MIT Biofuel Report

Prospects for Bi-Fuel and
Flex-Fuel Light-Duty Vehicles
An MIT Energy Initiative

Symposium
April 19, 2012
MIT Energy Initiative Symposium on Prospects for Bi-Fuel and Flex-Fuel Light-Duty Vehicles | April 19, 2012
Prospects for Bi-Fuel and

Flex-Fuel Light-Duty Vehicles
An MIT Energy Initiative

Symposium
April 19, 2012
about the report

Summary for Policy Makers
The April 19, 2012, MIT Energy Initiative Symposium addressed Prospects for Bi-Fuel and Flex-Fuel
Light-Duty Vehicles. The symposium focused on natural gas, biofuels, and motor gasoline as fuels
for light-duty vehicles (LDVs) with a time horizon of the next two to three decades. The important
transportation alternatives of electric and hybrid vehicles (this was the subject of the 2010 MITEi
Symposium1) and hydrogen/fuel-cell vehicles, a longer-term alternative, were not considered.
There are three motivations for examining alternative transportation fuels for LDVs: (1) lower life
cycle cost of transportation for the consumer, (2) reduction in the greenhouse gas (GHG) footprint of
the transportation sector (an important contributor to total US GHG emissions), and (3) improved
energy security resulting from greater use of domestic fuels and reduced liquid fuel imports. An
underlying question is whether a flex-fuel/bi-fuel mandate for new LDVs would drive development
of a robust alternative fuels market and infrastructure versus alternative fuel use requirements.
Symposium participants agreed on these motivations. However, in this symposium in contrast
to past symposiums, there was a striking lack of agreement about the direction to which the
market might evolve, about the most promising technologies, and about desirable government
action. This lack of consensus is itself instructive, pointing to the paucity of data in the public
domain about vehicle performance and emissions for a spectrum of fuel alternatives.
Despite this lack of consensus, there was agreement on two overarching realities. First, the
introduction of new technology will be greatly facilitated by backward compatibility with the
existing fuel and vehicle infrastructure. Second, the key factor is the extent that the stock of LDVs
on the road is compatible with a range of fuels.
The sessions dealt with different aspects of the engine/fuel system.
Prospects for bi-fuel vehicles The dramatic difference in the US of oil and natural gas prices
on an equal energy basis is a strong incentive for examining alternatives to the motor gasoline
internal combustion engine (ICE) (in todays configuration and with future improvements, e.g.,
turbocharging). Participants discussed a wide range of different vehicle/fuel combinations. Here
is a sample list:
Methanol
fueled LDVs
Compressed
Bi-fuel
Natural Gas (CNG) LDVs
natural gas and motor gasoline vehicles
Ethanol-gasoline
Tri-flex
flex fuel vehicles (FFVs) in range of E-15 to E-85
fuel vehicles fueled by gasoline/ethanol/methanol (GEM) mixtures.
For each of these alternatives there was considerable discussion (but little agreement) about
drivability, emissions, cost, and maintainability.
One session of the symposium was devoted to consumer choice as expressed in various countries,
e.g., ethanol in Brazil, Germany, and the United States, bio-diesel and propane-fueled vehicles in
Germany, and CNG in Asia and Europe. Unfortunately, the symposium did not hear an assessment
about how useful vehicle choice models, widely used by industry and government agencies,
were for projecting vehicle fleet composition among a wide range of vehicle/fuel options under
a variety of policy assumptions.
Some interesting tentative conclusions were inferred from the discussion offered here, but it is
important to stress that these conclusions are tentative and require both technical development
and verification from field experience.
1. It is possible to manufacture LDVs capable of operating on a wide range of GEM mixtures,
with a cost penalty under $1,000 per vehicle, possibly substantially less.
2. T
hese LDV engine/fuel combinations may comply with prevailing and anticipated air
emission standards over the wide range of expected driving conditions, e.g., start-stop,
summer-winter.
3. T
here is insufficient field experience with many LDV engine/fuel combinations that
are proposed.
4. A
ttention must be given to fuel efficiency, on-board electronic fuel controls, and
super charging to optimize vehicle performance with respect to emissions, cost, and
consumer satisfaction.
The lack of agreement about the most promising fuel/vehicle combinations from the point of view
of cost, sustainability, and environmental effects meant that no clear direction emerged for desirable
federal or governmental policy. Individual participants did propose a wide range of mechanisms:
fuel economy standards, vehicle fuel flexibility standards, and various mechanisms to subsidize
rapid deployment of fueling infrastructure. However, there also was recognition that the government
was neither in a position to select one preferred alternative vehicle direction nor to provide the
funds to subsidize multiple approaches.
There certainly are research, development, and deployment (R,D, & D) programs in alternative
fuel vehicles (AFVs) that deserve federal support although this topic was not sufficiently
addressed to either endorse present programs or offer many concrete suggestions. There was
general sympathy with the view that there was little field data for small fleets of AFVs and that a
suite of properly instrumented small fleet (less than 100 vehicles) demonstration projects could
yield valuable information about the cost, practicality, and consumer satisfaction of the tech
nology alternatives.
John Deutch
Institute Professor, MIT
MITEI Associates Program/Symposium Series

The MITEI Associates Program/Symposium Series is designed to bring together groups of energy experts
to examine, analyze, and report on critical and timely energy policy/technology issues with implications for
near-term actions. The centerpiece of the program is a one-day symposium in which invited experts, under
Chatham house rule, discuss the selected topic. Topical white papers, which are sent to the participants in
advance, are commissioned to focus and inform the discussion. The information from these white papers is
supplemented by work from graduate students, who generate data and provide background information.
Potential symposium topics are solicited from MITEI members and are provided to the Steering Committee
for consideration. Four MITEI Associate members Cummins, Entergy, Exelon, and Hess support the
program with a two-year commitment and serve on the Steering Committee.
After each symposium, a report is prepared and published, detailing the proceedings to include a range of
findings and a list of recommendations. Two students are assigned to each session. They serve as rapporteurs for the symposium and focus their masters theses on topics identified from the symposium.
This report is the fifth in the series, following Retrofitting of Coal-Fired Power Plants for CO2 Emissions
Reductions, Electrification of the Transportation System, The Role of Enhanced Oil Recovery in Accelerating
the Deployment of Carbon Capture and Sequestration, and Managing the Large-Scale Penetration of
Intermittent Renewables.
These reports are available electronically on the MITEI web site at http://web.mit.edu/mitei/research/
energy-studies.html. If you would like to receive a hard copy of one or more of the reports, please send an
email with your requested titles and quantities and your mailing address to askmitei@mit.edu.
MITEI extends its appreciation to these sponsors of the Symposium Series.
contents
2
About The Report / SUMMARY FOR Policy makerS
7 Section 1 OVERVIEW AND SUMMARY

19 Section 2 VEHICLE TECHNOLOGY AND MANUFACTURING PERSPECTIVES
Overview

Dedicated Mono-Fuel Vehicles

Bi-Fuel Vehicles

Flex-Fuel Vehicles

Vehicle Technology Modifications

Economics of FFVs

FFV/Fuel Combinations

Introduction of Methanol into FFVs

Environmental Performance of FFV/Fuel Combinations
35 Section 3 FUEL PRODUCTION AND DISTRIBUTION INFRASTRUCTURE PERSPECTIVE
Introduction
Gasoline
Ethanol

Natural Gas
Methanol

Pricing of Alternative Fuels Relative to Gasoline
55 Section 4 CONSUMER PERSPECTIVES

Modeling the Attributes of Consumer Behavior toward Bi-Fuel and Flex-Fuel Vehicles

Vehicle Performance

Backward Compatibility

Ease of Fueling

Cost Competitiveness

Lessons Learned from Alternative Fuels Experience in Other Countries

Summary of Alternative Fuels Experiences in Other Countries

Methanol Experience in China

Tri-Flex Fuel Vehicle Experience in Brazil: Use of CNG Instead of Methanol
65 Section 5 POLICY ISSUES

Historical Evolution of Alternative Fuels Legislation

Discussion of the Key Current Policies

Alternative Fuel Vehicle Credits in the Corporate Average Fuel Economy Program

Alternative Fuel Mandates

The Unsuccessful Effort to Incentivize the Introduction of Methanol Fuel

Are Alternative Fuels Needed in the US Transportation Fuel Market?

Possible Additional Areas for Government Intervention

Open Fuel Standard
Fuel Tax

Government Fleet Programs

Better Informing the Public

Federal R&D Support

CNG Refueling Programs

Vehicle and Fuel Certification Process
77 Endnotes
79
INDEX OF FIGURES
81
INDEX OF TABLES
82 Appendices

A. Glossary of Terms
B. Abbreviations / Acronyms

C. Comprehensive Summary of Vehicle-Fuel Options
D. Symposium Agenda
E. Symposium Participants
F. White Papers

1.The Evolving Fuels Context for Light-Duty Vehicles, Steven E. Koonin
2.Light-Duty Vehicles: A Discussion of the Evolving Engine, Vehicle,

and Fuel Requirements Context, John B. Heywood and Sun Jae, MIT
3.The Case for Bi-Fuel Natural Gas Vehicles, Michael D. Jackson, TIAX LLC
4.Evolution of Alcohol Fuel Blends towards a Sustainable Transport Energy Economy,

J.W.G. Turner and R.J. Pearson, Lotus Engineering, E. Dekker, bioMCN, B. Iosefa,
Methanol Institute, K. Johansson and K. ac Bergstrom, Saab Automobile
Powertrain AB
5.Prospects for Flexible- and Bi-Fuel Light-Duty Vehicles; Consumer Choice

and Public Attitudes, Ulrich Kramer and James E. Anderson, Ford Motor Company
6.Regulations and Incentives for Alternative Fuels and Vehicles
The MIT Energy Initiatives Symposium

on Prospects for Bi-Fuel and Flex-Fuel
Light-Duty Vehicles
S e c t i o n 1 O v e r v i e w a n d S u m m a r y
The 2012 MIT Energy Initiative symposium brought together experts and policy makers to discuss
prospects for alternative fuel technologies for LDVs. As its starting point, the symposium accepted
the proposition of a significant policy value in a greatly expanded alternative fuel market and
focused on the question of how it might be achieved.
The objective of the symposium was to examine alternative approaches that could lead to the
deployment of a large number of bi-fuel and flex-fuel vehicles. The symposium also examined
the relationships between AFV deployment and expansion of alternative fuel infrastructure
(the so-called chicken and egg problem). The symposium addressed this problem from the
vehicle perspective, assuming that adequate supplies of alternative fuels could be produced
to satisfy demand created by expansion of the AFV fleet. The scope of the symposium was
focused on AFV technologies using liquid fuels or natural gas. A previous MIT Energy Initiative
symposium examined prospects for electrification of the transportation sector.2
Rationale for the Symposium

Symposium organizers started from the premise that a greatly expanded alternative fuels market
would advance key US policy objectives. The rationale for organizing the symposium was built on
four fundamental premises:
1.
Alternative fuel solutions for the transportation sector are key to addressing major
policy challenges: The fuel mix for the LDV segment of the US transportation sector is
dominated by use of petroleum-based fuels; gasoline consumption accounted for 51% (by
volume) of total US petroleum demand in 2011.3 The scale of current petroleum usage in LDVs
poses significant challenges to national policy objectives.
2.
Energy and economic security: Nearly 8% of average household income is spent on
gasoline, and price volatility in gasoline markets is a recurring issue with policy makers.
Although physical quantities of petroleum imports have decreased from a peak in 2005, the
dollar value has increased. The United States spent $335 billion on foreign oil in 2011, an
increase of 84% from 2005 (the year of peak oil imports).4 The US oil import bill is estimated
to account for over half of the net trade deficit.5
3.
National security: Regardless of the source of physical supply, the price of petroleum is
set globally. Currently, 49% of US petroleum imports are from the Western Hemisphere, while
the remainder arrives via longer and less secure routes.6 The price of gasoline in the United
States tracks the price of crude oil, and crude oil prices are set in the global market not in the
United States. 79% of global conventional oil reserves are controlled by the Organization of
the Petroleum Exporting Countries (OPEC) cartel, which seeks to set global prices through
its control of production levels to maximize the revenue of its member countries. At $100 per
barrel, the value of OPEC-proved reserves is more than double the market capitalization of
all the worlds publicly traded companies combined.7 Global petroleum resources are concentrated in countries that may wish to leverage this petroleum-derived wealth in ways that
could constrain and limit US foreign policy options.
4.
Climate change: Gasoline-powered LDVs comprise nearly one-third of total net US GHG
emissions.8 To be effective, any climate policy will need to include measures to reduce GHG
emissions from this sector. However, reliance on policies that place a price on a carbon may
be insufficient to induce a shift in demand to lower carbon alternative fuels. For example,
a carbon price of $40 ton CO2 could result in significant changes in energy use in the electric
power sector, but it corresponds to only a $0.35 per gallon increase in the price of gasoline.9
Addressing these challenges is further complicated by the massive scale and tightly regulated
nature of the current LDV market and its associated petroleum-based fueling infrastructure.
Automobile manufacturing and petroleum refiners report that they account for about 10% of
US Gross Domestic Product (GDP) and 17 million jobs.10 Any changes in the outputs from these
sectors will require significant lead times, large-scale capital investment, and rigorous regulatory
reviews.

Achieving
a greatly expanded alternative fuels market requires significant levels

of AFV deployment in the mass market of LDVs: Addressing the major policy challenges
requires an informed discussion of the feasibility of alternative fuels in the mass market for
LDVs. The past three decades have seen growth in the number of AFVs primarily in the heavyduty vehicle (trucks and buses) and LDV fleet markets. This growth has been market driven,
assisted with modest federal incentives (both financial incentive and mandates on fleets). The
dramatic expansion of ethanol supply in LDVs has been driven by federal mandates that led to
the blending of ethanol with gasoline in low-level blends (current 10% ethanol, or E10).
In recent years, penetration of FFVs in the LDV market has been growing steadily. As of late
2011, over 9 million FFVs (ethanol (E85) and gasoline) were registered in the United States
(approximately 4% of all LDVs). New registrations of FFVs are at a pace of about 1 million/
year (yr), comprising about 17% of total new registrations annually.11 Energy Information
Administration (EIA) estimates, however, that only 0.6 million of these FFVs actually operate
on E85.12
While FFVs currently comprise about 4% of the LDV fleet, consumption of E85 in 2010 was 90
million gallons of gasoline equivalent (GGE), or about 1% of total gasoline consumption.13
Excluding ethanol in E10, total alternative fuel composition in 2010 is estimated at 457.8 million
gallons, or only 0.3% of the total gasoline market (excluding diesel and biodiesel). Figure 1
illustrates the composition of alternative fuels consumed in the United States in 2010, with
Figure 1 Consumption of Alternative Fuels by Fuel Type, 2010

(Excludes E10)
Electric [2]
1%
Liquefied
Natural Gas
(LNG)
6%
85% Ethanol
(E85)[1]
20%
Liquefied
Petroleum
Gas (LPG)
27%
Compressed
Natural Gas
(CNG)
46%
Methanol
(M85)
0%
Neat
Methanol
0%
Ethanol
(E85)[1]
0%
Hydrogen
0%
Total Consumption:
457.8 million GGE
Notes: [1] Excludes ethanol blended with gasoline (E10)

[2] E
xcludes electricity generated internally by hybrid
electric vehicles
Source: EIA Annual Energy Review. Available online at
http://www.eia.gov/renewable/afv/index.cfm
natural gas accounting for half

of consumption (on a GGE
basis). This figure addresses
alternative fuel use in FFVs
and AFVs. The total of nearly
0.5 billion GGE is significantly
smaller than the nearly 13 billion
gallons of ethanol currently
blended with gasoline for
consumption in conventional
vehicles with spark-ignition
engines.
he symposium was focused
T
on the LDV mass market, recognizing that there has been less
activity as well as greater challenges than in LDV fleets or in
medium and heavy-duty vehicle
markets. Light-duty vehicles
constitute 95% of total road
vehicles and they account for
64% of the total liquid fuel
consumption.

Increased
domestic natural gas resources create new market opportunities:

The development of domestic unconventional natural gas resources creates opportunities to
expand the role of natural gas in the transportation sector. The US natural gas resource base
has been estimated to support about 90 years of natural gas supply at current production
rates. The MIT Future of Natural Gas study estimated that a considerable portion of the shale
resource base can be produced economically at prices between $4/thousand cubic feet (mcf)
and $8/mcf.14 Discussion of AFV technologies was motivated in part by the desire to explore
vehicle technologies that could effectively utilize natural gas, either in the form of CNG or in
the form of liquid fuels derived from natural gas.
Framing the Scope of the Symposium

Several key questions were posed to frame the symposium discussion. These include:
1. W
hat are the most promising AFV technologies for use of liquid and gaseous fuels
(excluding electricity)?
2. How do the technology, performance, cost, and other characteristics of various alternative vehicle/fuel combinations compare with conventional gasoline-powered LDVs?
3. What are the perspectives of vehicle manufacturers, fuel distributors, consumers, and
policy makers regarding the difference in characteristics of alternative vehicle/fuel
combinations?
4. What are the barriers posed by the existing fueling infrastructure, and how can these
barriers best be overcome?
5. How will the different characteristics of AFVs (including fueling process) affect consumer preferences, and are the benefits of AFVs sufficient to stimulate consumer
demand?
6. Do the societal benefits of a robust alternative fuel LDV market justify government
policy intervention to stimulate deployment?
7. Is there a need for an alternative fuels program, or should the United States focus its
policy efforts solely on increasing automotive fuel economy?
Establishing a Conceptual Framework of Alternative LDVs and Fuels

Consideration of AFVs and fuels involves a complex set of interrelationships among governmental policies, vehicle manufacturing, the fuel supply, and consumer acceptance. Adopting the policy
challenges as the starting point, Figure 2 illustrates how the conceptual framework of vehicle technology options, fuel type options, infrastructure considerations, and alternative policy instruments
interact. The framework identifies four major classes of LDVs: (1) conventional vehicles, (2)
dedicated AFVs, (3) bi-fuel vehicles, and (4) FFVs. The principal fueling alternatives were then
identified for each vehicle category. The discussion of vehicle technologies centered on the use
of spark-ignition engines and compression-ignition engines using diesel fuel; other propulsion
systems such as hydrogen-powered fuel cell vehicles were not considered. Prospects for electric
vehicles (EV) and hybrid-electric vehicles (HEV)were discussed in depth in a previous MITEI
symposium.15
Figure 2 also identifies that the various combinations of alternative fuels/AFVs are linked to the
alternative fuel supply base and the availability of processing, distribution, and refueling infrastructure, whether it be integrated with the existing petroleum-based infrastructure or parallel to it.
10
Figure 2 The Interaction of Policy, Technical, and Economic Constraints on Alternative Vehicle-Fuel
Adoption, Deployment, and Use
Challenge
Dependence on petroleum-based fuels in the transportation sector challenges energy, economic and national security
policy objectives, and contributes significant quanities of GHG emissions, affecting climate change
Policy Goals
1. Enhance energy and economic security by diversifying supplies, mitigating energy price volatility and reducing
foreign oil payments
2. Enhance national security by reducing dependence on oil imports
3. Reduce GHG emissions
Decouple
alternative fuel
prices from
gasoline
Create greater price competition between

alternative fuels and gasoline, including potential
for fuel price arbitrage
Dedicated
alternative fuel
vehicles
Bi-fuel vehicles
FFVs
Drop-in blended
alternative fuels
(E5-E15, M5)
CNG
CNG/
gasoline
Gasoline
Manufacture
more fuel
efficient
vehicles
No change
Expand alternative
vehicle production
Expand alternative
vehicle production
Fuel Infrastructure
Impacts
No change
Continued
expansion of
alternative fuel
production
Expand CNG and

electric refueling
infrastructure
Expand CNG and

electric refueling
infrastructure
Consumer Acceptance
Comparable vehicle performance, acceptable trade-offs in vehicle functionality, and cost competitiveness
Policy Instruments
Fuel
Economy
Standards
Economic Considerations
Reduce petroleum demand and

petroleum price pressure
Vehicle Options
Conventional vehicles
Fuel Options
Gasoline
Vehicle Infrastructure
Impacts
RFS measures to
encourage more
alternative fuels
options
Electric
Electric/
gasoline
E85
M85
Expand alternative
vehicle production
Develop or expand
alternative fuel production;
transport and distribution;
and refueling infrastructure
CAFE credits, Open Fuel Standard, RFS measures, financial incentives
Source: MITEI.
The purpose of this framework was to establish a starting point for the symposium discussion
from a comprehensive perspective of the LDV market. The framework is driven by policy and
economic considerations. A recurring theme throughout the symposium discussion was whether
the various options for alternative vehicles and fuels not only be cost effective but also whether
they would exert downward price pressure if oil prices rose relative to other fuel feedstock.
This issue of price coupling is discussed in detail later in the report. Within this comprehensive
framework, the symposium focused on two alternative vehicle categories bi-fuel vehicles and
FFVs and three alternative fuel options natural gas, ethanol, and methanol.16 While the
symposium did not delve into other options, such as EVs or HEVs, the discussion was cognizant
that gaseous and liquid fuel alternative vehicles ultimately would have to compete with electricbased options for consumer acceptance and market share.

