Content uploaded by Max Åhman
Author content
All content in this area was uploaded by Max Åhman on Nov 16, 2017
Content may be subject to copyright.
Energy 26 (2001) 973–989 www.elsevier.com/locate/energy
Primary energy efficiency of alternative
powertrains in vehicles
Max A
˚hman
*
Department of Environmental and Energy Systems Studies, Lund University, Gerdagatan 13, SE-223 63,
Lund, Sweden
Received 23 June 2000
Abstract
This study considers the technical potential concerning the energy efficiency attainable for vehicles with
alternative powertrains within 10–20 years. The potential for electric vehicles (BEVs), hybrid electric
vehicles (HEVs) and fuel-cell electric vehicles (FCEVs) is assessed and compared with the potential
improvement in conventional vehicles with internal combustion engines (ICEVs). Primary energy efficiency
is the measure used in this study for comparison. The calculations of primary energy efficiency are based
on three different resources: fossil fuels, biomass, and primary electricity from wind, solar or hydropower.
This study shows that there is potential for doubling the primary energy efficiency using alternative power-
trains in vehicles such as BEVs, HEVs and FCEVs, compared with existing ICEVs. All vehicles with an
alternative powertrain have a higher potential for primary energy efficiency than vehicles with an improved
conventional powertrain. No “winner” amongst the alternative powertrains could be identified from a pri-
mary energy efficiency point of view. 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
Road traffic causes a number of environmental problems such as noise, congestion, and the
emission of NO
x
, CO, CO
2
, and particulate matter. According to a number of “business as usual”
prognoses, the growth in the demand for transport is expected to continue, which will exacerbate
the negative impact on the environment, see, for example [1,2].
The perhaps most difficult environmental problem to solve is the expected greenhouse effect,
caused primarily by the emission of CO
2
. Technical innovations, such as the catalytic converter
and improved fuels, have decreased the emission of VOC, NO
x
,SO
x
, and lead due to road trans-
* Tel.: +46-222-86-43; fax: +46-222-86-44.
E-mail address: max.ahman@miljo.lth.se (M. A
˚hman).
0360-5442/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved.
PII: S0360-5442(01)00049-4
974 M. A
˚hman / Energy 26 (2001) 973–989
port during the last 15 years while the limited improvement in vehicle fuel economy has been
offset by a growing demand for transportation [3].
A long-term solution for mitigating CO
2
emission in the transport sector would be the use of
renewable fuels instead of fossil fuels. Fuels and electricity from renewable sources are, however,
still relatively expensive and the supply is limited [4]. Even in optimistic scenarios the renewable
energy supply will be restricted [5]. The use of more energy-efficient vehicles is an important
step towards a technical solution of the CO
2
problem. Several new energy-efficient powertrains
are currently being investigated by scientists, governments, and car manufacturers.
The aim of this study was to assess the future possible primary energy efficiency attainable in
passenger cars with alternative powertrains. To be able to compare the system efficiency of
vehicles that use different energy carriers, the primary energy efficiency was used as a measure
for comparison. Primary energy efficiency takes into account all energy use “from the well to the
wheel”. The primary energy efficiency for energy chains based on fossil fuels, biomass, and
primary electricity from renewable sources, was compared in this study. Future efficiencies stated
in this study are assumed attainable if developments of key technologies are successful and energy
efficiency has high priority in development.
Other studies have compared alternative powertrains with conventional vehicles and analysed
the potential benefits regarding energy efficiency. Most of these studies, however, have compared
only one of the alternatives with the conventional powertrain. In Ecotraffic, Metz et al., Wang
and Deluchi, OECD/IEA, and Wabro and Wagner [6–10] the battery-powered electric vehicle
(BEV) is compared with the internal combustion vehicle (ICEV). The hybrid electric vehicle
(HEV) and the fuel-cell vehicle (FCEV) are compared with the conventional ICEV in Amann,
Cuddy and Wipke, Wang et al. [11–13] and in Lipman and DeLucchi [14], respectively.
When comparing powertrains using the same energy carrier (such as petrol) there is no need
to consider the primary energy efficiency [11,12]. When different energy carriers with varying
degrees of energy losses during fuel production and distribution are used, primary energy
efficiency analysis becomes necessary [6–8]. Some studies do not include renewable fuels in their
assessments [7] and other studies do not include all the alternative powertrains relevant today in
the comparison [6,8,14].
In this study, we focused on primary energy efficiency both from fossil and renewable
resources, and we included all of the currently most promising alternative powertrains. Another
important feature is that the future potential for all the different vehicle types and energy conver-
sion methods was investigated. This study does not show which energy efficiency will be attained
in the future but which energy efficiency could be attained.
2. Method
The powertrains studied included electric drivetrains
1
in BEVs, HEVs with an internal combus-
tion engine (ICE) and in FCEVs. These vehicles are probable future alternatives to conventional
1
The term drivetrain typically refers to the transmission system from engine output shaft to driven road wheels. In the term
alternative powertrain we include both the electric drivetrain, energy storage (e.g. batteries and hydrogen storage), and, possibly, a
prime mover (ICE or fuel-cell).
975M. A
˚hman / Energy 26 (2001) 973–989
ICEVs. The focus of the study was on powertrain technology potentially available within 10–20
years. The time scale was set to enable an assessment of potential without considering the time
needed for the development of different powertrains. The future development of the ICEV was
also assessed and compared with the alternative powertrains.
The calculations of primary energy efficiency were based on three different resources: fossil
fuels, biomass and primary electricity from wind, solar or hydropower. The definitions of power-
train efficiency, vehicle efficiency and primary energy efficiency are shown in Fig. 1. W
D
is the
primary energy, W
C
is the energy supplied to the vehicle, W
B
is the energy supplied to the power-
train, and W
A
is the useful energy at the wheels.
