable energy projects,
the purchase of RECs can help to increase the supply of re-
newable electricity (Mai et al. 2012).
Directly Invest in Battery Technology
Policies that support additional battery research and develop-
ment should be pursued in order to increase EV batteries’
eciency, lower their costs, and reduce the global warming
emissions attributable to them from their manufacture and at
their end of their service lives.
Congress should continue to fund federal battery re-
search programs in order to reduce battery costs and
RESs, as eective ways of decreasing the global warming
emissions from electricity generation and consequently
from EVs. Over the past 15 years, state-level RESs have prov-
en to be one of the most successful and cost-eective means
for driving renewable energy development in the United
States (Heeter et al. 2014; UCS 2013a). Currently, 29 states
and the District of Columbia have adopted some kind of RES.
Figure 9 shows the stringency and type (mandatory or volun-
tary) of each state RES. California recently expanded the na-
tion’s largest market for renewable energy by increasing its
RES to 50 percent by 2030. Earlier in 2015, Hawaii increased
its RES to require 100 percent renewables by 2045. Other
state governments should follow suit.
For its part, Congress should enact a national RES of at
least 30 percent by 2030. A recent UCS analysis found such a
policy would level the playing field across all states, stimulate
nearly $300 billion in new capital investments, and cut U.S.
11 See apps3.eere.energy.gov/greenpower/markets/marketing.shtml for additional information
12 See CRS 2015 for additional information.
FIGURE 8. Life Cycle Global Warming Emissions for a
Midsize BEV for Three Dierent Electricity Grid Mixes
Average
Midsize
Gasoline
Car
Midsize
84-mile-
range BEV
for Low
Renewable
Electricity
Grid
Midsize
84-mile-
range BEV
for High
Renewable
Electricity
Grid
Midsize
84-mile-
range BEV
for U.S.
Average
Electricity
Grid
0
10
30
20
50
40
70
60
OperationBattery Manufacturing
Vehicle Manufacturing
39%
reduction
42%
reduction
Life Cycle Global Warming Emissions
(tons CO
2
e)
Note: The low renewable and high renewable electricity grid mixes are based
on the electricity regions with the most and least fossil fuel electricity genera-
tion, respectively. These regions are RMPA and NYUP in the eGRID 2015 data
with regional electricity grid mixes equivalent to 890 gCO2e/kWh and 240
gCO2e/kWh at the consumer’s wall outlet. See Appendix A for more details.
27Cleaner Cars from Cradle to Grave
first-generation EVs have not reached the end of their service
lives, it is important to ensure there will be a ready market for
used batteries when their time comes.
This can be done, in part, by designing batteries to be
easily recycled, thus requiring little extra energy and result-
ing in modest levels of additional global warming emissions.
Currently there are only two companies, Retriev Technolo-
gies and Umicore, that can recycle lithium-ion batteries from
EVs, and they have entered partnerships with EV manufac-
turers to achieve that end. But the recycling—and reuse—of
used lithium-ion batteries are in their infancies; they would
benefit from more research and development, including pub-
lic-private collaborations and pilot projects, to help them be-
come more ecient and broader in scale.
Facilitate Electric Vehicle Accessibility
A 2013 survey conducted by UCS and the Consumers Union
found that 42 percent of American households, representing
increase EV aordability. Government investment in bat-
tery technology has already played a significant role in reduc-
ing battery costs. In 2007, lithium-ion batteries cost about
$1,000 per kWh, but by 2014 they were at $300 per kWh
(Nykvist and Nilsson 2015).
Several federal programs played important roles to make
this achievement a reality, mostly run by the DOE. Research
funded by the DOE’s Advanced Research Projects Agency-
Energy (ARPA-E) and Joint Center for Energy Storage Research
helped to modify batteries for EV use (JCESR 2015). ARPA-E
and the DOE’s Vehicle Technologies Oce are presently funding
research into novel battery chemistries, which have the potential
to greatly extend batteries’ range and durability, and funding
technology-transfer processes to expedite such improved bat-
teries’ commercial availability. (DOE 2014; DOE 2009).
Congress should fund programs that facilitate battery
recycling or reuse. Although today’s market for recycling
large lithium-ion batteries is limited, given that most of the
FIGURE 9. State Renewable Electricity Standards (Including the District of Columbia)
40% by 2017
15% by 2020
14.5% by 2019
27% by 2020
20.38+% by 2020
20% by 2022
25% by 2026
20% by 2020 (DC)
15% by 2025
12.5% by 2021
2% by 2021
Standard
Voluntary Goal
100% by
2045
24.8% by
2025
75% by 2032
SOURCE: DSIRE 2015.
29% by 2015
15% by
2020
15% by
2015
10% by
2015
26.5% by
2025
20% by
2020
105 MW
15% by
2021
25% by
2026
5,880 MW
by 2015
15% by
2015
30% by
2020
20% by
2020
15% by
2025
25% by
2025
50% by
2030
25% by
2025
20% by
2025
10% by
2015
10% by
2015
18% by
2021
12.5% by
2026
10% by
2015
10%
by
2025
28
nearly 42 million American homes with a vehicle, could
benefit today from using an electric vehicle (UCS 2013b). To
help EVs grow into this large potential market, their upfront
costs must be reduced.
Congress should protect the existing $7,500 federal EV
tax credit and reinstate the infrastructure tax incentive.
Osetting EV purchase prices through incentives such as the
$7,500 federal tax credit and additional state tax credits have
helped stimulate the markets for EVs across the country. In
California, for example, more than 3 percent of new vehicle
registrations were plug-in hybrid and battery-electric vehi-
cles in 2014 (CNCDA 2015). Governor Jerry Brown has also
set a goal of 1.5 million zero-emissions vehicles,
13
on the
state’s roads by 2025 (Oce of Governor Edmund G. Brown
Jr. 2012). California was an early adopter of state-level incen-
tives for EVs, and this policy influenced others—Connecticut,
Maryland, Massachusetts, New York, Oregon, Rhode Island,
and Vermont—to follow suit. These eight states’ governors
have signed an agreement establishing action plans in each
state that would put a total of 3.3 million zero-emissions ve-
hicles into service by 2025 (CARB 2013).
Ensuring that the federal tax credit remains in place will
help decrease the current cost dierences between EVs and
comparable gasoline-powered vehicles, but such policy sup-
port will not be needed indefinitely. As EV technology im-
proves, more models become available, and they are produced
at greater scales, costs will come down—while EV drivers
continue to save money in fuel costs. For example, we expect
that over the vehicle’s lifetime an EV driver will save nearly
$13,000 on fuel not purchased (Anair and Mahmassani 2012).
Similarly, reinstating the tax incentive to provide funding
for infrastructure such as charging stations and to reduce the
costs of installing infrastructure both at home and at work-
places could help overcome the barriers to easily accessible
charging discussed below.
Congress should support unifying guidance on charging
installations. At present there are three ways to charge EVs:
AC Level 1 and Level 2 chargers; and DC fast chargers. Each
type of charger replenishes the lithium-ion battery at dier-
ent rates. Typically, the Level 1 charger adds two to five miles
of range per hour, the Level 2 charger adds 10 to 20 miles of
range per hour, and the DC fast charger adds 50 to 70 miles of
range in 20 minutes (DOE 2015c). There also are various
types of connectors and plugs for EVs. The DC charging con-
nectors are not uniform across all vehicle manufacturers. Tes-
la has its own connector and charging infrastructure, which
13 Zero-emissions vehicles include plug-in hybrid electric vehicles, along with battery-electric cars and fuel cell cars.
