559

ANL/ESD/09-2
Well-to-Wheels Energy Use

and Greenhouse Gas Emissions
Analysis of Plug-in Hybrid Electric Vehicles
Energy Systems Division

About Argonne National Laboratory
Argonne is a U.S. Department of Energy laboratory managed by UChicago Argonne, LLC
under contract DE-AC02-06CH11357. The Laboratory’s main facility is outside Chicago,
at 9700 South Cass Avenue, Argonne, Illinois 60439. For information about Argonne
and its pioneering science and technology programs, see www.anl.gov.
Availability of This Report

This report is available, at no cost, at http://www.osti.gov/bridge. It is also available
on paper to the U.S. Department of Energy and its contractors, for a processing fee, from:
U.S. Department of Energy
Office of Scientific and Technical Information
P.O. Box 62
Oak Ridge, TN 37831-0062
phone (865) 576-8401
fax (865) 576-5728
[email protected]
Disclaimer
This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States
Government nor any agency thereof, nor UChicago Argonne, LLC, nor any of their employees or officers, makes any warranty, express
or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus,
product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific
commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply
its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of
document authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof,
Argonne National Laboratory, or UChicago Argonne, LLC.
ANL/ESD/09-2
Well-to-Wheels Energy Use

and Greenhouse Gas Emissions
Analysis of Plug-in Hybrid Electric Vehicles
By
A. Elgowainy, A. Burnham, M. Wang, J. Molburg, and A. Rousseau
Center for Transportation Research
Energy Systems Division, Argonne National Laboratory
February 2009
CONTENTS
ACKNOWLEDGMENTS ........................................................................................................ vii
ABSTRACT.............................................................................................................................. 1
1 INTRODUCTION............................................................................................................. 3
2 ELECTRICITY GENERATION MIX.............................................................................. 6
2.1 Factors Affecting Generation Mix for PHEV Charging .......................................... 6

2.1.1 Time of Day ................................................................................................ 7
2.1.2 Time of Year ............................................................................................... 8
2.1.3 Climate........................................................................................................ 8
2.1.4 Generation Mix ........................................................................................... 9
2.1.5 Vehicle and Charger Design Factors .......................................................... 10
2.1.6 Load Growth and Generation Expansion.................................................... 11
2.2 PHEV Grid Impact Studies ...................................................................................... 12
2.2.1 Background ................................................................................................. 12
2.2.2 Methods....................................................................................................... 13
2.3 Adoption of Marginal Mix in GREET..................................................................... 13
3 PSAT SIMULATION OF VEHICLE FUEL ECONOMY ............................................... 16
4 VEHICLE MILES TRAVELED SPLIT BY CHARGE-DEPLETING

VERSUS CHARGE-SUSTAINING OPERATION ......................................................... 21
5 GREET WELL-TO-WHEELS ENERGY USE AND GREENHOUSE GAS

EMISSIONS CALCULATIONS ...................................................................................... 24
5.1 Overview.................................................................................................................. 24
5.2 Results of Well-To-Wheels Simulation ................................................................... 25
6 IMPLICATIONS FOR FUTURE RESEARCH................................................................ 46
7 CONCLUSIONS ............................................................................................................... 47
8 REFERENCES .................................................................................................................. 48
APPENDIX: RECENT GRID IMPACT STUDIES................................................................. 51
iii
TABLES
1 Vehicle Technologies, Fuels, and Feedstock Sources....................................................... 4
2 Mix of Sources for Average Electric Generation among Regions in the

United States...................................................................................................................... 10
3 Distribution of U.S. Energy Consumption ........................................................................ 12
4 Generation Mixes for Recharging PHEVs for Use in GREET ......................................... 15
5 Vehicle Assumptions for PSAT Simulations .................................................................... 17
6 PSAT Electricity Use and Fuel Economy Results ............................................................ 18
7 Share of National VMT Available for Substitution by a PHEV Using

100% Grid Electricity in CD Mode until Depletion.......................................................... 21
8 Fuel Consumption Calculated from PSAT Simulated Fuel Economy Results ................. 25
FIGURES
1 NERC Regions from the Annual Energy Outlook 2007 ................................................... 6
2 Typical Summer Load Profile and Dispatch Scheme for Many U.S. Utilities.................. 7
3 Example of Hourly Marginal Fuels Data by Time of Day................................................ 8
4 SPP Daily Maximum and Minimum Electricity Demand Source..................................... 9
5 CD VMT on UDDS from PSAT Simulations ................................................................... 19
6 CD VMT on HWFET from PSAT Simulations ................................................................ 19
7 Average VMT for UDDS and HWFET on CD Operation ................................................ 20
8 National VMT Available for Substitution by a PHEV Using

100% Electric CD Mode ................................................................................................... 22
9 Fuel Consumption in CD Operation.................................................................................. 26
10 Fuel Consumption in CD and CS Operations ................................................................... 26
iv
FIGURES (CONT.)
11(a) WTW Total Energy Use for CD and CS Operations of PHEV 20 Using
the California Marginal Mix ........................................................................................... 28
11(b) WTW Fossil Energy Use for CD and CS Operations of PHEV 20 Using
11(c) WTW Petroleum Energy Use for CD and CS Operations of PHEV 20 Using
11(d) WTW GHG Emissions for CD and CS Operations of PHEV 20 Using

the U.S. Mix.................................................................................................................... 31
the U.S. Mix.................................................................................................................... 31
the U.S. mix .................................................................................................................... 32

the U.S. Mix.................................................................................................................... 32
the Illinois Marginal Mix ................................................................................................ 33

14 WTW Total Energy Use for CD and CS Operations of PHEV 20 Using

15(a) WTW Total Energy Use for Combined CD and CS Operations of PHEV 20
Using the California Marginal Mix................................................................................. 36
v
FIGURES (CONT.)
15(b) WTW Fossil Energy Use for Combined CD and CS Operations of

PHEV 20 Using the California Marginal Mix ................................................................ 37
15(c) WTW Petroleum Energy Use for Combined CD and CS Operations of

15(d) WTW GHG Emissions for Combined CD and CS Operations of

16 WTW Petroleum Energy Use for Combined CD and CS Operations as a

Function of AER ............................................................................................................. 38
17 WTW GHG Emissions for Combined CD and CS Operations as a

Function of AER Using the California Marginal Generation Mix ................................. 39

Function of AER Using the U.S. Generation Mix .......................................................... 40

Function of AER Using the Illinois Marginal Generation Mix ...................................... 40
20 WTW Total Energy Use for PHEV 20 Vehicle/Fuel Systems Using

Different Marginal Electricity Generation Mixes........................................................... 41
21 WTW Fossil Energy Use for PHEV 20 Vehicle/Fuel Systems Using

22 WTW Petroleum Energy Use for PHEV 20 Vehicle/Fuel Systems Using

23 WTW GHG Emissions for PHEV 20 Vehicle/Fuel Systems Using

24 Summary of WTW Petroleum Energy Use and GHG Emissions for Combined
CD and CS Operations Relative to Baseline Gasoline ICEV ......................................... 45
vi
ACKNOWLEDGMENTS
This study was supported by the U.S. Department of Energy, Assistant Secretary for
Energy Efficiency and Renewable Energy, the Hydrogen, Fuel Cells, and Infrastructure
Technologies Program, under contract DE-AC02-06CH11357. We thank Fred Joseck of the
Hydrogen, Fuel Cells, and Infrastructure Technologies Program for his input to this study, Phillip
Sharer and Sylvain Pagerit of Argonne National Laboratory for their support of PSAT
simulations for this analysis, and Anant Vyas and Danilo Santini for their input regarding vehicle
miles driven with grid power. We also thank Stanton Hadley for providing us with the raw data
of the marginal generation mix charts for each NERC region.
vii
viii
1
WELL-TO-WHEELS ENERGY USE AND GREENHOUSE GAS EMISSIONS

ANALYSIS OF PLUG-IN HYBRID ELECTRIC VEHICLES
Amgad Elgowainy, Andrew Burnham, Michael Wang,

John Molburg, and Aymeric Rousseau
ABSTRACT
Researchers at Argonne National Laboratory expanded the Greenhouse gases, Regulated

Emissions, and Energy use in Transportation (GREET) model and incorporated the fuel
economy and electricity use of alternative fuel/vehicle systems simulated by the Powertrain
System Analysis Toolkit (PSAT) to conduct a well-to-wheels (WTW) analysis of energy use and
greenhouse gas (GHG) emissions of plug-in hybrid electric vehicles (PHEVs). The WTW results
were separately calculated for the blended charge-depleting (CD) and charge-sustaining (CS)
modes of PHEV operation and then combined by using a weighting factor that represented the
CD vehicle-miles-traveled (VMT) share. As indicated by PSAT simulations of the CD operation,
grid electricity accounted for a share of the vehicle’s total energy use, ranging from 6% for a
PHEV 10 to 24% for a PHEV 40, based on CD VMT shares of 23% and 63%, respectively.
In addition to the PHEV’s fuel economy and type of on-board fuel, the marginal
electricity generation mix used to charge the vehicle impacted the WTW results, especially GHG
emissions. Three North American Electric Reliability Corporation regions (4, 6, and 13) were
selected for this analysis, because they encompassed large metropolitan areas (Illinois, New
York, and California, respectively) and provided a significant variation of marginal generation
mixes. The WTW results were also reported for the U.S. generation mix and renewable
electricity to examine cases of average and clean mixes, respectively. For an all-electric range
(AER) between 10 mi and 40 mi, PHEVs that employed petroleum fuels (gasoline and diesel), a
blend of 85% ethanol and 15% gasoline (E85), and hydrogen were shown to offer a 40–60%,
70–90%, and more than 90% reduction in petroleum energy use and a 30–60%, 40–80%, and
10–100% reduction in GHG emissions, respectively, relative to an internal combustion engine
vehicle that used gasoline. The spread of WTW GHG emissions among the different fuel
production technologies and grid generation mixes was wider than the spread of petroleum
energy use, mainly due to the diverse fuel production technologies and feedstock sources for the
fuels considered in this analysis.
The PHEVs offered reductions in petroleum energy use as compared with regular hybrid
electric vehicles (HEVs). More petroleum energy savings were realized as the AER increased,
except when the marginal grid mix was dominated by oil-fired power generation. Similarly,
more GHG emissions reductions were realized at higher AERs, except when the marginal grid
generation mix was dominated by oil or coal. Electricity from renewable sources realized the
largest reductions in petroleum energy use and GHG emissions for all PHEVs as the AER
increased. The PHEVs that employ biomass-based fuels (e.g., biomass-E85 and -hydrogen) may
not realize GHG emissions benefits over regular HEVs if the marginal generation mix is
dominated by fossil sources.
2
Uncertainties are associated with the adopted PHEV fuel consumption and marginal
generation mix simulation results, which impact the WTW results and require further research.
More disaggregate marginal generation data within control areas (where the actual dispatching
occurs) and an improved dispatch modeling are needed to accurately assess the impact of PHEV
electrification. The market penetration of the PHEVs, their total electric load, and their role as
complements rather than replacements of regular HEVs are also uncertain. The effects of the
number of daily charges, the time of charging, and the charging capacity have not been evaluated
in this study. A more robust analysis of the VMT share of the CD operation is also needed.
3
1 INTRODUCTION
The U.S. Department of Energy’s (DOE’s) Vehicle Technology Program examines the
precompetitive, high-risk research needed to develop the component and infrastructure
technologies necessary to enable a full range of affordable cars and light trucks that will reduce
the U.S. dependence on imported oil and minimize harmful vehicle emissions, without
sacrificing freedom of mobility and freedom of vehicle choice.1 Currently, plug-in hybrid electric
vehicles (PHEVs) are being developed for mass production by the automotive industry and have
been touted for their potential to reduce the transportation system’s petroleum dependence and
greenhouse gas (GHG) emissions by using off-peak excess electric generation capacity and
increasing the vehicle's energy efficiency. These vehicles are similar to regular hybrid electric
vehicles (HEVs), except that the battery uses electricity from the grid by being recharged
through a wall outlet. They share similar characteristics of regular HEVs, having an electric
motor and an on-board power unit, such as an internal combustion engine (ICE) or fuel cell (FC),
hereinafter referred to as “engine” for simplicity. The PHEV category can cover a wide variety
of options with respect to technical attributes, such as the battery chemistry, the amount of grid
electricity that can be stored in the battery, and the powertrain and fuel choices, which could
significantly impact the environment. In addition, the behavior of consumers, revealed by where
they live, when they charge, and how they drive, could also significantly affect the energy use
and emissions of PHEVs.
In the 1990s, PHEV prototypes were built in student competitions cosponsored by U.S.
automakers and DOE, while Japanese automakers introduced commercial HEVs that provided
significantly lower fuel consumption than similar internal combustion engine vehicles (ICEVs)
(Gaines et al. 2007). In 2001, as a response to these developments, both the Electric Power
Research Institute (EPRI) and DOE’s national laboratories began evaluating PHEVs (Graham et
al. 2001; Plotkin et al. 2001). Although these reports examined vehicles with nickel metal
hydride batteries, the recent interest in studying the effects of PHEVs has been spurred by
improvements in the energy density and cost of lithium-ion batteries.
While PHEVs offer the potential for significant reduction in the vehicle’s petroleum
energy use and GHG emissions, the significance of these benefits may not be fully realized due
to the upstream energy and emissions penalties associated with the electricity generation needed
to power the electric vehicle-miles-traveled (VMT) share. The implications of the upstream
marginal electricity generation mix as well as the PHEV’s powertrain technology, fuel source,
and all-electric range(AER) rating can be fully understood through a well-to-wheels (WTW)
assessment of energy use and GHG emissions, as provided by this analysis.
With funding from DOE, the Center for Transportation Research at Argonne National
Laboratory (Argonne) developed the Greenhouse gases, Regulated Emissions, and Energy use in
Transportation (GREET) model to estimate the full fuel-cycle energy use and emissions for
alternative transportation fuels and advanced vehicle systems (Wang 1999). In estimating the
1
U.S. DOE Plug-in Hybrid Electric Vehicle R&D Plan, available at: http://www1.eere.energy.gov/
vehiclesandfuels/pdfs/program/phev_rd_plan_02-28-07.pdf. Last accessed October 2008.
4
fuel-cycle energy use in British thermal units per mile (Btu/mi) and GHG emissions in grams per
mile (g/mi) for advanced vehicle technologies, including PHEVs, GREET tracks their
occurrences from the primary energy source to the vehicle's operation, which is known as a
“well-to-wheels” analysis. A WTW analysis is often divided into well-to-pump (WTP) and
pump-to-wheels (PTW) stages. The WTP stage starts with the fuel feedstock recovery, followed
by fuel production, and ends with the fuel available at the pump, while the PTW stage represents
the vehicle’s operation activities.
When analyzing the energy and emission implications of alternative fuels and advanced
vehicle technologies, a WTW analysis can provide important insight. In many cases, a
comparison is done for a particular vehicle technology by using different fuels and the same
powertrain system (with minor modification), or by using the same fuel/vehicle system with
different feedstock sources of fuel. However, to assess the impact of PHEVs, both the engine
fuel and the grid electricity powering the electric drive system must be examined.
The engine/fuel combinations examined in this analysis are: a spark ignition (SI) engine
using reformulated gasoline (RFG), an SI engine using a blend of 85% ethanol and 15%
reformulated gasoline (E85), a compression ignition engine using low-sulfur diesel (LSD), and a
fuel cell power system using gaseous hydrogen (H2). The feedstock sources considered are corn
and switchgrass for E85 and distributed natural gas (NG) steam methane reformation (SMR),
distributed electrolysis and switchgrass (gasification) for H2. Table 1 summarizes the vehicle
technologies and fuels considered in this analysis as well as the feedstock sources for these fuels.
TABLE 1 Vehicle Technologies, Fuels, and Feedstock Sources
Technology Fuel Feedstock