Powertrain efficiency, h
powertrain
⫽W
A
/W
B
Vehicle efficiency, h
vehicle
⫽W
A
/W
C
Primary energy efficiency, h
primary
⫽W
A
/W
D
.
The powertrain efficiency was calculated from the efficiencies of the different components
included in the powertrain. The component efficiencies are assumed future mean efficiencies
attainable over a normal drive schedule. To calculate the vehicle efficiency, the powertrain
efficiency was corrected for losses due to the power required for heating and for the benefits
when no energy is required during idling and the use of regenerative braking. The primary energy
efficiency included the energy used for energy extraction, conversion, distribution and storage.
The energy embodied in plants, buildings and vehicles was not included. Embodied energy typi-
cally accounts for only 7–8% of total life-cycle energy use today [52]. However, for future fuel
efficient vehicles this fraction could increase to between 14 and 18% for the different alternatives
in this study [52]. For electricity, a marginal perspective was used, which means that the efficiency
of the electricity supply was calculated as the efficiency of the supplementary electricity pro-
duction required by the system to supply the energy for the vehicles.
The alternative vehicles are assumed to have the same comfort, performance, and size as a
Fig. 1. Definitions of primary energy efficiency, vehicle efficiency, and powertrain efficiency.
976 M. A
˚hman / Energy 26 (2001) 973–989
conventional vehicle today. Looking at efficiency will, however, not give all the answers as the
future weight may differ between the alternatives (notably for the BEV). The potential for further
efficiency improvement by reducing vehicle weight, rolling resistance and aerodynamic drag is
discussed in Section 5.
3. Description of technologies
3.1. ICEVs with a conventional powertrain
The conventional powertrain consists of a fuel tank, an internal combustion engine, and a
transmission. A characteristic of the internal combustion engine is that maximum efficiency is
attained near the maximum load point. This makes the mean efficiency relatively low since
maximum power is very seldom used under normal driving conditions. The mean power required
in a US Federal Test Procedure (FTP) schedule is below 10 kW [15], while the maximum power
required is between 60 and 90 kW, depending on the size of the vehicle. The mean efficiency is
thus low, around 18% in an FTP schedule [6], compared with the maximum efficiency, which is
between 35 and 40% in a new engine today. Possible options for improving the mean efficiency
in the conventional powertrain are variable valve timing, shut-off during idling, higher com-
pression ratio and a continuously variable transmission, see [16].
3.2. The electric drivetrain and BEVs
The electric drivetrain is an essential part of BEVs, HEVs, and FCEVs and consists of a gener-
ator, an electric motor, and a transmission.
The battery has always been the weak link in BEVs. Low energy storage capacity compared
with petrol has restricted the driving range of BEVs. A battery in a BEV should store up to 30
kWh to afford the vehicle an acceptable range. In order to make BEVs commercially viable, the
United States Advanced Battery Consortium (USABC) argues that a BEV battery should be able
to store at least 150 Wh/kg [17]. The batteries used today are lead/acid (Pb/A), nickel–metal
hydride (NiMH), and lithium batteries which store 80–100 Wh/kg [18]. The only battery believed
to have the long-term potential to reach the USABC goal of 150 Wh/kg, is the lithium–polymer
battery, see, for example [18,19].
The cost of batteries is a major obstacle today. A NiMH battery costs somewhere between 200
and 350 US$/kWh [18], which means between 6000 and 10 500 US$ for a BEV battery package
of 30 kWh. The only battery expected to be able to attain the long-term goal set up by the USABC
(a cost of ⬍150 US$/kWh) is the Pb/A battery [18,20]. Some analysts argue, however, that the
lithium–polymer battery may also have the long-term potential to reach a cost close to 150
US$/kWh [18].
Pb/A batteries have been the dominating type of batteries in BEVs and HEVs so far, but the
future BEVs in this study are assumed to have a NiMH or lithium battery. Both batteries are
available today, but they differ in price and availability in favour of the NiMH battery. For future
HEVs, Pb/A batteries might well prove to be the best alternative in the medium term, considering
both technical performance and price [18].
977M. A
˚hman / Energy 26 (2001) 973–989
Both generators and electric motors have been greatly improved during the past 20 years. Due
to the development of advanced electronic control systems, the mean energy efficiency over a
normal drive schedule has increased both for generators and electric motors, see, for example
[21–23]. Today, only a one-speed reduction-gear is needed to manage all possible power and
speed requirements for an electric motor with an advanced control system [21–23]. The general
assumption of this paper is that the variations in loadand speedwill, to the greatestextent possible,
be handled by the electric drivetrain.
The potential energy efficiency of an electric drivetrain ranges between 65 and 75% (see Table
1). Current efficiencies are lower, around 57%. Future improvements in efficiency will result from
the implementation of an advanced control system together with a modern generator and electric
motor. The use of lithium–polymer batteries could also improve the electric drivetrain efficiency
substantially in the future. The electric drivetrain in HEVs may be more efficient than in BEVs
due to the potential for more efficient batteries.
2
Table 1
Assumed future possible component mean efficiencies over a normal drive schedule for the electric drivetrain
Electric drivetrain in Generator (%) Battery (%) Electric motor Transmission Total energy
the following and control (%) efficiency for the
configurations system (%) drivetrain (%)
h
BEV (NiMH battery) 85
a
80
a
86
b
98
g
57
today
BEV (NiMH battery) 92
b
81
c
89
f
98
g
65
potential
HEV (Pb/A battery) 92
b
90
d
89
f
98
g
72
potential
BEV/HEV (Li battery) 92
b
95
e
89
f
98
g
76
potential
a
The efficiencies of vehicles on the market today differ widely between manufacturers due to rapid development.