© Anne G. Blair © Creative Commons/Portland Development Commission (Flickr) © Creative Commons/mystuart (Flickr)
Solar-powered charging stations—such as those in ( from top to bottom)
Austin, Texas; Portland, Oregon; and Atlanta, Georgia—are popping up around
the country, oering EV drivers the opportunity to recharge with renewable
electricity.
29Cleaner Cars from Cradle to Grave
drivers know where to charge and also raise awareness of the
accessibility of chargers for potential EV buyers. The DOE
and Federal Highway Administration have developed such
signage through the Manual on Uniform Trac Control De-
vices, which defines standards that apply to all types of trac
signs, but the proposed EV signs have not yet been finalized
(DOE 2015e). Eorts to do so, and to implement these EV
standards, should proceed, with increased use of the signs
along interstates and other major roadways.
can only be used by Tesla owners. Nissan, Kia, and Mitsubishi
vehicles use a dierent type of connector, and BMW and
Chevrolet utilize yet another connector.
14
This situation can
make understanding charging dicult for potential EV drivers.
To date, state-by-state installation guidance has been
provided by the Society of Automotive Engineers and the
DOE, but streamlined and easier-to-understand guidance
would encourage more widespread installation. Similarly,
building codes across the United States vary, and some are
unclear on how EV chargers can be installed. Policy makers
could clarify these standards, and establish a federal standard,
to allow for more EV charging stations.
Congress should fund programs and partnerships for
more charging stations. The DOE is currently running a
workplace-charging challenge, which encourages employers,
through industry pledges, to provide charging access to their
stas. This is especially important for consumers, such as resi-
dents of multiunit dwellings, who may lack such access at
home (DOE 2015d). The DOE also oers valuable information
for these residents on overcoming such obstacles at home, and
it provides case studies on how the dwellers of apartment or
condominium buildings in various cities succeeded (DOE
2015d). Congressionally funded programs should give priority
to projects that install charging stations in locations with prox-
imity to many potential consumers and where the proposed
location is appropriate for the type of charging proposed.
Congress should support and adopt uniform EV charging
signage. As noted above, the many ways to charge an EV can
be confusing to consumers. Similarly, the chargers that are
available may be dicult for new drivers to find. Developing
uniform signage that is clearly displayed would help new EV
14 There are three types of connectors in the United States: SAE J1772, CHAdeMO, and Tesla (DOE 2015c).
© Creative Commons/Jason McHu (Flickr)
Clear, uniform signage is critical to helping EV drivers know where to charge, as
well as raising awareness of the accessibility of chargers for potential EV buyers.
30
Center for Resource Solutions (CRS). 2015. About us. Online at www.
resource-solutions.org/about.html, accessed June 1, 2015.
Cleetus, R., S. Clemmer, J. Deyette, S. Mullendore, and J. Richardson.
2014. The Clean Power Plan: A climate game changer. Cambridge,
MA: Union of Concerned Scientists. Online at www.ucsusa.org/
sites/default/files/attach/2014/10/Strengthening-the-EPA-Clean-
Power-Plan.pdf, accessed October 22, 2015.
Database of State Incentives for Renewables and Eciency (DSIRE).
2015. Renewable portfolio standard policies. Raleigh, NC: North
Carolina Clean Energy Technology Center. Online at http://
ncsolarcen-prod.s3.amazonaws.com/wp-content/uploads/2015/11/
Renewable-Portfolio-Standards.pdf, accessed October 22, 2015.
DeMorro, C. 2015. 99.99% of Nissan LEAF batteries still in opera-
tion. Clean Technica, March 25. Online at http://cleantechnica.
com/2015/03/25/99-99-nissan-LEAF-batteries-still-operation,
accessed June 1, 2015.
Department of Energy (DOE). 2015a. Fueleconomy.gov: The ocial
U.S. government source for fuel economy information. Online at
www.fueleconomy.gov, accessed October 22, 2015.
Department of Energy (DOE). 2015b. Green power markets. Online
at http://apps3.eere.energy.gov/greenpower/markets/pricing.shtml,
accessed October 22, 2015.
Department of Energy (DOE). 2015c. Developing infrastructure to
charge plug-in electric vehicles. Online at www.afdc.energy.gov/
fuels/electricity_infrastructure.html, accessed October 22, 2015.
Department of Energy (DOE). 2015d. EV Everywhere workplace
charging challenge. Online at www.energy.gov/eere/vehicles/
ev-everywhere-workplace-charging-challenge, accessed October
22, 2015.
Department of Energy (DOE). 2015e. Signage for plug-in electric
vehicle charging stations. Online at www.afdc.energy.gov/fuels/
electricity_charging_station_signage.html, accessed October 22,
2015.
Department of Energy (DOE). 2014. Advanced technology vehicles
manufacturing loan program. Online at www.energy.gov/sites/
prod/files/2014/05/f16/ATVM-Program-Application-Overview.
pdf, accessed October 22, 2015.
Department of Energy (DOE). 2009. DOE awards $8 billion in loans
for advanced vehicle technologies. EERE Network News, June
24. Online at http://apps1.eere.energy.gov/news/news_detail.cfm/
news_id=12594, accessed October 22, 2015.
Dunn, J.B., L.Gaines, J.C. Kelly, C. James, and K.G. Gallagher. 2014.
The significance of Li-ion batteries in electric vehicle life-cycle
energy and emissions and recycling’s role in its reduction. Energy
and Environmental Science 8:158–168. Online at http://pubs.rsc.
org/en/content/articlelanding/2014/ee/c4ee03029j, accessed
October 22, 2015.
American Recycler. 2009. Benefits of automotive reuse and recycling
study disclosed. Online at www.americanrecycler.com/1009/
benefits.shtml, accessed October 22, 2015.
Anair, D., and A. Mahmassani. 2012. State of charge: Electric vehicles’
global warming emissions and fuel-cost savings across the United
States. Cambridge, MA: Union of Concerned Scientists. Online at
www.ucsusa.org/sites/default/files/legacy/assets/documents/
clean_vehicles/electric-car-global-warming-emissions-report.pdf,
accessed October 22, 2015.
Argonne National Laboratory (ANL). 2014a. Greenhouse gases,
regulated emissions, and energy use in transportation Model.
GREET1_2014, released Oct 3. Online at https://greet.es.anl.gov/
main, accessed March 23, 2015.
Argonne National Laboratory (ANL). 2014b. Greenhouse gases,
regulated emissions, and energy use in transportation Model.
GREET2_2014, released Oct 3. Online at https://greet.es.anl.gov/
main, accessed March 23, 2015.
Bailie, A., S. Clemmer, and J. Deyette. 2015. Analysis of a 30 percent
by 2030 national renewable electricity standard. PowerPoint
presentation, May 12. Cambridge, MA: Union of Concerned
Scientists. Online at www.ucsusa.org/sites/default/files/
attach/2015/05/UCS_National-30-Percent-by-2030-RES-
Analysis.pdf, accessed October 22, 2015.
Burnham, A. 2012. Updated vehicle specifications in the GREET
vehicle-cycle model. Argonne, IL: Center for Transportation
Research, Argonne National Laboratory. Online at https://greet.
es.anl.gov/publication-update-veh-specs, accessed October
22, 2015.
California Air Resources Board (CARB). 2013. Governors announce
bold initiative to put 3.3 million zero-emission vehicles on the
road by 2025. Press release, October 24. Online at www.arb.
ca.gov/newsrel/newsrelease.php?id=520, accessed October
21, 2015.
California Center for Sustainable Energy (CCSE). 2013. What drives
California’s plug-in electric vehicle owners? Online at http://
energycenter.org/sites/default/files/docs/nav/transportation/cvrp/
survey-results/California_PEV_Owner_Survey_3.pdf, accessed
October 22, 2015.