Alternative
fuel vehicle categories: The framework identified the major AFV options,
which are further defined in Table 1.
11
Table 1 Types of Alternative Fuel Light-Duty AFVs17

Conventional fuel vehicle
Any vehicle engineered and designed to be operated using gasoline or a gasoline blend
containing ethanol or methanol that can be dropped-into the vehicle without need for
engine modifications.
Dedicated mono-fuel AFV
Any vehicle engineered and designed to be operated using a single source of alternative
fuel. This category includes battery electric vehicles (BEV) and dedicated natural gas
vehicles (NGV).
Bi-fuel vehicle
Any vehicle engineered and designed with two independent fuel systems, which can
be operated on either of the two fueling systems separately, but not in combined
operation simultaneously. This category includes gasoline/natural gas vehicles and
plug-in hybrid electric vehicles (PHEV).
FFV
Any vehicle engineered and designed to be operated on a single fueling system that
can accommodate mixtures of varying quantities of two or more liquid fuels that are
combusted together. This category includes vehicles that can operate on either gasoline
or E85 or gasoline and methanol (M85). This also includes vehicles with two liquid
fueling systems that can operate individually or simultaneously, employing up to three
liquid fuels (tri-flex fuel vehicles).
Source: MITEI.

Alternative
fuel options: In the introductory presentation, it was proposed that the

c ategories of AFVs be further categorized by their ability to use either drop-in or drop-out
alternative fuels. The distinction between the two was based on physical fungibility with
gasoline.
A drop-in fuel is one that can be blended with gasoline and used in conventional gasolinepowered vehicles without requiring vehicle modifications.18 The fuels compatibility with
the current gasoline distribution infrastructure is not a condition in defining the term. Thus,
for example, gasoline blended with ethanol or methanol up to a certain limit and used in
conventional spark-ignition gasoline vehicles without modification is considered a drop-in
fuel. For this purpose, the blending limit for ethanol with gasoline is 15% (E5-E15),19 and 5%
for methanol with gasoline (M5); gasoline blends with higher ethanol or methanol content
would require engine modifications.
By comparison, a drop-out fuel is one that either cannot be blended with gasoline, or if so,
requires modifications to the vehicle technology. These two types can be characterized as:
Physical drop-out fuel: a fuel that cannot be blended with gasoline, such as electricity
or CNG; and

Blendable drop-out fuel: a fuel that can be blended with gasoline but cannot readily be
used in a conventional gasoline-powered vehicle. Drop-out fuels include blends of ethanol
or methanol exceeding current Environmental Protection Agency (EPA) approved blending
limits (15% for ethanol, or 5% for methanol).
Some participants believed that the distinction between drop-in and drop-out fuels had
important implications for the pricing of alternative fuels as well as the question of whether
consumers would actually realize cost savings from alternative fuel purchases. These participants believed that consumers would realize the benefits of lower prices for alternative fuels
if the alternative were drop-out fuels, and not fully substitutable with gasoline. Others believed
that the distinction between drop-in and drop-out fuels was less important in determining
relative fuel prices as compared with the source of feedstock for the fuel. So, for example,
12
methanol produced from natural gas would be priced more favorably than gasoline regardless
of whether the methanol was used in a drop-in fuel (i.e., M5 blend) or a drop-out fuel (i.e., M85).
The issue of the degree of coupling between the price of alternative fuels and the price of gasoline was a recurring topic of discussion throughout the symposium, with no final consensus
position reached. The economics of price coupling are discussed in detail later in this report.
Key Issues Arising from the Symposium

The symposium discussion quickly revealed a more complex set of issues and a broad range of
views on the framing questions. Symposium participants developed a better understanding of
the key issues, but little consensus was achieved on specific conclusions and recommendations.
Participants agreed that the evolution of the large-scale flexible and bi-fuel LDV market required
large-scale quantities of competitively priced alternative fuels. But there was no consensus
among the participants on either the preferred alternative vehicle-fuel combination or the public
policy measures that would be most cost effective in effecting the transition. Some participants
questioned the fundamental need for alternative vehicles and fuels altogether and instead favored
fuel-efficiency standards as the preferred path to meeting the overarching policy challenges.

Vehicle/fuel
options: It became obvious at the beginning of the symposium that there was
no established taxonomy of alternative vehicle and fuel options. Working from the conceptual
framework in Figure 2, a detailed set of 11 possible combinations was developed. These are
described in Appendix B. While this provided a useful benchmark, most of the symposium
discussion centered on two principal combinations: bi-fuel vehicles capable of operation on
either gasoline or natural gas, and FFVs capable of operating on a wide range of blends of
gasoline, ethanol, or methanol.
The current price differential between natural gas and petroleum is exceptionally large by
historical standards. The historical rule-of-thumb price differential has been about 10-to-1
for a barrel of oil to 1,000 cubic feet (cf) of natural gas; at the time of the symposium, it was
over 20-to-1. This makes use of natural gas economically attractive in vehicles. This market
signal is currently incentivizing owners of heavy-duty vehicles in long-haul service to convert
diesel-powered trucks to natural gas.
What are the prospects for LDVs? Globally, almost 15 million LDVs have the capability to use
natural gas as a fuel. The majority of these vehicles have bi-fuel capability, allowing them to take
advantage of lower-cost natural gas where it is available. Bi-fuel vehicles powered by gasoline
and natural gas are similar to gasoline vehicles in engine design and capability with regard to
power, acceleration, and cruising speed. Due to the fuels gaseous nature and lower energy
content, however, NGVs require tank modifications that have different technical and economic
trade-offs. Symposium participants discussed the trade-offs involved in the choice of fuel
tank for CNG. For example, it was pointed out that the heavier but less expensive fuel tanks
(i.e., the Type I fuel tank) reduce overall driving range, fuel economy, and cargo capacity more
than the lighter but more expensive types (i.e., the Type IV fuel tank). Presenters noted that
while this trade-off was very important in dedicated mono-fueled NGVs, the issue of reduced
range was of less importance to consumers for the use of natural gas in a bi-fuel vehicle, since
the gasoline mode was always available as an option.
13
Currently, only one automaker (Honda) in the United States offers an original equipment manufacturer (OEM) vehicle that is a dedicated NGV; there are no OEM bi-fuel vehicles. Conversions of
gasoline-powered vehicles to dedicated NGVs have been incentivized for vehicle fleets through
tax credits (for dedicated NGVs only, but not bi-fuel) and alternative fuel mandates for fleet vehicles.
Larger-scale conversions in the light-duty fleet, either to dedicated NGVs or bi-fuel vehicles, have
been challenged by cost considerations, EPA certification requirements, and fueling availability.
Participants discussed a key impediment to bi-fuel vehicles, namely the lack of a CNG refueling
infrastructure. Current CNG refueling stations are located along interstate highways, serving the
long-haul heavy truck market, and at fleet operations centers providing central fueling to natural
gas fleet vehicles. Home refueling systems were discussed as a possible solution, but currently
the cost of such systems is an impediment, as the payback period may not justify the investment.
Participants generally agreed that a bi-fuel vehicle was valuable as an insurance policy or option.
The greater the fuel price volatility, for instance, the greater the opportunity for the owner to
exercise the option of fueling a vehicle with CNG instead of gasoline and arbitrage the prices. Or,
in the event that CNG refueling stations were scarce, the owner would not be forced to change
behavior and could continue to rely upon gasoline.
Flex-fuel vehicles have the distinct advantage of utilizing a wide range of blends of gasoline, ethanol,
or methanol. The technology for FFVs is well established. In the late 1980s and early 1990s, one
US automaker (Ford) manufactured gasoline-methanol FFVs for an extensive pilot program in
California. Fuel distributors generally opted for methyl-tertiary-butyl-ether (MTBE) to meet Clean
Air Act requirements for oxygenated fuels instead of methanol-gasoline blends. Subsequent
concerns regarding the environmental impacts of MTBE coupled with a continued requirement
(some would say unnecessary with advanced refining) for oxygenates, automakers then shifted
to gasolineethanol FFVs. The establishment of alternative fuels credits as part of the federal
Corporate Average Fuel Economy (CAFE) standards, combined with the mandate for ethanol
production as part of the federal Renewable Fuels Standard (RFS) has led to the manufacture and
sale of approximately 1 million FFVs annually.20
Owners of FFVs have the opportunity to utilize gasoline or E85 ethanol blend interchangeably,
enabling them to take advantage of price arbitrage among the fuels. While promising in theory,
FFVs have not gained significant acceptance in the United States due to the fact that the distribution of E85 is limited; where it is available, it is not always less expensive, and it has a lower
driving range.
Current FFVs are not certified to operate on gasoline-methanol blends exceeding M5. Presenters
at the symposium offered a new option: a GEM blend with similar stoichiometric properties to
E85, making it a potential substitute and possibly facilitating the introduction of methanol into
the current fleet of FFVs. While the combustion characteristics of GEM blends were similar to E85,
participants noted that other issues, such as fuel system engine materials and sensors and gauges
could be adversely affected by this alternative. Another option for methanol use involved minor
vehicle modifications to add a second tank. In the two-tank FFV, one tank could hold gasoline
and the second tank could hold methanol or ethanol or gasoline blends containing a high concentration of either alcohols.
14
The alcohol in the second tank is directly injected into the engine when needed to prevent engine
knock that would otherwise occur at high torque. This alcohol boosted gasoline engine concept would allow the engine to be optimized (e.g., higher compression ratio and a higher level of
turbocharging) for higher fuel efficiency and/or better performance. While it is still in the research
and development stage, this system represents a substantial technological advancement. Concerns
were however raised about methanol in general, centering on the experience with MTBE.

Price
coupling: There was considerable discussion throughout the symposium about the
potential for alternative fuels to offer cost savings to consumers. There was general agreement
among the participants that two conditions were necessary for greater market penetration of
alternative vehicles and fuels: 1) the price of alternative fuel needed to be lower than the price
of gasoline on an energy-equivalent basis, and 2) this price-spread, between gasoline and the
alternative fuel, had to be reliably sustained over time. Maintaining a spread between the
price of alternative fuels relative to the price of gasoline was referred to as the price decoupling.
If alternative fuels could be produced at a lower cost than the price of gasoline, and if the
alternative fuel supplies were prices at their marginal cost of production, then the price
behavior of alternative fuels would be decoupled from gasoline prices. If, on the other hand,
suppliers of alternative fuels priced their product at or near the price of gasoline, then alternative fuels would be price-coupled to gasoline. However, participants were in fundamental
disagreement regarding which alternative vehicle fuel options could satisfy these criteria,
currently or in the near term, and how this might be achieved.
Some participants noted that the prices of alternative liquid fuels typically were only
slightly discounted relative to gasoline prices, and that alternative fuel prices generally
tracked both the long-term prices and short-term price volatility of gasoline, as shown in
Figure 3. The only exception is natural gas. These participants concluded that drop-in
fuels employing blends of alternative fuels would generally be price-coupled to the price
of gasoline.
Other participants believed that the use of different feedstock could lead to greater price
decoupling, regardless of whether the fuel was drop-in or drop-out. For example, while
the price of corn-based ethanol was generally comparable to or higher than gasoline (on
an energy-equivalent basis), production of ethanol from another feedstock could result in
a lower cost. Similarly, natural gas to methanol conversion that took advantage of ample
domestic gas supplies combined with new large-scale conversion plants could achieve
lower methanol costs than the current US price for methanol imported in relatively small
quantities. Some participants noted that production of alternative fuels in large quantities
might not only be price-decoupled, but actually exert downward pressure on gasoline prices.
The differences between these two viewpoints centered on the question as to whether producers
and distributors of alternative fuels would be willing to pass through the lower cost of production
in the form of lower fuel prices, to establish market position, or whether gasoline, as the marginal
fuel, would set the market price regardless.
Other participants believed that the only way for consumers to achieve cost savings from alternative fuels was from a drop-out fuel vehicle in which there was no fuel relationship to petroleum;
this is currently limited to natural gas or electric vehicles. For example, a recent study examined
the statistical relationship between natural gas and petroleum prices and found that, in the short
term, there was an enormous amount of unexplained volatility in natural gas prices, and that,
over the long term, the relationship does not appear to be stable.21
15
Figure 3 Relationship among the Prices of Various Alternative Fuels
US Average Retail Fuel Prices

$5.00
$4.50
$4.00
Cost per GGE
$3.50
Propane
$3.00
E85
$2.50
B99/B100
$2.00
B20
$1.50
Gasoline
$1.00
Diesel
$0.50
CNG
4/1/00
7/1/00
10/1/00
1/1/01
4/1/01
7/1/01
10/1/01
1/1/02
4/1/02
7/1/02
10/1/02
1/1/03
4/1/03
7/1/03
10/1/03
1/1/04
4/1/04
7/1/04
10/1/04
1/1/05
4/1/05
7/1/05
10/1/05
1/1/06
4/1/06
7/1/06
10/1/06
1/1/07
4/1/07
7/1/07
10/1/07
1/1/08
4/1/08
7/1/08
10/1/08
1/1/09
4/1/09
7/1/09
10/1/09
1/1/10
4/1/10
7/1/10
10/1/10
1/1/11
4/1/11
7/1/11
10/1/11
1/1/12
4/1/12
7/1/12
10/1/12
1/1/13
$0.00
Source: Clean Cities Alternative Fuel Price Reports, http://www.afdc.energy.gov/fuels/prices.html
2006
1998
2002
1986
1990
1994
1978
1982
1962
1966
1970
1974
1950
1954
1958
1942
1946
1934
1938
1926
1930
In addition, the possibility of home refueling systems for both EVs and NGVs would mean that
consumers would be paying a price commensurate with the residential price of these fuels. In the
50need
50 to50
Loadcost
(MW)
case of natural gas, the cost of the home compressor also50would
be factored Net
in. The
of refueling with a home
and residential
natural gas
rates would yield
200 CNG compressor
300
400
500
600 a cost of 700
CNG fuel in the range of $3$5 per GGE; decoupling would be achieved but not necessarily a
lower fuel price.22 In the case of residential electricity, the cost of installing special charging equipment also needs to be considered.
Participants did not achieve a consensus view on how alternative fuels would be priced in the
2
marketplace
and in fact, had strongly competing views. This suggests the need for additional,
rigorous study and analysis of these issues.
1
2006
2002
1998
1994
1990
1986
1982
1978
1974
1970
1966
1962
1958
1954
1950
1946
1942
1938
1934
1930
choice: Participants discussed the attributes of AFVs and fuels that would affect
consumer choice. While participants drew some observations about consumer choices, they
0
noted that there were few available methods to model consumer behavior. Participants noted
that each of the economic models that could be used to understand and forecast consumer
behavior had particular limitations. Participants discussed several attributes of AFVs and fuels
that would impact consumer choice, including vehicle performance, functionality, ease of
fueling, safety, and most importantly, cost competitiveness.
1926

Consumer
Bi-fuel vehicles require vehicle modifications that can compromise certain key vehicle attributes that are important factors in consumer choice. These include cargo capacity, fuel
economy, and driving range. Consumers will need to carefully evaluate the trade-off of having
continued reliance on gasoline fueling and cost savings. Apprising consumers of the costs of
these trade-offs to inform their decisions and choices would likely require a significant public
education effort.
16
Flex-fuel vehicles utilizing gasoline and E85 mixtures are already on the road and appear to have
gained consumer acceptance. The introduction of methanol as a flex-fuel may pose challenges.
There are concerns about methanol toxicity (although the overall risk is similar to gasoline).
Methanol requires additional safety measures such as sufficient gasoline content to ensure flame
visibility and the addition of bitterants to deter ingestion, but it would cause fewer deaths by
energy release in a car crash. Some symposium participants believed that public perception of
methanols risks may be exaggerated, creating additional challenges for achieving public acceptance. The prospect for achieving sustainable cost savings could serve as a key countervailing
element in winning consumer acceptance. Others noted the lack of consumer utilization of the
E85 option.
Participants appreciated the experiences of alternative fuels deployment in other countries, but
questioned their application in the US market. On the whole, the adoption and deployment of
AFVs have been much more rapid internationally than domestically (including CNG use in Europe
and the introduction of methanol in China). Some of the drivers for these differences include
vehicle cost competitiveness, fuel backward compatibility, and strong government involvement in
building fuel distribution infrastructure. Participants noted that the key lessons learned from
experiences in other countries included:
1. Cost competitiveness is the most important requirement for new alternative fuels to
attract consumers at scale.
2. Backward compatibility of a vehicle greatly facilitates successful market penetration.
3. Sufficient fuel distribution infrastructure for alternative fuels is necessary for market
penetration at scale.
4. Bi-fuel capability is very important for AFVs that are not backward compatible to the
vehicle fleet and/or supported by a sufficient supply infrastructure.
5. The widespread availability of relatively low cost and easily adoptable retrofit kits can
significantly help to develop an alternative fuel market (e.g., Liquefied Petroleum Gas
(LPG) in Europe; CNG in Italy, Pakistan, India) because they allow the conversion of
used vehicles already in the fleet.
6. Most alternative fuels have shorter travel ranges than gasoline or diesel. Shorter travel
ranges should be compensated by other positive features of alternative fuels.
7. Incentives for sustainable alternative fuels are initially required if they have higher
production and/or distribution costs than gasoline/diesel (after tax) in order to be
affordable and cost competitive.
8. For any alternative fuel, there must be enough feedstock available to develop and
sustain the market in the long term while maintaining a competitive fuel price.
17