Conventional crude (82%) and
Reformulated gasoline
oil sand (18%)
Spark ignition
Corn
Ethanol
Herbaceous biomass (switchgrass)
Conventional crude (82%) and
Compression ignition Low-sulfur diesel
oil sand (18%)
Natural gas (SMR)
Fuel cell Hydrogen Electricity (electrolysis)
Herbaceous biomass (switchgrass)
A conventional gasoline ICEV and regular HEVs employing ICE and fuel cells are
compared with PHEVs using the same fuels to examine their relative benefits with respect to
energy use and GHG emissions. However, Santini and Vyas argued that regular HEVs and
PHEVs should be compared with ICEVs, but not to each other, since they will compete against
the ICEV in different niche markets (Santini and Vyas 2008). Regular HEVs are expected to be
more advantageous than PHEVs when operating at low average speeds and for shorter daily
driving distances (e.g., congested urban areas) in areas with a lower percentage of single-family
homes with garages. In contrast, PHEVs are expected to have an advantage over regular HEVs at
5
higher speeds in areas with less congestion (e.g., suburban areas) and a higher percentage of
single-family homes with garages available to recharge these vehicles.
Simulations for year 2020 with model year (MY) 2015 vehicles are chosen for this
analysis in order to address the implications of PHEVs within a reasonable timeframe after their
likely introduction in the next few years. The flexibility of GREET allows the user to modify key
assumptions when performing a WTW analysis. However, the challenge comes in finding
reliable data for inclusion in the model, especially for PHEVs that have not been commercially
produced. Therefore, external models and data are used to characterize these important
determinants of the WTW performance, such as the marginal electricity generation mix for
charging PHEVs, the fuel consumption and electricity use on a per-mile basis, and the VMT on
grid electricity. A recent study by Oak Ridge National Laboratory (ORNL) of region-specific
marginal generation mixes for PHEVs is adopted by this analysis to calculate the WTP energy
use and GHG emissions associated with the electric load from PHEVs. The Powertrain System
Analysis Toolkit (PSAT) is used to simulate the vehicle’s fuel economy and electricity use,
which are key inputs for the calculation of PTW energy use and GHG emissions. The following
sections provide an overview of the methodology used to obtain these determinants for inclusion
into the WTW analysis using GREET.
6
2 ELECTRICITY GENERATION MIX
A key factor in determining the environmental performance of PHEVs is the source of

the electricity used to charge the battery. One goal of this analysis is to gather projections of
regional generation mixes for a target year so that we can realistically examine how PHEVs will
perform in different markets. The type of power plants varies by region, so it is important to
examine these vehicles on a regional basis in order to better understand their effects.
A number of recent studies provided projections of the charging demand of PHEVs and
matched the projected demand to the estimates of available generation capacities. These studies
varied according to the regional scope and intent. Several nationwide studies produced results for
all North American Electric Reliability Corporation (NERC) regions (Figure 1), while other
studies were limited to specific regions. The generation mix at the time of charging became
increasingly uncertain as the time for large-scale PHEV deployment increased, but the large
current inventory of power plants, the availability of limited primary energy options for new
plants, and the trends in costs and regulations provided guidance for projecting future plant
inventories and their dispatch. By estimating the change in generating plant utilization due to
PHEV load, these studies estimated the effect of PHEV deployment on reserve margins, fuel use,
emissions, and costs.
2.1 FACTORS AFFECTING GENERATION MIX FOR PHEV CHARGING
The generation mix at the time of charging is a strong function of the time of day, time of
year, geographic region, vehicle and charger design, load growth patterns, and the associated
generation expansion in the years prior to the charging event of interest.
Region
1. ECAR
2. ERCOT
3. MAAC
4. MAIN
5. MAPP
6. NPCC‐NY
7. NPCC‐NE
8. FRCC
9. SERC
10. SPP
11. WECC‐NW
12. WECC‐RMP/ANM
13. WECC‐CA
FIGURE 1 NERC Regions from the Annual Energy Outlook 2007 (Source: EIA 2007) (see
Table 2 for definitions of abbreviations)
7
2.1.1 Time of Day
Figure 2, developed by Shelby and Mui, is an illustration of the diurnal peaks of demand
for a hypothetical summer day (Shelby and Mui 2007). Sharp summer peaks are caused by air-
conditioning demand, although such peaks typically occur in the late afternoon and early
evening. However, demand is at a minimum overnight when businesses are closed, lights are off,
and air-conditioning load is at its lowest (Hadley 2006).
As electricity demand increases, additional generating units are dispatched to meet the
load. When a PHEV charger is activated, it causes additional load on the marginal generator
(i.e., the last unit brought online). When that unit reaches full capacity, another unit is brought
online as the marginal unit, and so forth. Therefore, when a large number of PHEVs are added to
a system, several additional generation units may be required to meet the charging load.
Consequently, the energy use and emissions of those units are allocated to the PHEV charging
load. In an extensive interconnected region, transmission constraints can develop so that several
geographically separated generating units must operate at part load to meet an increasing
demand. Figure 3 displays an example of the marginal fuels during each hour of one day on the
entire PJM Interconnection.2 The PJM Interconnection includes parts of Region 1 (East Central
FIGURE 2 Typical Summer Load Profile and Dispatch Scheme for Many U.S. Utilities
(Source: Shelby and Mui 2007)
2
PJM Interconnection Marginal Fuel Type Data website, available at: ftp://ftp.pjm.com/pub/market/energy/
marginal-fuel-type/200802_Marginal_Fuel_Postings.csv. Last accessed October 2008.
8
Coal Natural Gas Light Oil Misc

100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour
FIGURE 3 Example of Hourly Marginal Fuels Data by Time of Day (Source: PJM
Interconnection Marginal Fuel Type Data website, available at:
ftp://ftp.pjm.com/pub/market/energy/marginal-fuel-
type/200802_Marginal_Fuel_Postings.csv
Area Reliability Coordination Agreement), Region 3 (Mid-Atlantic Area Council), and Region 9
(Southeastern Electric Reliability Corporation). The height of the bars represents the percentage
of contribution from each fuel.
2.1.2 Time of Year
Seasonal load variations also affect the mix of units brought on-line to meet the PHEV
charging demand. A typical trend is illustrated in Figure 4, which is based on operating data from
the Southwest Power Pool (SPP) for 2006 and 2007, and it shows two data traces: the minimum
and the maximum daily load (Roach et al. 2008). The annual pattern of relatively high summer
loads is typical for most of the United States, reflecting power demand for air-conditioning. In
some local areas, electric heating causes a winter peak or results in a more level annual load
pattern.
2.1.3 Climate
The SPP is a summer peaking system because of air-conditioning loads, which add to the
daytime peak. Electric heating loads tend to increase off-peak demands and may compete with
the off-peak charging of PHEVs during the winter season. These are some of the ways in which
regional climate affects the development of a generation mix and the nature of the generation
9
FIGURE 4 SPP Daily Maximum and Minimum Electricity Demand Source (Source:
Roach et al. 2008)
mix at the time of PHEV charging. A more subtle time-of-year effect, which is incorporated in
most generation expansion and load dispatching models, is the variation of power plant capacity
with ambient temperature. This also affects the availability of capacity for dispatching to meet
PHEV charging loads.
2.1.4 Generation Mix
Climate, fuel availability, population, industrial activity, local regulation, water

availability, pollution levels, and other regional characteristics have influenced the development
of each region’s specific power system, including the generation mix. As a result, the generation
mix varies substantially from region to region, as can be seen in Table 2, which shows the
projection for the 2020 average mix from the Energy Information Administration’s (EIA’s)
Annual Energy Outlook (AEO) 2008 for different regions in the United States.
10
TABLE 2 Mix of Sources for Average Electric Generation among Regions in the United States
(% in 2020) (Source: EIA 2008)
Natural
Region Coal Oil Gas Nuclear Other
1. East Central Area Reliability Coordination 83.9 0.5 5.7 9.0 0.9
Agreement (ECAR)
2. Electric Reliability Council of Texas (ERCOT) 45.2 0.5 37.3 11.9 5.1
3. Mid-Atlantic Area Council (MAAC) 44.1 1.2 5.5 34.2 15.0
4. Mid-America Interconnected Network (MAIN) 52.9 0.3 4.2 31.8 10.8
5. Mid-Continent Area Power Pool (MAPP) 73.1 0.4 1.8 12.3 12.4
6. Northeast Power Coordinating Council / NY 12.3 5.6 33.9 29.3 18.9
(NPCC-NY)
7. Northeast Power Coordinating Council / NE 17.5 1.8 43.0 21.9 15.8
(NPCC-NE)
8. Florida Reliability Coordinating Council (FRCC) 53.9 5.5 26.3 11.9 2.4
9. Southeastern Electric Reliability Corporation 49.9 0.6 11.9 33.0 4.6
(SERC)
10. Southwest Power Pool (SPP) 74.1 0.6 15.9 4.3 5.1
11. Western Electricity Coordinating Council / 28.8 0.1 6.4 3.2 61.5
Northwest Power Pool Area (WECC-NW)
12. Western Electricity Coordinating Council / Rocky 60.8 0.4 21.2 8.0 9.6
Mountain and AZ-NM-Southern NV Power Area
(WECC-RMP/ANM)
13. Western Electricity Coordinating Council / 13.0 0.0 40.4 17.3 29.3
California (WECC-CA)
2.1.5 Vehicle and Charger Design Factors
The vehicle design characteristic with the greatest influence on PHEV charging load is
the battery capacity, which is related to the AER and weight of the vehicle. It is most commonly
assumed that the charger will operate at normal household power levels, typically 110 V and no
more than 20 amps. A sport utility vehicle (SUV) type of PHEV may require larger batteries than
a compact or sedan type of PHEV. In order to charge these batteries in a reasonable length of
time, more charging current is required. This could be accomplished with a charger operating on
220 V at 30 amps. Single-phase, 220-V service is available to all residential customers, but
typically will require professional installation of additional circuit breakers, lines, and a
dedicated outlet. The benefit of reduced charging time comes at an additional cost of the higher
demand.
11
2.1.6 Load Growth and Generation Expansion
The inventory of units available for PHEV charging is slowly changing as old units retire
or are refitted with new environmental controls and as new units are constructed in anticipation
of increasing demand. Also, existing units may change place in the dispatch order as they age or
as new plants come on-line. Information on the dynamic nature of the generator inventory is
provided by recent inventory statistics from the EIA. In 2006, there were 986,000 megawatts
(MW) of generating capacity in the United States, including both utility and non-utility capacity.
Also, in 2006, there were 275 generators added, for a total of 13,152 MW of new capacity. At
the same time, 186 units retired, for a loss of about 3,500 MW, and net capacity revisions on
existing units represented a loss of about 700 MW of capacity (EIA 2007). Although commercial
introduction of PHEVs may occur as soon as 2010, it is likely to be one or two decades before a
substantial PHEV charging demand exists. Ideally, the generation mix applied at the time of
charging will reflect accumulated changes in the plant inventory.
Generation expansion planning, which optimizes changes to the generator inventory, is a

complex process that takes into account load growth projections, known and potential changes in
regulations, and the technical performance characteristics of current and future generator options.
The final inventory, one or two decades, or more in the future, would likely be substantially
different under carbon emission constraints than it would be in a business-as-usual case. While
the use of the current generation inventory might be useful as an indicator of the potential
capacity for PHEV charging, an understanding of the environmental trade-offs requires projected
generation expansion consistent with broad planning policies.
Generation expansion may also be influenced by the PHEV charging demand itself, and
this charging demand is likely to increase along with a general increase in transportation energy
demand. Thus, generation expansion projections become linked to projections of transportation
demand. In the AEO 2008, the EIA reference case is based on the historical (1980 to 2006)
growth rate for transportation energy use (EIA 2008). The revised growth rate leads to an
increase in transportation energy use from 28.2 quadrillion Btu (quads) in 2006 to 33.0 quads by
2030. The reduction in the rate of growth is due to higher fuel economy standards, higher fuel
prices, and slower economic growth, all of which lead to efficiency improvements and slower
growth in VMT.
Generation expansion planning must take into account both the extent of likely demand
growth and the daily and seasonal dynamic structure of the projected demand. Relatively
constant loads are best served by large base-load units with low fuel and variable costs. Daily
peak loads may best be served by units with low fixed (investment) costs, such as gas turbines.
However, lower fuel-cost options, including hydro, will be applied to peak loads if capacity is
available. The generation mix applied at a specific time is predicted by dispatch models rather
than by generation expansion models. The dispatch models match available capacity to the
dynamic load by using cost, emissions, or other criteria to optimize the system. Dispatch models
also take reliability and scheduled plant outages into consideration.
12
2.2 PHEV GRID IMPACT STUDIES
2.2.1 Background
Table 3 is an overview of U.S. energy consumption by fuel for 2007. It should be noted
that 91% of coal use is for electric power generation, and 51% of electric power is produced by
coal combustion. Petroleum has a virtual monopoly on transportation fuel, as the source of 96%
of the energy consumed by that sector. Automotive fuels account for 70% of the total petroleum
consumption. In total, the transportation energy demand is essentially equivalent to the primary
energy consumed for the combined residential and commercial electricity demand. Clearly, a
substantial shift from petroleum to electricity implies a substantial increase in the output of the
electricity sector, so the prospect of electrified transportation presents a great challenge to the
electric power infrastructure. Recognition of this challenge is one motivation for PHEV grid
impact studies.
TABLE 3 Distribution of U.S. Energy Consumption (quads)

(Source: EIA 2007)
Residential
Energy and
Source Transportation Industrial Commercial Electricity
Petroleum 27.8 9.5 2.0 0.8

Natural gas 0.6 8.0 8.0 7.0
Coal 0.0 1.9 0.0 20.7
Renewable 0.6 2.0 0.6 3.7
Nuclear 0.0 0.0 0.0 8.4
A second motivation is the need to understand the environmental trade-offs implicit in a

switch from petroleum to electricity. This very broad issue includes, for example, land-use issues
from minerals extraction, hazardous and radioactive waste disposal, and visual impairments from
stacks and facilities. The PHEV studies we have examined limited their environmental
considerations to air emissions — most prominently, emissions of GHGs.
Several studies have focused on the challenge of supplying electricity to a growing