California Energy Commission (CEC). 2015. California’s power con-
tent label. Online at www.energy.ca.gov/sb1305/power_content_
label.html, accessed October 22, 2015.
California New Car Dealers Association (CNCDA). 2015. California
auto outlook. Online at www.cncda.org/CMS/Pubs/Cal%20Cover
ing%202Q%2015%20Ver%202.pdf, accessed October 1, 2015.
Carlson, R, S. Salisbury, M. Shirk, and J. Smart. 2014. eVMT analysis
of on-road data from plug-in hybrid electric and all-electric ve-
hicles. PowerPoint presentation, October 2. Idaho Falls, ID: Idaho
National Laboratory. Online at http://avt.inl.gov/pdf/prog_info/
eVMTAnalysisResultsOct2014.pdf, accessed October 22, 2015.
[
]
31Cleaner Cars from Cradle to Grave
Environmental Protection Agency (EPA). 2009. The value of eGRID
and eGRIDweb to GHG inventories. Washington, DC. Online
at http://www3.epa.gov/statelocalclimate/documents/pdf/
TheValueofeGRID.pdf, accessed October 22, 2015.
Federal Highway Administration (FHWA). 2009. National household
travel survey. Washington, DC: Department of Transportation.
Online at http://nhts.ornl.gov/download.shtml, accessed October
22, 2015.
Gra Zivin, J.S., M.J. Kotchen, and E.T. Mansur. 2014. Spatial and
temporal heterogeneity of marginal emissions: Implications for
electric cars and other electricity-shifting policies. Journal of
Economic Behavior & Organization 107:248–268. Online at www.
sciencedirect.com/science/article/pii/S0167268114000808,
accessed August 28, 2015.
Gruber, P.W., P.A. Medina, G. Keoleian, S.E. Kesler, M.P. Everson, and
T.J. Wallington. 2011. Global lithium availability: A constraint
for electric vehicles? Journal of Industrial Ecology 15(5):760–776.
Online at http://onlinelibrary.wiley.com/doi/10.1111/j.1530-9290.
2011.00359.x/full, accessed August 28, 2015.
Hadley, S.W., and A. Tsvetkova. 2008. Potential impacts of plug-in
hybrid electric vehicles on regional power generation. Oak Ridge,
TN: Oak Ridge National Laboratory. Online at http://aprs.ornl.
gov/sci/ees/etsd/pes/pubs/Pub7922.pdf, accessed October
22, 2015.
Hawkins, T.R., B. Singh, G. Majaeu-Bettez, and A.H. Strømman. 2013.
Comparative environmental life cycle assessment of conventional
and electric vehicles. Journal of Industrial Ecology 17(1):53–64.
Online at http://onlinelibrary.wiley.com/doi/10.1111/j.1530-9290.
2012.00532.x/full, accessed August 28, 2015.
Hawkins, T.R., O.M. Gausen, and A.H. Strømman. 2012. Environmen-
tal impacts of hybrid and electric vehicles—A review. The Inter-
national Journal of Life Cycle Assessment 17(8):997–1014. Online
at http://link.springer.com/article/10.1007/s11367-012-0440-9,
accessed August 28, 2015.
Heeter, J., G. Barbose, L. Bird, S. Weaver, F. Flores-Espino, K.
Kuskova-Burns, and R. Wiser. 2014. A survey of state-level cost
and benefit estimates of renewable portfolio standards. NREL/
TP-6A20-61042. Golden, CO: National Renewable Energy Labo-
ratory. Online at www.nrel.gov/docs/fy14osti/61042.pdf, accessed
June 1, 2015.
Heeter, J., K. Belyeu, and K. Kuskova-Burns. 2014. Status and trends
in U.S. voluntary green power market (2013 data). Golden, CO:
National Renewable Energy Laboratory. Online at www.nrel.gov/
docs/fy15osti/63052.pdf, accessed October 22, 2015.
Hendrickson, T.P., O. Kavvada, N. Shah, R. Sathre, and C.D. Scown.
2015. Life-cycle implications and supply chain logistics of
electric vehicle battery recycling in California. Environmental
Research Letters 10(1). Online at http://iopscience.iop.org/
1748-9326/10/1/014011, accessed August 28, 2015.
IHS Automotive (IHS). 2015. Polk CY2014 new vehicle registration
data for EVs. Accessed March 11, 2015.
InsideEVs. 2015. Monthly plug-in sales scorecard. Online at www.
insideevs.com/monthly-plug-in-sales-scorecard, accessed October
22, 2015.
Dunn, J.B., L. Gaines, J. Sullivan, and M.Q. Wang. 2012. Impact of
recycling on cradle-to-gate energy consumption and greenhouse
gasoline emissions of automotive lithium-ion batteries. Envi-
ronmental Science and Technology 46(22):12704–12710. Online at
http://pubs.acs.org/doi/abs/10.1021/es302420z, accessed October
22, 2015.
Electric Power Institute and Natural Resources Defense
Council (EPRI and NRDC). 2015. Environmental assessment
of full electric transportation portfolio. Palo Alto, CA.
Online at www.epri.com/abstracts/Pages/ProductAbstract.
aspx?ProductId=3002006881, accessed October 22, 2015.
Elgowainy, A., J. Han, L. Poch, M. Wang, A. Vyas, M. Mahalik, and A.
Rousseau. 2010. Well-to-wheels analysis of energy use and green-
house gas emissions of plug-in hybrid electric vehicles. Argonne,
IL: Argonne National Laboratory. Online at www.transportation.
anl.gov/pdfs/TA/629.PDF, accessed October 22, 2015.
Environmental Protection Agency (EPA). 2015a. Green power loca-
tor. Washington, DC. Online at www.epa.gov/greenpower/pubs/
gplocator.htm, accessed October 22, 2015.
Environmental Protection Agency (EPA). 2015b. Clean Power Plan
for existing power plants. Washington, DC. Online at http://
www2.epa.gov/cleanpowerplan/clean-power-plan-existing-
power-plants, accessed October 22, 2015.
Environmental Protection Agency (EPA). 2015c. The emissions &
generation resource integrated database (eGRID): Tenth edition
with year 2012 data (eGRID20015 Version 1.0). Washington, DC.
Online at http://www2.epa.gov/energy/egrid, accessed October
12, 2015.
Environmental Protection Agency (EPA). 2015d. The emissions &
generation resource integrated database for 2012 (eGRID2012)
technical support document. Washington, DC. Online at http://
www2.epa.gov/energy/egrid2012-technical-support-document,
accessed October 20, 2015.
Environmental Protection Agency (EPA). 2014. Light-duty automo-
tive technology, carbon dioxide emissions, and fuel economy
trends: 1975-2014. Washington, DC. Online at www.epa.gov/
otaq/fetrends.htm, accessed June 1, 2015.
Environmental Protection Agency (EPA). 2013. Application of life-
cycle assessment to nanoscale technology: Lithium-ion batteries
for electric vehicles. EPA 744-R-12-001. Washington, DC.
Online at http://www2.epa.gov/sites/production/files/2014-01/
documents/lithium_batteries_lca.pdf, accessed October
22, 2015.
Environmental Protection Agency (EPA). 2012a. The emissions &
generation resource integrated database (eGRID): Eighth edition
with year 2009 data (eGRID2012 Version 1.0). eGRID2009_data.
xls file in eGRID files 1996–2012. Washington, DC. Online at
http://www2.epa.gov/energy/egrid, accessed October 22, 2015.