Assumptions based on [9,24].
b
Assumption based on: [25].
c
Assumption based on [18,26].
d
Pb/A batteries, which are efficient and easy to use in HEVs [12,18].
e
The performance of lithium–polymer batteries is difficult to validate because there is basically only one advanced
manufacturer, 3M/Hydro-Quebec. Adapted from [19,27,28].
f
Assumptions based on [21–23].
g
Assumed mean efficiency of a reduction gear. Based on: [9,12]. A reduction gear is also assumed in the electric
drivetrain in parallel HEVs in the future.
h
Multiplying these mean efficiencies is a simplification for calculating the system mean efficiency, but given the
uncertainties the figures represent good indicators on future possible mean efficiency.
2
Hybrid vehicles can use lead/acid batteries. Such batteries are very efficient if not charged to 100% [18].
978 M. A
˚hman / Energy 26 (2001) 973–989
3.3. HEVs with internal combustion engines
In the hybrid powertrain, an electric motor and a battery are combined with a heat engine and
a fuel tank. The heat engine, primary engine, can charge the battery, or take over the driving
from the electric drivetrain when the battery is discharged.
The primary engine and the electric drivetrain are either used in series or in parallel. In a series
configuration, all the energy must go through the electric drivetrain. In a parallel configuration,
part of the energy passes a mechanical drivetrain. The HEVs sold today are not pure series or
parallel configurations, but have increasingly integrated configurations.
Today, ICEs dominate as primary engines, see, for example Toyota Prius and Honda Insight.
With hybridisation, the ICE can be designed for the mean power of a normal driving schedule
instead of the maximum power required. This allows the engine to operate closer to its maximum
efficiency. We have not considered gas turbines or Stirling engines, as the potential for high
energy efficiency in passenger-car-sized engines seems to be low, see, for example [29,30].
The mean efficiency of the ICE in a hybrid configuration is given in Table 2. The ICE usually
used is a four-stroke direct injection (4SDI) engine, and the assumed ICE developed for future
use is either a compression ignition direct injection (CIDI) engine or a new engine type, combining
both CIDI and 4SDI features, e.g. the active thermal atmospheric combustion (ATAC) engine.
The parallel configuration assumed in this study is a genuine parallel configuration with as small
a battery pack as possible.
3.4. FCEVs
Fuel-cell technology is now being developed for automotive purposes. The most interesting fuel
cell for this application today is the proton exchange membrane (PEM) fuel cell. The advantage of
the fuel cell is the potential for high energy efficiency and zero tail pipe pollutants.
One disadvantage of the fuel-cell system is the requirement of pure hydrogen for fuel in the
cell. Hydrogen can be stored on board the vehicle, either as a liquid, in nanofibres or in hydrides,
or as compressed gas. Nanofibre technology is still at the basic research level and was not con-
sidered in this study. Liquid storage is the most energy-consuming alternative. About 50% of the
energy content is used to liquefy the hydrogen gas [4,31,32]. The energy efficiency of compressing
Table 2
Assumptions of mean primary engine energy efficiency in different configurations
Mean engine efficiency over a normal HEV series HEV parallel ICEV (%)
drive schedule configuration (%) configuration (%)
ICE, today (4SDI) ––18
a
ICE, future (CIDI or developed 4SDI) 40
b
36
b
24
c
a
Efficiency today. Adopted from [6].
b
Adapted from [12] assuming a diesel engine with 43% maximum efficiency and a 55%FUDS/45%FHDS. The
parallel engine will have to deal with greater variation in load and thus have a lower mean efficiency than the series
hybrid. Future engines will probably have the same efficiency, even if they are of another type (e.g. ATAC, 4SDI, GDI).
c
Adapted from [16] assuming variable valve timing, idle shut-off, higher compression ratio, etc.
979M. A
˚hman / Energy 26 (2001) 973–989
hydrogen gas to 350 bars is about 70–90%
3
[35]. Another alternative is to use a “hydrogen
carrier”, e.g. methanol or petrol, to provide the hydrogen for the cell. Methanol and petrol are
easy to store with normal vehicle technology, but this solution lowers the fuel-cell system
efficiency (see Table 3).
A fuel cell using methanol directly, without reforming it to hydrogen, is still in a very early
phase of development [36,37], and was thus not considered here.
The PEM fuel cell has different efficiency features compared with the ICE, making hybridis-
ation less interesting. The maximum efficiency of a fuel cell is attained at 25–50% of maximum
load [38], which gives no benefit from reducing maximum power with hybridisation. However,
there are other advantages in hybridisation of fuel-cell systems. The most obvious ones are the
possibility to use electricity from the battery during idling and to help FCEVs to start cold, and
the possibility of utilising regenerative braking [39,40].
In this study, hybridisation of the fuel-cell vehicle was assumed and the fuel cell was assumed
to be as small as possible due its high cost. This makes the efficiency lower than the maximum
possible. Other strategies may, however, be chosen by industry, see, the discussion in [40].
Assumptions regarding the efficiency of a PEM fuel-cell system are given in Table 3.
4. Efficiencies of different powertrains
Calculated efficiencies for the powertrains studied are given in Table 4. The highest efficiency
is achieved for the battery-powered electric vehicle. The two hybrid powertrains and the fuel-cell
powertrain fuelled with methanol have approximately the same efficiency. The fuel-cell powertrain
has about 20% higher efficiency when fuelled with pure hydrogen gas than when it is fuelled
with methanol. Furthermore, there is a considerable potential for improvement in the conven-
tional powertrain.