PHEV load by estimating the capacity of the existing generator inventory to supply additional
power, primarily off-peak. Other studies allow for growth of the power generator inventory,
recognizing that it may take several decades before PHEVs constitute a substantial new load and
that, in the intervening years, other power demands are likely to increase as well, older units will
retire, and the generation inventory may be quite different from that of today. A third-generation
expansion scenario also recognizes the possibility that the growing PHEV demand will add to the
demand growth rate, so that PHEVs will themselves eventually influence the generating options
available for charging.
13
2.2.2 Methods
The PHEV impacts on the grid, including fuel use and emissions, occur in the context of
the equilibrium between power demand and power generation. Equilibrium exists without the
PHEV load, and it is changed by the addition of that load. The difference between these two
states is the grid impact that we seek. So, in general, the grid impact studies identify a base case
and at least one PHEV case. The PHEV case is defined first by the extent of PHEV power
demand, which is the product of the vehicle market share held by PHEVs and the charging
requirements and timing for the PHEV fleet. None of the reviewed studies offer anything beyond
speculation or assumption as the basis of assumed PHEV market share. This is understandable,
given the uncertainties of battery and vehicle development, vehicle costs, fuel costs, and other
determinants of market share. Consequently, the most that can be expected of such studies is an
estimate of outcomes consistent with assumptions about the vehicle market (i.e., scenario
analysis of the PHEV market). Basically, what will the power generation inventory and
operations look like if the PHEV deployment assumptions are realized? However, some studies
have reversed this question and ask the following: What level of PHEV deployment could be
supported by an assumed power generator inventory?
The charging load depends on the PHEV market penetration and the charging
characteristics of various PHEV products. These charging characteristics include the voltage
(110 V or 220 V), the amperage, and the length of time required for charging. Like the market
share, these features are assumed rather than forecasted, but they are very narrowly constrained
by limitations of residential power systems and by practical battery capacities.
On the utility side, the charging load is met by available units on the margin at the time of
charging. The available units are selected from the full generator inventory by dispatch
algorithms of various complexities or by simple heuristics. The simple approaches examine the
dispatch order without the PHEVs and apply any remaining capacity from the marginal units to
PHEV charging. The net result of interest from our perspective is the distribution of charging
load among available generator types. This distribution or share is required by GREET to
estimate the associated upstream and downstream emissions. A review of several grid impact
studies can be found in the Appendix. We employed the most suitable study for our analysis,
which is explained in detail in the next section.
2.3 ADOPTION OF MARGINAL MIX IN GREET
The 2008 ORNL report by Hadley and Tsvetkova3 was found to be the most inclusive,
publicly available, source for providing region-specific default marginal generation mixes for
PHEVs. The report reflected AEO 2007 projections for generation capacity expansion and load
growth through 2020, and it also employed a region-specific dispatch model.The following is a
discussion of some of the major assumptions of that study, which addressed the following
3
Argonne National Laboratory Powertrain Systems Analysis Toolkit website, available at:
http://www.transportation.anl.gov/modeling_simulation/PSAT/index.html. Last accessed October 2008.
14
questions: How is the PHEV load determined, when is the charging taking place, and where is
the charging taking place?
What is the PHEV load?
Hadley and Tsvetkova used a PHEV penetration consistent with an EPRI base-case
assumption that the market penetration of PHEVs could be greater than 25% for the light-duty
vehicle market by 2020. New PHEV sales were assumed to start at 0% of the total vehicle sales
in 2010 and grow to 25% by 2020, then hold steady after that, with the vehicle retiring after
10 years. This assumption appears to be aggressive, but it fits with the goal of this analysis to
examine the effect of significant demand from PHEVs on the electric grid. The analysis assumed
four vehicle classes of PHEVs to be sold, all with a 20-mi AER, ranging from a compact sedan
(5.1-kWh battery) to a full-size SUV (9.3-kWh battery).
When is the charging taking place?
Two charging scenarios were examined: an “evening” case, where vehicles started
charging at 5 p.m., and a “night” case, where vehicles started charging at 10 p.m. Three charging
rates were evaluated (1.4 kW, 2 kW, and 6 kW), which, along with battery size, determined the
number of hours required for charging. For this analysis, the night case was chosen, even though
the true off-peak is probably close to midnight, because of its potential for lower electricity cost.
The 2-kW charging rate was chosen, which would likely avoid any additional cost required for
rewiring the household’s electrical system, as would likely be the case for the 6-kW charging
rate.
Where is the charging taking place?
The study by Hadley and Tsvetkova covered the 13 NERC regions identified in the AEO
2007 generation expansion plan. The regional power plant inventory for 2020 was taken from the
AEO 2007. That inventory reflected the necessary expansion to meet growth, anticipated unit
retirements, and fuel and technology choices based on capital costs, projected fuel costs, and
regulatory restrictions. Hadley and Tsvetkova determined the marginal electricity supply for
PHEVs from the AEO 2007 baseline projections. However, since the AEO 2007 does not
anticipate PHEV market penetration, PHEV charging demand is not incorporated in the
generation expansion planning. However, PHEV loads at the assumed vehicle penetration level
are not expected to have a significant effect on capacity expansion by 2020. As evidence of that,
a study by Kintner-Meyer et al. (Rousseau et al. 2004), which took a very broad look at the
ability of the existing U.S. capacity mix to serve PHEV load, estimated that up to 73% of the
current light-duty vehicle usage could be accommodated by the existing power infrastructure.
Thus, ignoring the possible effects of PHEV loads on generation expansion is a compromise that
is not likely to be a significant source of error under the current assumptions for PHEV
penetration and for the analysis year of 2020. For higher levels of PHEV penetration and a more
distant time horizon, the PHEV load should be included in the generation expansion plan. The
15
loading of generators to meet the demand pattern was developed with the Oak Ridge
Competitive Electricity Dispatch Model (ORCED). The ORCED determined which units will be
brought online or ramped up to meet the PHEV charging demand.
In this analysis, we focus on three regions — Region 4 (Illinois), Region 6 (New York),
and Region 13 (California) — that encompass large metropolitan areas and provide a significant
variation of marginal generation mixes. In addition, we examine a U.S. average generation case
as a baseline and a renewable case that represents the upper limit on benefits from PHEVs. These
five generation mixes are provided in Table 4. It should be noted that the selected NERC regions
for this analysis exhibit a significant variation of generation mix, which could also serve as
scenarios to predict the impact of employing PHEVs in regions with similar generation. The goal
of this analysis is to provide the results of these specific mixes as a guide to any region that has
similar generation. For example, a study that evaluates PHEV charging from a marginal mix that
relies mostly on the natural gas combined cycle (NGCC) technology may consider the WTW
results of this analysis for California. Similarly, a marginal mix that relies heavily on
conventional coal or residual oil for power generation may consider the WTW results of this
analysis for Illinois and New York, respectively.
This study is not meant to provide an interregional comparison or to impart a criticism of

the relative environmental performance of various regions. Thus, the regions and states
mentioned in this analysis should be viewed as short-hand labels for the underlying generation
mixes associated with them, since the results of this analysis directly reflect the impact of these
mixes.
TABLE 4 Generation Mixes for Recharging PHEVs for Use in GREET
Natural
Mix Coal Oil Gas Nuclear Other
U.S. Average 52.5 1.3 13.5 20.1 12.6

Illinois – Region 4 (MAIN) Marginal 75.2 0.0 24.7 0.0 0.1
New York – Region 6 (NPCC-NY) Marginal 3.4 67.2 29.4 0.0 0.0
California – Region 13 (WECC-CA) Marginal 0.0 0.0 99.0 0.0 1.0
Renewable 0.0 0.0 0.0 0.0 100.0
16
3 PSAT SIMULATION OF VEHICLE FUEL ECONOMY
The PSAT is a forward-looking tool designed to serve the requirements of automotive

engineering throughout the development process, from modeling to control (Rousseau et al.
2004).4 It uses the driver outputs to send commands to the different powertrain components in
order to follow a specified drive cycle, and it has been validated within 5% for several vehicle
powertrain configurations on a number of driving cycles (Pagerit 2006).
When analyzing the performance of PHEVs, the amount of electricity used by the vehicle
compared with the amount of fuel used by the engine is a key factor. The higher the amount of
energy storage (or capacity) the battery has, the less the engine power will need to be used.
Initially, the concept of a PHEV’s operation was to charge the battery to a high state-of-charge
(SOC) (e.g. 90%), then the vehicle would operate in a charge-depleting (CD) mode by using only
the stored electricity until it reached a low SOC (e.g. 30%). Once the battery reached the low
SOC threshold, it would operate in charge-sustaining (CS) mode, which is similar to the
operation of regular HEVs (Shidore 2007).This operation strategy allows the vehicle to operate
as a zero-emission vehicle in CD operation. However, the high cost of batteries required for
extended AER has led vehicle designers to rethink this control strategy and explore ways to
extend the VMT driven on the battery by using it more efficiently. A “blended” CD mode, which
intermittently turns on the engine during CD operation, increases the CD VMT range by utilizing
both electricity and engine fuel. For example, the blended mode operation increases the VMT
driven on a given amount of battery capacity by turning on the engine during high power
demands in the CD mode; otherwise a significant amount of the battery’s energy would be
drained if not supplemented by the engine. Thus, the blended mode operation could reduce the
initial size and cost of the PHEV battery, while providing a bridge between the current regular
HEVs and the future all-electric PHEVs as battery performance and cost are improved.
The PHEV electrical components (i.e., battery and electric machine, such as an electric
motor) were sized to be able to drive the Urban Dynamometer Driving Schedule (UDDS) cycle
electrically. The constraint to drive all-electrically imposes specific size limitations on the
battery and the electric machine, which also imply certain vehicle cost constraints, as mentioned
above. To minimize the cost of the electric powertrain in these hybrids, PSAT employed a
blended CD control strategy. In addition to lowering the power requirements for the battery and
electric machines, there has been interest in employing CD strategies to reduce fuel consumption
when the AER is exceeded. The batteries for each of the vehicles simulated with PSAT have
their energy capacity and power sized to reach their vehicle’s desired AER. Although the
batteries were sized to power the vehicle through the target AER, the vehicle can extend the CD
driving range by utilizing the engine during periods of the cycle when the road’s load power
demand is high. The CD extended range was constrained to within 20% (with a 10% tolerance)
of the rated AER by adjusting a vehicle’s control strategy parameter. This parameter was a
power threshold that determined when the engine should be turned on. When the power demand
exceeded this threshold, the engine was turned on. A study by Delorme et al. provides a detailed
4
Argonne National Laboratory Powertrain Systems Analysis Toolkit website, available at:
http://www.transportation.anl.gov/modeling_simulation/PSAT/index.html. Last accessed October 2008.
17
explanation of the assumptions and methodology of PSAT for evaluating the fuel economy of
advanced vehicle configurations (including ICEVs, HEVs, PHEVs, and electric vehicles for
model years 2010 to 2045) (Delorme 2008). The vehicle assumptions for the PSAT simulations,
which are incorporated in this study, are shown in Table 5.
Table 6 shows the electricity consumption and fuel economy results produced by PSAT
simulations of the UDDS and Highway Federal Emissions Test (HWFET) cycles for CD and CS
operations of different PHEVs assuming a MY 2015 midsize passenger car platform. Care
should be taken when interpreting the fuel economy of the engine in CD operation, as it
discounts the grid electric energy use during the same CD VMT distance. It should be noted that
the per-mile energy use from the engine and electric motor are additive in CD operation, since
the CD VMT is powered by the blended operation of both systems. Thus, the fuel economy data
for the on-board power unit (i.e., engine or fuel cell) in CD operation should always be
interpreted in conjunction with the CD electric consumption data shown in Table 6.
The CD VMT on UDDS and HWFET are shown in Figures 5 and 6, respectively. The
average VMT for UDDS and HWFET on CD operation are calculated and shown in Figure 7.
The fuel economy data for the engine in CD operation should be correlated with the actual CD
VMT range shown in Figures 5 and 6, since the engine could be intermittently employed by the
vehicle’s control strategy to charge the battery in CD operation. The charging of the battery
extends the VMT distance in CD mode beyond the rated AER and results in higher engine fuel
consumption (i.e., lower fuel economy) in CD operation.
TABLE 5 Vehicle Assumptions for PSAT Simulations

Vehicle Engine Fuel Cell Motor 1 Motor 2 Battery Frontal Drag Wheel
mass Power Power Power Power Power Area Coefficient Radius
(kg) (W) (W) (W) (W) (W) (m2) (m)
ICEV 1,515 102,109 n/a n/a n/a n/a 2.23 0.26 0.317
AER 0 1,563 82,530 n/a 60,134 49,474 26,748 2.23 0.26 0.317
AER 10 1,592 70,373 n/a 64,461 42,186 46,610 2.23 0.26 0.317
Gasoline ICE
AER 20 1,617 71,263 n/a 65,477 42,720 47,335 2.23 0.26 0.317
AER 30 1,646 72,257 n/a 66,594 43,316 48,093 2.23 0.26 0.317
AER 40 1,674 73,285 n/a 67,739 43,932 48,968 2.23 0.26 0.317
AER 0 1,615 71,247 n/a 63,656 59,626 27,886 2.23 0.26 0.317
AER 10 1,648 60,878 n/a 70,415 50,948 48,465 2.23 0.26 0.317
Diesel ICE AER 20 1,676 61,671 n/a 71,526 51,612 49,279 2.23 0.26 0.317
AER 30 1,707 62,521 n/a 72,547 52,323 50,076 2.23 0.26 0.317
AER 40 1,734 63,314 n/a 73,954 52,987 50,978 2.23 0.26 0.317
AER 0 1,546 88,115 n/a 61,139 58,712 26,748 2.23 0.26 0.317
AER 10 1,569 75,099 n/a 62,991 50,040 46,103 2.23 0.26 0.317
E85 ICE AER 20 1,597 76,101 n/a 64,064 50,707 46,884 2.23 0.26 0.317
AER 30 1,627 77,944 n/a 65,338 51,935 47,718 2.23 0.26 0.317
AER 40 1,653 79,107 n/a 66,612 52,710 48,503 2.23 0.26 0.317
AER 0 1,530 n/a 72,857 90,726 n/a 29,024 2.23 0.26 0.317
AER 10 1,552 n/a 59,568 94,424 n/a 49,509 2.23 0.26 0.317
H2 FC AER 20 1,583 n/a 60,396 95,992 n/a 50,554 2.23 0.26 0.317
AER 30 1,615 n/a 61,017 97,654 n/a 51,804 2.23 0.26 0.317
AER 40 1,650 n/a 62,735 99,333 n/a 53,423 2.23 0.26 0.317
18
TABLE 6 PSAT Electricity Use and Fuel Economy Results (Wh/mi for CD electric operation, and
miles per gasoline-equivalent gallons (mpgge) for CD and CS engine operations)a
ICEV AER 0 AER 10 AER 20 AER 30 AER 40
Engine Regular CD CD CS CD CD CS CD CD CS CD CD CS
Type Cycle Hybrid Electric Engine Engine Electric Engine Engine Electric Engine Engine Electric Engine Engine
Gasoline UDDS 27.6 45.6 148.1 132.4 47.1 141.3 122.3 46.9 174.1 184.3 46.6 165.1 153.4 46.2
ICE
HWFET 34.0 39.7 107.8 78.3 41.1 136.9 103.9 41.0 158.2 134.5 40.6 168.0 152.6 40.2
E85 UDDS 42.9 146.1 125.5 44.4 141.2 118.4 44.2 172.6 179.7 43.8 164.3 148.6 43.4
ICE
HWFET 37.5 106.3 73.8 38.9 136.9 99.3 38.8 156.8 126.2 38.3 167.0 144.4 37.9
Diesel UDDS 49.4 151.4 138.1 50.0 144.7 127.5 49.7 179.7 191.3 49.3 169.7 158.7 48.9
ICE
HWFET 43.0 110.2 84.1 43.8 140.3 112.2 43.6 163.3 145.7 43.2 172.6 164.5 42.9
H2 UDDS 59.4 157.7 132.6 59.5 154.2 123.4 58.8 156.2 120.7 58.1 181.8 142.7 57.3
FC
HWFET 62.3 229.4 1514.4 61.5 224.0 601.5 60.9 170.1 189.6 60.3 184.7 225.4 59.7
a
PSAT results incorporated EPA mpg-based adjustment formulae to reflect the on-road fuel economy but made no adjustment to
electricity use.
Since the control parameters in PSAT have been designed to achieve a CD range within
20% of the rated AER, some VMT distances are greater than others, as shown in Figures 5–7.
For example, the gasoline PHEV produced a longer CD range in the HWFET cycle than the CD
range of the corresponding fuel cell PHEV at AER 10. This is because the gasoline engine is
employed significantly during the HWFET cycle, resulting in a relatively low electric energy
consumption of 107.8 Wh/mi for the AER 10 case, while the electricity consumption for the
corresponding H2 FC PHEV is higher, at 229.4 Wh/mi. This indicates that the fuel cell is not
significantly employed on that cycle, and hence the observed high fuel economy of
1,514.4 mpgge for the H2 FC in CD operation.
19
UDDS
50 SI Gasoline SI E85 CI Diesel FC H2
45
40
35
CD Distance (mi)
30
25
20
15
10
5
0
AER 10 AER 20 AER 30 AER 40
FIGURE 5 CD VMT on UDDS from PSAT Simulations
HWFET
45
40
35
CD Distance (mi)
30
25
20
15
10
5
0
FIGURE 6 CD VMT on HWFET from PSAT Simulations