Government
policy role: Current governmental policies provide incentives for both the
manufacture of AFVs and the establishment of alternative fuel distribution facilities.
Symposium participants acknowledged that there has been progress in the medium- and
heavy-duty vehicles arena, principally because these can be operated as centrally managed
and fueled fleets.
Participants acknowledged that there has been limited alternative fuel penetration of the LDV
market both the numbers of AFVs and the availability and volumes of alternative fuels are
insufficient to bootstrap each other sufficient to have a material impact on the market. Even
with improvement in these areas, the long fuel savings payback time for standard driving would
remain a challenge. Some participants argued that gasoline should be allowed to compete
with other alternative fuels in order to lower prices, while others noted that federal assistance
in infrastructure development was crucial to enable the fuels to effectively compete.
A threshold issue was whether government policy should rely principally on increasingly
stringent fuel economy standards as the policy mechanism of choice for meeting the set of
public goods articulated earlier.
Some participants noted that proposed new CAFE standards will have significant benefits, and
that government policy should not complicate these standards by simultaneously promoting
the deployment of AFVs. Other participants noted that there are limits to efficiency standards,
including: the documented problem of the rebound effect (i.e., a portion of the fuel savings
from increased efficiency is offset by an increase in vehicle miles driven); financial cost
avoidance by maintaining older, less efficient vehicles in service for longer periods; and
diminishing fuel economy improvements as CAFE standards become more stringent. These
participants believed that government policies to promote the use of AFVs and fuels could
complement a national strategy of increased fuel economy in the LDV fleet. Such measures
could provide policy signaling that could stimulate market-driven efforts as well.
Participants who favored an increased government role also believed that government policy
should not seek to identify particular winners and losers among various alternative vehicles
and fuels options. Participants discussed the merits of an Open Fuel Standard that would
provide a relatively low-cost policy to assure large-scale manufacturing of various types of
AFVs dedicated mono-fuel, bi-fuel, and flex-fuel. In addition, participants discussed the
importance of public education programs to assist consumers in making sound choices
among competing AFVs and fuels. Finally, participants discussed increased federal investments in RD&D and innovation in order to enhance the technological capabilities and cost
effectiveness of AFVs and fuels. Participants noted that current policies that focused on extant
vehicle technologies may be placing too much emphasis on lowhanging fruit.
Guide to the Report

The remaining sections of the report discuss these issues in greater detail. The sections are
organized from the perspective of the three major players in the alternative fuels market the
vehicle technology and manufacturing perspective, the fuel distribution perspective, and the
consumer perspective and the government policy perspective.
18
S e c t i o n 2 V e h i c l e T e c h n o l o g y a n d M a n u f a c t u r i n g
P e r s p e c t i v e s
Overview
The key characteristics defining the distinction among dedicated, mono-fuel, bi-fuel, and flex-fuel
vehicles with spark-ignited engines are the number of fuel tanks, the fuel delivery components,
control systems, and the changes in engine control settings. For purposes of the symposium
discussion, several alternative configurations were considered as shown in Table 2. Specific
vehicle and operational modifications for each configuration are discussed in further detail in this
section.
Table 2 Configurations of AFVs and Fueling Systems
Mono-fuel
(dedicated)
Bi-fuel
Flex-fuel
Number of tanks
One
Two
One
Two
Fuel combustion
Only one fuel
Each fuel separately
Fuels
simultaneously
Fuels simultaneously
or separately
Liquid fuels
Yes
No
Yes
Yes
Liquid fuel mixtures
No
No
Yes
Yes
Gaseous fuels
Yes
Yes
No
No
Fuel types
Gasoline
CNG
Tank 1: Gasoline
Tank 2: CNG
Optimized for E85
Tank 1: Gasoline
Tank 2: Ethanol, methanol
Source: MITEI.
Dedicated Mono-Fuel Vehicles

In 2009, the United States had a total of about 245 million vehicles on the road, of which about
235 million were light-duty cars passenger cars, sport utility vehicles (SUV), and light trucks
(the remainder were buses and medium- and heavy-duty trucks).23 Only 4% were vehicles with
either dedicated fuels (other than gasoline or diesel fuel) or flex-fuel systems.
Natural Gas Vehicles
Other than gasoline and diesel, the only significant dedicated mono-fuel vehicles are NGVs. Currently,
there are about 116,000 NGVs in the United States.24 One estimate is that over 80% of these vehicles
are predominantly dedicated CNG fleet vehicles that are unable to operate on gasoline.25 (This pattern
is very different from the global market, where most NGVs are bi-fuel vehicles). The deployment of
dedicated NGVs has largely been driven by government policy:
Purchases
of dedicated NGVs have been eligible for federal tax credits (bi-fuel vehicles
are not); 26 and
Many
dedicated NGVs were purchased by state governments and alternative fuel provider
fleets to comply with the requirements of the Energy Policy Act of 1992.27
19
In the United States, dedicated NGVs typically operate on CNG. (Some long-haul heavy-duty trucks
are being converted to operate on LNG. Dedicated NGVs can either be offered by OEMs or NGV
capability can be retrofitted to conventional gasoline-powered vehicles.
Currently, Honda is the only OEM offering a NGV, the Honda GX model. Market penetration of
dedicated NGVs in the United States has been primarily aftermarket conversions of gasoline
vehicles by small volume manufacturers. Conversions have been concentrated in specific models,
because of EPA certification requirements established under the Clean Air Act. A small volume
manufacturer must obtain EPA certification for each make and model to be converted. The cost
of obtaining an EPA certification has been estimated to be as much as $200,000 per vehicle make
and model.28 This cost is then amortized over the number of vehicles converted to operate as
NGVs. It is estimated that a certified conversion by a small vehicle manufacturer costs an additional $10,000 compared to the price of a comparable gasoline-powered vehicle.
Engine Design and Vehicle Performance
Figure 4 illustrates the modifications to enable a spark-ignition, gasoline-fueled vehicle to operate
on natural gas. The hardware modifications are designed to deliver comparable vehicle performance
(although with considerably less range with CNG). Because CNG has a higher octane rating than
gasoline, engine controls can be optimized for greater performance and fuel economy. However,
to maximize performance potential, the engine cylinder compression ratio would need to be
increased, which is typically not implemented in engines originally manufactured for gasoline
operation. Increasing the compression ratio to improve fuel economy with CNG would prevent
acceptable gasoline operation (because of engine knock).
Figure 4 Modifications for CNG-Dedicated Vehicles
Source: http://www.mijoautogas.co.in/cng-mixer-system-lambda-control-stystem.htm
20
Economics of Dedicated NGVs

Participants discussed the fact that NGVs have a higher initial vehicle cost than conventional
gasoline-powered LDVs as shown in Table 3. Several different perspectives were provided on the
magnitude of this cost premium. Data from Honda, which is the only producer of dedicated NGVs
in the United States, shows the retail price for the Honda Civic GX model is $26,305, while its
equivalent gasoline model is $18,360 a premium of about $8,000 in retail price. However,
presently Honda is offering a $3,000 fuel card upon purchase of its GX model, effectively making
the cost premium about $5,000.
Table 3 Cost Comparison of Dedicated NGV and Conventional Gasoline-Powered Vehicle
Vehicle Make and Model
MSRP
(Miles per Gallon (MPG)
Honda Civic GX (dedicated CNG)
$26,305
28
Honda Civic LX (dedicated gasoline)
$18,360
29
Source: Honda.com
The cost premium for NGV conversions by small volume manufacturers is higher, with estimates
of $10,000 or more per vehicle.29
Bi-Fuel Vehicles
The most common type of bi-fuel vehicle is one that can operate on either gasoline or CNG.
Currently, bi-fuel vehicles are primarily in other countries than the United States. It is estimated
that there are more than 14.8 million vehicles worldwide that can operate on natural gas.30 The
majority of these vehicles are
bi-fuel. The geographical distri
Figure 5 Global Distribution of Mono-Fuel and Bi-Fuel
bution of natural gas capable
NGVs, 2010
vehicles is predominantly in
Russian
Africa
developing countries in Latin
Federation & C.I.S.
0.9%
7.7%
America, Asia-Pacific, and to some
extent, the Middle East as shown
in Figure 5.
Asia-Pacific
35.6%
Latin
America
and the
Caribbean
39.4%
Europe
3.0%
North America
2.2%
Middle East
11.3%
Regional NGV Distribution 2010

Source: J. Seisler, Clean Fuels Consulting working paper to TIAX on
International Perspective NGV Market Analysis: Light- and Medium-Duty
Vehicle Ownership and Production, April 2011.
21

Bi-fuel
vehicle technology and fueling systems: Bi-fuel vehicles operate with a conventional spark-ignition engine on either gasoline or CNG stored in two separate tanks, but are
not combusted simultaneously. This requires modifications in the vehicle design to accommodate two tanks a gasoline tank and a new CNG-optimized tank to withstand pressures
of the gas an engine that can operate between the two fuels, and a fuel processing system
(e.g., fuel regulator, injector, engine management, manual switch) that can switch fuel operations. The fuel system technology to support the CNG mode is similar to that in a dedicated
NGV. The additional requirement is for an engine control system that allows switching
between fuels, with the ability to modify engine settings to optimize engine performance for
either fuel.
A CNG bi-fuel vehicle has a second fueling system, fuel tank, and fuel delivery system
completely separate from the conventional gasoline fueling system. The modifications to a
gasoline vehicle necessary to achieve bi-fuel generation are shown in Figure 6. The four basic
types of CNG fuel tanks are illustrated in Table 4. Each of the four meet the same performance
and safety requirements, such as resistance to temperature extremes (-40F to +185F),
multiple fills (pressure changes), cargo spillage, vibration, vehicle fires, corrosion, and collision.
There are considerable differences, however, in the choice of material, weight, and cost.
Weight is a critical parameter. For LDVs, fuel consumption is reduced by 0.6%0.9% for every
3% increase in weight.31
Table 4 Various Types of CNG Fuel Tanks
Tank Design
Material
Type 1
All metal (aluminum or steel)
Type 2
Metal liner partially reinforced by

c omposite wrap (glass or carbon fiber)
around middle (hoop wrapped)
Type 3
Metal liner reinforced by composite

wrap around entire tank (full wrapped)
Type 4
Plastic gas-tight liner reinforced by

composite wrap around entire tank
(full wrapped)
Cost
Weight
Least expensive
Heaviest
Most expensive
Lightest
Source: http://www.cleanvehicle.org/technology/CNGCylinderDesignandSafety.pdf
22
Figure 6 Components to Convert and Operate Conventional

Vehicles with CNG
The CNG fuel delivery system is

illustrated in Figure 6.
There currently are no US OEM
bi-fuel vehicles and a relatively
limited number of aftermarket
conversions to bi-fuel operation.
In contrast, the majority of NGVs
in other countries are bi-fuel
vehicles that have been after
market conversions. These con
versions have been motivated
by market forces, i.e., a relatively
short payback period resulting
from a combination of low-cost
conversion kits and installation
as well as having fewer emission
controls and no OEM certification
requirement.32 However, with
continued higher gasoline prices,
Europe has moved steadily toward
OEMs, which currently have at
least 12 OEM bi-fuel vehicle
models.33 As European OEMs
expand their bi-fuel vehicle offerings, further market segmentation
is taking place. Some bi-fuel
vehicle models have larger CNG
and small gasoline tanks, intended
primarily for natural gas use, while
others have larger gasoline tanks
with slightly smaller CNG tanks.
The European Union (EU) currently
classifies bi-fuel vehicles with
gasoline tanks less than 15 liters
as mono-fuel, even though these
vehicles have bi-fuel capability.34
Note: Attached to the fuel tank [1] is the regulator [2], which reduces tank
pressure from 3,600 psi to 125 psi. Fuel is then fed to a parallel fuel rail
[3] and to new, secondary injectors plugged into an adapter [4]. A wiring
harness [5] plugs into the factory engine-control unit and intercepts
throttle information, sending it to a new fueling computer [6], which
slightly alters the data and passes it to the CNG injectors [7] through
a parallel wiring harness [8].
Source: http://www.popularmechanics.com/cars/how-to/maintenance/
should-you-convert-your-car-to-natural-gas-2
23
CNG bi-fuel vehicles are similar to dedicated gasoline mono-fuel vehicles with regard to power,
acceleration, and cruising speed. Due to the lower energy density of CNG relative to gasoline,
and the additional weight associated with the CNG fuel tank, CNG bi-fuel vehicles have a shorter
driving range, lower fuel economy, and less cargo capacity. Consumer perspectives on these
trade-offs are discussed in detail in Section 4.
Economics of CNG Bi-Fuel Vehicles
Domestically, OEMs have produced only CNG-dedicated fuel vehicles, and not bi-fuel vehicles.
As described earlier, the present cost premium in the United States for a CNG-dedicated vehicle
is currently about $5,000, reflecting the effects of incentives relative to a comparably equipped
gasoline vehicle. Information presented to symposium participants showed a cost premium in
Europe of approximately 3,500 (USD 4,500).35 Symposium participants believed that bi-fuel
vehicles could be offered by OEMs in the United States for a comparable premium. Participants
also discussed that CNG bi-fuel capability could be retrofitted to gasoline vehicles. Conversions
of gasoline vehicles to dedicated NGV use cost about $10,000; 36 participants believed that conversions to bi-fuel operation would be about the same. Some participants noted that aftermarket
conversion costs were significantly lower in other countries, for example, the cost of conversion
in Singapore is reported to be about $2,500.37
Flex-Fuel Vehicles
Conceptually, FFVs can operate with a mixture of more than one liquid fuel. The United States
currently has 9 million registered FFVs on the road, representing about 4% of all LDVs.38 These
vehicles are capable of operating on either gasoline or E85 or mixtures of the two. Symposium
participants considered a broader range of alternates, including tri-flex fuel vehicles capable
of operating on gasoline, ethanol, or methanol in various combinations. For tri-flex fuel mode,
participants considered two alternative fueling options: 1) a single tank operation with up to
three blended fuels (gasoline, ethanol, and methanol) that are combusted simultaneously; and
2) a two-tank system where one tank contains gasoline and the other tank contains a blend of
high concentrations of either ethanol or methanol with gasoline as a supplemental fuel. This
arrangement is also referred to as dual-fuel operation, where the two fueling systems could
be operated either standalone or simultaneously. Presently, FFVs on the road in the United States
do not use the two-tank system.
Flex-fuel vehicles on the market today are optimized to operate on gasoline or E85, and not
on methanol blends, which technically makes them bi-flex fuel vehicles. However, symposium
participants did receive a presentation on a proposed ternary mixture of gasoline, ethanol, and
methanol (GEM fuel) that has the same stoichiometric properties of E85. Such a mixture may
provide an option for introducing methanol into existing FFVs originally designed for operation
with only gasoline or ethanol.
24
Vehicle Technology Modifications39

Flex-fuel vehicles are designed to operate with much higher concentrations of alcohols (ethanol
and methanol) in fuel mixtures than the current gasoline content of 10% ethanol. Alcohol fuels
have several fuel properties that differ from gasoline. Alcohol fuels (1) have lower energy density;
(2) are electrically conductive; (3) are more corrosive to metal, rubber, and plastic materials;
(4) have higher oxygen content, affecting the stoichiometry of combustion; and (5) have higher
evaporative emissions. These factors drive the nature and types of technology modifications
in FFVs.
The modifications can be illustrated by consideration of current FFVs which are designed for
bi-flex fuel operation with either gasoline or E85, as shown in Figure 7.
Figure 7 Special Features of the FFV 40
Engine calibration updates:
Fueling and spark advance
calibrations directed by vehicle
computer to control combustion,
enable cold start, and meet
emissions requirements
Fuel system electrical connections and

wiring: Must be electrically isolated and
made of materials designed to handle
ethanols increased conductivity and
corrosiveness (if exposed to fuel)
Internal engine parts: Piston

rings, valve seats, valves, and
other components must be
made of ethanol-compatible
materials that are designed to
minimize corrosion and the
cleansing effects of alcohol
fuels, which can wash
lubrication from parts
Fuel identifier system:
Automatically senses the
composition of the fuel and
adjusts engine for varying
ethanol-gasoline blends
Fuel injection system: Must be
made of ethanol-compatible
materials and designed for
higher flow to compensate for
ethanols lower energy density
Fuel pump assembly: In-tank

components must be made from
ethanol-compatible materials and
sized to handle the increased fuel
flow needed to compensate for
ethanols lower energy density
Fuel filler assembly:

Includes anti-siphon and
spark-arrestor features
Fuel rail and fuel lines: Must

be made of ethanol-compatible
materials with seals, gaskets,
and rubber fuel hoses rated
for ethanol use
Fuel tank: Must be made of

ethanol-compatible materials and
designed to minimize evaporative
emissions from ethanol
Source: Flexible Fuel Vehicles, Alternative Fuels Data Center
25

Fuel
tank: Because of ethanols lower energy density, the driving range for a given size fuel
tank is lower for ethanol operation than gasoline operation. To compensate for this difference,
fuel tanks may need to have larger capacity to provide a comparable driving range relative to
gasoline-only operation. Tanks also need to be fabricated from ethanol-resistant materials,
which can include special coatings to existing tank materials. The design of the tank should
minimize evaporative emissions. In addition, the fuel filler assembly should have anti-siphon
and spark-arresting features.

Fuel
delivery system: The fuel sender and the fuel pump materials need to be alcohol
compatible and the pump needs to be designed for higher flow rates and pressures to compensate for the lower energy density. Fuel lines and fuel rails, including seals, gaskets, and
rubber hoses, should be made of ethanol-compatible materials, such as stainless steel, and
be designed for higher pressures. Fuel injectors should utilize materials that are corrosion
resistant and should be designed for higher injection pressures. Electrical connections and
wiring should be isolated from and made of materials that are unaffected by the increased
electrical conductivity of alcohols.