Environmental Protection Agency (EPA). 2012b. Regulatory impact
analysis: Final rulemaking for 2017–2025 light-duty vehicle
greenhouse gas emission standards and corporate average
fuel economy standards. Washington, DC. Online at www.epa.
gov/oms/climate/documents/420r12016.pdf, accessed October
2, 2015.
32
Swezey, B., and L. Bird. 2001. Utility green-pricing programs: What
defines success? Golden, CO: National Renewable Energy Labo-
ratory. Online at www.nrel.gov/docs/fy01osti/29831.pdf, accessed
October 22, 2015.
Tamayao, M.A., J.J. Michalek, C. Hendrickson, and I.M. Azevedo.
2015. Regional variability and uncertainty of electric vehicle life
cycle CO2 emissions across the United States. Environmental Sci-
ence and Technology 49(14):8844–8855. Online at http://pubs.acs.
org/doi/abs/10.1021/acs.est.5b00815, accessed August 28, 2015.
Tesla Motors. 2015. Tesla Model S. Online at www.teslamotors.com/
models, accessed October 22, 2015.
Tesla Motors. 2014. Gigafactory. Blog post, February 26. Online at
www.teslamotors.com/blog/gigafactory, accessed March 25, 2015.
Umicore. 2015. Umicore battery recycling. Online at www.
batteryrecycling.umicore.com/UBR, accessed October, 22, 2015.
Union of Concerned Scientists (UCS). 2015. Clean Fuels for Wash-
ington. Cambridge, MA. Online at www.ucsusa.org/sites/default/
files/attach/2015/02/Clean-Fuels-Washington.pdf, accessed
October, 22, 2015.
Union of Concerned Scientists (UCS). 2013a. How renewable
electricity standards deliver economic benefits. Cambridge,
MA. Online at www.ucsusa.org/assets/documents/clean_energy/
Renewable-Electricity-Standards-Deliver-Economic-Benefits.pdf,
accessed October, 22, 2015.
Union of Concerned Scientists (UCS). 2013b. New survey finds many
households can use electric vehicles. Press release, December 11.
Cambridge, MA. Online at www.ucsusa.org/news/press_release/
ev-survey-0384.html, accessed October, 22, 2015.
Union of Concerned Scientists (UCS). 2012. Half the oil plan. Cam-
bridge, MA. Online at www.ucsusa.org/clean-vehicles/clean-fuels/
half-the-oil-how-it-works, accessed October, 22, 2015.
Voelcker, J. 2015. Toyota Prius “coming soon”: 55-MPG hybrid to
be sportier. Green Car Reports, June 15. Online at www.
greencarreports.com/news/1098719_2016-toyota-prius-coming-
soon-55-MPG-hybrid-to-be-sportier, accessed June 20, 2015.
WardsAuto. 2015. U.S. light vehicle sales, December 2014. Online
at www.wardsauto.com/datasheet/us-light-vehicle-sales-
december-2014 (subscription required), accessed January
14, 2015.
Washington State Department of Commerce (WA DOC). 2015. Fuel
mix disclosure. Online at www.commerce.wa.gov/Programs/
Energy/Oce/Utilities/Pages/FuelMix.aspx, accessed October
22, 2015.
Weis, A., P. Jaramillo, and J. Michalek. 2014. Estimating the po-
tential of controlled plug-in hybrid electric vehicle charging
to reduce operational and capacity expansion costs for electric
power systems with high wind penetration. Applied Energy
115:190–204. Online at www.sciencedirect.com/science/article/pii/
S0306261913008374, accessed August 28, 2015.
Zackrisson, M., L. Avellan, and J. Orlenius. 2010. Life cycle assess-
ment of lithium-ion batteries for plug-in hybrid electric ve-
hicles—Critical issues. Journal of Cleaner Production 18(15):1519–
1529. Online at www.sciencedirect.com/science/article/pii/
S0959652610002167, accessed August 28, 2015.
Joint Center for Energy Storage Research (JCESR). 2015. Homepage.
Online at www.jcesr.org/about, accessed October 22, 2015.
Mai, T., R. Wiser, D. Sandor, G. Brinkman, G. Heath, P. Denholm,
D.J. Hostick, N. Darghouth, A. Schlosser, and K. Strzepek, 2012.
Exploration of high-penetration renewable electricity futures.
Volume 1 of Renewable Electricity Futures Study. NREL/
TP-6A20-52409-1, p.210. Golden, CO: National Renewable
Energy Laboratory. Online at www.nrel.gov/analysis/re_futures,
accessed October 22, 2015.
Majeau-Bettez, G., T.R. Hawkins, and A.H. Strømman. 2011. Life
cycle environmental assessment of lithium-ion and nickel metal
hydride batteries for plug-in hybrid and battery electric vehicles.
Environmental Science and Technology 45(10):4548–4554. Online
at http://pubs.acs.org/doi/abs/10.1021/es103607c, accessed August
28, 2015.
Nealer, R., and T.P. Hendrickson. 2015. Review of recent lifecycle
assessments of energy and greenhouse gas emissions for electric
vehicles. Current Sustainable/Renewable Energy Reports 2(3):66–
73. Online at http://link.springer.com/article/10.1007%2Fs40518-
015-0033-x, accessed October 22, 2015.
National Mining Association (NMA). 2015. 40 common minerals
and their uses. Online at www.nma.org/index.php/minerals-
publications/40-common-minerals-and-their-uses, accessed June
1, 2015.
Nissan. 2015. Vehicle Assembly Plant and Battery Plant— Smyrna,
Tennessee. Fact sheet. Online at www.nissannews.com/en-US/
nissan/usa/channels/Plant-Fact-Sheets/releases/vehicle-
assembly-plant-smyrna-tennessee, accessed March 25, 2015.
Notter, D.A., M. Gauch, R. Widmer, P. Wager, A. Stamp, R. Zah, and
H.-J. Althaus. 2010. Contribution of Li-ion batteries to the envi-
ronmental impact of electric vehicles. Environmental Science and
Technology 44(17):6550–6556. Online at http://pubs.acs.org/doi/
abs/10.1021/es903729a, accessed August 28, 2015.
Nykvist, B., and M. Nilsson. 2015. Rapidly falling costs of battery
packs for electric vehicles. Nature Climate Change 5:329–332.
Online at www.nature.com/nclimate/journal/v5/n4/full/
nclimate2564.html, accessed August 28, 2015.
Oce of Governor Edmund G. Brown Jr. 2012. Executive Order B-16-
2012. Sacramento, CA. Online at https://www.gov.ca.gov/news.
php?id=17472, accessed October 22, 2015.
Retriev Technologies. 2015. Lithium ion. Online at www.retrievtech.
com/recycling/lithium-ion, accessed June 1, 2015.
Saxena, S., C. Le Flonch, J. MacDonald, and S. Moura. 2015. Quantify-
ing EV battery end-of-life through analysis of travel needs with
vehicle powertrain models. Journal of Power Sources 282:265–
276. Online at www.researchgate.net/publication/272200824_
Quantifying_EV_Battery_End-of-Life_through_Analysis_of_
Travel_Needs_with_Vehicle_Powertrain_Models (subscription
required), accessed August 28, 2015.
SGL Automotive Carbon Fibers LLC (SGL). 2015. Moses Lake carbon
fiber production. Online at www.sglacf.com/en/production/
moses-lake-usa.html, accessed October 22, 2015.
33Cleaner Cars from Cradle to Grave
each one based on the emissions intensities of the plants it
contains. This methodology ignores imports and exports of
electricity between subregions, which harms the accuracy of
the regional emissions estimates. Therefore, the 26 subre-
gions are recommended by eGRID’s designers as the level of
disaggregation best suited for estimating electricity use–
related emissions, as they achieve the best balance between
the precision gained by disaggregation and the accuracy lost
by omitting imports and exports (EPA 2009).