20
Combined
45
40
35
CD Distance (mi)
30
25
20
15
10
5
0
FIGURE 7 Average VMT for UDDS and HWFET on CD Operation

21
4 VEHICLE MILES TRAVELED SPLIT BY CHARGE-DEPLETING VERSUS

CHARGE-SUSTAINING OPERATION
Graham et al. (2001) discussed two methods for evaluating the potential of PHEVs to
replace miles driven by gasoline with miles driven by electricity. The mileage weighted
probability (MWP) method by EPRI and the utility factor (UF) method by the SAE J1711
subcommittee were both developed by using the 1995 National Personal Transportation Survey
(NPTS) to calculate the “average” VMT displaced by an all-electrical PHEV that is fully charged
and discharged once per day. The MWP method resulted in a lower potential for electric mile
substitution than the UF method. Vyas et al. investigated these results but were unable to find
how the MWPs were developed (Vyas et al. 2007). When the 2001 NHTS data became available,
Vyas et al. updated the UF and examined the blended-mode strategy, which was not considered
in the original calculations. The UF partitioned the average national miles driven into VMT that
could be met by the PHEV’s CD mode and VMT that exceeded the rated CD range.
Table 7 shows the share of national VMT contributed by vehicles traveling various
ranges per day and the maximum percentage of VMT that could be substituted by all-electric
operation of a PHEV. If a PHEV has an AER rating equal to or larger than the daily VMT, it
could travel all those miles on electricity. However, if the vehicle is driven longer than the AER,
only the first miles driven up to the AER can be electrified. Figure 8 shows a curve fitted to these
results. Furthermore, if the PHEV does not operate all-electrically in CD mode and employs
some type of blended-mode strategy, the miles to deplete the battery will be extended beyond the
AER rating. When a PHEV that operates under a blended CD mode travels a distance shorter
than or equal to its rated electric range, the battery will not be depleted and fewer miles will be
displaced by electricity, as compared with a PHEV that uses 100% electricity in the CD mode.
TABLE 7 Share of National VMT Available for Substitution by a PHEV Using 100% Grid
Electricity in CD Mode until Depletion
Daily Travel One Charge/Day – % “Electric” VMT by PHEV Type

Range of VMT Share in
Vehicle NHTS 2001 10 EV mi 20 EV mi 30 EV mi 40 EV mi 60 EV mi
Up to 10 mi 3.3% 3.3% 3.3% 3.3% 3.3% 3.3%

10-20 mi 8.1% 5.3% 8.1% 8.1% 8.1% 8.1%
20-30 mi 10.0% 3.9% 7.9% 10.0% 10.0% 10.0%
30-40 mi 10.0% 2.8% 5.7% 8.5% 10.0% 10.0%
40-60 mi 16.8% 3.4% 6.7% 10.1% 13.5% 16.8%
Over 60 mi 51.8% 4.5% 8.9% 13.4% 17.9% 26.7%
PHEV sum 100.0% 23.2% 40.6% 53.4% 62.8% 74.9%
22
% VMT on Electricity (Utility Factor)

100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
0 10 20 30 40 50 60 70 80 90 100
All Electric Range (mi)
FIGURE 8 National VMT Available for Substitution by a PHEV Using 100% Electric
CD Mode
When estimating the potential of national savings in petroleum energy use and GHG
emissions, an estimate of the electrifiable VMT share based on Figure 8 is complicated further
by the following issues, according to Santini and Vyas (2008):
• Slow fleet turnover (~7-8%/year) will require time to accomplish large-scale

change.
• PHEVs will not be purchased by everyone.
• PHEVs will likely complement rather than displace HEVs, thus expanding the
long-term hybrid drivetrain market (i.e., PHEVs may not become a universal
powertrain).
• Various control strategies for utilizing the engine and the electric machine
could result in a myriad of extended VMT shares driven in CD mode.
• PHEVs will vary in their AER capability and will have different
configurations of the electric machine, battery, and engine.
• PHEVs purchased with a nominal range capability (AER rating) will not
exactly realize that rated value in practice.
• Batteries for PHEVs may be charged more than once every day.
23
Due to the above issues and the methodological differences in estimating the VMT
displaced by electricity, our analysis employed the UF method to evaluate the share of VMT
driven in CD mode, based on the AER of the vehicle and data in Figure 8. Furthermore, because
of the uncertainties in estimating that share and in order to simplify the analysis, the rated AER
(rather than the extended miles driven in CD operation, as shown in Figure 7) was used to
determine the UF. Then the UF was used to combine the WTW results of the CD and CS
operations, as explained in Section 5.
24
5 GREET WELL-TO-WHEELS ENERGY USE AND GREENHOUSE GAS EMISSIONS

CALCULATIONS
5.1 OVERVIEW
To perform WTW energy and GHG emissions calculations in GREET, the PSAT (on-
road adjusted) fuel economy results for different fuel/vehicle systems were processed for
inclusion in GREET. The first step in the processing of PSAT simulation results was to convert
the electricity use and the fuel economy values of the engine (ICE or fuel cell) to per-mile fuel
consumption in consistent units (e.g., Btu/mi), as shown in Table 8. The electricity consumption
at the wall outlet was calculated from the grid electricity use in CD operation by assuming a
charger efficiency of 85%. The average fuel consumption of the engine in the CD and CS
operational modes was calculated based on weighting factors of 55% and 45% for the fuel
consumption in UDDS and HWFET driving cycles, respectively.5 Thus, Table 8 includes three
types of fuel consumptions for each PHEV system: grid electricity consumption in CD operation,
engine fuel consumption in the blended CD operation, and engine fuel consumption in the CS
operation. The first two columns in Table 8 represent the fuel consumption of the corresponding
conventional gasoline ICEV and regular HEV (AER 0) systems, respectively. They are provided
to allow the comparison of fuel consumption between the existing and future powertrain systems.
The data in Table 8 are plotted in Figures 9 and 10 for different fuel/vehicle systems.
Figure 9 reveals two qualitative features of the PSAT fuel consumption results for PHEV
powertrains that use blended mode operation: the ICEs consume more (fuel) energy than the
electric motor at the lower AER range, while the opposite trend is observed for the fuel cell. It
should be noted that the conversion efficiency of the electric energy to mechanical energy
(powering the wheels) is several times higher than the conversion efficiency of fuel energy in the
engine, since the electric energy has already been upgraded in the upstream process of power
generation. The impact of this issue will become evident in the WTW results presented in
Section 5.2. Figure 9 also reveals the effect of the control strategy on the contribution of the
engine relative to that of the electric motor in blended CD operational mode. Such an effect is
evident in Figures 5 and 6 at AER 30, where the fuel consumption of the fuel cell exceeded the
electricity consumption of the electric motor, thus significantly extending the distance in CD
operation for the H2 FC PHEV 30. The observed buckling shown in Figure 9 for the H2 FC
PHEV 30 is mainly due to the control strategy parameters in PSAT, which are tuned to obtain a
CD range within 20% of the rated AER. The 20% (±10%) allowance in the CD range may allow
additional usage of the engine (or fuel cell) at the expense of the electric motor, which impacts
the trend of fuel and electricity consumption in CD operation.
5
In this analysis we used the PSAT 55% city/45% highway weighting for calculating composite fuel economy,
which is consistent with the EPA Fuel Economy Guide (EPA 2006). However, the GREET model employs a
43% city/57% highway weighting for calculating composite fuel economy values, which is being used by EPA to
calculate the light-duty automotive fuel economy trends for 2008 and newer model year vehicles (EPA 2008).
25
TABLE 8 Fuel Consumption Calculated from PSAT Simulated Fuel Economy Results (Btu/mi)
AER 0 AER 10 AER 20 AER 30 AER 40
Regular CD CD CS CD CD CS CD CD CS CD CD CS
Fuel ICEV Hybrid Electric Engine Engine Electric Engine Engine Electric Engine Engine Electric Engine Engine
Gasoline 3,790 2,680 520 1,135 2,590 560 1,010 2,600 670 725 2,620 670 750 2,645
E85 2,840 515 1,200 2,740 560 1,050 2,750 665 760 2,780 665 780 2,810
Diesel 2,470 535 1,080 2,435 575 955 2,450 690 680 2,470 685 710 2,490
Hydrogen 1,890 760 510 1,895 745 595 1,915 650 795 1,940 735 670 1,960
Figure 10 shows the differences in fuel consumption in CS and CD operational modes for
various PHEV powertrains. The markers shown on the vertical axis represent the fuel
consumption of the gasoline ICEV and the regular HEVs (AER 0) to compare the fuel
consumption of these powertrains with those of PHEV systems. Figure 10 indicates that the
vehicle’s energy consumption in the CD operation is much lower than that in the CS operation,
mainly due to the implication of the electric energy use in the CD operation, as discussed above.
Overall, the vehicle’s energy consumption trend exhibits a small change with increasing AER for
both CS and CD operations.
5.2 RESULTS OF WELL-TO-WHEELS SIMULATION
The WTW analysis of PHEVs in GREET is separated into three distinct parts: grid
electricity use in CD operation, fuel use in CD operation, and fuel use in CS operation. It should
be noted that the combined operation of the electric motor and engine contribute to the VMT in
CD blended mode; thus their per-mile energy use and emissions must be added to properly
characterize the PHEV CD operation. The data shown in Table 8 represent only the energy use in
the PTW (vehicle operation) stage. The PTW GHG emissions are calculated based on the carbon
content of the fuel and the engine's emissions characteristics. The electricity use by the vehicle
does not produce any GHG emissions, since all emissions have already occurred upstream of the
vehicle during the electric power generation and transmission stage (WTP). Thus, the WTP
energy use and emissions must be calculated to account for their occurrences during electricity
generation and transmission, as well as during fuel production and transportation to the vehicle's
point of use. For each of the WTP and PTW stages, GREET calculates total energy use, fossil
energy use (combining petroleum, natural gas, and coal), petroleum energy use, and carbon
dioxide (CO2)-equivalent GHG emissions. The GHG emissions calculation combines CO2,
methane (CH4), and nitrous oxide (N2O) with their global warming potentials, which are 1, 25,
and 298, respectively, as recommended by the latest Intergovernmental Panel on Climate Change
for a 100-year time horizon (IPCC 2008).
26
Grid and On-board Fuel Consumption in CD mode

1300
PHEV SI Gasoline
1200
PHEV SI E85
PHEV CI Diesel
1100 PHEV FC H2
Fuel Consumption [Btu/mi]
1000
On-board
900
Grid
800
700
600
500
400
0 10 20 30 40
All Electric Range [mi]
FIGURE 9 Fuel Consumption in CD (blended mode) Operation
CD and CS Mode Fuel Consumption

4000
Baseline (GV)
PHEV SI Gasoline
3500 PHEV SI E85
PHEV CI Diesel
PHEV FC H2
3000
Fuel Consumption [Btu/mi]
2500
2000
CS
1500 CD
1000
500
0
0 10 20 30 40
FIGURE 10 Fuel Consumption in CD (blended mode) and CS Operations

27
The vehicle technologies and fuels considered in this analysis, as well as the feedstock
sources for these fuels, are provided in Table 1. The selected vehicle platform is the mid-size
vehicle, and the examined AERs for PHEV technologies are AER 10, 20, 30, and 40. The
marginal electricity generation mixes considered in this WTW analysis include those in NERC
regions 4, 6, and 13 (representing Illinois, New York, and California, respectively), as well as
electricity generation from the U.S. average mix and renewable sources. As shown in Table 4,
the California marginal mix is almost entirely powered by natural gas, which is a fuel of low
carbon intensity, while the marginal mixes in Illinois and New York are dominated by coal and
oil, respectively, which are fuels of higher carbon intensity. The WTW results of this analysis
should be correlated to the underlying generation mix, rather than to the specified region or state,
as discussed above.
Figures 11(a–d) show the WTW energy and GHG emissions results for various PHEV
technologies at AER 20, utilizing the California (NERC Region 13) marginal mix for charging
the vehicle overnight. As shown in Table 4, the marginal generation mix for that region is almost
entirely from natural gas (99%) — the majority of which (83%) is provided by the NGCC
technology. GREET calculates an average efficiency of 53% for the marginal electricity
generation from NG in California for the year 2020 and assumes 8% losses for electricity
transmission and distribution activities. It should be noted that the emission rates during the
vehicle's operation will deteriorate over time. Thus, the data of the lifetime mileage midpoint for
a typical MY vehicle should be applied for the simulation. Since, on average, the midpoint for
U.S. light-duty vehicles is about five years, the fuel economy values in GREET are based on a
MY five years earlier than the calendar year targeted for simulation. Therefore, fuel economy
values of MY 2015 vehicles are employed in the simulations of calendar-year 2020.
Two stacked bars for CD and CS and operations are shown in Figures 11(a–d) for each
vehicle technology. The stacked bar on the left represents the CD blended mode operation and
consists of four components, which are (from bottom to top) the vehicle’s (PTW) fuel and
electricity use, followed by the upstream (WTP) stages of electricity generation and fuel
production, respectively. The stacked bar at the right represents the CS operation of the vehicle
and consists of the engine’s fuel consumption, followed by the upstream stage of the fuel
production, from bottom to top, respectively.
Figure 11(a) shows the WTW total energy use for CD (blended mode) and CS operations
of different PHEV 20 technologies using the California marginal mix. The total energy includes
fossil energy (e.g., petroleum, natural gas, and coal) and non-fossil energy (e.g., nuclear and
renewables). Of interest is the second component from the bottom in the stacked CD bar of
Figure 11(a), which represents the amount of electricity purchased from the grid to charge the
batteries of PHEVs. Although electric energy use is expected to dominate the CD operation, it is
remarkable that the electric energy use appears small relative to the fuel energy use in that mode
of operation. However, it should be noted that the contribution of electric energy to power the
wheels through the electric motor is several times higher than that of the fuel energy through the
engine. Thus, most of the energy that reaches the wheels is provided by the electric motor in the
CD operation. Figure 11(a) also shows that the CD operation provides significant energy savings
compared with the CS operation for all vehicle technologies using the California marginal mix.
28
Figure 11(b) shows that fossil energy use exhibits a trend similar to that of total energy
use, except for E85 and hydrogen from herbaceous biomass (switchgrass), where the CS
operation consumes less fossil fuel compared with the CD operation. This is attributed to the
biomass renewable energy that dominates the total energy embedded in ethanol and hydrogen
fuels for CS operation, as opposed to the natural gas that dominates the electricity generation for
CD operation.
Figure 11(c) shows the petroleum energy use for the different PHEV 20 technologies.
The electricity use in the CD operation reduces petroleum use relative to CS operation for RFG,
LSD, and E85 PHEVs. The E85 PHEV exhibits lower dependence on petroleum energy than
RFG and LSD PHEVs due to the high percentage of bio-ethanol in the blend. All hydrogen
PHEV systems almost eliminate the dependence on petroleum energy sources.
As expected, the WTW GHG emissions shown in Figure 11(d) exhibit a similar trend to
that of fossil energy use for all PHEV fuel/vehicle systems. The negative GHG emissions shown
for the biomass-based fuels represent the CO2 sequestered from the atmosphere by the biomass,
which is deducted from the top of the GHG emissions bars to calculate the net WTW GHG
emissions for these fuels, as shown by the vertical arrows. It should be noted that the use of
biomass-based fuels in PHEVs produces higher GHG emissions in CD operation compared with
CS operation, even with the efficient and low carbon intensity marginal generation mix of
California. Thus, PHEVs that use fuels produced from biomass sources and operate in CD mode
may generate less GHG emissions relative to CS operational mode only if the source of
electricity is non-fossil (e.g., nuclear, biomass, or renewable energy sources).
9,000
Total Energy Use [Btu/mi]
WTP (Fuel Production) PHEV20 (Model Year 2015 in CA)