Other
engine components: Internal engine parts, including valves and piston rings, should
be designed to withstand the corrosiveness and cleaning effects on metals. Lubricant specifications also may require changes. Engine controllers need additional software capability and
sensor systems need to be able to continuously sense the incoming alcohol/gasoline composition and adjust air-fuel mixtures and spark timing for optimal performance.
Figure 8 Illustration of Dielectric Flex-Fuel Sensor (by Duralast)
Source: autozone.com, available at

http://www.autozone.com/autozone/parts Duralast-Flex-FuelSensor/_ /N-8veh0?itemIdentifier=910217_0_0_
26
A key element of FFV technology

is the flex-fuel sensor, which
monitors fuel composition and
signals the powertrain control
module (PCM) to adjust engine
operation (e.g., air-fuel ration
and ignition timing) accordingly.
The commonly used sensor is
an oxygen sensor that can infer
the alcohol-gasoline composition
based on the oxygen content of
the blend. An alternative technology isa dielectric sensor that can
measure electrical conductivity of
the fuel, with higher conductivity
associated with higher concentrations of alcohol in the fuel blend.
An example of a dielectric sensor
is shown in Figure 8.
Economics of FFVs
Current FFVs, which are certified to operate on either gasoline or E85 (and mixtures of each), do
not carry a cost premium relative to comparably equipped gasoline mono-fuel vehicle models.
From the outset, FFVs (designed for methanol as the flex-fuel) were generally sold without a price
premium relative to comparable gasoline-only versions, despite higher production costs for
engineering, tooling, materials, and controls.41, 42 In 1983, Ford sold a test fleet of Ford Escorts to
California at a price premium of $2,200.43 However, it was believed that the price premium would
disappear at higher production volumes, as experience in Brazil with production of ethanol
vehicles had shown.44 So the actual cost premium is unknown. Some symposium participants
noted that ethanol flex-fuel capability was essentially provided without incremental cost to
consumers because the costs to OEMs were minor. Others suggested that the costs were
absorbed by the OEMs in return for the benefits garnered by OEMs from the CAFE credits for
producing vehicles with alternative fuel capability. The provisions of the CAFE regulations affecting
credits for AFVs are discussed in detail in Section 5.
Symposium participants also discussed the feasibility of operating gasoline mono-fuel vehicles
on alcohol fuels, without vehicle modification. They pointed out that many current gasoline
mono-fuel vehicles are mechanically capable of operating with alcohol fuels such E85, but are
not operationally optimized for them. Such operation would produce inaccurate readings of fuel
gauges and the speedometer due to the lower energy density of alcohol fuels. Some participants
also noted that operation of conventional vehicles with ethanol would not be feasible on a longterm basis, because the vehicles would sustain damage over time to fuel lines, seals, and valves,
among other areas, due to the corrosive properties of alcohol fuels.
Participants also discussed aftermarket conversion of gasoline mono-fuel vehicles to flex-fuel
capability. Conversions to flex-fuel operation would require vehicle modifications in three areas:
engine, tank design, and fuel processing system. While some participants believed aftermarket
conversions were feasible, some necessary parts are not readily available in the aftermarket.
Also, there can be challenges in modifying engine controllers to be able to manage flex-fuel
operation, depending upon the degree of flex-fuel operation.45
FFV/Fuel Combinations
Current FFVs are bi-flex vehicles designed to operate on gasoline, ethanol E85, or mixtures.
Symposium participants discussed the possibility of flex-fuel operation with methanol. Current
gasoline mono-fuel vehicles and gasoline/ethanol FFVs are EPA-certified to accept methanolgasoline blends not to exceed 5%, as shown in Table 5. Because of the RFSs for ethanol, gasoline
distributors use ethanol almost exclusively, thereby resulting in little or no use of M5 blends.46
27
Table 5 Approved Methanol Gasoline Blends with Requirements for Co-Solvent Alcohols and Additives
Market
Region
Introduction
Year
Maximum
Volume %
Methanol
Minimum
Volume %
Co-Solvent
Maximum
Wt %
Oxygen
Corrosion
Additives
Europe
EC Directive
1985
3.00
Methanol
3.7
United States
Sub Sim*
1979
2.75
Methanol
2.0
United States
Fuel Waiver
1981
4.75
Methanol
3.5
Required
United States
Fuel Waiver
1986
5.00
2.5
3.7
Required
China, Shanxi
M15
Standard
2007
15.00
~7.9
Required
For Water
Tolerance
*US EPAs Substantially Similar Regulation for commercial gasolines.

Source: Methanol Gasoline Blends, Methanol Blending Technical Product Bulletin, Methanol Institute.
Participants noted that many of the vehicle modifications needed to permit operation with E85
would also support use of higher blends of methanol (e.g., up to M85). However, additional
vehicle modifications would be needed to address characteristics of methanol that differ from
ethanol, such as the potential for higher levels of evaporative emissions.
Table 6 compares the physical and chemical properties of various alternative fuels relative to
conventional gasoline.
Table 6 Comparison of Fuel Properties
Gasoline
CNG
Ethanol
Methanol
n-Butanol
Chemical Structure
C4to C12
CH4(83% 99%)
C2H6(1% 13%)
CH3CH2OH
CH3OH
C 4H9OH
Physical State
Liquid
Compressed Gas
Liquid
Liquid
Liquid
Main Fuel Source
Crude Oil
Underground
reserves
Corn, grains, or
agricultural
waste (cellulose)
Natural gas, coal,

or woody biomass
Corn, biomass,
cellulose, yeast
Energy Density
32 MJ/L
19.6 MJ/L
16 MJ/L
29.2 MJ/L
Specific Energy
2.9 MJ/kg air
3.0 MK/kg air
3.1 MJ/kg air
3.2 MJ/kg air
Heat of Vaporization
0.36 MJ/kg
0.92 MJ/kg
1.2 MJ/kg
0.43 MJ/kg
Pump Octane
Number*
84-93 (a)
120+ (b)
110 (c)
112 (c)
96
Research Octane
Number** (RON)
91-99
130
108.7
108.6
92-103
Motor Octane
Number (MON)
81-89
120
89
92
78
Energy Content
116,090 Btu/
(Lower Heating Value) gal (d)
20,268 Btu/lb (d)
76,330 Btu/gal
for E100 (d)
57,250 Btu/gal (d)
110,000 Btu/gal
Energy Content
124,340 Btu/
(Higher Heating Value) gal (d)
22,453 Btu/lb (d)
84,530 Btu/gal
for E100 (d)
65,200 Btu/gal (d)
Energy Contained in
Various Alternative
Fuels as Compared
to One Gallon of
Gasoline
100%
5.66 pounds or
126.67 cf of CNG
has 100% of the
energy of one
gallon of gasoline.14
1 gallon of E85
has 77% of the
energy of one
gallon of
gasoline.^
1 gallon of
methanol has
49% of the energy
of one gallon
of gasoline.
Air-Fuel Ratio
14.6
14.2
9.0
6.4
11.1
28
Table 6 Comparison of Fuel Properties (continued)

Gasoline
CNG
Ethanol
Methanol
n-Butanol
Anti-Knock Index
(AKI)
99.15
98.65
83-97
FuelingStations
(Private and Public)
104,845
988
2,498
Energy Security
Impacts
Manufactured
using oil, of
which nearly
50% is
imported (e).
CNG is domestically produced.

The United States
has vast natural
gas reserves.
Ethanol is
produced
domestically. E85
reduces lifecycle
petroleum use by
70% per passenger vehicle and
E10 reduces
petroleum use
by 6.3% (f).
Methanol is
domestically
produced,
sometimes from
renewable
resources.
Butanol is
domestically
produced,
sometimes
from renewable resources.
High-pressure
tanks require
periodic inspection
and certification.
Corrosive and
hygroscopic.
Special lubricants
may be required.
Practices are very
similar, if not
identical, to those
for conventionally
fueled operations.
Corrosive and
hygroscopic.
Special lubricants
must be used as
directed by the
supplier and
M85-compatible
replacement parts
must be used.
Toxicity at the
rate of 20
grams per liter.
Distillation
technology is
expensive.
Maintenance Issues
* Pump octane number is the average of the research octane number and motor octane number.
** Octane as tested in a single-cylinder octane test engine operated under less severe operating conditions.
Octane as tested in a single-cylinder octane test engine at more severe operating conditions.
According to the AFDC, cubic feet units were not given because there were infinite combinations of temperature
and pressure and their effect on fuel density. Instead, fuels were dispensed by Coriolis flow meters, which track fuel
mass and report fuel dispensed on a GGE basis.
Energy comparisons are given in percent energy content on a gallon-to-gallon basis unless other units are given.
^ According to the AFDC, the ethanol content of E85 is usually lower than 85% for two reasons: 1) fuel ethanol
contains 2% 5% gasoline as a denaturant and 2) fuel ethanol content is lowered to 70% in the winter in cold
climates to facilitate cold starts. When the actual composition of E85 is accounted for, the lower heating value of
E85 varies from 82,970 Btu/gal to 89,650 Btu/gal, which is 72% to 77% the heat content of gasoline.
Sources:
(a) Petroleum Product Surveys: Motor Gasoline, Summer 1986, Winter 1986/1987. National Institute for Petroleum and
Energy Research.
(b) K. Owen and T. Coley. 1995. Automotive Fuels Reference Book: Second Edition. Society of Automotive Engineers,
Inc., Warrendale, PA.
(c) J. Heywood. 1988. Internal Combustion Engine Fundamentals. McGraw-Hill Inc. New York.
(d) Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (GREET) Model, version 1.7. 2007. Input
Fuel Specifications. Argonne National Laboratory. Chicago, IL.
(e) Energy Information Administration. Monthly Energy Review. Summary for 2006.
(f) M. Wang. 2005. Energy and Greenhouse Gas Emissions Impacts of Fuel Ethanol. Presentation to the NGCA
Renewable Fuels Forum, August 23, 2005. Argonne National Laboratory. Chicago, IL.
29
The data reveal a mixed picture as the relative advantages of alternative fuels compared to
conventional gasoline. For example:
The
most commonly cited comparison shows that CNG and alcohol fuels have lower
energy density than gasoline, resulting in a shorter driving range for a comparable
volume of fuel.
Alternative
fuels generally have a higher octane rating than gasoline. A higher octane
rating results in less likelihood of engine knock and enables engines to be set for
higher compression ratios, which in turn, leads to better performance and higher fuel
economy. This fuel economy gain can partially offset the lower range due to lower
energy density.
Ethanol
and methanol have a higher heat of vaporization, which is the energy required
to transform a given quantity of a substance from a liquid into a gas at a given pressure
(usually atmospheric). A fuel with a high latent heat of vaporization can create engine
difficulties in cold conditions, namely, a cold start.
Although
methanol and ethanol operate at lower air-fuel ratios than gasoline, the ratio
for each is set at a level close to the stoichiometric ratio of oxygen-to-carbon for that
particular fuel, so the differences in values reported in the table do not necessarily infer
superiority. Operation of a spark-ignition engine at a stoichiometric fuel/air ratio
enables use of the highly effective three-way catalyst for vehicle emissions control.
Ethanol and methanol also are hydroscopic and corrosive, potentially causing damage to metals
and polymers used in fuel-handling systems and engine components. The hydroscopic nature of
ethanol and methanol also pose challenges for bulk fuel transport and distribution, which is
discussed in Section 3.
Introduction of Methanol into FFVs
Use of methanol in FFVs is more challenging than use of ethanol. While methanol possesses
many similar properties with ethanol, there are also some significant differences. For example,
methanol contains soluble and insoluble contaminants which increase the fuels corrosiveness.
As an alcohol fuel, methanol is hygroscopic, where it will absorb water vapor from the atmosphere, thereby diluting the fuel. Water contaminants can suppress engine knock and can also
cause separation of methanol-gasoline blends.
As noted by participants, methanol is currently EPA-certified for use in methanol-gasoline blends
of 5% or less and the certification is further limited to blends in which ethanol is not present. The
current RFS has led to almost universal use of ethanol in gasoline, thus effectively blocking
low-level methanol blends from the market.
Symposium participants discussed several alternative approaches for FFV operation with methanol.
One approach was a two-tank flex-fuel system, in which the vehicle contained a second tank
holding alcohol (either ethanol or methanol) or a high-concentration alcohol-gasoline blend, with
a parallel fuel handling and injection system to separately inject the alcohol into the combustion
chambers in parallel with the primary fuel (either gasoline or a low concentration ethanol-gasoline
blend). The alcohol is directly injected into the engine when needed to prevent knock at high
30
torque, enabling operation at a high compression ratio and with a higher level of turbocharging,
thereby providing greater efficiency and performance.
Another novel idea presented to symposium participants was the possibility of a three-part
fuel mixture of gasoline/ethanol/methanol (or GEM) that could be a replacement for E85 in
current FFVs.

Two-tank
system: A two-tank FFV would have one tank containing gasoline as the primary
fuel and the other tank containing a blend of either ethanol or methanol with gasoline as a
supplemental fuel, as illustrated in Figure 9.
Figure 9 Concept of a Two-Tank Design for the FFV
1st Tank
Gasoline (primary fuel source)
2nd Tank
Ethanol + gasoline
OR
Methanol + gasoline
Source: MITEI.
The rationale for this two-tank

system was to take advantage
of the potential boost or ondemand octane enhancement
a vehicle could obtain by using
a mix of a small amount of
alcohol fuel in a separate tank;
the high intrinsic octane of the
alcohol along with the effective
cooling provided by direct
injection removes the knock
limit on gasoline operation
and increases gasoline fuel
efficiency. Another benefit of
this two-tank system is that it
can resolve the evaporative
emissions issue. Evaporative
emissions only occur when the
methanol concentration is low.
If methanol is stored in a fuel
tank as pure methanol (M100)
or as a high-concentration
blend, evaporative emissions
are minimized.

The
GEM fuel blend: The GEM fuel blend was presented as a novel approach that would
enable the introduction of large quantities of methanol into the LDV market by taking advantage of the FFVs currently on the road. According to the white paper in this document by
Turner et al., the GEM blend has the same stoichiometric properties as that of E85 and, as a
result, the difference between the new GEM blend and E85 is indistinguishable to a current
FFV designed for the latter. They also argued that producing new FFVs that can run on the
GEM blend is not highly challenging since current AFVs have been already tested with M100.
However, other participants were skeptical of the reported results and noted that the fuel blend
would still have impacts on the vehicles performance. Details of the GEM fuel blend are
highlighted in Figure 10.

The
GEM fuel: Symposium participants from Lotus Engineering presented that there were
fuel blends of gasoline/ethanol/methanol (or GEM blend) that can be produced in such a
way that the blends have the same stoichiometric properties as that of E85 and, as a result,
the differences between the GEM blends and E85 are indistinguishable to current FFVs. The
stoichiometric relationship can vary, as shown in Figure 10.
31
Figure 10 Combinations of GEM Fuel Blends
Aiding Transition: GEM9.7 Blend Concentrations

Blend D
Blend C
Blend B
Blend A Straight E85
Fraction of gasoline /methanol in blend
100%
90%
Gasoline
80%
70%
Volumetric energy content is constant
60%
Octane numbers are constant
50%
Latent heat varies by 2% across all such blends
40%
Methanol
30%
Ethanol
Therefore true potential for a drop-in solution
20%
10%
0%
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
Ethanol fraction/(%)
Source: J.W.G. Turner, R.J. Pearson, et al., Evolution of Alcohol Fuel Blends Towards a Sustainable Transport Energy
Economy, Lotus Engineering, Symposium White Paper, 2012.
Figure 10 highlights four possible GEM blends, whose properties are compared in more detail in
Table 7. These blends were selected for detailed engine testing. Note that Blend A is the same as
commercial E85.
As shown by the table, Blends C, D4, and D show nearly the same characteristics as those of Blend
A (E85); the values of stoichiometric AFR (air-fuel ratio), LHVs and octane numbers (RON and MON)
are almost identical. Some participants argued that producing vehicles that can run on the GEM
blend is not highly challenging since current AFVs have been already tested with M100 (pure
methanol). However, other participants were skeptical of the reported results and noted that the
fuel blend would still have impacts on the vehicle performance that could vary with alternative
GEM blend composition.
Lotus also performed NOx emissions testing of the selected GEM blends. All GEM blends produced
10%15% lower amounts of NOx than gasoline. Furthermore, the amount of NOx produced from
the blends was less than 20% of the legal maximum, which is substantially lower than the normal
engineering target of 50%.
Lotus concluded that GEM blends can become a true drop-in fuel for current FFVs. As the number
of these vehicles on the road is increasing, this becomes a potential strategy for the introduction of
methanol into the fuel supply in a manner that does not require further vehicle modifications.
32
Table 7 GEM Ternary Blend Fuels Used in the FFV Tests

Fuel
GEM Component Ratios
Stoichiometric AFR
Blend A
Blend C
Blend D4
Blend D
G15 E85 M0
G37 E21 M42
G40 E10 M50
G44 E0 M56
9.69
9.71
9.65
9.69
Density (kg/L)
0.781
0.769
0.767
0.765
Gravimetric LHV (MJ/kg)
29.09
29.56
29.46
29.66
Volumetric LHV (MJ/L)
22.71
22.71
22.60
22.69
1,627.9
1,623.9
1,613.9
1,620.2
RON (to ASTM D2699)
107.4
106.4
105.6
106.1
MON (to ASTM D2700)
89.7
89.3
89.0
89.0
Carbon Intensity (gCO2/L)
Notes:
Blend A G15 E85 M0 is a test fuel representing Straight E85.
Blend B G29.5 E42.5 M28 splits the ethanol available for E85 across twice the total volume of fuel.
Blend C G37 E21 M42 splits the ethanol available for E85 across four times the total volume of fuel. Methanol is twice
the volume of ethanol; alcohol is approximately twice gasoline.
Blend D G44 E0 M56 methanol-gasoline equivalent of Straight E85. Extreme of range of ternary blends at 9.7:1
stoichiometric AFR.
Source: J.W.G. Turner and R.J. Pearson, et al., Evolution of Alcohol Fuel Blends Towards a Sustainable Transport
Energy Economy, Lotus Engineering, Symposium White Paper, 2012.
Environmental Performance of FFV/Fuel Combinations

For alcohol fuels made from certain feedstock materials (sugar-based ethanol, used as a major fuel
in Brazil; ethanol and methanol from cellulosic material, not yet made in any significant quantities)
FFV/fuel alternatives generally are more environmentally beneficial than dedicated gasoline
mono-fuel vehicles in terms of GHG emissions and conventional tailpipe emissions. Methane
leakage and evaporative emissions, however, were noted by participants as areas of concern for
further investigation.

Greenhouse
gas emissions: Lifecycle assessments of the GHG emissions associated with

the use of alternative fuels in FFVs include upstream emissions (fuel production, processing,
and distribution), as well as downstream emissions (tailpipe and evaporative emissions).
The comparison in Figure 11 shows that ethanol and butanol derived from various forms
of biomass generally have lower GHG emissions than gasoline.
For methanol, the lifecycle (i.e., well-to-wheels) GHG emissions from natural gas-to-methanol
are slightly lower than gasoline. However, the GHG emissions could be somewhat higher than
that of gasoline if emissions from methane are included.47 Several participants noted that the
lifecycle emissions of methanol could be reduced through increased energy efficiency from
methanol engines that can take advantage of very high octane.
Estimates of GHG emissions from various tri-FFV configurations, including GEM fuel blends,
were not available for symposium participants to review.
33
Figure 11 Percent Change in GHG Emissions Relative to Gasoline

150
114.7%
100
l
no
50
c
si
o
ul
ll
Ce
a
th
el
e
di
Bi
u
eo
as
H2
ol
n
ha
ity
ic
El
r
ct
CN
-33.0% -27.8%
-50
LN
rn
ol
et
Co
LP
-21.8% -21.0% -19.1%
-48.0%
n
ha
et
-7.6%
L
CT
6.7%
1.8%
&
/C
TL
se
ie
CT
/o
C&
-68.2%
-100
-90.8%
Note: All estimates derived from Argonne National Lab GREET Model v2.7.
Source: Carmine Difiglio, Background: US Alternative Fuel Policies and Methanol, US DOE, July 2011.