The electric drivetrain offers the benefits of no fuel use during idling and the possible use of
regenerative braking. But there is also a disadvantage in that electric power is needed for heating,
since the heat loss from the ICE are too small to cover the demand for interior heating of the
Table 3
Assumed future PEM fuel-cell system mean efficiency
Mean fuel efficiency, hybrid Hydrogen gaseous Hydrogen liquid Methanol (%) Petrol (%)
configuration (%) (%)
PEM fuel-cell system
a
47 47 47 47
Reformer efficiency
b
––85 80
Total fuel-cell system energy 47 47 40 38
efficiency
a
Adapted from [27,30,39] with an assumed maximum efficiency of 55% and the efficiency curve in [40].
b
Adopted from [30].
3
With figures from [33], the energy efficiency is 73% assuming compression from 7 to 35 MPa. DeLuchi and Ogden claim a
higher efficiency, 91% [34]. The figures are uncertain and depend partly on the assumptions regarding electricity generation.
980 M. A
˚hman / Energy 26 (2001) 973–989
Table 4
Future powertrain efficiencies
Powertrain efficiency Primary engine 5-speed Electric drivetrain Total powertrain
(%)
a
transmission (%) (%)
b
energy efficiency
(%)
Battery-powered powertrain ––65 65
Hybrid powertrain parallel 36 92 68
c
30
d
Hybrid powertrain series 40 –72 29
Fuel cell powertrain 40 –72 29
methanol fuelled
Fuel cell powertrain 47 –72 34
hydrogen fuelled
Conventional developed 24 92 –22
Conventional today 18 92 –16
a
Based on Table 2.
b
Based on Table 1.
c
Adapted from Table 1 with the exception that the transmission is assumed to be mechanical.
d
Assuming a 55% city driving schedule with electric drivetrain and 45% highway driving schedule with mech-
anic drivetrain.
vehicle. All these features influence the total vehicle efficiency (see Table 5). To calculate the
effects on vehicle efficiency, the energy sinks for braking, idling and heating were estimated
according to Amann [11] and DeCicco and Ross [16]. The difference between vehicle and power-
train efficiency is small, but a small relative improvement for vehicles using electric drivetrains
compared with vehicles using conventional powertrains can be identified (see Table 5).
The primary energy efficiencies, when energy carriers based on fossil resources are used, are
given in Table 6. The fossil fuels are produced from crude oil or natural gas. Marginal fossil-
based electricity generation from coal was assumed in the short term and, in the long term, new
marginal electricity from natural-gas plants was assumed. The reason for this is that power from
new natural-gas plants is produced at a lower cost than in new coal-fired plants [41].
With new electricity production, based on natural gas, the BEV has the highest primary energy
efficiency of the alternatives studied. The advantage with regard to the emission of CO
2
is even
higher due to lower carbon content per unit energy than in coal or oil. If the electricity is generated
from coal, the primary energy efficiency for BEVs is lower than for HEVs and FCEVs.
HEVs with advanced heat engines are twice as efficient as conventional vehicles today. FCEVs
have lower efficiencies than HEVs due to high conversion losses from natural gas to hydrogen
or methanol.
An important option for CO
2
mitigation is the use of biomass as a renewable energy source
[41]. The primary energy efficiencies for vehicles using energy carriers based on biomass are
given in Table 7. The BEV has the highest potential for primary energy efficiency. One reason
for this, apart from the high vehicle efficiency, is the fact that liquid and gaseous fuels are pro-
duced from biomass with relatively low efficiency. The various HEVs and FCEVs all have simi-
lar efficiencies.
Finally, the primary energy efficiencies for vehicles using primary electricity from solar, wind,
981M. A
˚hman / Energy 26 (2001) 973–989
Table 5
Vehicle efficiency calculated as consumed energy at the wheels divided by the total energy supply to vehicle, see Fig. 1
Vehicle Consumed Energy out Energy to Energy Energy Total Vehicle
energy at from powertrain
b
used for during energy efficiency
f
the wheels powertrain
a
extra loads
c
idling
d
supplied to (%)
vehicle
BEV 100 95 146 17 0 163 61
HEV parallel 100 95 327 14 0 340 29
HEV series 100 95 327 14 0 340 29
FCEV methanol 100 95 327 14 0 340 29
FCEV hydrogen 100 95 279 14 0 293 34
ICEV developed 100 100 452 10 27
e
489 20
ICEV today 100 100 625 10 75 700 14
a
5% of consumed energy at the wheels is saved due to regenerative braking which assumes that 25% of the braking
energy is regenerated. This relatively low figure is due to traffic safety.
b
Calculated as (energy out from powertrain)/(powertrain efficiency based on Table 4).
c
Assumed to be 10% of the useful energy at the wheels according to [16]. Consideration taken of the fact that some
heat losses can be used for heating for the FCEV and the HEV.
d
Energy during idling assumed to be 12% of “energy to powertrain”. Adapted from [11,16].
e
50% of energy used during idling can be saved, according to [16].
f
Calculated as (consumed energy at the wheels)/(total energy supplied to vehicle).
Table 6
Primary energy efficiency with a fossil primary energy source
Vehicle Primary energy Energy carrier Primary Distribution
b
Vehicle Primary
energy to (%) efficiency
e
energy
energy (%) efficiency
carrier
a
(%)
BEV Coal today Electricity 40
c
93 61 23
BEV Natural gas future Electricity 55
d
93 61 31
HEV parallel Crude oil Diesel 95.3 99.8 30 28
HEV series Crude oil Diesel 95.3 99.8 29 28
FCEV Natural gas Hydrogen (350 bar) 85 86 34 25
FCEV Natural gas Methanol 72 99.6 29 21
ICEV developed Crude oil Petrol 91.5 99.8 20 19
ICEV today Crude oil Petrol 91.5 99.8 14 13
a
Adopted from [42,43].
b
Adopted from [42,43].
c
Relates to Danish coal power with an efficiency ranging between 36% and 47%.
d
Adopted from [44].
e
From Table 5.