8,000
WTP (Electricity Generation)
7,000
PTW (Grid Electricity Use)
6,000
PTW (Engine Fuel Use)
5,000
4,000
3,000
2,000
1,000
0
CD
CD
CD
CS
CS
CS
CS
CS
CD
CD
S
CS
CD
,C
,C
R,
G,
G,
s,
s,
s,
D,
D,
rn,
s,
rn,
is,
MR
sis
as
as
as
as
LS
LS
RF
RF
SM
ly s
Co
Co
oly
dS
om
iom
iom
om
CI
CI
SI
SI
tro
5-
5-
ted
ctr
.Bi
ute
.Bi
H.B
H.B
E8
E8
lec
u
Ele
lH
lH
tib
tib
SI
dE
SI
5-
5-
t ra
tra
Dis
Dis
ted
E8
E8
ute
en
en
bu
-
-
SI
SI
H2
H2
tib
-C
-C
sti
Dis
H2
FC
FC
H2
Di
FC
FC
-
-
H2
H2
FC
FC
FIGURE 11(a) WTW Total Energy Use for CD (blended mode) and CS Operations of
PHEV 20 Using the California Marginal Mix
29
Fossil Energy Use [Btu/mi]

9,000
PTW (On-board) PHEV20 in CA (Model Year 2015)
8,000
PTW (Grid Electricity)
7,000
WTP (Grid Electricity)
6,000
WTP (On-board)
5,000
4,000
3,000
2,000
1,000
0
D
CD
CD
S
CS
CS
CS
CS
CD
S
D
CD
CS
,C
,C
,C
,C
,C
R,
D,
D,
s,
s,
s,
,
s,
,
FG
FG
R
rn
sis
rn
sis
as
as
as
as
LS
LS
SM
SM
Co
Co
R
ly
oly
om
iom
iom
om
CI
CI
tro
SI
SI
5-
te d
5-
ted
c tr
.Bi
.Bi
H.B
H.B
E8
E8
c
u
u
Ele
Ele
lH
lH
tib
tib
SI
SI
5-
5-
a
a
Dis
Dis
ted
te d
nt r
ntr
E8
E8
bu
u
Ce
-
Ce
-
SI
SI
H2
H2
tib
sti
-
-
Dis
FC
H2
FC
H2
Di
FC
FC
-
-
H2
H2
FC
FC
FIGURE 11(b) WTW Fossil Energy Use for CD (blended mode) and CS Operations of
Petroleum Energy Use [Btu/mi]
9,000
8,000
7,000
PTW (Grid Electricity Use)
6,000
PTW (Engine Fuel Use)
5,000
4,000
3,000
2,000
1,000
0
D
CD
S
CD
CD
CS
CS
CS
S
D
D
CS
S
D
,C
,C
,C
,C
,C
,C
,C
D,
D,
R,
s,
s,
R,
s,
FG
FG
rn
ss
sis
rn
sis
as
as
as
LS
LS
SM
SM
Co
Co
ma
R
oly
oly
om
iom
iom
CI
CI
SI
SI
5-
5-
ted
ted
o
ctr
c tr
.Bi
.Bi
H.B
H.B
E8
E8
ibu
bu
Ele
Ele
lH
lH
SI
SI
sti
ist
a
a
85
85
ted
te d
nt r
ntr
Di
-D
E
u
u
Ce
-
Ce
SI
SI
H2
H2
tib
tib
-
-
Dis
Dis
FC
H2
FC
H2
FC
-
FC
-
H2
H2
FC
FC
FIGURE 11(c) WTW Petroleum Energy Use for CD (blended mode) and CS Operations
of PHEV 20 Using the California Marginal Mix
30
400
PTW (Engine Emissions)
GHG Emissions [g/mi]
200
, CD
S
D
S
S
S
D
CS
CD
CD
CD
D
sis, C
R, C
rn, C
FG, C
ss, C
rn, C
D, C
G, C
D, C
as s,
MR,
ss,
s,
lysis
mas
d SM
CI L S
SI RF
C I LS
5 -Co
5 -Co
ioma
ioma
ctroly
SI R
.B iom
ctro
uted
H .Bio
tibute
5 -H.B
S I E8
5 -H.B
SI E8
d Ele
d E le
tr al H
istib
l
2- D is
entr a
SI E8
SI E8
ti bute
tibute
2- D
- C en
2- C
FC H
FC H
-200
2- D is
2- Dis
2
FC H
FC H
FC H
FC H
FIGURE 11(d) WTW GHG Emissions for CD (blended mode) and CS Operations of
To study the effect of the electricity marginal generation mix on the WTW analysis of
PHEV technologies, two sets of graphs, Figures 12(a–d) and Figures 13(a–d), have been
generated for the U.S. average mix and the Illinois marginal mix, respectively. The significance
of the marginal generation mixes mostly lies in the electricity generation (WTP stage). Close
examination reveals that the only difference between the graphs in Figures 11(a–d) for the
California mix and the corresponding graphs in Figures 12 and 13 for the U.S. and Illinois mixes
is the size of the electricity WTP component in CD operation (the third component from the
bottom of the CD bars).
The electricity WTP energy use and GHG emissions increase successively as the
marginal mix becomes less efficient and dominated by a larger share of coal, such as the cases of
the U.S. and Illinois mixes, respectively. The U.S. and Illinois generation mixes do not affect
petroleum energy use because they incorporate little or no petroleum sources in their portfolio.
The reduction in total energy consumption in CD relative to CS operation progressively
diminishes as the electricity generation mix changes from the California to the U.S. and Illinois
mixes, respectively. Furthermore, the WTW GHG emissions advantage of CD over CS operation
disappears by moving from the California to the U.S. generation mix, and it is even reversed by
moving to the Illinois marginal generation mix, thus surrendering the potential GHG emissions
benefit of PHEVs (except for the case of hydrogen when produced via electrolysis). In other
words, the improved energy efficiency and GHG emissions of PHEVs over regular HEVs could
be entirely negated by the energy penalty and GHG emissions associated with the electricity
generation in power plants. Such implications underscore the significance of the employed
electricity generation mix for charging PHEVs.
Fossil Energy Use [Btu/mi] Total Energy Use [Btu/mi]
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
SI SI
RF RF
G, G,
CD CD
SI SI
RF RF
G, G,
CS CS
CI CI
LS LS
D, D,
WTP (On-board)
PTW (On-board)
WTP (On-board)
PTW (On-board)
CD CD
CI CI
LS LS

D, D,
CS CS
PHEV 20 Using the U.S. Mix

SI SI
E8 E8
5- 5-
Co Co
rn rn
,C ,C
SI
E D SI
E D
85 85
- Co - Co
rn rn
,C ,C
S S
SI SI
E8 E8
5- 5-
H. B H.
iom Bi om
SI as SI as
E8 s, E8 s,
31
5- CD 5- CD
H. B H. B
iom iom
as as
s, s,
CS CS
FC FC
H2 H2
-D - Di
ist sti
ibu bu
FC te d FC te d
H2 SM H2 SM
-D R, - Dis R,
is t CD tib CD
PHEV20 - US Mix (Model Year 2015)
ibu
ted u te d
SM SM
FC R, FC R,
H2 CS H2 CS
-D - Dis
is t tib
ibu ute
FC te d FC
H2 H2 dE
-D
Ele
ctr - Di lec
ist oly sti tro
ibu sis bu lys
ted ,C te d is,
Ele D Ele CD
ctr ctr
oly oly
sis sis
,C , CS
FC S FC
H2 H2
-C - Ce
en ntr
FC
tra
lH FC a lH
H2 .B H2 .Bi
-C iom - Ce o ma
en as
s, ntr s s,
t ra
lH CD a lH CD
.Bi .Bi
om o ma
as s
s, s,
CS CS
GHG Emissions [g/mi] Petroleum Energy Use [Btu/mi]
-200
0
200
400
600
800
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
SI
S I RF RF
G, C G,
D CD
SI
SI R F RF
G, C G,
S CS
WTP (On-board)
PTW (On-board)
CI
CI LS LS
D,
WTP (On-board)
PTW (On-board)
D, CD CD

CI
LS
CI L S D,
D, C
S CS

SI
of PHEV 20 Using the U.S. mix

E8
SI E8 5-
Co
5 -Co rn
rn, C ,C
D SI
E8 D
5-
SI E8 Co
5 -Co rn ,C
rn, C
S S
SI
E 85
- H.B
SI E8 iom
5 - H.B SI as
ioma
ss, C E 85 s,
32
D - CD
H.B
SI E8 iom
5 -H.B
ioma as
ss, C s,
S CS
FC
H2
- Dis
tib
FC H
2 - Dis FC u ted
tibute H2 SM
d SM - Dis R,
R, C CD
D tib
FC H
2- Dis u te d
tibute SM
d SM FC R,
R, C H2 CS
S - Di
sti
bu
FC ted
FC H H2 Ele
2- Dis
tibute - Dis c tr
d Ele tib oly
c troly u sis
FC H sis, te d , CD
2- Dis CD Ele
tibute ctr
d Ele oly
ctroly sis
sis, C FC , CS
S H2
- Ce
nt r
FC H FC a lH
2- Ce H2 .Bi
o
ntral - Ce ma
H.B ioma ntr s s,
ss, C a lH CD
FC H D .Bi
2 - Ce o
ntral ma
H.B ioma s s,
ss, C CS
S
Fossil Energy Use [Btu/mi] Total Energy Use [Btu/mi]
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
SI SI
RF RF
G, G,
CD CD
SI SI
RF RF
G, G,
CS CS
CI CI
LS LS
D, D,
WTP (On-board)
PTW (On-board)
WTP (On-board)
PTW (On-board)
CD CD
CI CI
LS LS

D, D,
CS CS
SI SI
E 85 E8
- Co
5-
Co
rn rn
,C ,C
SI
E8 D SI
E8 D
5- 5-
Co Co
rn rn
,C ,C
S S
SI SI
E8 E8
PHEV 20 Using the Illinois Marginal Mix

5- 5-
H.B H.
Bio
SI
iom
as SI m as
E8 s, E8 s,
33
5- CD 5- CD
H.B H.
iom Bio
as m as
s, s,
CS CS
FC FC
H2 H2
- Dis - Dis
tib tib
FC u te d FC u ted
H2 SM H2 SM
- -
PHEV20 in IL (Model Year 2015)

Dis R, Dis R,
tib CD tib CD
ute u te d
dS SM
FC MR FC R,
H2 ,C H2 CS
- S - Dis
Dis
tib tib
FC u ted FC u te d
H2 Ele H2 Ele
- Di c - Dis c tr
sti tro oly
bu ly sis
tib
u sis
te d , CD
ted , CD
Ele Ele
c tr c tro
oly
sis ly sis
, CS FC , CS
FC
H2 H2
- Ce - Ce
nt r ntr
FC a lH FC a lH
H2 .Bi H2 .Bi
- Ce om - Ce om
as nt r as
ntr
a s, a s,
lH CD lH CD
.Bi .Bi
om o ma
as s
s, s,
CS CS
GHG Emissions [g/mi] Petroleum Energy Use [Btu/mi]
-200
0
200
400
600
800
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
SI
S I RF RF
G, C G,
D CD
SI
SI R F RF
G, C G,
S CS
WTP (On-board)
PTW (On-board)
CI
CI LS LS
D,
WTP (On-board)
PTW (On-board)
D, CD CD

CI
LS
CI L S D,
D, C CS
S
SI
E8
5-
SI E8 Co
5 -Co rn
rn, C ,C
D SI
E D
85
SI E8 - Co
5 -Co rn ,C
rn, C
S S
SI
E 85

-
H. B
SI E8 iom
of PHEV 20 Using the Illinois Marginal Mix
5 - H.B SI as
ioma
ss, C E 85 s,
34
D -H CD
SI E8 . Bi
5 -H.B om
ioma as
ss, C s,
S CS
FC
H2
- Di
sti
FC H bu
2 - Dis FC te d
H2 SM
tibute -
d SM Dis R,
R, C tib CD
FC H D u
2- Dis ted
tibute SM
d SM FC R,
R, C H2 CS
S - Di
sti
bu
FC te d
FC H H2 Ele
2- Dis
tibute - Dis c tr
d Ele tib oly
c troly u sis
FC H sis, ted , CD
2- Dis CD Ele
tibute c tro
d Ele ly sis
ctroly
sis, C FC , CS
S H2
- Ce
ntr
FC a lH
FC H
2- Ce H2 .Bi
ntral - Ce om
H.B nt r as
ioma a s,
ss, C lH CD
FC H
2 - Ce D .Bi
om
ntral
H.B as
ioma s,
ss, C CS
S
35
GREET calculates the weighted average energy use and GHG emissions of CD and CS
operational modes by using the VMT share in each mode. The UF at the rated AER of the PHEV
(Figure 8) combines the PHEV’s average fuel consumption (AFC) in CD and CS operational
modes according to the following formula:
AFCcombined = (AFCGrid + AFCEngine)CD*UF + AFCCS*(1-UF). (1)
The UF for PHEV 20 is 40%, as shown in Table 7. The UF serves as a weighting factor
to average the CD and CS WTW energy use and emissions of PHEVs. Thus, the combined AFC
is always bounded by the height of the CD and CS AFC, as shown in Figure 14. A UF of 100%
yields a combined AFC identical to the CD AFC, which signifies pure CD operation, while a UF
of 0% yields a combined AFC identical to the CS AFC, which signifies pure CS operation
(similar to the operation of a comparable regular HEV.
9,000
8,000
7,000
6,000 PTW (Grid Electricity Use)
5,000 PTW (Engine Fuel Use)