Evaporative
emissions: Evaporative emissions are typically caused by gasoline vapors that

Biodiesel
escape from storage tanks or fuel lines. They are unburned fuel vapors that are comprised
of volatile organic compounds. Evaporative emissions are potentially a much greater source
of hydrocarbons than tailpipe emissions. Evaporative emissions highly contribute to groundNGCC
Sub Coal
Super Coal
level ozone concentrations.
Cost of Electricity ($/MWh)
Mixtures of gasoline and methanol generally have higher levels of evaporative emissions.
The level of evaporative emissions is higher with blends containing low concentrations of
methanol; as the methanol content of the blend increases, the level of evaporative emissions
decreases.
COE without Dispatch
COE
withestablished,
Dispatch
Control systems for evaporative
emissions are
well
and new vehicles currently
are required to have evaporative emission control systems. The system consists of a canister
of charcoal that captures vapors created in the fuel tank and releases them to the engine
intake manifold. Current evaporative emission control systems are designed for gasoline
vehicles. Therefore, using blends of methanol and gasoline will almost certainly result in
canister saturation and higher evaporative emissions.48 Thus, higher capacity systems may
be required for FFVs.

Conventional
tailpipe emissions: Because of their oxygen content, fuel blends with

ethanol and methanol generally have lower carbon emissions than conventional gasoline.
NOx emissions also are lower. Data presented at the symposium showed that NOx emissions
were generally lower for fuels with ethanol or methanol blends relative to current NOx emission standards for LDVs.
34
S e c t i o n 3 F u e l P r o d u c t i o n a n d D i s t r i b u t i o n
Infrastructure Perspective
Introduction
This section summarizes the symposium discussion from the viewpoint of fuel production and
distribution infrastructure. It examines the issues related to production capacity to serve the LDV
market, distribution systems to move alternative fuel products from production and processing
locations to markets, and the infrastructure issues associated with vehicle fueling.
Figure 12 provides a graphic snapshot comparison of the current infrastructure for gasoline,
natural gas, and ethanol as a starting point for discussion. There are several overarching takeaway messages, summarized below, that are discussed in more detail in the following sections:

CNG
an extensive pipeline infrastructure, but insufficient number of refueling stations.

Worthwhile noting that the stations installed are mostly located at interstate highways.

Ethanol
concentrated and limited production facilities, and a wide area for fuel distribution.

Methanol
US methanol is presently imported (produced from natural gas in other countries).

Potentially, the United States could become a large producer of methanol from domestic
natural gas, requiring new large-scale production facilities.
Figure 12 Comparison of Fueling Infrastructure for Various Alternative Fuels

US Crude Oil Pipelines
US Natural Gas Pipelines & Compressor Stations
US Ethanol Production Plants
Legend
= Interstate Pipeline
= Intrastate Pipeline
= Compressor Station
US Gasoline Refueling Stations
US CNG Refueling Stations
US Ethanol Refueling Stations
Natural Gas
Fueling Stations
Interstate Highways
Source: Oil diagrams: EIA, Elisheba Spillers white paper; natural gas: EIA and NREL (http://www.eia.gov/pub/oil_gas/
natural_gas/analysis_publications/ngpipeline/ngpipeline_maps.html); ethanol: NRELs TransAtlas maps (http://maps.
nrel.gv/transatlas).
35
As the predominant fuel for LDVs, the gasoline distribution infrastructure is highly developed and
responsive to customer demand. It provides a starting point for comparison with the infrastructure
requirements for large-scale deployment of alternative fuels. Natural gas has an extensive pipeline
infrastructure capable of reaching a very large segment of the LDV market, but the current infrastructure of refueling stations is limited. Moreover, the location of these stations is primarily on
interstate highways, designed to serve fleets and long-haul heavy-duty vehicles rather than the
LDV market. The ethanol infrastructure also is highly developed, with production concentrated in
the major corn and graining growing regions of the country. The infrastructure of E85 refueling
stations is generally dispersed in a pattern that matches the current density of FFVs in the market.
By comparison, there is virtually no current methanol infrastructure in the United States. The
United States is currently a net importer of methanol, and very limited amounts are used for
LDVs, due to current regulatory mandates and certification requirements that favor the use of
ethanol over methanol.
A more detailed discussion of fuel supply and infrastructure issues, as considered by symposium
participants, is provided in the sections that follow, organized by fuel type.
A Note on Comparability among Fuels

US consumers are used to understanding and comparing fuel economy and fuel prices
on a volumetric basis i.e., MPG and $ per gallon (gal) respectively. When comparing
conventional gasoline with alternative fuels, the volumetric comparison can be misleading,
due to key differences in fuel properties such as energy density (i.e., how much energy is
contained in a unit volume of liquid fuel), performance (i.e., amount of power output from
an engine optimized for an alternative fuel), and price. For purposes of the symposium, the
following conversion factors were used where appropriate in order to arrive at comparable
estimates of gallons of GGE shown in Table 8. The retail prices are shown in Table 9.
Table 8 Conversion Factors
Unit of Measure
Gallon Equivalent
BTUs/Unit
Gasoline (regular)
gal
1.00 gal
114,100
Ethanol (E85)
gal
1.39 gal
81,800
Methanol (M85)
gal
1.74 gal
65,400
CNG
cf
126.67 cf
Propane (LPG)
gal
1.35 gal
84,300
Diesel #2
gal
0.88 gal
129,500
Biodiesel (B20)
gal
0.90 gal
127,250
Ethanol (E100)
gal
1.50 gal
76,100
Methanol (M100)
gal
2.01 gal
56,800
Biodiesel (B100)
gal
0.96 gal
118,300
900
Source: Alternative Fuels and Advanced Vehicles Data Center (AFDC) Quarterly Report, January 2012.
A comparison of January 2012 prices also is noted, in both actual physical units and in
terms of price per gallon of GGE.
36
Table 9 January 2012 Overall Average US Retail Fuel Prices

Retail Price
Retail Price per Gasoline

Gallon Equivalent (GGE)
Retail Price per

Million Btu
(based on GGE)
Gasoline
$3.37/gal
$3.37
$29.23
CNG
$2.13/GGE
$2.13
$18.49
Propane (LPG)
$3.08/gal
$4.26
$36.93
Ethanol (E85)
$3.14/gal
$4.44
$38.50
Methanol
$1.34/gal
Diesel
$3.86/gal
$3.46
$30.00
Biodiesel (B20)
$3.95/gal
$3.61
$31.24
Biodiesel (B99/B100)
$4.20/gal
$4.14
$35.84
Note: The price shown for methanol is the contract price of $1.34/gal reported by Methanex. This is equivalent
to $2.69/GGE. The methanol contract price is more comparable to the spot wholesale price of gasoline.
In January 2012, the gasoline wholesale spot price as $2.82/gal for New York harbor and $2.77/gal for the
US Gulf Coast.
Source: Alternative Fuels and Advanced Vehicles Data Center (AFDC) Quarterly Report, January 2012.
Figure 13 illustrates historical trends in the prices of various alternative fuels. Note that the liquid
fuels generally follow the same pattern of price volatility as gasoline. Natural gas does not. The
issue of price coupling is discussed in more detail later in the report.
Figure 13 Comparison of US Average Retail Fuel Prices per GGE
US Average Retail Fuel Prices

$5.00
$4.50
$4.00
Cost per GGE
$3.50
Propane
$3.00
E85
$2.50
B99/B100
$2.00
B20
$1.50
Gasoline
$1.00
Diesel
$0.50
CNG
4/1/00
7/1/00
10/1/00
1/1/01
4/1/01
7/1/01
10/1/01
1/1/02
4/1/02
7/1/02
10/1/02
1/1/03
4/1/03
7/1/03
10/1/03
1/1/04
4/1/04
7/1/04
10/1/04
1/1/05
4/1/05
7/1/05
10/1/05
1/1/06
4/1/06
7/1/06
10/1/06
1/1/07
4/1/07
7/1/07
10/1/07
1/1/08
4/1/08
7/1/08
10/1/08
1/1/09
4/1/09
7/1/09
10/1/09
1/1/10
4/1/10
7/1/10
10/1/10
1/1/11
4/1/11
7/1/11
10/1/11
1/1/12
4/1/12
7/1/12
10/1/12
1/1/13
$0.00
Source: AFDC, January 2012.
50
200
300
50
400
50 50
500
Net Load (MW)

600
06
98
02
86
90
94
78
82
62
66
70
74
50
54
58
42
46
34
38
26
30
700
37
Gasoline
The US gasoline market is a large, efficient, and mature industry. As a point of departure for
discussing the scale needed for an alternative fuels infrastructure, it was noted that the current
US petroleum infrastructure consisted of 55,000 miles of crude oil pipelines, feeding 150 refineries. Gasoline product from these refineries is transported through another 95,000 miles of refined
product pipelines and many local delivery trucks, supplying approximately 160,000 gasoline
refueling stations.49 By comparison, it was reported that one major petroleum company estimated that a FFV market using methanol would require that 10% of current gasoline refueling
stations be equipped with methanol refueling capability.50
As Table 10 illustrates, the total number of refueling stations for all types of alternative fuels
constitutes about 14% of the total number of gasoline refueling stations; excluding electricity, the
number of alternative liquid and gaseous alternative refueling stations constitutes only 4% of
gasoline stations.
Table 10 Comparison of the Number of Refueling Stations in the US
Biodiesel
CNG
E85
Electric
Hydrogen
LNG
LPG
696
1,190
2,583
15,192
58
66
2,776
Note: Includes both public and private refueling stations, as of Dec 31, 2012.
Source: Alternative Fueling Station Counts by State, AFDC http://www.afdc.energy.gov/fuels/stations_counts.html
Ethanol
Supply
In 2010, US consumption of ethanol was 13,189 million gallons, most of which was from domestic
production [Figure 14].
Figure 14 US Production, Consumption, and Trade* of Fuel Ethanol
14,000
Million Gallons Ethanol
12,000
10,000
8,000
6,000
4,000
Net Imports
2,000
Production
Consumption
2011
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
*Trade includes small changes in stock

Source: AFDC, 2012.
38
U.S. Average Retail Fuel Prices
50
50
50 50
Net Load (MW)
200
300
400
500
600
700
In 2011, ethanol production was about 14 billion gallons.51 According to the Renewable Fuels
Association, out of this total production, only 67.4 million gallons, or 0.47%, were from non-corn
feedstock materials, including brewery/beverage waste, milo/wheat starch, waste sugars, wood
waste, cheese whey, potato waste, and sugar cane.
The cost of production varies depending on the choice of feedstock material, which affects both
the cost of raw materials as well as the cost of processing. A 2006 US Department of Agriculture
(USDA) study provided a comparison of these costs on an equivalent basis, as summarized in
Table 11. Notably, producing ethanol from US raw or refined sugar is significantly higher than
from other domestic feedstock crops, particularly corn, though Brazil still has the lowest ethanol
production costs.
Table 11 Ethanol Production Costs from Various US Feedstock Materials
Summary of Estimated Ethanol Production Costs (Dollars per Gallon)*
Cost Item
US Corn
Wet
Milling
US Corn
Dry
Milling
US Sugar
Cane
US Sugar
Beets
US
Molasses
US Raw
Sugar
US Refined
Sugar
Brazil
Sugar
Cane
EU Sugar
Beets
Feedstock
Costs**
0.40
0.53
1.48
1.58
0.91
3.12
3.61
0.30
0.97
Processing
Costs
0.63
0.52
0.92
0.77
0.36
0.36
0.36
0.51
1.92
Total Cost
1.03
1.05
2.40
2.35
1.27
3.48
3.97
0.81
2.89
*Excludes capital costs.

**Feedstock costs for US corn wet and dry milling are net feedstock costs; feedstock costs for US sugar cane and sugar
beets are gross feedstock costs.
Excludes
Average
transportation costs.
of published estimates.
Source: US Department of Agriculture, The Economic Feasibility of Ethanol Production from Sugar in the United States,
July 2006.
Table 11 shows that the cost of

ethanol production from most
feedstock materials was less than
the price of gasoline during this
same period, which ranged from
$2.65/gal to $3.24/gal. Although
there were significant differentials,
the price of ethanol, in the form of
E85, was slightly higher than
gasoline (on an energy-equivalent
basis). In fact, on an energyequivalent basis, the price of E85
has been slightly higher than
gasoline since 2000, except for
a brief period in early 2009, as
shown in Figure 15. Consequently,
the lack of any significant price
discount for E85 has probably
contributed to the low levels
of E85 consumption in the
United States.52
4.50
30
4.00
20
3.50
10
3.00
2.50
-10
2.00
-20
1.50
-30
E85 Cost Benefit, %
Retail Price, $/Gallon E0 Equivalent
Figure 15 Historical Relationship between E85

and Gasoline Prices
-40
1.00
2000
2002
E85
2004
2006
Gasoline
2008
2010
2012
E85 Benefit
Source: Ulrich Kramer and James Anderson, Symposium White Paper.
50
50
50 50
Net Load (MW)
/1/00
/1/00
/1/00
/1/01
/1/01
/1/01
/1/01
/1/02
/1/02
/1/02
/1/02
/1/03
/1/03
/1/03
/1/03
/1/04
/1/04
/1/04
/1/04
/1/05
/1/05
/1/05
/1/05
/1/06
/1/06
/1/06
/1/06
/1/07
/1/07
/1/07
/1/07
/1/08
/1/08
/1/08
/1/08
/1/09
/1/09
/1/09
/1/09
/1/10
/1/10
/1/10
/1/10
1/1/11
4/1/11
7/1/11
0/1/11
/1/12
/1/12
/1/12
/1/12
/1/13
200
300
400
500
600
700
39
Production of ethanol has been encouraged and subsidized by the government for decades.
Through fiscal year 2010, the EIA reported that the US ethanol fuel industry had received approximately $5.68 billion in Volumetric Ethanol Excise Tax Credit (VEETC).53
Ethanol Transport Infrastructure
While most US ethanol plants are concentrated in the Midwest, gasoline consumption is highest
along the coastlines. The population of FFVs capable of using E85, while more concentrated in the
Midwest, also exhibits a greater population density on the coasts [Figure 16].
Due to its high oxygen content and solvent properties, ethanol is corrosive and tends to absorb
water and impurities when transported through pipelines, which currently only distribute less
than 10% of fuel ethanol. As illustrated in Figure 17, over 90% of ethanol production is transported
by rail or truck from production facilities to gasoline storage terminals, where it is splash blended
with gasoline.
The significant ramp-up in production and consumption has caused consideration of the need for
dedicated ethanol pipelines, specifically designed to suit the chemical characteristics of ethanol.
One such pipeline in current operation is the Central Florida Pipeline Project. POET LLC and
Magellan Midstream Partners have proposed to construct a new dedicated ethanol pipeline
connecting the Midwestern and Northeastern states [Figure 18].
Figure 16 US Ethanol Production Facilities and Areas of FFVs
Production Facilities Ethanol-Producing

Ethanol Plants (mgal/yr)
Vehicle Density Flex-Fuel
Less Than 50
>139 vehicles/5 sq miles
Between 50 and 100
91139 vehicles/5 sq miles
Between 100 and 150

Between 150 and 200
Greater than 200
45.591 vehicles/5 sq miles

Note: Shaded areas on the map denote the density of registrations of FFVs.
Source: National Renewable Energy Laboratory (NREL), 20092012, available at http://maps.nrel.gov/transatlas
40
Figure 17 Schematic of US Rail and Truck Ethanol Distribution System
Source: AFDC, 2012. Available at http://www.afdc.energy.gov/afdc/fuels/ethanol_production.html
Figure 18 Proposed Dedicated Ethanol Pipeline
Legend
Proposed Dedicated Ethanol Pipeline
Ethanol Plant
Ethanol Plant (Under Construction)
Proposed Receipt Location
50 Mile Radius
Proposed Distribution Location
Source: AFDC, 2012. Available at http://www.afdc.energy.gov/afdc/fuels/ethanol_production.html
41
Ethanol Fueling Infrastructure

Currently, there are 2,498 E85 refueling stations in the United States. Figure 19 shows the location
and density of refueling stations correlated with the location of FFVs.
According to the Alternative Fuels Data Center (AFDC) and a 2008 National Renewable Energy
Laboratory (NREL) report, US gasoline stations generally only have an average of 3.3 tanks. To
provide E85 fueling capability, a gasoline station could either install an additional tank or convert
an existing tank. A new tank costs on average $71,735 (median $59,153), while converting an
existing tank is an average of $21,031 (median $11,237) [Table 12].
Figure 19 US Ethanol Refueling Stations and Areas with FFVs
Vehicle Density Flex-Fuel

>139 vehicles/5 sq miles
91139 vehicles/5 sq miles
Existing E85 Stations

Planned E85 Stations
Source: NREL, 20092012 (http://maps.nrel.gov/transatlas).
42
Table 12 Cost of Adding E85 Fueling Capability to Existing Gasoline Stations