Transmission Loss Factors
The eGRID emissions rates do not account for transmission
and distribution losses between the power plant and the
household. To calculate emissions per unit of energy used
(rather than energy produced), we followed eGRID’s recom-
mendation (EPA 2015d) to increase the emissions rates using
grid loss factors found in the data files in EPA 2015c, shown
here in Table A-2 (p. 36). There are five grid loss factors that
vary by regions called interconnect power grids, and each
eGRID subregion is given a grid loss factor based on the inter-
connect power grid to which it belongs (EPA 2015c).
Upstream Emissions Factors
The eGRID subregion emissions rates include only those
emissions produced at the plant generating the electricity,
and they exclude upstream emissions resulting from the
mining and transport of the power plant feedstock (EPA
2015d). Therefore we calculated a feedstock emissions rate
for each subregion; this rate depends on which fuel types
the corresponding power plants use. Each fuel type has a
unique upstream emissions rate, which we obtained from a
publicly available life cycle emissions model called GREET
(Greenhouse Gases, Regulated Emissions, and Energy Use in
Transportation), developed by the Argonne National
Laboratory (ANL 2014a; ANL 2014b).
15
The percentage of
generation from each fuel type in a subregion was then
obtained from eGRID2015 (EPA 2015c).
[
]
The global warming emissions we attribute to operating an
electric vehicle (EV) today are those that result from the pro-
duction of electricity needed to charge the vehicle. We factor
in emissions created by power plants when generating the
electricity, and also emissions that result from obtaining and
transporting the fuel used in these plants.
Power Plant Emissions
The electricity generation-related emissions values used in
our analysis come from the U.S. Environmental Protection
Agency’s (EPA’s) Emissions & Generation Resource Integrat-
ed Database (eGRID), which is a comprehensive source of
emissions data for every power plant in the United States that
provides its generation data to the government. We used to
the most up-to-date version of eGRID available—eGRID
2015v1.0—which contains plant emissions and generation
data from the year 2012 and subregion organization from the
year 2012 (EPA 2015c). The mixes of generation sources for
each region are shown in Table A-1.
The subregions are groups of plants organized by the
EPA, based on Power Control Areas and North American Re-
liability regions (EPA 2015d). These groupings reflect which
power plants serve which households, and they reasonably
approximate the grid mix of electricity used by those house-
holds. The global warming emissions rates for electricity gen-
eration for each of the 26 regions analyzed in the report come
from eGRID2015 (EPA 2015c).
The level of disaggregation of the eGRID subregions al-
lows for more precise calculation of plants’ emissions intensi-
ties than a national average, as regional variations in grid mix
are taken into account. For this reason, eGRID was chosen over
other data sources that had the same detailed plant informa-
tion but fewer subregions. The actual grid mix of a household’s
electricity is specific to the individual utilities serving each
household, but specific grid-mix data are not readily available
for most utilities and therefore were not used in the study.
eGRID’s methodology treats the subregions as closed sys-
tems, calculating the emissions intensity of generation for
Operation Emissions Modeling
15 GREET1 2014 was used. Feedstock emissions factors come from Table 9: Fuel-cycle energy use and emissions of electric generation: Btu or grams per mmBtu of
electricity available at user sites (wall outlets), in the Electricity tab.
34
Grid
Region
Acronym
Grid Region
Name % Coal
% Natural
Gas
% Other
Fossil % Nuclear
%
Biomass % Hydro
% Wind,
Solar,
Geothermal
AKGD
ASCCAlaska
Grid
AKMS
ASCC
Miscellaneous
ERCT ERCOTAll
FRCC FRCCAll
HIMS
HICC
Miscellaneous
HIOA HICCOahu
MROE MROEast
MROW MROWest
NYLI
NPCCLong
Island
NEWE
NPCCNew
England
NYCW
NPCCNYC
Westchester
NYUP
NPCCUpstate
NY
RFCE RFCEast
RFCM RFCMichigan
RFCW RFCWest
SRMW SERCMidwest
SRMV
SERC
Mississippi
Valley
SRSO SERCSouth
SRTV
SERC
Tennessee
Valley
SRVC
SERCVirginia
Carolina
SPNO SPPNorth
SPSO SPPSouth
CAMX
WECC
California
NWPP
WECC
Northwest
RMPA WECCRockies
AZNM
WECC
Southwest
TABLE A-1. Mix of Generation Sources for Each Grid Region
35Cleaner Cars from Cradle to Grave
for each subregion with the corresponding grid loss–adjusted
feedstock emissions rate.
Emissions Rate Assumptions and Results by
Subregion
The regional grid mix and estimated emissions intensity for
all eGRID subregions, with adjustments for upstream emis-
sions and grid losses, are shown in Tables A-1 and A-2.
Conversion of g/kWh to MPG
ghg
To translate electricity-related emissions intensity into driv-
ing-related emissions intensity (measured as gasoline miles-
per-gallon equivalent, or MPG
ghg
), we multiplied the EPA
emissions intensity values (gCO
2
e/kWh) from Table A-2 and
the EV average eciency values (kWh/mile) from Table 2,
resulting in a gCO
2
e/mile estimate. Then we used the GREET
carbon intensity of gasoline (ANL 2014a) and divided by the
gCO
2
e/mile estimate to get the estimated MPG
ghg
for each
region. This figure is an electric vehicle equivalent to the
MPG of a gasoline-powered vehicle: vehicles with the same
MPG
ghg
will produce the same amount of global warming pol-
lution for each mile traveled, regardless of fuel type.
For each subregion, the fuel-type emissions rates are
multiplied by the share of generation they represent in that
subregion; the sum of these products is the subregion’s feed-
stock emissions rate. Most fuel types in GREET correspond
directly to a fuel type in eGRID, but there were a few excep-
tions. A very small share of generation in eGRID subregions
corresponds to a fuel type labeled “generic fossil”; for this fuel
type, the emissions rate from GREET for natural gas was cho-
sen as a conservative guess, given that its upstream emissions
value is higher than those of coal and oil (the other two fossil
fuels with known feedstock emissions rates in GREET). An
even smaller share of generation in eGRID subregions comes
from unknown sources; for this category of fuel type, the
feedstock emissions rate is the generation-weighted average
of the upstream emissions rates for the other fuel types.
GREET has already built a uniform grid loss factor into
these feedstock emissions rates. But to keep the loss factors
consistent with those applied to power plant emissions, we
back GREET’s grid loss factor out of the feedstock emissions
rates. We then apply eGRID’s power plant grid loss factors to
each subregion’s feedstock emissions rate.
The totals of electricity generation-related global warm-
ing emissions for each eGRID subregion were computed by
summing the grid loss–adjusted power plant emissions rates
36
eGrid
Subregion
Acronym
Direct Emissions
at Power Plants
(gCO
2
e/kWh)
Transmission
Loss Multiplier
Emissions
from Power
Plants After
Transmission Loss
(gCO
2
e/kWh)
Upstream
Emissions After
Transmission
Loss (gCO
2
e/
kWh)
Emissions
Intensity from
Electricity
Generation
(gCO
2
e/kWh)
AKGD
AKMS
ERCT
FRCC
HIMS
HIOA
MROE
MROW
NYLI
NEWE
NYCW
NYUP
RFCE
RFCM
RFCW
SRMW
SRMV
SRSO
SRTV
SRVC
SPNO
SPSO
CAMX
NWPP
RMPA
AZNM
Note: Upstream emissions are those associated with the extraction and transportation of feedstocks for electricity generation. Emissions intensity is based on
2012 generation data. Sales-weighted average based on where EVs are sold today (480 gCO2e/kWh) is used to calculate the average operation emissions for EVs
(IHS 2015).