4,000
3,000
2,000
1,000
0
D
D
S
S
S
CD
S
CD
S
CS
CS
S
CD
CS
CS
S
D
D..
,C
,C
&C
,C
,C
,C
&C
&C
,C
,C
,C
&C
,C
D&
s,
R,
rn,
&
s,
rn,
,C
FG
FG
SD
SD
MR
ss
ss
sis
sis
as
as
CD
CD
CD
SM
CD
CD
Co
ma
ma
Co
sis
R
oly
,C
oly
dS
om
om
CI
CI
SI
SI
G,
5-
R,
D,
5-
s,
rn,
ted
oly
Bio
ss
ctr
ctr
ute
.Bi
i
.Bi
as
H. B
LS
RF
SM
E8
E8
Co
ma
ctr
ibu
Ele
H.
Ele
om
lH
lH
tib
CI
SI
SI
SI
5-
te d
5-
5-
Ele
Bio
ist
t ra
tra
Dis
.Bi
te d
ted
E8
E8
E8
-D
ibu
H.
en
ted
en
lH
ibu
ibu
-
SI
SI
SI
5-
H2
H2
-C
is t
-C
t ra
ibu
ist
is t
E8
-D
FC
H2
FC
H2
en
ist
-D
-D
SI
H2
-C
FC
FC
-D
H2
H2
FC
H2
H2
FC
FC
FC
FC
FIGURE 14 WTW Total Energy Use for CD (blended mode) and CS Operations of
Figures 15(a–d) show the WTW results for various PHEV technologies at AER 20,
utilizing the California (NERC Region 13) marginal mix by combining the CD and CS results
and applying a UF of 40%. Figure 15(a) shows a relatively small grid electricity energy use by
PHEV 20 with respect to its total vehicle energy use. On average, the grid electricity energy
share is 6%, 12%, and 24% of the total PHEV energy use for AER 10, 20, and 40, using a UF of
23%, 40%, and 63%, respectively. The small share of electricity use is due to the significant
amount of fuel use by the engine in the CD blended mode of operation. The fuel use in CS
operation further dilutes the share of grid electricity, as implied by the above equation. However,
it is expected that, on a Btu/mi basis, a larger fraction of the electric energy would power the
PHEV wheels in CD operation, compared with the fuel energy, due to the much lower energy
conversion efficiency of the engine relative to the electric motor, as discussed above.
36
Figure 15(b) indicates that PHEVs using fuels from bio-feedstock sources — such as
hydrogen from switchgrass and E85 from corn and switchgrass — consume less fossil energy
relative to other PHEV fuels. Figure 15(c) shows that PHEVs that employ hydrogen as a fuel
almost eliminate the dependence on petroleum energy. The PHEVs that employ E85, which is
blended with a small percentage of gasoline, demonstrate a small dependence on petroleum
energy. The PHEVs that employ hydrogen and E85 from switchgrass exhibit the least GHG
emissions, followed by PHEVs that employ E85 from corn and hydrogen from SMR, as shown
in Figure 15(d). The PHEVs that employ hydrogen produced via electrolysis exhibit the highest
fossil energy use and GHG emissions, despite the high efficiency and low carbon intensity of the
California marginal generation mix. This suggests that PHEVs that use hydrogen produced via
electrolysis may provide GHG emissions benefits over other PHEVs only if the electricity is
generated from non-fossil sources.
Figure 16 shows the WTW petroleum energy use as a function of the PHEV’s AER. The
AER 0 represents the regular HEVs. As expected, the petroleum energy use decreases
significantly with a corresponding increase in AER for petroleum-based fuels (e.g., RFG and
LSD), due to the displacement of petroleum fuels with electricity generated from non-petroleum
sources. A PHEV 40 that uses either RFG or LSD provides a 60% reduction in petroleum energy
use compared with a conventional gasoline ICEV. It should be noted that the trends shown in
Figure 8 are insensitive to the marginal generation mix as long as the mix of fuels for electricity
generation is from non-petroleum sources, such as the case of the California, Illinois, and
U.S. mixes (see Table 4). The reduction of petroleum energy use with the increase in AER is less
significant for the E85 PHEVs because of the small share of gasoline in the E85 blend. All
hydrogen PHEVs are nearly independent of petroleum energy, since the feedstock sources of the
hydrogen fuel are non-petroleum based. For all AER ratings, including AER 0 (regular HEV),
7000
PTW (On-board)
PHEV20 in CA (Model Year 2015)

6000 PTW (Grid Electricity)

5000
WTP (On-board)
4000
3000
2000
1000
0
G
s
rn
sis
as
as
LS
RF
SM
Co
ly
iom
om
CI
tro
SI
te d
85
.Bi
H.B
c
u
E
Ele
lH
tib
SI
5-
a
Dis
ted
nt r
E8
Ce
-
SI
H2
tib
-
Dis
FC
H2
FC
-
H2
FC
FIGURE 15(a) WTW Total Energy Use for Combined CD and CS Operations of PHEV
20 Using the California Marginal Mix (UF=40%)
37
7000
Fossil Energy Use [Btu/mi]


5000
WTP (On-board)
4000
3000
2000
1000
0
FG
SD
s
rn
sis
as
s
SM
Co
ma
R
ly
iom
CI
tro
SI
5-
te d
o
.Bi
H.B
E8
le c
bu
lH
dE
SI
5-
s ti
a
nt r
E8
Di
ute
Ce
-
SI
H2
tib
-
Dis
FC
H2
FC
-
H2
FC
FIGURE 15(b) WTW Fossil Energy Use for Combined CD and CS Operations of PHEV
7000

5000
WTP (On-board)
4000
3000
2000
1000
0
G
s
rn
sis
as
as
LS
SM
RF
Co
oly
iom
om
CI
SI
5-
te d
c tr
.Bi
H. B
E8
Ele
lH
tib
SI
a
85
Dis
te d
ntr
E
bu
Ce
-
SI
H2
sti
-
FC
H2
Di
FC
-
H2
FC
FIGURE 15(c) WTW Petroleum Energy Use for Combined CD and CS Operations of
PHEV 20 Using the California Marginal Mix (UF=40%)
38
400
GHG Emissions [g/mi] WTP (Grid Electricity)
300
WTP (On-board)
200
100
-100
G
s
rn
sis
as
s
LS
RF
SM
Co
ma
ly
iom
CI
SI
tro
5-
te d
io
H. B
.B
E8
c
bu
Ele
lH
SI
5-
s ti
a
te d
nt r
E8
Di
Ce
-
SI
H2
tib
-
Dis
FC
H2
FC
-
H2
FC
FIGURE 15(d) WTW GHG Emissions for Combined CD and CS Operations of PHEV
4500
4000
3500 ICEV SI Gasoline

PHEV SI Gasoline
3000 PHEV CI Diesel
PHEV SI Corn E85
PHEV SI Biomass E85
2500
PHEV FC Biomass
PHEV FC SMR
2000 PHEV FC Electrolysis
1500
1000
500
0
0 10 20 30 40
FIGURE 16 WTW Petroleum Energy Use for Combined CD and CS Operations as a Function of
AER
39
the petroleum use is significantly reduced relative to the gasoline ICEV. The UF for combining
the CD and CS petroleum energy use is 23%, 40%, 53%, and 63% for PHEV 10, 20, 30, and 40,
respectively.
Figures 17, 18, and 19 show the WTW GHG emissions as a function of AER for the
California, U.S., and Illinois marginal generation mixes, respectively. The PHEVs that employ
fuels produced from biomass sources (e.g., E85 and hydrogen produced from switchgrass)
exhibit a proportional increase in GHG emissions with increasing AER because of the significant
contribution of fossil fuels to the electricity generation in the California, U.S., and Illinois mixes,
as shown in Figures 17–19. Thus, PHEVs that employ biomass-based fuels (e.g., biomass-E85
and -hydrogen) may not realize GHG emissions benefits over regular HEVs if the marginal
generation mix is dominated by fossil sources.
Figure 17 shows a significant decrease in the WTW GHG emissions, with a

corresponding increase in AER for all PHEVs that use the California marginal generation mix
(except for fuels produced from biomass sources), due to the displacement of fossil fuels by the
highly efficient electricity generation with low carbon intensity. Figure 19 shows an opposite
trend for the Illinois mix because of its high dependence on carbon-intensive coal for the
marginal generation. All PHEVs, regardless of the AER rating, significantly reduce the GHG
emissions relative to a conventional gasoline ICEV, except for PHEVs that employ hydrogen
produced via electrolysis from carbon-intensive electricity generation, as shown in Figures 18
and 19 for the U.S. and Illinois marginal generation mixes, respectively.
400
350
300
250
200
150
100
ICEV SI Gasoline PHEV FC SMR

50 PHEV FC Electrolysis PHEV FC Biomass
PHEV SI Gasoline PHEV SI Corn E85
PHEV CI Diesel PHEV SI Biomass E85
0
0 10 20 30 40
FIGURE 17 WTW GHG Emissions for Combined CD and CS Operations as a Function of AER
Using the California Marginal Generation Mix
40
700
PHEV FC Electrolysis PHEV FC Biomass
600 PHEV SI Gasoline PHEV SI Corn E85
500
400
300
200
100
0
0 10 20 30 40
FIGURE 18 WTW GHG Emissions for Combined CD and CS Operations as a Function of

AER Using the U.S. Generation Mix
900
800
700
600
PHEV FC Electrolysis PHEV FC Biomass
500 PHEV SI Gasoline PHEV SI Corn E85
400
300
200
100
0
0 10 20 30 40
FIGURE 19 WTW GHG Emissions for Combined CD and CS Operations as a Function of

AER Using the Illinois Marginal Generation Mix
41
Figures 20–23 show side-by-side comparisons of the WTW results of analyses for the
alternative PHEV 20 systems using U.S., California, Illinois, and renewable marginal generation
mixes. The WTW results combine the CD and CS operations, on the basis of a UF of 40% for
AER 20. The first bar on the left represents the WTW result of a conventional gasoline ICEV,
which is provided as a baseline for comparison with the alternative PHEV fuel/vehicle systems.
The graphs show a similar trend across all PHEVs for the different generation mixes. The only
exception is the disappearance of the fossil, petroleum, and GHG emissions bars for the PHEV
that employs hydrogen when produced from renewable electricity via electrolysis. In general, the
energy use and GHG emissions decline progressively in the following order of marginal
generation mixes: Illinois, U.S., California, and renewable. As shown in Figure 22, petroleum
energy use is insensitive to the marginal generation mixes because the main sources for these
mixes are non-petroleum fuels (e.g., coal, natural gas, and renewables). Figure 23 shows that all
PHEV systems provide GHG emissions benefits over the conventional gasoline ICEV, except for
PHEVs that employ hydrogen via electrolysis when the electricity is produced from a less-
efficient and more carbon-intensive generation mix (e.g., U.S. and Illinois).
7000
6000
W TW Total E nergy B tu/m i
5000
4000
3000
2000
1000
0
ICE V S I Gas oline
P HE V CI Dies el
P HE V CI Dies el
P HE V CI Dies el
P HE V CI Dies el
P HE V S I G as oline
P HE V S I Corn E 85
P HE V S I B iom as s E 85
P HE V F C G .H2 S M R
P HE V FC G.H2 E lec troly s is

P HE V F C G .H2 B iom as s


US Average California Illinois Renewable
FIGURE 20 WTW Total Energy Use for PHEV 20 Vehicle/Fuel Systems Using
Different Marginal Electricity Generation Mixes
W TW Petroluem Energy B tu/mi W TW Fossil Energy B tu/m i
0
1000
2000
3000
4000
5000
6000
7000
0
1000
2000
3000
4000
5000
6000
7000
ICE V S I Gas oline ICE V S I G as oline
P HE V S I Gas oline P HE V S I G as oline
P HE V S I Corn E 85 P HE V S I Corn E 85
US Average
P HE V S I B iom as s E 85 P HE V S I B iom as s E 85
US Average
P HE V CI Dies el P HE V CI Dies el
P HE V FC G.H2 S M R P HE V FC G .H2 S M R
P HE V FC G.H2 E lec troly s is P HE V F C G .H2 E lec troly s is
P HE V FC G.H2 B iom as s P HE V F C G.H2 B iom as s

California
California
42


Illinois
Illinois
Renewable
Renewable
FIGURE 21 WTW Fossil Energy Use for PHEV 20 Vehicle/Fuel Systems Using
FIGURE 22 WTW Petroleum Energy Use for PHEV 20 Vehicle/Fuel Systems Using
43
700
600
WTW GHG g/mi
500
400
300
200
100
0
ICEV SI Gasoline
PHEV SI Corn E85
PHEV SI Biomass E85
PHEV SI Corn E85
PHEV SI Biomass E85
PHEV SI Corn E85
PHEV SI Biomass E85
PHEV SI Corn E85
PHEV SI Biomass E85

PHEV SI Gasoline
PHEV SI Gasoline
PHEV SI Gasoline
PHEV SI Gasoline
PHEV CI Diesel
PHEV CI Diesel
PHEV CI Diesel
PHEV CI Diesel
PHEV FC G.H2 SMR
PHEV FC G.H2 SMR
PHEV FC G.H2 SMR
PHEV FC G.H2 SMR