Scenario
New tank, new or retrofit
dispenser(s)
Convert existing tank, new

or retrofit dispenser(s)
Cost
Source*
Mean: $71,735
Median: $59,153
NREL Survey
$50,000 $200,000
NACS
$50,000 $70,000
DOT, EPA, DOE
>$50,000
NEVC
<$62,407
DAI
Mean: $21,031
Median: $11,237
NREL Survey
$19,000 $30,000
DAI
$5,000 $30,000
DOT, EPA, DOE
$2,500 $25,000
NEVC
Description
Major Variables
Affecting Cost
Includes new
storage tank, pump,
dispenser(s), piping,
wiring, excavation,
and concrete work
Dispenser needs,
excavation, concrete
work, sell backs,
canopy, tank size,
location, labor price,
regulations
Tank cleaning, replace

non-compatible
components in piping
and dispensers
Dispenser needs,
number of
non-compatible
components,
location, labor
price, regulations
*NREL estimates based on invoices and cost estimates provided by grant administrators, station owners, and project
managers for 120 E85 fueling stations, of which 84 were new tank installations and 36 were conversions of existing
tanks. The range of costs for a new tank was between $7,559 and $247,600 and for conversion of an existing tank,
$1,736 to $6,800. NREL notes that the lowest-cost tank conversions may have taken shortcuts and are not recommended because of concerns about safety and materials.
Source: AFDC, March 2008. Available at http://www.afdc.energy.gov/afdc/pdfs/42390.pdf
Natural Gas
Supply
In 2011, natural gas supply and demand reached record levels, with 23 trillion cubic feet (tcf) of
domestic dry gas production and total consumption of 24.4 tcf.54 The average wellhead price was
$3.95/mcf, and the natural gas price at city gate locations was the lowest (in inflation-adjusted
terms) in a decade.55
The US natural gas resource base has been estimated at about 2,100 tcf, including shale gas and
Alaska natural gas.56 This corresponds to about 90 years of natural gas supply at current production rates. The potential supply base of shale gas is very large, and may not yet be fully characterized. The MIT Future of Natural Gas study estimated that a considerable portion of the shale
resource base can be produced economically at prices between $4/mcf and $8/mcf.
The current supply outlook suggests that domestic natural gas resources could support a significant alternative fuels infrastructure, either in the form of CNG or through conversion to methanol.
For example, it was estimated that operating 50% of the current LDV fleet on CNG would increase
current natural gas demand by about one-third.57
Transport Infrastructure
The United States has a robust and mature interstate and intrastate transportation system,
consisting of 300,000 miles of transmission pipelines [illustrated in Figure 20] and 1.9 million
miles of distribution lines.58
43
Figure 20 US Natural Gas Pipeline Compressor Stations Illustration, 2008
Legend
= Interstate Pipeline
= Intrastate Pipeline
= Compressor Station
Source: Energy Information Administration, Office of Oil & Gas, Natural Gas Division, Natural Gas Transportation
Information System.
Changes in the geographical pattern of natural gas production (e.g., increased production from
the Marcellus gas shale region) as well as changes in the geographical pattern of demand for
natural gas likely will require additions to the pipeline system. However, the processes for planning, regulatory approvals, and financing of new natural gas pipeline infrastructure are well
established and not likely to pose a barrier to increased use of natural gas in AFVs.
Fueling Infrastructure
The current fueling infrastructure for CNG has evolved around the two principal sources of
vehicle demand: heavy-duty trucks in long-haul interstate transport and inner-city fleets mainly
of trucks and buses. This pattern is illustrated in Figure 21.
Consequently, the current CNG fueling infrastructure is limited and concentrated along the
interstate highway system. It was designed to serve centrally fueled fleets of LDVs, trucks, and
buses and longer-haul heavy-duty vehicles rather than the light-duty market.
The Clean Cities program has been working to promote expansion of this network so that it can
support the broader LDV market.59 Current proposals to expand the CNG refueling infrastructure
are illustrated in Figure 22.
44
Figure 21 US CNG Refueling Stations and Interstate Highways
Natural Gas
Fueling Stations
Interstate Highways
Source: NREL, February 2010.
Figure 22 US CNG Existing and Proposed Refueling Stations and Clean Cities Coalitions
Source: AFDC, April 2012 and NREL, 2012. Available at http://maps.nrel.gov/transatlas
45
This proposed expansion will make an important contribution to removing current barriers to
CNG refueling.
Bi-fuel vehicles operating on CNG require a high-pressurized compressor station for natural gas,
and special nozzles to ensure a tight seal during the refueling process. Earlier refueling station
designs used nozzles that required training to use, but recent nozzle designs more closely resemble
those used to pump gasoline. There are two types of CNG refueling stations: fast fill and time fill
(described in Figures 23 and 24). The different terms refer to the capacity of storage tanks and the
throughput of gas compressors.
Fast-fill stations typically have a large storage capacity of CNG available for rapid refueling. The
natural gas is compressed to pressures in the range of about 4,000 pounds per square inch (psi)
and held in storage for refueling. In the refueling process, the vehicle tanks are pressurized to a
level of about 3,500 psi. Fast-fill stations are necessary for non-fleet LDVs. These vehicles gen
erally arrive at the refueling station randomly and need to be refueled quickly. For a 20-gallon
equivalent tank, refueling can take about 5 minutes, which is similar to a gasoline refueling
experience. The equipment needed for fast-fill stations is about the size of a parking space.
By comparison, time-fill stations are designed for fleets. In this case, vehicles refuel at a central
refueling location overnight. Time-fill stations typically have a relatively small amount of buffering storage. Instead, the refueling operation is directly linked to the compressor, and refueling
times are linked to compressor throughput. Depending on the number of vehicles, compressor
size, and the amount of buffer storage, refueling can take from several minutes to several hours.
One advantage of time fill is that the user can choose the time to refuel vehicles; electricity
needed for running the compressor can cost less at off-peak hours.60
Figure 23 Illustration of a CNG Fast-Fill Fueling Station
Source: Compressed Natural Gas Fueling Stations, AFDC http://www.afdc.energy.gov/fuels/natural_gas_

cng_stations.html#fastfill
46
Figure 24 Illustration of a CNG Time-Fill Fueling Station
Source: Compressed Natural Gas Fueling Stations, AFDC http://www.afdc.energy.gov/fuels/natural_gas_cng_stations.

html#fastfill
Participants discussed that bi-fuel vehicles may have less demanding requirements for refueling
than dedicated NGVs. For example, it may be acceptable to fill bi-fuel vehicle tanks to lower
pressures, reducing fill times or allowing for lower-rated (and less expensive) compressors. Also,
improvements in compressor technology may allow for faster fill rates with lower temperature
buildup.
The cost for CNG refueling stations depends upon the size of stations and the types of natural
gases (CNG, LNG, or both) that the stations offer. Whether a station is a fast-fill or a time-fill
station also affects the cost. According to a 2010 report by US DOE Pacific Northwest National
Laboratory, a CNG refueling station can cost from $400,000 to $2 million as shown by Table 13.
Table 13 Cost for CNG Refueling Station
Refueling station type
Maximum Capacity
Maximum Capacity,
GGE
Estimated Cost
CNG, small
<500 scfm
4.0 gge/min
$400,000
CNG, medium
5002000 scfm
4.015.8 gge/min
$600,000
CNG, large
>2000 scfm
>15.8 gge/min
$1,700,000
Source: US DOE, Pacific Northwest National Laboratory, Issues Affecting Adoption of Natural Gas Fuel in Light and
Heavy-Duty Vehicles, PNNL-19745, September 2010.
47
Participants also discussed the possibility of at-home CNG refueling capability. An at-home
refueling capability would create a more level playing field between CNG bi-fuel vehicles and
PHEVs. For a period of time, Honda, which produces and sells a dedicated NGV (the Accord GX),
also marketed a home CNG refueling appliance called Phill, through a separate company
(Fuelmaker). The appliance is now being marketed by the Italian Company BRC Gas Equipment.
Phills costs were about $4,500 and depending on the customers residential gas rate, and installation, operating, and maintenance costs, the resulting cost of CNG could be in the $3 to $5 per
GGE. Phill was a relatively low pressure (0.5 psi) CNG refueling system, requiring about 8 hours
to fill a CNG tank. Anecdotal comments on Hondas experience of selling both the GX and Phill
in Southern California indicated that once customers became accustomed to the existing CNG
fueling infrastructure, they did not see the value of the additional investment for home CNG
refueling. Consumers in other regions where the density of CNG refueling stations is lower may
have a greater interest in at-home refueling capability. DOE, through the Advanced Research
Projects Agency Energy (ARPA-E), recently awarded grants for development of low-cost
at-home refueling systems. Development of a cost-effective at-home refueling system could
represent a disruptive technology that could significantly impact the demand for both dedicated NGVs as well as CNG bi-fuel vehicles.
Methanol
Supply
Methanol can be produced from several feedstock materials, including natural gas, coal, and
biomass. In 2009, US demand for methanol was 1.85 billion gallons, of which about 90% was
used for chemicals production.61 86% of US methanol demand is imported, mainly from the
Caribbean and South America. There currently is limited domestic production of methanol;
the largest four facilities, shown in Table 14, total 329 million gallons of production, the bulk
of domestic supply.
Table 14 US Methanol Production (2009) (Millions of Gallons)
Production
Feedstock
71
coal
203
NG
CF Industries, Woodward. OK
40
NG
Praxair, Geismar, LA
15
NG
Eastman Chemical, Kingsport, TN

LaPorte Methanol/Lyondell, Deer Park, TX
Source: L. Bromberg and W.K. Cheng, Methanol as an Alternative Transportation Fuel in the US: Options for
Sustainable and/or Energy-Secure Transportation, Sloan Automotive Laboratory, Massachusetts Institute of Technology,
November 2010.
Natural gas is an ideal feedstock for the production of methanol, and symposium participants
assumed that large-scale use of methanol would require new domestic natural gas to methanol
conversion facilities. There is considerable global experience in large-scale natural gas to
methanol conversion, mainly as a feedstock for chemical production, and the conversion process
is relatively efficient. Figure 25 shows that natural gas can be converted into a variety of liquid
products, including methanol. A range of liquid fuels can be produced from natural gas by
thermochemical conversion to a synthesis gas followed by catalytic conversion to the liquid fuel.
These fuels include methanol, ethanol, mixed alcohols (methanol, ethanol, and others), and
diesel. Methanol can in turn be converted into gasoline or into dimethyl ether (DME), a cleanburning fuel for diesel engines.
48
Figure 25 Conversion of Natural Gas to Alternative Fuels
Natural Gas
Reformer
Synthesis Gas
Catalyst
Methanol
Mixed Alcohols
DME
Diesel
Ethanol
Gasoline
Source: MITEI 2011, The Future of Natural Gas: An MIT Interdisciplinary Study, 2011.
Methanol imported into the United States, mainly from the Caribbean and South America, has
been priced comparable to gasoline over the period 2005 to 2010 on an energy-equivalent basis.
Currently, the wholesale price cost of methanol (on an energy-equivalent basis) is comparable to
the wholesale price of gasoline. According to Methanex, the contract cost of methanol in January
2012 was equivalent to $2.70/GGE. At the same time, the spot price for gasoline was $2.82/gal for
New York Harbor conventional gasoline and $2.77/gal for US Gulf Coast conventional gasoline.
The coupling of methanol prices to gasoline and to ethanol is illustrated in Figure 26.
Figure 26 Normalized Costs of Liquid Fuels, E85, Gasoline at the Gas Station, and Estimated Costs
of Methanol at the Station
$6.00
$5.00
Cost, $/GGE
$4.00
$3.00
$2.00
Gasoline
$1.00
E85
Methanol
Apr-12
Nov-10
Jul-09
Feb-08
Oct-06
May-05
Jan-04
Sep-02
Apr-01
Dec-99
Jul-98
$0.00
Source: L. Bromberg and W.K. Cheng, Methanol as an Alternative Transportation Fuel in the US: Options for
Sustainable and/or Energy-Secure Transportation, Sloan Automotive Laboratory, Massachusetts Institute of Technology,
November 2010.
50
200
300
50
400
50 50
500
Net Load (MW)

600
700
1994
1998
2002
2006
1990
1970
1974
1978
1982
1986
1926
1930
1934
1938
1942
1946
1950
1954
1958
1962
1966
1 Initiative Symposium on Prospects for Bi-Fuel and Flex-Fuel Light-Duty Vehicles | April 19, 2012
MIT Energy
49
Construction of state-of-the-art methanol plants in the United States and use of US natural gas at
present prices could provide methanol at a significantly lower cost than gasoline. With deployment of new plants, using existing technology, methanol could be produced from US natural gas
at a cost less than the 2010 US gasoline price of around $2.30/gal (excluding the tax). Table 15
shows an illustrative projection of methanol production costs (for a large state-of-the-art plant
with an ROI appropriate for large-scale deployment of well-established technology). It is based on
a 67% energy conversion efficiency of natural gas into methanol and a contribution of amortized
capital and operating costs of $0.50/GGE of methanol production.62, 63 Under these assumptions,
the spread between gasoline price and methanol cost is around $1.00/GGE. The cost advantage
of methanol at the fueling station is reduced by around $0.10/GGE due to higher cost per unit
energy of transporting methanol to fueling stations. The production cost of methanol at this
assumed natural gas price would be lower than the cost of corn-based ethanol by more than
$1.00/GGE.64 The issue about the price of methanol from a large-scale domestic natural gas to
methanol conversion industry was sharply debated by symposium participants. While some
believed that a significant fraction of the cost savings would be passed along by producers to
consumers, others believed that methanol prices would continue to be coupled to gasoline
prices. The issue of price coupling is discussed further in Section 4.
Table 15 Illustrative Methanol Production Costs, Relative to Gasoline (Excluding Taxes) at $2.30 per
Gallon
Natural Gas Price
Methanol Production Cost,

per GGE
Cost Reduction Relative

to Gasoline, per GGE
$4/MMBtu
$1.30
$1.00
$6/MMBtu
$1.60
$0.70
$8/MMBtu
$2.00
$0.30
Source: MITEI, The Future of Natural Gas: An Interdisciplinary MIT Study, 2011.
While the economics of natural gas to methanol conversion appear promising, the market for
methanol as an alternative fuel in the transportation sector faces a number of challenges. They
include the financial risk for private investment in US methanol production plants. The demand
for methanol as a transportation fuel could be reduced by a decline in oil prices and domestic
natural gas prices are volatile. In addition, incentives are lacking for building methanol capability
into vehicles and incurring the costs of additional infrastructure, such as pumps in fueling stations. It is likely that some form of government assistance would be necessary to facilitate this
option at large scale.
Transport Infrastructure
Since methanol is generally produced overseas, it is often shipped through ocean tankers, the
largest of which is used by Methanex, a world leader in methanol production. As a liquid at
standard temperature and pressures, methanol is fairly easy to transport and has been successfully transported through pipelines in Canada.* Though there is currently no nationwide pipeline
network, studies have suggested that only minimal changes of the current infrastructure would
be required, namely by compartmentalizing the fuel from other hydrocarbon products in the
pipeline or converting existing pipelines to dedicated methanol use.65
*In both demonstrations, 4000 tons of methanol were shipped: the first through the 1146km long Trans Mountain crude
oil pipeline, and the second through the ~3000km long Cochin LPG pipeline.
50
Fueling Infrastructure
Currently there are few M85 stations in the United States. Since methanol is a hazardous chemical and reacts strongly to moisture, it requires a secondary containment unit made of methanolcompatible materials, liners, new dispensers, and filters to ensure health and fire safety.
Underground storage tanks cost approximately $50,000.*
Pricing of Alternative Fuels Relative to Gasoline

The cost advantages of alternative fuels relative to gasoline suggest opportunities for marketdriven penetration of alternative fuels, if the appropriate AFVs are available. A major focus of
discussion among symposium participants is whether this cost differential could be garnered by
consumers, thus creating demand for bi-fuel and flex-fuel vehicles. Since the incremental costs
of AFVs are relatively small, it may be feasible that some manufacturers would respond to this
consumer demand and begin offering more choices in bi-fuel and flex-fuel vehicles. There also
may be appropriate role for policy makers in establishing a new mandate such as an Open Fuel
Standard that would further encourage the manufacture of AFVs. This scenario could overcome
the chicken and egg problem and lead to a pathway for significant penetration of alternative fuels
in the LDV market.
The key issue is the sharing of the cost advantages of alternative fuels between producers and
consumers. If alternative fuels are priced at or near petroleum-based fuels, the cost-saving
advantages to consumers of switching to alternative fuels are greatly diminished. Thus, symposium participants noted that in order for these vehicles to successfully compete in the LDV market,
alternative fuels need to be priced below the price of gasoline in order to introduce consumers to
synthetic fuels.
Assuming that the cost of production for an alternative fuel is less than that of gasoline, consumers can benefit from switching if the alternative fuel is priced at its marginal cost of production,
which is decoupled from the price of gasoline. In this scenario, consumers could arbitrage
between the price of alternative fuels and gasoline. If suppliers set the price of alternative fuels at
a level comparable to gasoline (i.e., price coupling), consumers would realize little if any benefit
from fuel switching. The likelihood of these two options was the topic of lively debate among
symposium participants, and ultimately proved inconclusive. Some symposium participants
believed that alternative fuel suppliers would pass along lower production costs in the form of
lower prices for alternative fuels. Other participants believed that price decoupling could be
realized only with alternative vehicle options that were completely divorced from the gasolinefueled spark-ignition engine.
The issue of fuel price coupling is illustrated in Figure 27, which compares two possible pricing
scenarios.
*According to the MIT Bromberg report, this figure may vary, as a California program promoting methanol use
subsidized this cost.
51
Figure 27 Scenarios for Pricing of Alternative Fuels Relative to Gasoline

SCENARIO A
Price
SCENARIO B
Fuel Demand
Price
Price
Fuel Demand
Alternative Fuel
Supply
Gasoline Price
Alternative Fuel
Price
Alternative Fuel
Supply
Arbitrage
Opportunity
Fuel Price
Supply of Gasoline
Q alternative
fuel
Q fuel demand
FUEL MARKET
Quantity
Quantity
GASOLINE MARKET
Q gasoline
Q fuel demand
Quantity
FUEL MARKET
Source: MITEI.
Scenario A illustrates a market that allows for consumers to arbitrage between the price of
alternative fuels and gasoline. When the supply of an alternative fuel that satisfies the current fuel
demand can be produced and supplied at a cost less than the price of gasoline and assuming that
vehicle technologies allow consumers to easily switch between fuels, it allows for the existence
of two fuel prices, one for gasoline and one for the alternative fuel, in the market. The existence
of two fuel prices enables consumers to arbitrage the price difference in the short run, which is
represented by the green wedge. Over the long run, a significant expansion of supply of lowerpriced alternative fuels may exert downward pressure on the price of gasoline and ultimately the
two prices may converge. Some participants believed that large-volume production of alternative
fuels from low-cost feedstock such as natural gas will enable large volumes of alternative fuel
gasoline blends to be offered in the market under conditions that would allow for price arbitrage.
Scenario B illustrates a market in which the price of all fuels is set by the market clearing price for
gasoline. When the supply of an alternative fuel cannot be supplied at a cost lower than the price
of gasoline and assuming that vehicle technologies allow consumers to easily switch between
fuels, two fuels can still exist in the market. However, the quantities sold will differ, and only one
fuel price will exist. This one fuel price will be set by the marginal fuel, or gasoline. Thus, the
quantity of the alternative fuel supplied will be up to the amount that can be produced up to the
price of gasoline, and the remaining fuel demand will be supplied by gasoline. If suppliers are
able to exercise price coupling, consumers may not see any price advantage, nor would they have
the opportunity for price arbitrage. Thus, the prospect for cost savings is eliminated and the
incentive for the consumer to purchase a bi-fuel or flex-fuel vehicle is greatly diminished.
Historical market data show that the price of alternative fuels has been at or very near the market
price for gasoline.
52
Figure 28 illustrates a market with two alternative fuels (C and D), competing with gasoline. It is
important to note that if the alternative fuels C and D are ethanol and methanol, then the supply
of the alternative fuel is a blend of only the alcohol fuels. If blended with gasoline, then the
competitiveness of the fuel will be a function of the percentage of gasoline in the blend.
Figure 28 Possible Price Arbitrage under Conditions of Large Volumes of Alternative Fuel Blends
Price
Price
Fuel
Demand
Supply of
Alternative
Fuel C
Fuel Price
Supply of
Alternative
Fuel C+D Blend
Supply of Gasoline
Q fuel C
Quantity
GASOLINE MARKET
Supply of
Alternative
Fuel D
Q fuel D Q gasoline
Q fuel demand
Quantity
FUEL MARKET
Source: MITEI.
Some symposium participants believed that this was the key issue affecting the success of
alternative fuels, and argued strongly that a viable alternative fuels market for LDVs could occur
only if the alternative fuel price was decoupled from the price of petroleum-based fuels. Table 16
summarizes how the various vehicle-fuel options achieve price decoupling or do not achieve
price decoupling.
53
Table 16 Conditions for Price Decoupling with the Vehicle-Fuel Options

Fuel Options
Corresponding
Vehicle Options
Fuel Efficiency
Infrastructure
Needed
Conditions for
Price Decoupling
Benefits Other Than