TABLE A-2. Emissions Intensity from Electricity Generation by Region in 2012
37Cleaner Cars from Cradle to Grave
We used the average curb weight of 3,000 pounds and
average fuel economy of 29 MPG to calculate the manufactur-
ing- and operation-related global warming emissions of a
gasoline vehicle comparable with the smaller 84-mile-
range BEV.
We applied the same methods as above to find five gasoline
vehicles comparable in footprint with a Tesla Model S, the car
on which the 265-mile-range full-size BEV is based—note that
the Tesla Model S footprint is 54 square feet (see Table B-2).
We used the average curb weight of 4,300 pounds and
average fuel economy of 21 MPG to model the manufacturing-
and operation-related global warming emissions of a gasoline
vehicle comparable with a 265-mile-range BEV.
Methods for Modeling Manufacturing
Emissions
For all of our vehicle manufacturing modeling eorts, we
used the 2014 versions both of GREET 1 (a fuel cycle model)
and GREET 2 (a vehicle cycle model) (ANL 2014a; ANL
2014b). In the absence of specific details about the gasoline
model and BEV model material composition, we used the
[
]
This appendix details the calculation of global warming emis-
sions from vehicle manufacturing. We describe the selection
of comparable gasoline vehicles, modifications to existing
manufacturing modeling, battery manufacturing sensitivities,
the impact of the electricity source on manufacturing emis-
sions, and the battery lifetime and replacement impacts.
Gasoline Vehicle Selection
In addition to estimating the global warming emissions from
battery-electric vehicles (BEVs), we chose comparable gaso-
line vehicles and then determined their manufacturing-
related global warming emissions. We examined a range of
gasoline vehicles that are similar to the representative
BEVs we modeled—an 84-mile-range midsize BEV and a
265-mile-range full-size BEV.
Table B-1 shows the five midsize cars we chose because
of their similarity in vehicle footprint to a Nissan LEAF, the
midsize BEV whose representative we modeled—note that
the Nissan LEAF has a footprint of 45 square feet. We used
data for the most popular version of each vehicle make and
model (WardsAuto 2015).
Manufacturing Emissions Modeling
TABLE B-1. Midsize Gasoline Vehicles Comparable with the Nissan LEAF
Make Model Engine
Fuel Economy
(MPG)
Curb Weight
(lb.)
Footprint
(sq. ft.)
Mazda -or-Doori LI
Ford Focus(Hatchback) LI
Mitsubishi LancerSportback LI
Volkswagen Golf LTurboI
Kia Forte
LTurboand
LI
Average(harmonic
forfueleconomy)
Note: The Kia Forte5 numbers shown are sales-weighted averages of vehicles oered with two types of engines.
SOURCE: DOE 2015A.
38
Comparing our results with other battery literature
(Dunn et al. 2014; EPA 2013; Dunn et al. 2012; Notter et al.
2010), the emissions (kilograms of CO
2
per kilogram of bat-
tery weight) depend on the battery chemistry. These esti-
mates are on the lower end of the spectrum for battery-
production global warming emissions because they derive
from process-level analyses. The alternative approach—top-
down methods, which refer to how the battery production
model default values related to the supply chain global warm-
ing emissions of each material. Using the GREET framework,
the percentage of each material in the vehicles was scaled by
total vehicle weight.
Specifications of Modifications in GREET
The material composition of the vehicles is an important
component of modeling vehicle manufacturing because the
material’s life cycle global warming emissions vary by mate-
rial type. For example, manufacturing wrought aluminum
produces, by weight, more than twice the global warming
emissions of steel. To complicate this relationship for dier-
ent parts of the vehicle, the replacement ratio varies. Because
we had limited data on the actual composition of the vehicle
models, we used the best available data in the GREET model.
We modeled the midsize and full-size gasoline car as an
average gasoline car; the midsize 84-mile-range BEV as an
average battery-electric vehicle; and the full-size 265-mile-
range BEV as a lightweight battery-electric vehicle. Moreover,
we modeled the full-size 265-mile-range BEV with more alu-
minum and no carbon fiber–reinforced plastics to more accu-
rately represent what we know industry is providing today
(Tesla Motors 2015). However, for the majority of the materi-
als we used the GREET default values because further infor-
mation on the specific models was not available. Table B-3
shows the composition by percentage of the vehicle weight
(without the battery or fluids), as defined by GREET; note,
however, that the values in Table B-3 are dierent from the
GREET model defaults for the full-size BEV.
The modifications made for the battery specifications are
detailed in Table B-4.
TABLE B-2. Full-size Gasoline Vehicles Comparable with the Tesla Model S
Make Model Engine
Fuel Economy
(MPG)
Curb Weight
(lb.)
Footprint
(sq. ft.)
Hyundai Equus LV
Chrysler RWD LV
Mercedes SRWD LV
Porsche Panamera LV
Audi A LV
Average(harmonic
forfueleconomy)
SOURCE: DOE 2015A; WARDSAUTO 2015.
TABLE B-3. Composition of Vehicles by Material Type
Material
Gasoline
Vehicle
Midsize
BEV
Full-size
BEV
Steel
CastIron
Wrought
Aluminum
CastAluminum
CopperBrass
Glass
AveragePlastic
Rubber
GlassFiber–
reinforcedPlastic
Other
Note: Excludes batteries and fluids.
39Cleaner Cars from Cradle to Grave
supply chain manufacturing dierences of the cathode chem-
istries. Assembly of the batteries remains the same across the
battery chemistries.
Electricity Mix for Battery and Vehicle
Manufacturing
We investigated the impact that electricity-related global
warming emissions could have on vehicles’ manufacturing
alone—i.e., the electricity used to power manufacturing equip-
ment—in light of the fact that dierent sources of electricity
result in higher or lower global warming emissions. The elec-
tricity mixes with more fossil fuels cause higher global warm-
ing emissions for vehicle manufacturing. We investigated the
impacts of a future electricity grid with 80 percent renewable
energy sources, and we concluded that vehicle manufactur-
ing-related global warming emissions could be reduced by up
to 45 percent, as opposed to a grid with 90 percent coal power,
as is the case today in China. These impacts are similar for
gasoline and BEV. Compared with today’s average U.S. grid
mix, manufacturing emissions could be reduced by just over
30 percent if the United States generated 80 percent of its
electricity from renewable sources.
Vehicle and Battery Lifetime
The most common way to compare the life cycle impact of
one vehicle with that of another is to measure emissions per
distance driven over their lifetimes. BEVs tend to create more
energy is assessed—results in higher estimates because the
scope of the assessment is larger (Hawkins et al. 2013; Ma-
jeau-Bettez, Hawkins, and Strømman 2011; Zackrisson, Avel-
lan, and Orlenius 2010). These values are dependent on
battery chemistry.