PHEV FC G.H2 Electrolysis

PHEV FC G.H2 Biomass

US Average California Illinois Renewable
FIGURE 23 WTW GHG Emissions for PHEV 20 Vehicle/Fuel Systems Using Different
Marginal Electricity Generation Mixes
Figure 24 summarizes the most significant WTW results for all considered PHEV
fuel/vehicle systems, AER ratings, and marginal generation mixes by calculating the per-mile
ratio of the petroleum energy use and GHG emissions of the PHEVs relative to those of the
baseline conventional gasoline ICEV. It is worthwhile to provide a few guidelines to facilitate an
easy interpretation of Figure 24. The reference point for comparison with all PHEVs is (1,1),
which represents arbitrary units for the baseline conventional gasoline ICEV’s petroleum energy
use and GHG emissions. The color of the marker represents a particular PHEV fuel/vehicle
technology, while the size of the marker represents the AER rating of that PHEV (a smaller
marker for PHEV 10 and a larger marker for PHEV 40). The shape of the marker represents the
marginal generation mix used for recharging the batteries of PHEVs.
Figure 24 notes that the WTW results for the combined CD and CS operations employ
23% and 63% UF for the PHEV 10 and PHEV 40, respectively. All PHEV/grid mix technology
combinations that fall inside the frame bounded by the two points (0,0) and (1,1) provide a
reduction in per-mile petroleum energy use and GHG emissions relative to the conventional
gasoline ICEV. Conversely, all technology combinations that lie outside that frame represent an
increase in petroleum energy use or GHG emissions, or both. The closer the marker is to the
vertical coordinate, the less dependent the technology is on petroleum energy. Similarly, the
closer the marker is to the horizontal coordinate, the lower the GHG emissions are from the
technology.
The markers for a particular PHEV technology are connected from PHEV 10 to PHEV
40. The position of the PHEV 40 marker relative to that of PHEV 10 indicates the relative
change in petroleum energy use and GHG emissions as the AER increases from 10 mi to 40 mi.
Quantitatively, the relative change in petroleum energy use and GHG emissions with AER can
44
be represented by the respective horizontal and vertical components of a vector extending from
the PHEV 10 marker to the PHEV 40 marker.
Figure 24 indicates that all PHEV/grid mix technologies provide a significant reduction
in petroleum energy use and GHG emissions, except PHEVs powered by hydrogen produced via
electrolysis, where the electricity mix is dominated by oil or coal. For example, using the
U.S. average, New York, or Illinois marginal generation mix for hydrogen production via
electrolysis creates the only outliers in Figure 24 due to the high percentage of oil or coal in these
mixes. However, using renewable generation of electricity for hydrogen production via
electrolysis entirely eliminates petroleum use and GHG emissions. Thus, the implication of the
marginal generation mix resides in the electricity generation stage (WTP). In general, the
electricity WTP energy use and GHG emissions increase progressively as the marginal mix
becomes less efficient and dominated by a larger share of oil or coal. It should be noted that use
of the U.S. or Illinois generation mix leads to a reduction in petroleum energy use, since these
mixes incorporate insignificant petroleum sources in their portfolio. The following discussion
focuses on PHEVs with a significant potential for petroleum energy savings and GHG emissions
reduction.
Figure 24 shows three distinct zones of petroleum energy use and GHG emissions for
PHEVs powered by petroleum, E85, and hydrogen fuels. The PHEVs that employ petroleum
fuels, E85, and hydrogen offer a 40–60%, 70–90%, and more than 90% reduction in petroleum
energy use, respectively, compared with the conventional gasoline ICEV. The corresponding
reductions in GHG emissions for PHEVs that employ petroleum fuels, E85, and hydrogen are
30–60%, 40–80%, and 10–100%, respectively. For the same fuel, the spread of the WTW GHG
emissions among the different fuel production technologies and grid mixes is much higher
compared with the spread of petroleum energy use. This is particularly true for E85 and
hydrogen because of the diverse production technologies and feedstock sources considered for
these fuels in this analysis.
Overall, more petroleum energy savings are realized at a higher AER, except when an
oil-intensive grid mix is used. Similarly, more GHG emissions reductions are realized at a higher
AER, except when an oil- or coal-intensive grid mix is used. (In Figure 24, notice the trend from
the smaller to larger diamond- and circular-shaped markers for most PHEVs). The U.S. mix
provides a slight reduction in GHG emissions as the AER increases for PHEVs that employ
petroleum and corn-E85 fuels, but significantly increases the GHG emissions for PHEVs
powered by biomass-E85 and SMR- and biomass-hydrogen fuels for the same increase in AER.
(In Figure 24, see the trend of connected disc-shaped markers). Certainly, PHEVs that use
electricity from renewable sources would realize the most reduction in petroleum energy use and
GHG emissions as the AER rating increases. (In Figure 24, see the trend of connected square-
shaped markers). Using the California marginal mix for PHEV charging provides a significant
reduction in petroleum energy use as well as GHG emissions, except for biomass-based fuels
(e.g., biomass-E85 and biomass-hydrogen). (In Figure 24, notice the trend of connected
triangular-shaped markers). These favorable characteristics of PHEVs in California are attributed
to the highly efficient NGCC technology that dominates its marginal mix.
45
2.0
color/pattern of marker = fuel/vehicle type
shape of marker = electricity generation mix
1.8 size of marker = AER rating
1.6
Baseline (GV)
1.4
GHG Emissions (relative to GV)
PHEV SI Gasoline
PHEV CI Diesel
PHEV SI Corn-E85
PHEV SI H. Biomass-E85 [ VMTCD / VMTtotal ]PHEV10 = 23%
1.2 PHEV FC Distributed SMR-H2 [ VMTCD / VMTtotal ]PHEV40 = 63%
PHEV FC Central H. Biomass-H2
PHEV FC Distributed Electrolysis-H2
1.0
H2
Petroleum Fuels Small marker for PHEV10
Large marker for PHEV40
0.8
E85
0.6
AER 0 (Regular HEV)
0.4 US Ave. Mix
CA Mix
NY Mix
0.2
IL Mix
Renewable
0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
Petroleum Use (relative to GV)
FIGURE 24 Summary of WTW Petroleum Energy Use and GHG Emissions for Combined
CD and CS Operations Relative to Baseline Gasoline ICEV
The isolated markers in Figure 24 represent the regular HEVs (AER 0). The positions of
these isolated markers relative to the baseline conventional gasoline ICEV marker indicate the
reduction in petroleum energy and GHG emissions due to the (grid-independent) hybridization
technology (CS operation) of these vehicles. In addition, the position of these markers relative to
the PHEV markers represents the change in relative petroleum energy use and GHG emissions
because of the partial displacement of VMT from the CS operation of the regular HEV to the CD
operation of the PHEV. The displaced CS VMT in this case is represented by the UF (23% for
PHEV 10, and 63% for PHEV 40). For a carbon-intensive generation mix, such as that of
Illinois, Figure 24 shows that PHEVs produce more WTW GHG emissions compared with
regular HEVs for most fuels. Such implication becomes more pronounced as the AER increases
from 10 mi to 40 mi, especially for E85 and hydrogen fuels, which highlights the significance of
the electricity generation mix for charging PHEVs.
46
6 IMPLICATIONS FOR FUTURE RESEARCH
The WTW results documented in this report are influenced by Argonne’s PSAT
simulation results for the per-mile electricity use and fuel consumption of alternative vehicle
technologies. The WTW results are also influenced by the ORNL predictions of the marginal
electricity generation mix for PHEVs charging in different U.S. regions. Further investigation
and research are required in these two significant areas to better understand how the penetration
of PHEVs into the transportation market will be able to transfer miles to electricity. For example,
various configurations of the electric machine, battery, and engine, as well as the various control
strategies for the combined operation of the electric motor and engine, could significantly affect
the performance in CD operational mode. The lack of an approved testing standard for rating
various PHEV configurations adds to these complications.
The market penetration of the PHEVs, their total electric load, and their role as
complements rather than replacements of regular HEVs are also uncertain. In addition, various
generation expansion paths, which determine available marginal units, should be included to
represent policy options and other factors in grid expansion. The effects of the number of daily
charges, the time of charging, and the charging capacity have not been evaluated in this study. A
more robust analysis of the VMT share of the CD operation is also needed.
47
7 CONCLUSIONS
GREET incorporated PSAT simulation of the fuel economy and electricity use of PHEVs
to perform a WTW energy use and GHG emissions analysis. The WTW results were separately
calculated for the CD and CS modes of PHEV operation, and then combined by using a UF that
represented the CD VMT share. Based on PSAT simulations of the blended CD mode of
operation, grid electricity accounted for a share of the total energy use of the vehicle, ranging
from 6% for PHEV 10 to 24% for PHEV 40, by using a UF of 23% and 63%, respectively.
The electricity generation mix significantly impacted the WTW results, especially GHG
emissions. Three NERC regions (4, 6, and 13) were selected for this analysis because of their
significance. These regions represented marginal generation mixes dominated by coal, oil, and
natural gas, respectively. Results were also reported for the U.S. generation mix and renewable
electricity to examine cases of “average” and “clean” mixes, respectively. The PHEVs that
employed petroleum fuels, E85, and hydrogen, with an AER between 10 mi and 40 mi, were
shown to reduce petroleum energy use by 40–60%, 70–90%, and more than 90%, and GHG
emissions by 30–60%, 40–80%, and 10–100%, respectively, compared with those of a
conventional gasoline ICEV. The spread of the WTW GHG emissions among the different fuel
production technologies and grid generation mixes was wider than the spread of petroleum
energy use, mainly due to the diverse fuel production technologies and feedstock sources for the
fuels considered in this analysis.
In addition, PHEVs offered more savings of petroleum energy use than regular HEVs.
More petroleum energy savings were realized as the AER increased, except for the case of a
marginal grid mix dominated by oil fuel. Similarly, more GHG emissions reductions were
realized as the AER increased, except when the marginal grid mix was dominated by coal or oil.
Electricity from renewable sources realized the most reduction in petroleum energy use and
GHG emissions for all PHEVs as the AER increased. The PHEVs that employ biomass-based
fuels (e.g., biomass-E85 and -hydrogen) may not realize GHG emissions benefits over regular
HEVs if the marginal generation mix is dominated by fossil sources.
48
8 REFERENCES
Delorme, A., A. Rousseau, and S. Pagerit, 2008, “Fuel Economy Potential of Advanced
Configurations from 2010 to 2045,” Proceedings of Les Rencontres Scientifiques de l’IFP —
Advances in Hybrid Powertrains, Nov. 25–26.
EIA: Energy Information Administration
EIA, 2008, “Annual Energy Outlook 2008 with Projections to 2030,” DOE/EIA-0383 (2008),
U.S. Department of Energy, Washington, D.C., June.
EIA, 2007, “Electric Power Annual 2006,” DOE/EIA-0348(2006), U.S. Department of Energy,
Washington, D.C., Oct.
EPA: Environmental Protection Agency
EPA, 2008, “Light-Duty Automotive Technology and Fuel Economy Trends: 1975 Through
2008,” EPA420-R-08-015, U.S. EPA Office of Transportation and Air Quality, Washington,
D.C., Sept.
EPA, 2006, “Fuel Economy Labeling of Motor Vehicle Revisions to Improve Calculation of
Fuel Economy Estimates,” EPA420-R-06-017, U.S. EPA Office of Transportation and Air
Quality, Washington, D.C., Dec.
Gaines, L., A. Burnham, A. Rousseau, and D. Santini, 2007, “Sorting Through the Many Total-
Energy-Cycle Pathways Possible with Early Plug-in Hybrids,” Proceedings of the Electric
Vehicle Symposium 23, Anaheim, CA, Dec. 2–5.
Graham, R., et al., 2001, “Comparing the Benefits and Impacts of Hybrid Electric Vehicle
Options,” 1000349, Electric Power Research Institute, Palo Alto, CA, July.
Hadley, S, 2006, “Impact of Plug-in Hybrid Vehicles on the Electric Grid,” ORNL/TM-
2006/554, Oak Ridge National Laboratory, Oak Ridge, TN, Oct.
IPCC: International Panel on Climate Change
IPCC 2008, 2008, Climate Change 2007: Technical Summary, Cambridge University Press,
Cambridge, UK.
Pagerit, S., A. Rousseau, and P. Sharer, 2006, “Fuel Economy Sensitivity to Vehicle Mass for
Advanced Vehicle Powertrains,” SAE Paper 2006-01-0665, SAE World Congress, Detroit, MI,
April.
49
Plotkin, S., D. Santini, A. Vyas, J. Anderson, M. Wang, J. He, and D. Bharathan, 2001, “Hybrid
Electric Vehicle Technology Assessment: Methodology, Analytical Issues, and Interim Results,”
ANL/ESD-02-2, Center for Transportation Research, Argonne National Laboratory, Argonne,
IL, Oct.
Roach, C., S. Rein, and K. Gottshall, 2008, “2007 State of the Market Report Southwest Power
Pool, Inc.” Boston Pacific Company, Inc., Washington, DC, April.
Rousseau, A., P. Sharer, and F. Besnie, 2004, “Feasibility of Reusable Vehicle Modeling:
Application to Hybrid Vehicles,” SAE paper 2004-01-1618, SAE World Congress, Detroit, MI,
March.
Santini, D., and A. Vyas, 2008, “How to Use Life Cycle Analysis Comparisons of PHEVs to
Competing Powertrains,” presented at the 8th International Advanced Automotive Battery and
Ultracapacitor Conference, Tampa, FL, May 12–16.
Shelby, M., and S. Mui, 2007, “Plug-in Hybrids: A Scenario Analysis,” presented at the 86th
Annual Meeting of the Transportation Research Board, Washington, D.C., Jan. 22.
Shidore, N., T. Bohn, M. Duoba, H. Lohse-Busch, and P. Sharer, 2007, “PHEV ‘All Electric
Range’ and Fuel Economy in Charge Sustaining Mode for Low SOC Operation of the JCS
VL41M Li-Ion Battery Using Battery HIL,” Proceedings of the Electric Vehicle Symposium 23,
Anaheim, CA, Dec. 2–5.
Vyas, A., D. Santini, M. Duoba, and M. Alexander, 2007, “Plug-In Hybrid Electric Vehicles:
How Does One Determine Their Potential for Reducing U.S. Oil Dependence?” Proceedings of
the Electric Vehicle Symposium 23, Anaheim, CA, Dec. 2–5.
Wang, M., 1999, “GREET 1.5 — Transportation Fuel-Cycle Model,” ANL/ESD-39, Center for
Transportation Research, Argonne National Laboratory, Argonne, IL, Aug.
50
51
APPENDIX: RECENT GRID IMPACT STUDIES
Some recent studies on plug-in hybrid vehicle (PHEV) grid impact are summarized
below. The summaries identify the assumed PHEV market share, the generator inventory, and
the resulting dispatch to meet the charging load.
A.1 LEMOINE ET AL. (2008)
This is a study of the capability of existing generators to support PHEV charging in the
California Independent System Operator (CAISO) region. The current generator inventory is
used to meet the charging load for arbitrary levels of PHEV penetration. The study concludes
that the current grid mix is capable of supporting a PHEV fleet much larger than what might
reasonably be anticipated in the current planning period. The analysis estimates how many
vehicles could economically charge at various gasoline price levels in the range of $0.50 to
$3.00 per gallon.
A.1.1 PHEV Performance and Market Assumptions
The performance assumptions were adapted from a 2002 Electric Power Research
Institute (EPRI) study (Duvall et al. 2002). The PHEVs are compact sedans with an all-electric
range (AER) of 20 mi, an on-board fuel economy of 52.7 mi per gallon gasoline equivalent, and
electrical consumption of 249 watt-hours per mile (Wh/mi), including charging losses. Rather
than postulate a market share and growth for PHEVs, this analysis examines the grid impacts
associated with a fleet of 1 million, 5 million, and 10 million PHEVs under several charging
scenarios and for an electricity demand determined by the relative cost of gasoline and
electricity.
A.1.2 Charging Demand and Aggregate Load
All scenarios assume a 1.2 kW demand from each charger as sport utility vehicles
(SUVs) are charged for longer periods rather than at higher power. Three scenarios are
evaluated: optimal charging at times of minimum demand, evening charging that has some
overlap with peak hours, and twice-per-day charging. These charging demands are aggregated
for 1, 5, and 10 million PHEV scenarios in the CAISO region.
A.1.3 Generating Capacity Mix
The current capacity mix for the CAISO region is assumed with no generation expansion.
52
A.1.4 Generating Capacity Dispatch
Rather than explicitly represent or model unit dispatch, this analysis overlays charging
demand on the system load curves to obtain a graphical representation of the impact on peak
loads. Any increase in peak load is an indication that additional capacity would be required.
Under optimal charging (timed to fill the valleys in the load curve), no additional capacity is
required.
A.1.5 Key Results
If the timing of charging is controlled by appropriate mechanisms, even the current