982 M. A
˚hman / Energy 26 (2001) 973–989
Table 7
Primary energy efficiency with biomass as primary energy source
Vehicle Primary energy Energy carrier Primary Distribution/ Vehicle Primary
energy to storage
c
(%) efficiency
f
energy
energy (%) efficiency
carrier
a,b
(%)
(%)
BEV Biomass Electricity 45
c
93 61 25
HEV parallel Biomass Methanol 63 99.6 30 19
HEV series Biomass Methanol 63 99.6 29 18
FCEV Biomass Hydrogen (350 bar) 69 86 34 20
FCEV Biomass Methanol 63 99.6 29 18
ICEV developed Biomass Methanol 63 99.6 22
d
14
ICEV today Biomass Methanol 63 99.6 15
e
10
a
Adopted from [42,43].
b
Adopted from [45].
c
Adopted from [42,43].
d
Efficiency is assumed to be 10% higher than for petrol. Source: [46].
e
Efficiency is assumed to be 6% higher than for petrol. Source: [46].
f
From Table 5.
or hydro power are given in Table 8. In this case, only BEVs and FCEVs were considered.
Hydrogen for the fuel cell is produced through the electrolysis of water. The efficiency of produc-
ing primary electricity is set to 100% for all the alternatives.
There is a substantial energy loss when converting electricity to hydrogen and back to electricity
again in the FCEV. For this reason, the most efficient alternative would be to use the electricity
directly in a BEV. Hydrogen is, however, easier to store than electricity and in a long-term
scenario, using solar power, hydrogen might be the safest and most practical energy carrier.
Table 8
Primary energy efficiency with solar, wind or hydropower as primary energy source
Vehicle Primary energy Energy carrier Primary Distribution/ Vehicle Primary
energy to storage (%) efficiency
d
energy
energy (%) efficiency
carrier (%) (%)
Solar, wind or
BEV Electricity 100 93
a
61 57
hydropower
Solar, wind or
FCEV Hydrogen (350 bar) 90
c
86
b
34 26
hydropower
a
Adopted from [43].
b
Adopted from [42].
c
Adopted from Ogden and Nitsch [47] who assumed that the efficiency today (1994) of 70–75% can be increased
to 85–90% in the future.
d
From Table 5.
983M. A
˚hman / Energy 26 (2001) 973–989
4.1. Sensitivity analysis
The results presented above are based on a number of assumptions regarding the future energy
efficiency of components in the powertrain and the possibility to integrate them with continued
high mean efficiency over a normal drive schedule. This makes the conclusions sensitive to uncer-
tainties regarding the successful development of components and control techniques for the elec-
tric drivetrain. The most speculative component assumptions that influence the future potential
are associated with batteries and fuel cells, but assumptions regarding the development of ICEs
could also be uncertain. The possibility for the electric drivetrain to handle load variations with
the assumed high mean efficiency is also a crucial assumption. Different advances in development
regarding the efficiencies of the most important and uncertain components are given as best,
assumed, or worst cases in Table 9.
The effects on primary efficiencies, calculated with the highest and lowest component
efficiencies in Table 9 for each powertrain, are shown in Fig. 2. The results in the figure are
based on primary energy from biomass.
For the conventional powertrain, no further development of the current status is assumed in
the worst case. The best possible development for an ICEV is equal to the assumed potential
given in Table 8, since no further improvement beyond this level seems likely today. For all
alternative powertrains, the development of a highly efficient electric drivetrain is crucial. The
best possible development for the BEV is based on an optimistic potential for a 95% efficient
lithium–polymer battery. For the HEV, the development of higher efficiency for small diesel
engines is important, and for the FCEV, of course, the development of the fuel cell itself.
The conclusions that can be drawn from Fig. 2 are that the uncertainty in component develop-
ment will not change the fact that ICEVs have the lowest potential for primary energy efficiency
and BEVs have the highest potential, regardless of the assumptions made concerning component
development as long as they are reasonable. The difference between alternative and conventional
powertrains could, however, be smaller if the electric drivetrain proves less efficient than assumed.
The relationship between HEVs and FCEVs could also change with different assumptions. All
Table 9
Different advances in development regarding energy efficiency of powertrain components
Uncertainty about future efficiency Best possible Assumed Worst possible
efficiency (%) efficiency (%) efficiency (%)
Electric drivetrain
a
BEV 76 65 57
HEV and FCEV 76 72 64
Fuel cells
b
48 47 42
ICEs
c
42 40 35
a
Best possible development includes a lithium–polymer battery. The “assumed and worst possible mean efficiencies”
are both for BEV (65 and 57%) and for a HEV (72 and 64%).
b
The assumed fuel cell efficiency is optimistic, but possible in the long term. Not much potential above that assumed
is considered possible.
c
High efficiency ICEs are available today. Incremental improvements can be made upon the assumed efficiency,
but not much more.
984 M. A
˚hman / Energy 26 (2001) 973–989
Fig. 2. The effects of uncertainty in component development on primary energy efficiency. Calculations were made
assuming the primary energy to be based on biomass.
the selected components can develop independently, but links exist between components used in
the lithium–polymer batteries and in the PEM fuel-cell system. Development of the conventional
powertrain will have positive effects on the potential of the HEV, since the ICE is used in both
hybrid and conventional powertrains.
5. Other means of reducing energy use in future vehicles
Reducing weight, rolling resistance, and aerodynamic drag can also play a major part in improv-
ing the energy efficiency of passenger cars. The effects of such measures are briefly discussed
here. The importance of different measures in reducing energy use is given in Table 10. The
division between the three different road loads is assumed to be one third each of the total road
load, based on Amann [11] and DeCicco and Ross [16].