Price Decoupling
Conventional
gasoline vehicle
Drop-in fuels
E5E15
M5
Biodiesel up to 100%
Conventional
gasoline vehicle
Ethanol production
and distribution
Price decoupling
not possible
Consumer surplus
Conventional
gasoline vehicle
Methanol production
and distribution
Price decoupling
not possible
Consumer surplus
Conventional
diesel vehicle
Biodiesel production
Price decoupling
not possible
Fuel arbitrage
Conventional
FFV
FFV/ethanol
production/ethanol
distribution
Substantial increase
in ethanol production
Fuel arbitrage
FFV for methanol/

methanol
production/
methanol
distribution
Substantial increase in
methanol production
Market supply and

demand curves of
ethanol + methanol meet
where the price is lower
than that of gasoline
Consumer
flexibility from
multiple fuel
options up to
a certain point
Consumer
flexibility from
multiple fuel
options up to
a certain point
Drop-out fuels
Blendable
drop-out
fuels
Ethanol
(E16E85)
Methanol
(M6M85)
Physical
drop-out
fuels
FFV for methanol
The demand curve for

ethanol and the supply
curve of ethanol meet at
a point where the price
of ethanol is lower than
the price of gasoline
The demand curve for

methanol and the
supply curve of
methanol meet at a
point where the price of
methanol is lower than
the price of gasoline
Tri-flex fuel
(Gasoline +
ethanol +
methanol)
Tri-flex fuel
vehicle
Tri-flex fuel vehicle/

ethanol and
methanol production
and distribution
CNG
Dedicated
CNG vehicle
CNG dedicated
vehicle/CNG
distribution
Bi-fuel vehicle
Bi-fuel vehicle/CNG
distribution
Dedicated
electric vehicle
EV/recharging
station
The prices of electricity

and gasoline are not
coupled
Hybrid electric
vehicle
EV/recharging
station
The prices of electricity

and gasoline are not
coupled
Electricity
Fuel arbitrage
Fuel arbitrage
Fuel arbitrage
Fuel arbitrage
Source: MITEI.
54
S e c t i o n 4 C o n s u m e r P e r s p e c t i v e s
Modeling the Attributes of Consumer Behavior toward Bi-Fuel and Flex-Fuel Vehicles
Successful deployment of bi-fuel and flex-fuel vehicles ultimately will depend on consumer
demand. To predict how consumers will react to new vehicles and fuel options in the market, two
methods are generally used. The first method is to conduct consumer surveys. So far, a number
of surveys have been done in many countries including the United States and the results have
shown that consumers care about prices, safety, and power. The second method of analysis is
through the development of economic models based on past consumer data. That is, to take all
the vehicles that have been purchased, study various attributes of those vehicles, and try to figure
out which attributes are being valued in terms of the additional price people are willing to pay for
those attributes. The results from this method also show that consumers value cheap, safe, and
powerful cars.
Participants expressed strong reservations regarding the limitations of the use of surveys and
models in terms of forecasting consumer attitudes about purchases of bi-fuel and flex-fuel
vehicles. Several people pointed out that, contrary to the general conclusions from economic
models, consumers do not always look for the most cost-effective vehicles. There exists a
demand for highly priced cars and traditional economic models do not very well account for this
phenomenon. Others pointed out that another fundamental limitation of the use of models comes
from the fact that models use past data sets to predict consumer attitudes toward a new technology that did not exist before. As the bottom line, participants agreed that it is necessary to
understand the limitations of choice models before they are implemented.
Despite these limitations, participants generally agreed that there is no practical alternative to
models. Therefore, the discussion should be focused on how to improve the modeling of consumer behavior. For example, it was argued that choice models can account for factors that lead
customers to choose pricey vehicles. Whether it is the make of the vehicle or any other attribute,
anything that has a utility to customers can be taken into account in the model.
After reviewing a number of global and domestic case studies, symposium participants noted
that consumers primarily valued the following attributes when making their purchasing decision:
Vehicle performance
Vehicle functionality
Ease of refueling
Cost competitiveness
Backward compatibility
Safety
Using these attributes, and drawing upon the technical information presented at the symposium,
a summary matrix comparison was constructed [Table 17] of the various alternative vehicle/
alternative fuel alternatives.
55
Table 17 Comparison of Vehicle-Fuel Options from a Consumer Perspective

Attributes Consumers Value
for Alternative Vehicles
Bi-fuel Vehicle
FFV
Comparable to gasoline vehicle; less

prone to knocking
Comparable to gasoline vehicle, with

appropriate engine modifications
Lower fuel economy due to fuel tank

weight and size
Lower fuel economy on volumetric

basis, but not necessarily on an energy
basis
Vehicle Functionality
Trunk space reduced due to larger tank
Potentially no compromises
in vehicle design
Ease of Refueling
Proximity to CNG refueling stations
Availability of alternative fuel stations
T ime required for refueling

(e.g., high-speed filling systems
of 45 minutes)
No change in fueling process

(same as conventional vehicle)
Vehicle Performance
Possibility of home refueling

(e.g., Phill home compressor systems)
Vehicle cost premium
Fuel cost premium
Fuel savings*
Concerns with pressurized gas
Safety
Toxicity concerns
*These savings would be reduced if refueled with a Phill home compressor system. Depending on the customers
residential gas rate and installation, operating, and maintenance costs, the resulting cost of CNG could be $3 $5/GGE.
Source: MITEI.
Vehicle Performance
Symposium participants were in general agreement that current alternative vehicle technologies,
both bi-fuel and flex-fuel vehicles, were well optimized to deliver equivalent vehicle performance
relative to conventional gasoline-powered LDVs. Therefore, while this is an important attribute in
consumer acceptance, it did not appear to be a significant differentiator among the various
alternative vehicle/alternative fuel options.
Because their octane ratings exceed that of gasoline, CNG, ethanol, and methanol offer comparable engine and vehicle performance to conventional gasoline vehicles. However, because of
their higher heat of vaporization, ethanol and methanol could have more issues with cold start
capability. This issue can be avoided by using an appropriate blend with gasoline, e.g., the M85
blend makes cold starts possible in most climates.66 Participants expressed some uncertainty as
to how consumers would react to the fact the CNG, ethanol, and methanol might appear to offer
lower fuel economy (on a volumetric basis) due to lower energy density. However, it was believed
that consumers would be able to understand that comparisons of actual energy efficiency would
be different, and because of lower fuel costs, the cost per mile would actually be lower in the
case of the alternative fuels. Vehicle range or reductions in trunk space (if fuel tanks were
enlarged) represented another possible area of concern.
Backward Compatibility
Backward compatibility in a vehicle refers to the capability of a vehicle to operate on conventional
fuels as well as alternative fuels. Symposium participants agreed that bi-fuel, flex-fuel, and hybrid
vehicles are similarly attractive in that they share this advantage of backward compatibility with
conventional gasoline, and could potentially attract consumers who value this particular kind of
fuel flexibility. Participants described this value as similar to that of an insurance policy or an
option. If there are few refueling stations or the price of gasoline remains significantly higher,
56
there are cost advantages to switch between fuels. In case studies abroad where some of these
vehicles are more widely used, particularly bi-fuel and flex-fuel vehicles, backward compatibility
and more broadly, fuel flexibility is a desirable vehicle attribute, particularly when gasoline
prices are volatile and alternative fuel prices remain low, or when there is uncertainty in refueling
availability.
Ease of Refueling
Ease of fueling includes several factors: availability of refueling stations, length of time to refuel,
and operational safety of the refueling process. As described in Section 3, the gasoline refueling
infrastructure is well developed. The comparison with alternative fuel refueling stations [Table 18]
availability of E85 refueling stations does not appear to be a constraint to use of ethanol in FFVs.
Table 18 Comparison of the Number of Refueling Stations in the US
Biodiesel
CNG
E85
Electric
Hydrogen
LNG
LPG
696
1,190
2,583
15,192
58
66
2,776
Note: Includes both public and private refueling stations, as of Dec 31, 2012.
Source: Alternative Fueling Station Counts by State, AFDC, http://www.afdc.energy.gov/fuels/stations_counts.html
The availability of CNG refueling stations does pose a challenge; while there are a large number
of stations, they are currently located to conveniently serve centrally fueled fleets and vehicles
that travel primarily on the interstate highway system. However, symposium participants stated
that ease of fueling with CNG would be much less of an issue with a bi-fuel vehicle than with a
dedicated NGV. An owner of a bi-fuel vehicle would not be forced to change behavior, especially
in cases in which range or resultant drive routes might be impacted. Instead, drivers of such
vehicles can selectively take advantage of the lower operating cost and greener footprint of
natural gas, knowing that there is no walk home factor that threatens their convenience or
safety should travel take them beyond natural gas pumps. Despite the shorter range when compared to gasoline vehicles, drivers of CNG bi-fuel vehicles have greater range and greater fuel
flexibility relative to other mono-fuel vehicles such as EVs, hydrogen vehicles, and dedicated
mono-fueled CNG vehicles.
As liquid fuels, ethanol and methanol would have comparable refueling times with gasoline. In
contrast, CNG requires the gas to be compressed. The refueling time at a fast-fill public refueling
station, operating with high-pressure storage tanks, is about 45 minutes, comparable to refueling times with gasoline. Home refueling times vary depending on the compressor system; Phill,
the home compressor system marketed by Honda, is capable of refueling in eight hours (natural
gas is delivered at about .5 psi, to reach an ultimate pressure of 3,600 psi).
Safety issues associated with the use of alternative fuels are also a source of concern for consumers. The risk associated with ingestion of methanol is higher than with gasoline; unlike
gasoline, methanol does not cause vomiting if ingested, and can cause serious health effects at
low levels of ingestion.67 While there was not a single case of accidental poisoning by methanol
reported in California in the 1980s, participants agreed that sufficient safety measures are
needed. For example, a very small amount of bitterant can be added to methanol in order to
let the consumers know that it should not be ingested.
57
Although participants agreed that the use of methanol as a transportation fuel will not pose a
significant threat to human health, they also acknowledged that the public perception of danger
of methanol might not be rigorously based on technical knowledge. In this regard, achieving
public acceptance may well be decoupled from technical verification of the safety, and in fact,
acquiring public acceptance can be much more challenging than technical verification.
Cost competitiveness is a key determinant of consumer behavior. Some participants discussed
whether this in fact is the single most important attribute in consumer behavior, noting that most
consumers will choose the least expensive fuel even if the price difference with the second least
expensive fuel is very small. Making an assessment of the most cost-competitive choice among
competing bi-fuel vehicles and FFVs requires a comparison of the fuel cost savings to the consumer relative to the initial cost-premium on the vehicle and recurring maintenance costs. The
fuel cost savings are a function of the retail price of the alternative fuel and the overall fuel efficiency of the AFV. Because fuel cost savings are a critical element of this assessment, the issue
of price coupling, discussed in Section 2, is particularly important. If the price of the alternative
fuel is coupled with the price of gasoline, or if the consumer perceives a possible coupling, then
bi-fuel and flex-fuel vehicles offer no cost savings, and the consumer will make the decision on
vehicle type based on other factors. Because bi-fuel and flex-fuel vehicles may not be as attractive in other ways less trunk space, fewer refueling stations, longer refueling times, performance uncertainties, safety concerns consumers will continue to prefer conventional
gasoline-only vehicles. Conversely, if the consumer believes that a bi-fuel or flex-fuel vehicle
offers the possibility of fuel price arbitrage or a measure of insurance against price volatility,
then cost competitiveness is more likely to become the deciding factor in vehicle selection.
Cost competitiveness can be analyzed in several ways. One approach is to estimate the payback
time (undiscounted) by comparing initial cost premium to annual fuel cost savings. Another
approach is to compare actual monthly cash flows, which is possible in cases where the purchase
price is largely financed. Both approaches require that, in the case of the bi-fuel or flex-fuel
vehicle, the assessment considers the likely pattern of consumption of the alternative fuel options
relative to the likely proportion of continued gasoline use. The value of a bi-fuel or flex-fuel
vehicle as a hedge against gasoline price volatility, proposed by some participants as an option
value or insurance value, was supported in concept by symposium participants, but is not readily
quantifiable.

Payback
estimate for bi-fuel vehicles: Bi-fuel vehicles are most amenable to payback
analysis because they have significant vehicle price premiums while offering the most
significant fuel cost savings potential. In recent years, natural gas prices have become largely
decoupled from petroleum prices, and with the surge of shale gas production, natural gas
prices have been significantly lower than gasoline prices, on an energy-equivalent basis.
This difference is shown in Figure 29.
An analysis of CNG conversions in other countries shows that periods of strong CNG vehicle
market penetration occurred when the payback period was less than 3 years.68 For LDVs,
meeting this condition requires a combination of a price spread of $1.50/GGE, vehicles in high
mileage service (35,000 miles/yr), and an initial cost premium of less than $5,000.69
58
Figure 29 Natural Gas and Gasoline Prices, 20042012

$12.00
$10.00
$8.00
Propane
E85
$6.00
B99/B100
B20
$4.00
Gasoline
$2.00
Diesel
CNG
$0.00
04
05
06
07
08
09
10
11
12
Adjusted US Natural Gas Wellhead Price ($/GGE)

US All Grades All Formulations Retail Gasoline Prices ($/gal)
Source: cngprices.com
Gasoline Price Volatility and Consumer Response

In the United States, gasoline prices fluctuate by season and by region. The price is also
affected by short-run changes in commodity prices. It was shown that fluctuations in
gasoline price causes substitution in vehicle usage and choice. First, when the gasoline
price goes up, people with multiple vehicles spend more time driving more fuel-efficient
vehicles rather than less-efficient ones. It is rare to find people who choose to reduce their
total driving time; rather, they increase the use of high MPG vehicles. This point is confirmed by a study by Granger and Miller as shown by the Table 19. As the gasoline price
increases (Y-axis), people prefer higher MPG vehicles.
Table 19 The Effect of $1 Increase in the Gasoline Price
Manufacturer Price Change (000s)
$1.0
$ 0.5
$ 0.0
$ 0.5
$ 1.0
$ 1.5
10
20
30
40
50
60
MPG
Source: Symposium presentation, drawn from Granger and Miller.
50
50
50 50
59
Participants discussed possible approaches to reduce the payback period for bi-fuel vehicles. One
suggestion was to consider a bi-fuel vehicle as analogous to a PHEV. Under this scenario, the cost
of the bi-fuel system could be lowered by reducing the size of the CNG tank, which is the largest
single item affecting the bi-fuel vehicle price premium. A study indicated 67% of all US drivers
drive fewer than 40 miles a day,70 and a storage tank 10 times smaller than the ones in vehicles
currently sold would be sufficient to fuel 40 miles.71 To extend this concept to PHEVs, a home
refueling system for CNG also would be required. This would result in an estimated payback
period of about seven years for a low-mileage vehicle.72 Another approach to lowering initial
costs is to reduce the design pressure for a CNG tank. This would reduce tank and compressor
costs. However, some symposium participants believed that this would result in an unacceptable
reduction in vehicle range.
Symposium participants also generally agreed that there was an additional value proposition to
bi-fuel vehicles, which could be described in two ways: 1) as an option value, and 2) as an insurance policy. As with any option, the value of a vehicle capable of operating two different fuels
increases with uncertainty, specifically, fuel price volatility. Alternatively, a bi-fuel vehicle could
also be viewed as an insurance policy, which is valuable for those who desire to use an alternative fuel, but are not confident in finding close refueling stations, or for the buyer who does not
want to be forced to change behavior.
Cost competitiveness of FFVs: The cost competitiveness assessment for FFVs using gasoline
or E85 is simpler. Currently, FFVs do not have a price premium, although flex-fuel capability is
available in only a limited number of models. Vehicle manufacturers are generally offering FFVs
in larger-size class LDVs, in SUVs, and trucks, because they can maximize the value of the alternative fuels CAFE credits in larger vehicles. Looking at the cost comparison among alternative fuels,
there is no cost savings. Because of the RFS requirements, gasoline distributors currently purchase
over 99% of total ethanol supply for blending into E10 gasoline. Because of the mandate, ethanol
producers have no incentive to price ethanol lower than gasoline. In fact, insome markets,
ethanol producers may command a small premium, as evidenced by historical price trends.73
This is possible in situations where gasoline distributors have limited access to ethanol supplies
needed to meet RFS requirements. So there is no cost advantage to either alternative on either
the vehicle price or the fuel price. What advantages do occur in the market may more likely be
due to the effects of federal and state government financial incentives. Nonetheless, consumers
may be motivated by the ability to hedge fuel prices against potential future gasoline price volatility (assuming that prices for ethanol do not remain completely coupled).
While there is no significant current market for FFVs and methanol, there is the potential for a
cost-competitive FFV/methanol combination in the future. Participants noted that methanol can
be produced from natural gas at costs significantly below gasoline (on a GGE basis), providing
anopportunity for the introduction of methanol fuels into FFVs on a cost-competitive basis.
Inaddition, methanol can be produced from coal and biomass, providing even greater flexibility
in methanol supply and pricing. Participants were not able to develop an estimate of the potential
cost savings in this area. The cost of methanol FFVs capable of operation on M85 or M100 was
unknown, although participants believed it would not be much higher than for ethanol. In addition, the economics of large-scale methanol production have yet to be demonstrated in the
United States. Finally, there is no current fuel distribution infrastructure in place for methanol.
While participants generally believed that the prospects for methanol FFVs were attractive, some
pointed out that methanol, if used in flex-fuel rather than bi-fuel or dedicated mono-fuel vehicles,
could become subject to price coupling with gasoline. China, for its energy security, has adopted
an ambitious methanol policy. Chinese methanol is produced from coal and a blend of methanol
and gasoline is being widely distributed as transportation fuel. Although background contexts are
different in China and the United States, Chinas methanol case is an important experience to
observe for the United States.
60
Lessons Learned from Alternative Fuels Experience in Other Countries