We conducted a sensitivity analysis, based on battery
chemistry, around the default GREET values for a 28 kWh
battery. If the battery chemistry were changed, the global
warming emissions for the midsize battery could decrease by
18 percent or increase by 45 percent, and the full-size battery
global warming emissions could decrease by up to 43 percent,
when using alternative battery chemistries for the same 28
kWh output (see Table B-5). The dierences are due to the
TABLE B-4. Lithium-ion Battery Specifications for
Midsize and Full-size BEVs
Midsize BEV
Battery
Full-size BEV
Battery
Chemistry
Lithiumnickel
manganese
cobaltoxide
Lithiumcobalt
oxidevia
hydrothermal
process
SpecificEnergy
(Whkg)
Weight(kg)
Size(kWh)
TABLE B-5. Global Warming Emissions Changes Based on Battery Chemistry for 28 kWh Battery
Global Warming Emissions Change
from Modeled BEV Battery Chemistries (%)
Battery Chemistry Midsize BEV Battery Full-size BEV Battery
LithiumManganeseOxide(LMO) - -
LithiumNickelManganeseCobaltOxide(NMC) -- -
LithiumIronPhosphate(LFP)viaHydrothermalProcess - -
LithiumIronPhosphate(LFP)viaSolidStateProcess - -
LithiumCobaltOxide(LCO)viaHydrothermalProcess --
LithiumCobaltOxide(LCO)viaSolidStateProcess -
AdvancedLithiumNMC(LMR-NMC)withGraphite - -
AdvancedLithiumNMC(LMR-NMC)withSilicon -
40
unwilling to drive long distances very often, given the fre-
quent need for stopping to “fill up” (Plotz 2015; Carlson et al.
2014). Table B-6 shows the annual distances assumed to driv-
en during each year of the modeled vehicles’ lifetimes.
We explored the possibility that the midsize 84-mile-
range BEV might not reach the full 135,000 miles on one lithi-
um-ion battery, due to required or voluntary battery
replacement. We analyzed the impacts of one, two, and 1.5
battery replacements, with the latter representing a mix of
some BEVs needing only one and others needing two battery
replacements. Figure B-1 shows the results of this analysis for
the midsize vehicle. An important finding is that despite high-
er global warming emissions for more battery replacements,
the total emissions are still lower than those of a comparable
gasoline vehicle. The total life cycle global warming emis-
sions increase by 6 percent and 11 percent for 1.5 and 2 batter-
ies, respectively, compared with only one battery for the
lifetime of the vehicle.
global warming emissions when they are manufactured than
gasoline vehicles because of the former’s battery production.
However, gasoline vehicles give o more global warming
emissions per mile while being driven. This creates a tradeo:
greater upfront emissions from the manufacturing of BEVs
versus higher emissions from the use of gasoline vehicles be-
cause of their engines’ combustion.
For this study we assumed a lifetime of 179,000 miles,
both for gasoline and long-range battery-electric vehicles,
based on the National Household Travel Survey (FHWA
2009) data for the first 15 years of a vehicle’s lifetime. How-
ever, we posited an exception for the 84-mile-range BEV and
comparable gasoline car—that total mileage would be
135,000—75 percent of the mileage of the 265-mile-range
BEV. This dierence is due to “range limitations” of a car
with a more modest-sized battery: its driver would likely be
TABLE B-6. First 15 Years of Vehicle Lifetime in Mileage
Annual Mileage
Vehicle
Age
Gasoline
Car
84-mile-
range BEV
265-mile-
range BEV
TotalMiles
SOURCE: EPA 2012B.
FIGURE B-1. Life Cycle Global Warming Emissions
from the Manufacturing and Operation of Gasoline and
Battery-Electric Vehicles with More Than One Battery
Replacement
Midsize
Average
Gasoline
Car
Midsize
84-mile-
range BEV
(1 Li-ion
Battery)
Midsize
84-mile-
range BEV
(1.5 Li-ion
Batteries)
Midsize
84-mile-
range BEV
(2 Li-ion
Batteries)
0
10
30
20
50
40
60
OperationBattery Manufacturing
Vehicle Manufacturing
Life Cycle Global Warming Emissions
(tons CO
2
e)
Note: This figure assumes battery production scales linearly and only battery
manufacturing emissions are increased, not total vehicle manufacturing emis-
sions. We only show the results for the midsize 84-mile BEV, because the re-
sults are similar for the full-size 265-mile BEV.
41Cleaner Cars from Cradle to Grave
potential for recycling to reduce emissions and increase re-
source utilization.
Lithium-ion Battery Reuse
Various pilot projects are under way to reuse lithium-ion bat-
teries for other purposes after the vehicle is retired. From a life
cycle perspective, this would eectively spread the impacts of
the battery over a longer battery lifetime. But without more
information on the nature of the battery second use, and on
how long it is used, we are unable to estimate the emissions
savings this would have on the life cycle of the battery. There-
fore for this report we assume that all global warming emis-
sions are attributed to the first use of the battery—in the vehicle.
[
]
We assume that BEVs will be disposed of, with the exception
of their lithium-ion batteries, in similar ways as their gasoline
counterparts. This section describes some of the details of the
recycling possibilities, related assumptions in the manufac-
turing modeling, and second use of the vehicle battery after it
is retired.
Recycling of Vehicle Materials
Table C-1 shows the recycling assumptions for each material
in GREET, the Argonne National Laboratory’s vehicle manu-
facturing emissions model. When these assumptions are
changed to use all virgin materials, the global warming emis-
sions from the vehicles’ manufacture increase by about 15 to
20 percent. Note that “recycling rates” refer to what is used in
the manufacturing of the vehicle, not to what is recycled at
the end of the vehicles’ lifetime; although a large percentage
of the vehicle is recycled, not all the materials reenter the au-
tomotive industry.
Lithium-ion Battery Recycling
There are only two companies that currently have the capa-
bility to recycle the large lithium ion batteries required for
BEVs—Retriev Technologies and Umicore (Retriev Technolo-
gies 2015; Umicore 2015)—and there are three dierent ap-
proaches to the recycling: hydrometallurgy, pyrometallurgy,
and direct physical recycling. It is expected that recycling can
reduce battery manufacturing–related energy consumption
by 10 to 17 percent, depending on the manufacturing process
and battery type (Hendrickson et al. 2015; EPA 2013). Our
results do not incorporate this energy savings because
there are sparse data on how industry is currently recycling
batteries or intends to do so in the future, but there is
Disposal, Recycling, and Reuse
TABLE C-1. Recycling Rates of Metals for All Vehicles
Modeled
Virgin
Material
Recycled
Material
Steel
WroughtAluminum
CastAluminum
Lead
Nickel
Magnesium
Note: These recycling rates are developed by GREET, and left unmodified for
the vehicle modeling presented in this report.
SOURCE: ANL 2014A; ANL 2014B.
42
plant emissions for the year 2012. In eGRID, the U.S. Environ-
mental Protection Agency (EPA) assembled global warming
and other emissions data from thousands of power plants op-
erating across the country. The eGRID subregion emissions
rates include only those emissions produced at the plant gen-
erating the electricity, and they exclude transmission and up-
stream emissions resulting from the mining and transport of
the power plant feedstock (EPA 2015d). To account for the
transmission emissions, we increased the emissions rates us-
ing grid loss factors found in eGRID2015 (EPA 2015c) for the
transmission losses. To account for the upstream emissions,
we used a feedstock emissions rate obtained from
GREET1_2014 (ANL 2014a).
Marginal Emissions Estimation
An alternative approach involves “marginal” emissions inten-
sity, which is estimated by identifying which power plants, or
types of power plants, are likely to be deployed or increase
their output when new electricity demand is added to the
electricity grid above and beyond the demand that already
exists. In this type of analysis, the electricity consumed by an
additional load, such as a newly purchased EV or even an ex-
tra television set, would be assigned a dierent emissions in-
tensity from electricity used by existing electric loads—e.g.,
a light fixture in your home. A variety of analyses have used
various marginal emissions approaches to evaluate the poten-
tial impacts of increasing amounts of EV charging on future
emissions of the electricity grid (EPRI and NRDC 2015;
Tamayao et al. 2015; Gra Zivin, Kotchen, and Mansur 2014;
Elgowainy et al. 2010; Hadley and Tsvetkova 2008). These
marginal emissions analyses can be broken into two dierent
categories: short-term and long-term.