Apart from vehicle technology, energy use also depends strongly on vehicle speed. Rolling
resistance is proportional to speed, the energy invested in kinetic energy is proportional to the
square of the speed, and the aerodynamic drag is proportional to the cube of the speed. This
means that a reduction in speed of 1% would give a theoretical reduction in driving energy (road
load) of 1.98% (see Table 10).
Table 10
Changes in total energy use for vehicles as a result of changes in speed, weight, rolling resistance, and aerodynamic drag
Effect of road-load factors Speed. 1% Weight. 1% Rolling coefficient. Drag coefficient. 1%
on total energy use reduction gives: reduction gives: 1% reduction gives: reduction gives:
Rolling resistance 1% reduction 1% reduction 1% reduction No reduction
Aerodynamic drag 2.97% reduction No reduction No reduction 1% reduction
Kinetic energy (braking) 1.98% reduction 1% reduction No reduction No reduction
Total road load (theoretical) 1.98% reduction 0.66% reduction 0.33% reduction 0.33% reduction
985M. A
˚hman / Energy 26 (2001) 973–989
Considering the medium-term scope for improvements in road load factors, the total road load
can be reduced by 36% for future ICEVs compared with standard ICEVs today (see Table 11).
For electric drivetrains the scope is less, 22% for a BEV and 29% for a HEV, due to the heavy
batteries. Note that the reductions in road load are assumed to be introduced while preserving
both the comfort and performance of the vehicle. Much larger reductions in road load might be
possible in the longer term, see, for example [48].
6. Total potential reductions in primary energy use
Based on the road–load reductions given in Table 11, and the primary energy efficiency
improvement given in Table 7, the total potential in percent, to reduce primary energy use can
be calculated according to Eq. (1):
⌬W
T
⫽(100⫺⌬W
RL
)⫻h
c
/h
n
(1)
where ⌬W
RL
(%) is the potential for reducing the road load (see Table 11). h
c
is the primary
energy efficiency for conventional vehicles today, and h
n
is the primary energy efficiency for
vehicles with alternative powertrains (see Table 7).
A combination of improved powertrain efficiency and reduced vehicle road load would enable
future BEVs to run on only 38% of the fossil primary energy of a conventional vehicle today.
Table 11
The road-load factors today and the medium-term scope for improvements
Energy use Today Medium-term Potential change Impact of potential
potential change on road load
Weight 1400 kg
BEV 1300 kg
a
⫺7% ⫺5%
FCEV 1200 kg
b
⫺14% ⫺9%
HEV 1150 kg
c
⫺18% ⫺12%
ICEV 1000 kg
d
⫺29% ⫺19%
Rolling resistance 0.011 0.008
e
⫺27% ⫺8.9%
Aerodynamic drag 0.4 0.3
f
⫺25% ⫺8.25%
Total reduction of road load
BEV ⫺22%
FCEV ⫺26%
HEV ⫺29%
ICEV ⫺36%
a
Assuming a general weight reduction to 1000 kg+300 kg battery. Adopted from [18,30].
b
Assuming a general weight reduction to 1000 kg+200 kg extra for fuel cell and battery. Adopted from [18,30].
c
Assuming a general weight reduction to 1000 kg+150 kg extra battery. Adopted from [18,30].
d
Assuming a general weight reduction to 1000 kg. Adopted from [30].
e
A reduction down to 0.0085–0.0065 is possible, according to [12,16]. The Partnership for a New Generation of
Vehicles (PNGV) hopes to reduce rolling resistance by 20% compared with a standard car today [11].
f
A reduction of aerodynamic drag affects the appearance of the vehicle, hence the scope of improvement is limited.
PNGV hopes to reduce aerodynamic drag by 20% compared with a standard car today [11].
986 M. A
˚hman / Energy 26 (2001) 973–989
Future HEVs would require 37%, FCEVs 40%, and the developed ICEV 45% of the primary
energy consumed compared with an ICEV today.
The PNGV aims at developing a future vehicle that uses only 34%
4
of the fuel consumed in
a standard car today. In this study, the total potential to reduce vehicle fuel consumption was
calculated for the HEV and FCEV fuelled with petrol or methanol [Eq. (2)].
⌬W
F
⫽(100⫺⌬W
RL
)⫻h
c2
/h
n2
, (2)
where ⌬W
F
is the total reduced vehicle fuel consumption and the vehicle efficiency of an ICEV
today (h
c2
) and the vehicle efficiency of a HEV or FCEV (h
n2
) are adopted from Table 5. The
results show that fuel consumption will decrease to 33–34% of that of a conventional ICEV today.
The road load reductions in Table 11 and the powertrain technologies assumed in Eq. (2) are
comparable to those targeted by the PNGV. Despite the uncertainties, it is reasonable to conclude
that it will be possible to reach the goal set by the PNGV without changing the size or performance
of the vehicle. The development of new powertrains requires, however, dedicated efforts to
improve efficiency in order to attain the potential anticipated in this study.
7. Discussion
This study shows that there is substantial potential for improving the primary energy efficiency
for vehicles using alternative powertrains. However, considerable improvements in performance
and reductions in cost are necessary to make the alternative powertrains commercially competitive.
The common denominator for the alternative powertrains is the electric drivetrain. This paper
assumes that a highly developed and efficient electric drivetrain can be attained in the stated time
frame, 10–20 years. Most likely, the optimistic efficiencies assumed in this paper will not be
attained for a long time. Other performances, such as cost, will be prioritised first in development.