CAISO generating capacity could meet the charging demand from millions of PHEVs. However,
on-peak or on-shoulder charging would result in a total demand that exceeds the available
capacity. These results are interesting, but they do not provide information applicable to the
Greenhouse gases, Regulated Emissions, and Energy use in Transportation (GREET) model
enhancement, since they do not identify the nature of marginal units used to meet the charging
demand. Also, our preference is for a generating mix that represents 2020 or some time at which
significant PHEV penetration is more likely.
A.2 MILLER (2007)
This is a limited analysis of a possible synergy between nuclear generation and PHEV
charging. This analysis concludes that if the current coal capacity in Ontario were replaced with
nuclear, the off-peak energy available from the new nuclear capacity could supply more than
one-third of the Ontario light-duty vehicle miles traveled (VMT) if PHEVs or electric vehicles
(EVs) were used.
Vehicles are assumed to be small light-duty vehicles that obtain 90% of their energy
requirement from the grid.
In Ontario, the generation mix includes run-of-river hydro, dammed hydro, nuclear, coal,
and natural gas. The availability of natural river resources, including the Niagara Falls, has long
defined and will continue to define the Ontario capacity mix. Continued operation or expansion
of nuclear capacity will be determined by several factors, such as policy decisions and public
acceptance. This study assumes that new nuclear capacity is an acceptable generation expansion
option.
53
The base load is served first by nuclear and then by run-of-river hydro, such as the Beck
plant on the Niagara Falls. Variable loads are served by dammed hydro, coal, and natural gas.
A.2.4 Key Results
One goal of large-scale PHEV deployment is a reduction of carbon dioxide (CO2)

emissions. This goal is best served if a low carbon generation alternative is used for charging.
Nuclear is one such option, but nuclear plants must be run continuously. This makes them a poor
candidate for off-peak valley filling operation. However, if nuclear capacity is employed to
displace the existing coal capacity used for variable load, it would be available off-peak to serve
a substantial PHEV charging demand. The author estimates that one-third of Ontario’s light-duty
VMT could be met with this new nuclear capacity.
A.3 DUVALL ET AL. (2007)
This is a sensitivity analysis intended to identify the possible net environmental impacts
of the broad distribution of PHEVs by 2050. High, medium, and low PHEV market-share cases
are evaluated under future generation mixes that are high, medium, and low with respect to
carbon emissions. All cases resulted in GHG reductions, but the range of reductions varies by a
factor of four between the best and worst combinations of PHEV market share and generation
mix; the range through 2050 is 3.4 to 10.3 billion metric tons of CO2. The results are based on a
marginal approach, which estimates the change in emissions from a base case to a PHEV
penetration scenario.
Overall, the vehicle inventory includes internal combustion engine vehicles, HEVs, and
PHEVs with three AER options: 10 mi, 20 mi, and 30 mi. The onboard power for these vehicles
is either diesel or gasoline. The PHEV technology is applied only to light-duty vehicles and to
heavy-duty vehicles with less than a 19,500-pound gross vehicle weight rating. The HEV fuel
consumption is 35% less than a comparable conventional vehicle, while the PHEV fuel economy
in charge-sustaining mode is assumed to be the same as the fuel economy of a comparable HEV.
By the year 2050, the low, medium, and high penetration assumptions result in a 20%,
62%, and 80% market share of new vehicle sales, respectively. In the non-PHEV case, the
market share of regular HEVs increases to 30%, 63%, and 75%, respectively, as a baseline for
each of these scenarios.
54
Vehicle charging is assumed to match a demand profile where charging ramps up after
6 p.m., maintains a sustained peak from 11 p.m. until 3 a.m., and declines after 3 a.m. to a
minimum at 8 a.m. A modest mid-day charging increase extends from 11 a.m. through 3 p.m.
Utilities may use controls to impose charging restrictions on large fleets.
The National Electric System Simulation Integral Evaluation (NESSIE) model,

developed by EPRI, is used to simulate generation expansion and generation dispatch at five-
year intervals from 2010 to 2050. The National Energy Modeling System (NEMS), developed by
the Energy Information Administration (EIA), is used to obtain fuel prices and emission
allowances. The generating unit inventory at year 2030 resulting from the NESSIE model
simulations is very similar to that reported in the AEO 2006, which used the NEMS Electricity
Market Module for generation expansion and dispatch. The analysis is nationwide, with regional
disaggregation based on the NEMS regions. However, only aggregate national results are
provided in the report.
The NESSIE model is also used for dispatch of the expanded capacity. Details are not
discussed in the report. It is noted that, since charging is largely off-peak, base load coal units
will provide most of the charging power.
A.3.5 Key Results
The regional generation mixes were not given explicitly in this report, so we were unable
to apply the results to GREET. However, if regional results were available, this study would
provide a useful validation of other dispatch modeling results.
A.4 KINTNER-MEYER ET AL. (2007)
This analysis seeks to estimate the maximum PHEV fleet size that could be charged by
filling the so-called “valley” in the diurnal load curve with regards to 12 North American
Electric Reliability Corporation (NERC)-based regions.
55
The AER assumed was a PHEV 33, which was based on a national annual average daily
VMT of 33 mi per day per vehicle. The electrical consumption of these PHEVs ranged from
260 Wh/mi for a compact sedan to 460 Wh/mi for a full-size SUV.
The vehicle charging rate was not specified. However, the battery capacity for the
PHEVs ranged from 8.6 kWh to 15.2 kWh, depending on vehicle class. The aggregate load is
actually calculated from the available capacity to fill the load curve valley for each region. This
aggregate load divided by the battery capacity yields the number of vehicles.
The analysis is restricted to the existing generation and transmission and distribution
infrastructure. This yields a conservative capacity result, since it allows for no generation
expansion.
All charging is simply applied to fully utilize available capacity, so no dispatch modeling
was used.
A.4.5 Key Results
The analysis concludes that as much as 73% of U.S. light-duty VMT could be supported
by the existing electric power infrastructure, with an overall reduction in net GHG emissions.
This analysis is not directly applicable to GREET calibration, but it shows the potential for
PHEVs to reduce petroleum use.
A.5 HADLEY AND TSVETKOVA (2008)
In this regional analysis, the distribution and use of PHEVs are based on region-specific
populations of vehicle types and AEO 2007 projections of total sales of light-duty vehicles. The
make-up of the generator fleet used for charging is also region-specific. The 13 regions are based
on the 10 NERC regions within the continental United States (as defined before 2006), with the
Western Electricity Coordinating Council split into three regions for this analysis and the
Northeast Power Coordinating Council split into two regions. The PHEV market penetration
assumptions yield a PHEV inventory for 2020 and 2030. The charging load for these vehicles is
met by an expanded generation inventory from EIA projections. Electricity costs and emissions
56
from power production are projected for each region for a base case with no PHEVs and for the
case with PHEVs. The emissions and costs associated with the PHEVs can be taken as the
difference between these values. Fuel use and emissions for the PHEVs are compared in this
analysis with those from charge-sustaining HEVs, which are regarded as the alternative vehicles
being displaced by the plug-in technology.
National new vehicle sales projections from the AEO 2007 were distributed among
regions and vehicle types (all light duty) according to current vehicle registrations. The PHEV
market share is assumed to increase from zero before 2010 to 25% of new car sales by 2020 and
hold steady after that. With these assumptions and expected vehicle retirement at 10 years, the
fleet inventory is constructed. The AER is assumed to be 20 mi.
Several charging rates are evaluated (1.4 kW, 2 kW, and 6 kW). The 6 kW requires a
240-V, 30-A circuit. The aggregate load is based on the assumption that each PHEV will require
daily charging of a battery capacity sized for 20 mi of operation. This corresponds to a charge
from 20% to 100% of the battery packs. These packs range in capacity from 5.1 kWh for a
compact sedan to 9.3 kWh for a full-size SUV. Because the number of PHEVs of various types
is known from the market assumptions, the aggregate daily demand can be calculated for each
region. Several alternative charging strategies are analyzed. “Evening” charging starts at 6 p.m.
or 7 p.m. One-half of the PHEV fleet begins charging at each of these hours. “Night” charging
starts at 10 p.m. or 11 p.m.
The regional generator inventory for 2020 and 2030 is taken from the AEO 2007, and
that inventory reflects necessary expansion to meet growth, anticipated unit retirements, and fuel
and technology choices on the basis of capital costs, projected fuel costs, and regulatory
restrictions. However, since the AEO 2007 does not anticipate PHEV market growth, PHEV
charging demand is not incorporated in the generation expansion planning.
The PHEV charging demand has been superimposed on the demand patterns from the
AEO 2007 projections. This changes the diurnal load pattern by adding load during the evening
or nighttime hours selected for charging. The loading of generators to meet this new demand
pattern is developed with the Oak Ridge Competitive Electricity Dispatch Model. This model
determines which units will be brought on-line or ramped up to meet the PHEV charging
demand and, consequently, what associated emissions will result.
57
A.5.5 Key Results
Our interest is in obtaining an estimate of the generation mix employed on the margin to
meet the PHEV generating load. GREET uses that mix to estimate the electricity generation
portion of the WTW results for PHEV operation. From that perspective, the results of this study
are very informative. Results for six cases, including three charge rates (1.4 kW, 2 kW, and
6 kW) and two charging times (evening and night), are presented. It should be noted that the
high-charge-rate case (6 kW) results in some unserved load with evening (5 p.m. or 6 p.m.)
charging. This is a reflection of the facts that the original generation expansion included in the
AEO 2007 did not anticipate PHEV charging demand and that the charging start time occurs
during a high load period. Similar results are presented for each of the 13 regions.
A.6 REFERENCES
Duvall, M., E. Knipping, M. Alexander, L. Tonachel, and C. Clark, 2007, “Environmental

Assessment of Plug-in Hybrid Electric Vehicles; Volume 1: Nationwide Greenhouse Gas
Emissions,” 1015325, Electric Power Research Institute, Palo Alto, CA, July.
Duvall, M., et al., 2002, “Comparing the Benefits and Impacts of Hybrid Electric Vehicle
Options for Compact Sedan and Sport Utility Vehicles,” 1006892, Electric Power Research
Institute, Palo Alto, CA, July.
Hadley, S., and A. Tsvetkova, 2008, “Potential Impacts of Plug-in Hybrid Electric Vehicles on
Regional Power Generation,” ORNL/TM-2007/150, Oak Ridge National Laboratory, Oak Ridge,
TN, Jan.
Kintner-Meyer, M., K. Scheider, and R. Pratt, 2007, “Impacts Assessment of Plug-in Hybrid
Vehicles on Electric Utilities and Regional U.S. Power Grids; Part 1: Technical Analysis,”
Pacific Northwest National Laboratory, Richland, WA, Nov.
Lemoine, D., D. Kammen, and A. Farrell, 2008, “An Innovation and Policy Agenda for
Commercially Competitive Plug-in Hybrid Electric Vehicles,” 014003, Environmental Research
Letters Volume 3(1), Feb.
Miller, A., 2007, “A Historic Perspective on the Future Cost of Off-Peak Electricity for EVs,”
Plug-in Highway PHEV2007 Conference, Winnipeg, MB, Canada, Nov.
58
Energy Systems Division
Argonne National Laboratory
9700 South Cass Avenue, Bldg. 362
Argonne, IL 60439-4815
www.anl.gov
A U.S. Department of Energy laboratory

managed by UChicago Argonne, LLC

559

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

559

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

559

Uploaded by

Copyright:

Available Formats

ANL/ESD/09-2

Well-to-Wheels Energy Use

Energy Systems Division

Availability of This Report

Well-to-Wheels Energy Use

ACKNOWLEDGMENTS ........................................................................................................ vii

2 ELECTRICITY GENERATION MIX.............................................................................. 6

2.1 Factors Affecting Generation Mix for PHEV Charging .......................................... 6

3 PSAT SIMULATION OF VEHICLE FUEL ECONOMY ............................................... 16

4 VEHICLE MILES TRAVELED SPLIT BY CHARGE-DEPLETING

5 GREET WELL-TO-WHEELS ENERGY USE AND GREENHOUSE GAS

6 IMPLICATIONS FOR FUTURE RESEARCH................................................................ 46

APPENDIX: RECENT GRID IMPACT STUDIES................................................................. 51

1 Vehicle Technologies, Fuels, and Feedstock Sources....................................................... 4

2 Mix of Sources for Average Electric Generation among Regions in the

3 Distribution of U.S. Energy Consumption ........................................................................ 12

4 Generation Mixes for Recharging PHEVs for Use in GREET ......................................... 15

5 Vehicle Assumptions for PSAT Simulations .................................................................... 17

6 PSAT Electricity Use and Fuel Economy Results ............................................................ 18

7 Share of National VMT Available for Substitution by a PHEV Using

1 NERC Regions from the Annual Energy Outlook 2007 ................................................... 6

3 Example of Hourly Marginal Fuels Data by Time of Day................................................ 8

4 SPP Daily Maximum and Minimum Electricity Demand Source..................................... 9

5 CD VMT on UDDS from PSAT Simulations ................................................................... 19

6 CD VMT on HWFET from PSAT Simulations ................................................................ 19

7 Average VMT for UDDS and HWFET on CD Operation ................................................ 20

8 National VMT Available for Substitution by a PHEV Using

9 Fuel Consumption in CD Operation.................................................................................. 26

10 Fuel Consumption in CD and CS Operations ................................................................... 26

11(d) WTW GHG Emissions for CD and CS Operations of PHEV 20 Using

12(d) WTW GHG Emissions for CD and CS Operations of PHEV 20 Using

13(d) WTW GHG Emissions for CD and CS Operations of PHEV 20 Using

14 WTW Total Energy Use for CD and CS Operations of PHEV 20 Using

15(b) WTW Fossil Energy Use for Combined CD and CS Operations of

15(c) WTW Petroleum Energy Use for Combined CD and CS Operations of

15(d) WTW GHG Emissions for Combined CD and CS Operations of

16 WTW Petroleum Energy Use for Combined CD and CS Operations as a

17 WTW GHG Emissions for Combined CD and CS Operations as a

18 WTW GHG Emissions for Combined CD and CS Operations as a

19 WTW GHG Emissions for Combined CD and CS Operations as a

20 WTW Total Energy Use for PHEV 20 Vehicle/Fuel Systems Using

21 WTW Fossil Energy Use for PHEV 20 Vehicle/Fuel Systems Using

22 WTW Petroleum Energy Use for PHEV 20 Vehicle/Fuel Systems Using

23 WTW GHG Emissions for PHEV 20 Vehicle/Fuel Systems Using

WELL-TO-WHEELS ENERGY USE AND GREENHOUSE GAS EMISSIONS

Amgad Elgowainy, Andrew Burnham, Michael Wang,

Researchers at Argonne National Laboratory expanded the Greenhouse gases, Regulated

TABLE 1 Vehicle Technologies, Fuels, and Feedstock Sources

Technology Fuel Feedstock

2 ELECTRICITY GENERATION MIX

A key factor in determining the environmental performance of PHEVs is the source of

2.1 FACTORS AFFECTING GENERATION MIX FOR PHEV CHARGING

2.1.1 Time of Day

Coal Natural Gas Light Oil Misc

2.1.2 Time of Year

2.1.4 Generation Mix

Climate, fuel availability, population, industrial activity, local regulation, water

2.1.5 Vehicle and Charger Design Factors

2.1.6 Load Growth and Generation Expansion

AFCcombined = (AFCGrid + AFCEngine)CDUF + AFCCS(1-UF). (1)