The short-term approach looks at how the electricity
grid responds instantaneously to a new load, such as when an
EV is plugged in. In this approach, the emissions from plug-
ging in the EV are tied specifically to how the grid would re-
spond to the new load, all other factors being fixed. Increases
in electricity demand are met through increasing generation
output at a power plant that is operating at less than full
[
]
Electricity is produced using a mix of generation units that
vary in size, fuel, and eciency. This mix of generation varies
over both long and short time scales, as demand for electric-
ity, availability, and fuel costs are always changing. Emissions
attributable to electricity use are linked directly to this gen-
eration mix; the emissions from electricity vary by region,
time of year, and time of day. Because of the complexity of the
electricity grid and how it operates, as well as the inability to
track specific electricity generation to a specific end use, mul-
tiple methods have been developed to estimate the emissions
from electricity use. For this analysis, an average emissions
approach is used, in which the emissions from electricity pro-
duction are averaged over all of the electricity generating
units in an entire electricity grid region for a year. The ratio-
nale for this choice is described below, as along with a discus-
sion of alternative approaches.
Average Emissions Estimation
To estimate electric-vehicle emissions from plugging into the
electricity grid, the average global warming emissions inten-
sity (i.e., emissions emitted for each net kilowatt-hour of elec-
tricity delivered) is calculated by region. This method of
averaging emissions intensity treats all of the electricity pro-
duced and consumed in the region equally. That is, no matter
how much electricity you use or when you use it, your elec-
tricity is assumed to be just as clean (or dirty) as anyone else’s
in the same region. In essence, this approach assumes that
any additional electricity needed to power an EV would come
from the same mix of sources that generate electricity to meet
all other current demands. (The import and export of elec-
tricity across regions are not accounted for in the average
emissions approach.) Using the averaging approach allows
changes in the underlying generation mix to be captured
when estimating future years’ emissions. However, as de-
scribed below, it does not reflect the short-term changes in
the electricity grid that may result from a new electricity load
being added to the grid.
The data used in this report to calculate regional global
warming emissions intensities were based on actual power
Average vs. Marginal Electricity Global
Warming Emissions
43Cleaner Cars from Cradle to Grave
electricity grid to be evaluated, including power plant
retirements, new electricity generation, and changes in EV
demand. Importantly, it also allows for the evaluation of poli-
cies to reduce emissions from both the transportation and elec-
tricity sectors, as well as for estimates of the cumulative impact
on emissions from both sectors. For example, one could use
this approach to examine the impact of the increased EV de-
ployment triggered by implementation of the EPA’s Clean
Power Plan.
While a long-term marginal emissions approach doesn’t
tell us what the emissions are from EVs today, it is an impor-
tant tool for assessing the impacts of transportation and en-
ergy policies designed to reduce emissions—of deploying
more EVs while also deploying cleaner electricity sources.
But because this approach requires modeling both of the
transportation and energy sectors, and of the specific changes
to the electricity grid that might occur under various future
scenarios, it was outside the scope of this report’s analysis.
Why We Used Average Emissions Estimation
The goal of this analysis was to identify the typical global
warming emissions of the mix of electricity sources used to
charge EVs on today’s power grid, as well as to evaluate how
that mix changes over time and compares with past and pos-
sible future electricity grids. Therefore, we used the average
emissions intensity of the electricity, essentially treating all
electricity on the grid at a given time as a shared resource
available to all electricity consumers. This approach does not
capture the very short-term marginal emissions impact on the
grid from plugging in a new EV, but it does reflect changes
that are occurring in non-marginal load generation around
the country. The average emissions approach also allows for
comparison with future and past emissions analyses and cap-
tures the impact of ongoing changes to the electricity grid as a
whole resulting from regulatory policy and other factors. In
other words, as consumers buy EVs today, the trajectory of
the grid and the global warming emissions over the life of the
vehicles should be taken into account.
output—typically, a natural gas or coal power plant. These
types are considered the marginal generation sources. In con-
trast, sources such as nuclear, hydro, wind, and solar are rare-
ly “on the margin” because they have limited ability to vary
output. These electricity sources provide non-marginal
generation.
This short-term marginal emissions approach can provide
a more precise snapshot of how the grid responds to a new
load during a short amount of time, and it quantifies the net
emissions change during that period. Carrying out the same
type of analysis in future years could produce very similar re-
sults, regardless of changes to non-marginal load generation.
For example, if over some time period 25 percent of electricity
generation in a region moved to renewable sources, fossil fuel
power plants may still be the only electricity sources on the
margin responding to instantaneous increases in demand for
electricity. So an EV powered on a grid with no renewables and
one with 25 percent renewables might still have the same
emissions profile using this type of marginal emissions analy-
sis. In addition, new electricity demand will eventually lead to
changes in the source of electricity production. Over time a
large number of EVs will create significant demand that will
need to be met through either greater energy eciency, in-
creased utilization of existing sources, new electricity genera-
tion, or very likely a combination of all three. A short-term
marginal analysis only considers increased utilization of exist-
ing generating resources, though researchers have been look-
ing into a more consequential approach with new capacity as a
consideration (Weis, Jaramillo, and Michalek 2014).
The long-term marginal emissions approach, or a conse-
quential life cycle approach, evaluates how the electricity grid
responds over a longer time period. This approach can esti-
mate what would happen to the grid without new electricity
load being added and then contrast this outcome with what
would happen under a the new load. For example, an analysis
could estimate electricity demand between 2015 and 2030
assuming no EVs and then what demand would look like
with several million EVs added (EPRI and NRDC 2015). This
type of modeling approach allows long-term changes in the
: www.ucsusa.org - © NOVEMBER 2015
NATIONAL HEADQUARTERS
Two Brattle Square
Cambridge, MA 02138-3780
Phone: (617) 547-5552
Fax: (617) 864-9405
WASHINGTON, DC, OFFICE
1825 K St. NW, Suite 800
Washington, DC 20006-1232
Phone: (202) 223-6133
Fax: (202) 223-6162
WEST COAST OFFICE
500 12th St., Suite 340
Oakland, CA 94607-4087
Phone: (510) 843-1872
Fax: (510) 843-3785
MIDWEST OFFICE
One N. LaSalle St., Suite 1904
Chicago, IL 60602-4064
Phone: (312) 578-1750
Fax: (312) 578-1751
The Union of Concerned Scientists puts rigorous, independent science to work to solve our planet’s most pressing problems. Joining with citizens across
the country, we combine technical analysis and eective advocacy to create innovative, practical solutions for a healthy, safe, and sustainable future.
: www.ucsusa.org/EVlifecycle
Electric vehicles (EVs) are a critical part of the American trans-
portation future given their potential to dramatically cut oil use
and global warming emissions—especially when charged by a
clean-electricity grid. Based on our calculations that weighted
where EVs were sold in 2014, along with updated power plant
emissions data, driving an EV in the United States produced
global warming emissions similar to a gasoline vehicle that
gets 68 miles per gallon, on average. And over its lifetime—from
manufacturing to operation to disposal—a battery-electric vehicle
(BEV) cuts emissions just over 50 percent relative to a compa-
rable gasoline car.
To reach their full potential, EVs must account for a larger
share of vehicle sales while the electricity grid shifts from coal to
low-carbon renewable sources. This report presents the compre-
hensive results of comparing the global warming emissions of
BEVs with their gasoline counterparts in the United States today,
accompanied by recommendations on how to increase their envi-
ronmental benefits over the next 30 years.
Cleaner Cars from
Cradle to Grave
How Electric Cars Beat Gasoline Cars on
Lifetime Global Warming Emissions
On average, driving an electric vehicle produces
lower global warming emissions than the most
fuel-ecient gasoline car on the market today.