No battery today meets the USABC and PNGV performance targets for BEVs regarding specific
energy, specific power, or cost, which are necessary for commercialisation. The BEV is totally
dependent on the development of batteries for successful introduction on to the market. With the
NiMH battery, an energy efficiency of 81% is possible today, but the potential specific energy
for the NiMH battery is too low (80–100 Wh/kg) to be a long-term option to the BEV. The
reported potential of 95% efficiency and a specific energy of 150 Wh/kg for lithium–polymer
batteries is difficult to validate. According to Kuller [28], at the leading lithium–polymer battery
manufacturer (3M/Hydro-Quebec), they have prototype batteries today that meet these targets, see
also [19]. The development of lithium–polymer batteries with the stated performance is probably
necessary to make the BEV competitive. It has been argued that the lithium–polymer battery
suffers from inherently low specific power, and is thus not an alternative for HEVs [18], but
recently, lithium–polymer batteries with high specific power have been demonstrated and might
be a future alternative to HEVs as well [19].
The technology providing the assumed maximum efficiency of 43% for ICEs exists today. A
4
The PNGV (Partnership of a New Generation of Vehicles) aims at increasing the fuel economy from 27.5 mpg today up to 80
mpg [30].
987M. A
˚hman / Energy 26 (2001) 973–989
potential problem facing the HEV with a heat engine is the emissions of NO
X
, hydro carbons
and particulate matter. A higher compression ratio, and thus higher efficiency, also means a risk
of increased emissions of small particles [49]. New engines will have to be characterised by both
low emission and high efficiency. The HEV is not dependent on battery development to the same
degree as the BEV. HEVs already exist on the market. If the targets for specific power or cost
are not be met, it is possible to compromise and sacrifice some of the potential efficiency improve-
ment by allowing the ICE to suffer greater variation in load. This lowers the mean efficiency,
but at the same time also the demands on specific power and energy capacity on the battery.
The time perspective for the introduction of FCEVs is longer than for HEVs. Series production
of the current FCEV will begin, at the earliest, in 2004 [50], but series production of fully
developed fuel cells will probably have to wait until 2010. The fuel cell still requires much
development in order to bring costs down to a competitive level and to attain the assumed
maximum efficiency of 55% used in our calculations and requires to much platinum in the catalyst
in order to reach the long-term goal of a cost of 47 US$/kW set by PNGV [30]. The assumptions
in this paper of 55% maximum efficiency and 47% mean efficiency are high, but possible in a
10 year perspective, see, for example [30,38,39,51]. However, 55% maximum efficiency requires
that the fuel-cell system be optimised. Apart from the fuel cell stack, the system also includes
compressors, pumps, etc. [51]. A system based on hydrogen requires a new refuelling infrastruc-
ture and complicated fuel storage on board the vehicle. So far, car manufacturers prefer to reform
methanol on board the vehicle instead, see conclusions in [40]. This does not require a totally
new infrastructure, but lowers the energy efficiency and increases the vehicle weight.
8. Conclusions
This comparative assessment shows that there is potential to double the primary energy
efficiency using electric drivetrains in vehicles, such as BEVs, HEVs or FCEVs, compared with
present ICEVs. All vehicles with an alternative powertrain have the potential for higher primary
energy efficiency than vehicles with an improved conventional powertrain.
BEVs seems to have a small advantage over HEVs and FCEVs from a primary energy efficiency
point of view, but given the uncertainties, all alternative powertrains have approximately the same
primary energy efficiency when supplied with fossil or biomass fuels. No outright winner amongst
the alternative powertrains could be identified from a primary energy efficiency point of view. If
the energy carrier is renewable primary electricity, e.g. in a solar photovoltaics future, the BEV
will have a higher primary energy efficiency than the FCEV using hydrogen from electrolysis.
Acknowledgements
This work was sponsored by the Swedish Transport and Communications Research Board
(KFB). Valuable comments and criticism were made by Dr Bengt Johansson and Professor Lars
J. Nilsson, both at the Department of Environmental and Energy Systems Studies at Lund Univer-
sity, and three anonymous reviewers.
988 M. A
˚hman / Energy 26 (2001) 973–989
References
[1] WEC. Global transport and energy development. The scope of change. London: World Energy Council, 1998.
[2] IEA/OECD. World energy outlook. Paris: IEA/OECD, 1998.
[3] Schipper L, Marie-Lilliu C. Carbon dioxide emissions from IEA countries. Stockholm: KFB, 1999.
[4] IEA-AFIS. Automotive fuels for the future—the search for alternatives. Paris: OECD/IEA, 1999.
[5] Johansson B. Will Swedish biomass be sufficient for future transportation-fuel demands? Energy
1996;21(12):1059–69.
[6] Ecotraffic. The life of fuels. Stockholm, 1992.
[7] Metz N, Sporckmann B. Comparison of the energy consumption and emissions of future drive trains for passenger
cars at the year 2000. In: Electric Vehicle Symposium 14, Orlando (FL), 1997.
[8] Wang Q, Deluchi MA. Impacts of electric vehicles on primary energy consumption and petroleum displacement.
Energy 1992;17(4):351–66.
[9] OECD/IEA. Electric vehicles: technology, performance and potential. Paris: OECD, 1993.
[10] Wabro R, Wagner U. Energieverbrauchsanalyse von elektro-Pkw-stand der technik und ausblick. ATZ Automobil-
technische Zeitshcrift 1998;100(2).
[11] Amann CH. The stretch for better passenger car fuel economy: a critical look Part 1 and 2. Automotive Engineering
International 1998:305–13, 71–6.
[12] Cuddy MR, Wipke KB. Analysis of the fuel economy benefit of drivetrain hybridization. Society of Automotive
Engineers 1997;101–9.
[13] Wang M et al. Total energy-cycle energy and emissions impacts of hybrid electric vehicles. In: Electric Vehicle
Symposium 14, Orlando (FL), 1997.
[14] Lipman TE, DeLucchi MA. Hydrogen fueled vehicles. International Journal of Vehicle Design 1996;17:562–89.
[15] Reuyl JS, Schuurmans PJ. Policy implications of hybrid-electric vehicles. Final report to NREL, NEVCOR Inc.,
Stanford (CA), 1996.
[16] DeCicco J, Ross M. An updated assessment of the near-term potential for improving automotive fuel economy.
Washington (DC): American Council for an Energy Efficient Economy (ACEEE), 1993.
[17] NRC. Effectiveness of the United States advanced battery consortium as a government–industry partnership. Wash-
ington (DC): National Research Council, 1998.
[18] Rand DAJ, Woods R, Dell RM. Batteries for electric vehicles, vol. 1. Exeter: Research Studies Press, 1998.
[19] St-Pierre C et al. Lithium polymer battery for electric vehicle and hybrid vehicle applications. In: Electric Vehicle
Symposium 16, Beijing (China), 1999.
[20] Lipman T. The cost of manufacturing electric vehicle batteries. Davis: ITS, University of California, 1999.
[21] Kopf C et al. Development of a multifunctional high power system: meeting the demands of both a generator
and traction drive system. In: Electric Vehicle Symposium 14, Orlando (FL), 1997.
[22] Lutz JF, Patel BR. Performance comparison of a 100 kW brush less PM and induction motor system. In: Electric
Vehicle Symposium 14, Orlando (FL), 1997.
[23] Sato Y et al. New control methods for high performance PM motor drive system. In: Electric Vehicle Symposium
14, Orlando (FL), 1997.
[24] STOSEB. Framtidens bila—En energi-och miljo
¨studie. STOSEB: Stockholm, 1994.
[25] BRUSA, 20/4 1999 http://www.rol.ch/BRUSA/fenglish/frame.htm
[26] Corrigan DA et al. Higher power ovonic nickel–metal hydride batteries for electric and hybrid vehicles. In: Electric
Vehicle Symposium 14, Orlando (FL), 1997.
[27] Georen P. Personal communication, Department of Applied Electrochemistry, KTH (Sweden), 1999.
[28] Kuller D. Personal communication, 3M/Hydro-Quebec (Canada), 1999.
[29] Sjo
¨blom B, Rehn J, Volvo Aero Corporation Sweden. The Volvo high speed generation hybrid drive and associated
combustion system. Energy Conversion & Management 1997;38(1013):1225–35.
[30] NRC. Review of the partnership for a new generation of vehicles, Fourth report. Washington (DC): National
Research Council, 1998. p. 122.
[31] Peschka W. The status of handling and storage techniques for liquid hydrogen in motor vehicles. International
Journal of Hydrogen Energy 1987;12(11):753–64.
989M. A
˚hman / Energy 26 (2001) 973–989
[32] Shelef M, Kukkonen CA. Prospects of hydrogen-fueled vehicles. Progress Energy Combustion Sciences
1994;20:139–48.
[33] James BD et al. Onboard hydrogen storage in fuel cell vehicles. In: Electric Vehicle Symposium 14, Orlando
(FL), 1997.
[34] DeLuchi MA, Ogden JM. Solar-hydrogen fuel-cell vehicles. Transportation Research 1993;27A(3):255–75.
[35] Ogden JM et al. Hydrogen energy systems studies. Princeton, NJ: Center for Energy and Environmental Studies,
Princeton University, 1995.
[36] Cleghorn SJC et al. PEM fuel cells for transportation and stationary power generation applications. International
Journal of Hydrogen Energy 1997;22(12):1137–44.
[37] McNicol BD, Rand DAJ, Williams KR. Direct methanol–air fuel cells for road transportation. Journal of Power
Sources 1999;83:15–31.
[38] Kordesh K, Simader G. Fuel cells and their applications. Weinheim: VCH, 1996.
[39] Borroni-Bird CE. Fuel cell commercialisation issues for light-duty vehicle applications. Journal of Power Sources
1996;61:33–48.
[40] Kalhammer FR et al. Status and prospects of fuel cells as automobile engines. Sacramento (CA): State of California
Air Resources Board (CARB), 1998.
[41] IPCC. Climate change 1995—impacts, adaptations and mitigation of climate change: scientific-technical analyses.
New York: Cambridge University Press, 1996.
[42] Katofsky RE. The production of fluid fuels from biomass. Princeton (NJ): Center for Energy and Environmental
Studies, 1993.
[43] Blinge M et al. Livscykelanalys (LCA) av drivmedel Bilagor. Stockholm: KFB, 1997.
[44] Energikomissionen, Omsta
¨llning av energisystemet. Energikommisionen: Stockholm, 1995.
[45] Gustavsson L. et al., Reducing CO
2
-emissions by substituting biomass for fossil fuels. Energy 1995;20:1097–113.
[46] Olsson LO. Fordonskostnadens pa
˚verkan vid anva
¨ndande av biodrivmedel. Stockholm: KFB, 1996.
[47] Ogden JM, Nitsch J. Solar hydrogen. In: Johansson TB et al, editor. Renewable energy. Washington (DC): Island
Press, 1994.
[48] Lovins AB. Hypercars: the next industrial revolution. In: Electric Vehicle Symposium 13, Osaka (Japan), 1996.
[49] Egnell R. Personal communication, Department of Combustion Engines, LTH (Sweden), 1999.
[50] Daimler–Chrysler. http://www3.daimlerchrysler.com, Date:17/5, 1999.
[51] Johansson K. Personal communication, Department of Chemistry, KTH (Sweden), 1999.
[52] Weiss MA, Heywood JB, Drake EM, Schafer A, AuYeung FF. On the road in 2020—a life-cycle analysis of
new automobile technology. Energy Laboratory Report #MIT EL 00-003. Cambridge (MA): MIT, 2000