In many cases, understanding what attitudes the public or consumers have toward a specific
product or technology is critical for forecasting its success in the market. Taking into account the
fact that both tri-flex fuel vehicles and bi-fuel vehicles are at their nascent stages at the moment,
it is essential to discuss the anticipated public attitudes toward these new types of vehicles.
As a starting point for the discussion, Ulrich Kramer and James Anderson from Ford Motor
Company provided an analysis on various cases of the use of non-conventional transportation
fuels around the world. They looked at the introduction of four different types of non-conventional
transportation fuels in more than eight countries. The fuel types and countries analyzed are
summarized in Table 20.
Table 20 Alternative Fuel Experience in Other Countries
Fuel Type
Countries
Ethanol
Brazil, United States, Sweden, Germany (E85 and E10, respectively)
Biodiesel
Germany
LPG
Europe
CNG
Asia and Europe (Pakistan, India, Germany, and Italy)
From these case studies, Kramer et al. drew conclusions that helped symposium participants to
understand how consumers make choices when new types of transportation fuels are introduced
into the market.
The discussion below summarizes several basic requirements for the introduction of alternative
transportation fuels at a significant scale into the market, as suggested by Kramer et al.74
1. Cost competitiveness
Cost competitiveness is the most important requirement for new alternative fuels to attract
consumers at scale. Typically, alternative fuel systems require extra investment. Competitive
costs should guarantee a reasonable chance to recover any upfront costs within the first few
years.
Table 21 Minimum Cost-Benefit Recommended for Different Alternative Transportation Fuels
Consumer On-Cost
Minimum Benefit/Cost
Recommended
FFV
5%
OEM LPG vehicle
2,0002,500 (USD 2,6003,300)
CNG vehicle
3,500 (USD 4,500)
40%
50% 70%
Case Countries
Brazil, Germany
Germany
Pakistan, India
61
Another important requirement for any alternative fuels introduced is that the prices are stable
and reliable in the long term. Fuel prices should remain without significant fluctuations even in
the case of rising demand, for example. This point is clearly proven in the cases of Germany and
Brazil. To maintain the price within an acceptable range, sufficient feedstock and fuel production
capacity are necessary.
In order to guarantee the stability of prices of alternative fuels, government actions are usually
needed in the long run. A striking example is illustrated by the Germans B100 case. As the government began to reduce incentives for B100 due to the increase in governmental tax loss, B100s
market success reversed. Therefore, any government policy for high penetration of alternative
fuels into the market should be run for the long term.
2. Backward compatibility
Backward compatibility of a vehicle refers to the vehicles capacity to use existing fuels and
infrastructure. Backward compatibility greatly facilitates successful market penetration. If a
vehicle is backward compatible, the minimum amount of fuel cost-benefit can be smaller (a fuel
cost-benefit of 5%15% was sufficient for B100s case in Germany).
3. Distribution infrastructure
A sufficient number of fuel distribution outlets for alternative fuels is necessary to support
large-scale market penetration of alternative fuels. However, the high upfront cost for infrastructure development can lead to slow growth in the expansion of alternative fueling stations.
Sweden provides a good example of government policy intervention to expand ethanol fueling
stations (In 2006, a law was passed that required all fuel stations above a certain size to offer at
least one alternative fuel).
4. Vehicle capability for two fuels (bi-fuel vehicle, mono-fuel vehicle, or FFV)
If vehicles can run on two different fuels, it is beneficial to consumers because they can always
choose the more cost-competitive fuel among the two options. This is observed in the Brazilian
case where E100-dedicated vehicles failed while FFVs succeeded.
Moreover, the lower the extra cost for the second fuel capability, the earlier the payback period
for the second fuel system installation cost, which can increase the probability of the market
success. While these vehicles can have improved performance thanks to the second fuel, they
generally show less-efficient performance when they run on a single fuel compared to mono-fuel
vehicles.
5. Retrofit kits
Since retrofit kits allow the conversion of existing vehicles, they can greatly help expand
alternative vehicles markets (e.g., LPG in Europe; CNG in Italy, Pakistan, and India). Especially,
conversion of vehicles is particularly attractive in markets that are cost sensitive.
As in the case of LPG in Europe, retrofit kits are usually cheaper than purchasing OEM-produced
vehicles. However, they may be of lower quality, with fewer upgrade options. For instance, OEM
vehicles typically have valve and valve seat insert upgrade options whereas retrofit kits do not
offer such things. As a result, retrofitted vehicles can have less durability due to the increased
valve seat wear caused by poorer lubricity of gaseous and alcohol fuels. Therefore, when introducing alternative vehicles, conversion of vehicles with retrofit kits should be carefully considered.
62
6. Sufficient fuel energy density

Generally speaking, a longer travel range of a vehicle is desired by any consumer. Reduction in
the travel range can be compensated for cost benefit. Most alternative fuels have shorter travel
ranges than gasoline or diesel. To increase travel ranges, larger fuel storage spaces are used at
the expense of having smaller interior space. Shorter travel ranges are particularly problematic
for gaseous fuel-based vehicles and even more so for BEV.
7.Incentives
In general, sustainable fuels (non-fossil, renewable, and GHG-reducing) are currently more
expensive than gasoline or diesel fuel. At least in the beginning, incentives are required to make
these sustainable fuel options more cost competitive. Increases in oil prices can be helpful in this
sense, but development of large-scale fuel distribution infrastructure usually justifies extra
incentives. Governments may use various policy instruments that account for the differences of
the situations that each sustainable fuel faces.
8.Scale
In order to maintain the market in the long term, feedstock for any alternative fuel should be
sufficient, allowing a competitive fuel price. When this condition is not met, price increase of the
alternative fuel is inevitable. Especially when the market solely relies on a dedicated-fuel vehicle,
the market will eventually collapse for that vehicle, as witnessed in the E100-dedicated vehicles
case in Brazil. Therefore, alternative fuel supply needs be scalable with future demand in the long
term (e.g., B100 case in Germany).
63
Summary of Alternative Fuels Experiences in Other Countries

Methanol Experience in China
Methanol production in China has been growing steadily since the 1980s and it has been
accelerated in recent years as China plans to use methanol as an alternative transportation
fuel. In 2010, Chinas methanol production capacity reached 12.8 billion gallons/yr and this
is expected to reach 16.6 billion gallons/yr by 2015.75 In 2010, China consumed 7.6 billion
gallons of methanol; this is about 40% of global consumption of methanol.
A major source of Chinese methanol production comes from Chinese coal reserves.
Methanol can be produced from gasification of coal. China has 114.5 billion tons of coal,
which corresponds to 4.9% of global reserves. Methanol production from coal in China is
mainly driven by energy security concerns, not by a reduction in GHG emissions. Although
the Chinese government is trying to develop nationwide standards for methanol as transportation fuel, there is not a uniform opinion on the policy in China.
Tri-Flex Fuel Vehicle Experience in Brazil: Use of CNG Instead of Methanol
In Brazil, there are tri-flex fuel vehicles that run on CNG instead of methanol. These vehicles
have two separate tanks: one for a liquid gasoline and ethanol blend, and the other for
gaseous CNG. These tri-flex fuel vehicles are becoming popular not only because the price
of natural gas is low, but also because the government discounts the annual vehicle registration fee for car owners who convert their cars to include a natural gas fueling system.
(In Rio de Janeiro, there is a 75% reduction in annual vehicle registration fee, in So Paulo
it is 25%.)76 The cost of conversion ranges from R$1,200 to R$2,500 (USD 5801,208) in
Brazil; in the United States the price starts around USD 5,000.
These tri-flex fuel vehicles proved successful for some groups of drivers (e.g., taxi drivers
and individuals with short daily travel ranges). However, due to the short fuel mileage,
these vehicles have not been popular with long-distance drivers.
64
S e c t i o n 5 P o l i c y I s s u e s
Symposium participants discussed the issue of whether governmental policy intervention was
warranted to enable effective competition of alternative fuels in the LDV market. The discussion
also addressed questions regarding market failures caused by externalities and imperfect information, and the potential role of various policy instruments (standards and regulation, financial
incentives, and Research and Development (R&D) funding) in correcting the market imperfections.
Historical Evolution of Alternative Fuels Legislation

The federal government has intervened in alternative fuels issues in a significant way for a variety
of different policy reasons for at least the past three decades. Prompted by the oil crises of the
1970s and the growing awareness of the energy security and environmental issues raised by oil,
alternative fuels (specifically renewable fuels such as ethanol) were first advocated to reduce oil
import dependence. In the 1980s, leading up to the enactment of the Clean Air Act of 1990,
alternative fuels were advocated as a strategy to reduce urban air emissions. Over the past
decade, there has been interest in the expanded use of alternative fuels derived from biomass
as a measure to reduce net CO2 emissions in the transportation sector. Figure 30 illustrates the
enactment of statutory authority affecting alternative fuels.
Figure 30 Timeline of Federal Alternative Fuels Legislation
1970 Amendment
of Clean Air Act
Required EPA
to set up a
renewable fuel
program
Energy Policy Act

of 1992
Mandated 75% of
new light-duty vehicles
(LDVs) acquired
by certain federal fleets
to be alternative fuel
vehicles (AFV) by
FY2000
Interagency Commission
on Alternative Motor
Fuels called for
nationwide infrastructure for ethanol
and methanol
Energy Independence
and Security Act
Amended RFS (RFS2)
requirements for
ethanol, tightened CAFE
standards for LDVs,
established Advanced
Technology Vehicles
Manufacturing (ATVM)
program
RFS2: EPA mandated
36 billion gallons of
renewable fuel be
blended into gasoline
by 2022
American
Recovery and
Reinvestment Act
Extended and
reinstated alternative
fuel tax credits
1970
1992
2007
2009
1988
2005
2008
2011
Alternative Motor
Fuels Act
Established
CAFE standards for
alternative fuels
and the Interagency
Commission
on Alternative
Motor Fuels
Energy Policy Act

(EPAct) of 2005
Established tax
incentives and
loan guarantees
for GHG reducing
technologies, and
mandated blending
levels for biofuels
Emergency
Economic
Stabilization Act
Limits on ethanol
blends in
conventional
and FFVs
Tax credits
for ethanol
production and
blending expire
RFS1: EPA
mandated 7.5
billion gallons
of renewable-fuel
to be blended into
gasoline by 2012
Source: MITEI.
65
The United States has employed a variety of policy tools to stimulate the market for alternative
vehicles and fuels, including incentives, mandates, tax credits, loan guarantees, and fleet demonstration programs. As illustrated in Figure 31, Government intervention has occurred in all stages
of alternative fuel production and distribution as well as in AFV manufacturing and sales.
Policy mechanisms stimulating production of alternative fuels began with the Clean Air Act of
1970, which established the first renewable fuels program. The Energy Tax Act of 1978 created an
excise tax credit for ethanol in gasoline, which was expanded several times in the 1980s and
subsequently phased down beginning in the 1990s. In addition, legislation in the 1990s and early
2000s established a tax credit for ethanol producers, the VEETC, which expired at the end of 2011.
Measures to mandate the use of alternative fuels began with the Clean Air Act of 1990, which
established a minimum oxygen requirement for reformulated gasoline. Federal mandates were
restructured and greatly expanded with the enactment of the Energy Policy Act of 2005, which
established the RFS. The Energy Independence and Security Act of 2007 further expanded the
RFS mandate.
Figure 31 Policy Instruments Affecting All Elements of the Alternative Fuels Supply Chain
Policy Mechanisms for Alternative Fuels and Vehicles
Fuels
Supply
Chain
Entities
Vehicles
Policies
Feedstock suppliers
Land owners
Production comapnies
Mandates
Tax credits
Refineries
Processors
Mandates
Pipelines
Railroads
Transporters
Distributers
Tax credits
Mandates (e.g., RFS)
Policies
Entities
Supply
Chain
Production
Conversion
Distribution
Consumers
Consumption
Gas taxes
Tax credits
Mandates
Certifications
Tax credits
Loan guarantees
Components suppliers
Vehicle manufacturers
Manufacturing
Parts suppliers
Dealers
Repair shops
Tax credits
Mandatory
purchases
Sales &
Servicing
Fleet owners
Consumers
Customer
Source: MITEI.
66
Policy mechanisms stimulating production of AFVs by auto manufacturers began with the
Alternative Motor Fuel Act of 1988, which authorized credits for AFVs with the CAFE program. In
1989, the federal government initiated purchases of AFVs for the federal fleet. The Energy Policy
Act of 1992 mandated the purchase of AFVs by certain federal and state government fleets. The
Energy Policy Act of 2005 added new provisions for AFV acquisition, tax incentives for the development of alternative fuel infrastructure, and mandated alternative fuel use in AFVs. The Energy
Independence and Security Act of 2007 extended and expanded CAFE standards, including
changes in the provisions for AFV credits.
Discussion of the Key Current Policies

Symposium participants discussed two policy measures in detail: credits for AFVs as part of the
CAFE standards and the RFS programs.
Alternative Fuel Vehicle Credits in the Corporate Average Fuel Economy Program
Symposium participants focused on the role of CAFE standards in relation to alternative fuels and
AFVs. In addition, there was some discussion as to provisions in the new CAFE standards that
had the effect of encouraging the use of alternative fuels as a compliance option for meeting
CAFE requirements.
Under the CAFE program, for vehicles through model year 2011, FFVs received special consideration that assisted manufacturers in meeting the CAFE goal of 27.5 MPG. For the purposes of
calculating fuel economy, FFVs were assumed to operate on alternative fuels 50% of the time,
and each gallon of alternative fuel was assigned a value equivalent to 0.15 gallons of gasoline.
The combination of these two factors allowed the calculated MPG of a FFV (for CAFE compliance
purposes) to be increased by a factor of 6.67 relative to its actual MPG value using gasoline
100% of the time. For example, a FFV at 15 MPG when operated with gasoline would be credited
at 100 MPG for purposes of CAFE compliance.
Changes to the CAFE program in the 2007 Energy Independence and Security Act, as well as
changes to harmonize CAFE with CO2 tailpipe standards, have resulted in the phase-out of favorable conversion factors for FFVs. Beginning in model year 2016, manufacturers have to compute
MPG for FFVs based on actual consumption of alternative fuels. Thus, vehicle manufacturers
have a strong incentive to encourage owners of FFVs to actually use alternative fuel blends in
those vehicles, in order to develop a database for future CAFE compliance purposes.
In May 2010, President Obama set goals for the next round of CAFE standards, affecting model
year 20172025 LDVs. The proposed standard is much more stringent, calling on manufacturers
to increase average efficiency to 54.5 MPG by 2025. To help manufacturers achieve these stringent standards, some flexibility for determining MPG values for FFVs will be reinstated.
Participants noted that while the CAFE provisions for FFVs will become increasingly less favorable after model year 2012, CAFE provisions for EVs, PHEVs, and fuel cell vehicles are more
favorable. For model years 2012 and beyond, these vehicles will be rated solely on the basis
of downstream fuel use and emissions essentially zero. The Obama Administration CAFE
proposals for the 20172125 model years are more generous, allowing manufacturers to take
67
credit for two vehicles for each EV or fuel cell vehicle produced, and credit for 1.6 vehicles for
each PHEV produced. Several participants noted that these changes substantially tilt the policy
playing field toward electrification of the LDV market relative to the use of alternative liquid fuels
or CNG.
Alternative Fuel Mandates
Federal alternative fuel mandate have driven the growth of ethanol production over a 30-year
period from nearly zero to almost 14 billion gallons annually, as illustrated in Figure 32.
The Energy Policy Act of 2005 included a Renewable Fuel Standard (RFS1), which mandated a
minimum amount of alternate fuels to be blended into gasoline beginning in 2006. The Energy
Independence and Security Act of 2007 extended and greatly expanded the RFS mandate to
36billion gallons by 2022 (now known as RFS2). Total gasoline consumed in the US was approximately 133 billion gallons in 2011, thus the RFS mandate was equivalent to about 10% of total
gasoline consumed in 2011, increasing to almost 27% by 2022.
Figure 32 Growth of US Ethanol Production in Response to Federal Mandates
14
Billion Gallons
12
EISA 2007
10
8
6
Energy Policy Act 2005
Clean Air Act 1990
4
2
2009
2007
2005
2003
2001
1999
1997
1995
1993
1991
1989
1987
1985
1983
1981
Source: US Department of Agriculture, US on Track to Become Worlds Largest Ethanol Exporter In 2011,
http://www.fas.usda.gov/info/iatr/072011_ethanol_iatr.pdf
EISA 2007
RFS1 also created an incentive for refiners to utilize cellulosic biomass, providing them with 2.5
renewable fuel credits for cellulosic ethanol. This means50
that each
ethanol
50 gallon
50 50of cellulosicNet
Load (MW)
counts as 2.5 gallons of renewable fuel toward the RFS mandate. RFS1 also set a floor on the
300
400
500 which could
600
quantity of cellulosic200
ethanol to be included
in the overall
RFS mandate,
be adjusted 700
based on EPA estimates of cellulose ethanol production and total gasoline demand. The cellulose
ethanol floor in RFS1 was set at 250 million gallons in 2013. RFS2 substantially increased the
mandated floor for cellulose ethanol so that by 2022, half of RFS requirement would be satisfied
by cellulosic ethanol. The projected RFS2 mandate, along with its composition, is illustrated in
Figure
33.
2
68
2006
2002
1998
1994
1990
1986
1982
1978
1974
1970
1966
1962
1958
1954
1950
1946
1942
1938
1934
1930
1926
Figure 33 RFS2 Mandate and Composition

40
Historical Biofuels
Production
Mandated Use
Billion Gallons
30
20
Biodiesel
Unspecified Advanced Biofuels
10
Cellulosic Biofuel
Cornstarch Ethanol
Actual Biofuel Production
1995
2000
2005
2010
2015
2020
Source: Congressional Research Service, January 2012. Available at

http://www.nationalaglawcenter.org/assets/crs/R40155.pdf
Mandates are a hidden form of subsidy. In 2010 the Congressional Budget Office estimated that
the costs of the RFS2 mandate ranged from $1.78/gal for corn50
ethanol
50to $3/gal
50 50for cellulosicNet Load (MW)
ethanol, and that the implicit cost per ton of CO2 reduced is $750/metric ton for ethanol and $275/
ton for cellulosic ethanol.
2006
1998
2002
1986
1990
1994
1978
1982
1974
1966
1970
1946
1950
1954
1958
1962
1926
1930
1934
1938
1942
The impact of the RFS on gasoline prices appears to be small, at least up to the date of the
symposium. While some have speculated that the RFS increases average gasoline prices, EPA
estimates that it has reduced average prices by about 2.5 cents/gal, the equivalent of $10.25/yr for
an average vehicle owner (i.e., a 20-MPG vehicle driven 10,000 miles/yr).
2
40
30
20
69
2006
2002
1998
1994
1990
1986
1982
1978
1974
1970
1966
1962
1958
1954
1950
1946
1942
1938
1934
1930
1926
Symposium participants discussed the fact that ethanol did not have a clear fuel advantage over
other available biofuels based on its chemical properties. Rather, its primary advantage was in
1
having a more developed fuel production infrastructure resulting from its history of receiving
federal financial incentives as well as the RFS mandate. However, the prospects for further
expansion
of ethanol supply are being challenged by the possibility of limitation on the degree
0
of ethanol blending with gasoline.
The Unsuccessful Effort to Incentivize the Introduction of Methanol Fuel

The focus on ethanol as the principal alternative liquid fuel ignores the history of efforts by
the federal government to incentivize the introduction of methanol fuel. This effort originated in 1989, when the George H.W. Bush Administration, as part of a comprehensive
Clean Air Act rewrite, proposed to establish a mandatory methanol vehicle program in the
most polluted cities. The president also signed an Executive Order to mandate the purchase of methanol vehicles for the federal fleet.
The Clean Air Act of 1990, enacted by Congress one year later, established a reformulated
gasoline program with a 2% oxygen requirement in lieu of the methanol vehicle mandate.
Nonetheless, it was assumed at the time that the oxygenate would be either methanol or
ethanol, and notwithstanding the tax incentives for ethanol production, the cost of methanol was highly competitive with ethanol.
In the implementation of the oxygenate requirement, refiners opted to use MTBE as the
oxygenate of choice. In order to comply with the 2% oxygen requirement, refiners blended
11% by volume of MTBE into gasoline. Refiners and fuel distributors preferred MTBE
because it did not have certain problematic characteristics of ethanol and methanol, principally that is was not hydrophilic and was not corrosive. Thus, it could be blended into
gasoline at the refinery, avoiding the need for a dual distribution system and splash blending at storage terminals.
Environmental issues arose when spills of reformulated gasoline led to migration of MTBE
into groundwater. MTBE is highly miscible with water and malodorous. This issue caused a
number of states to ban the use of MTBE. Even so, the federal oxygenate requirement
remained in statute, and refiners turned to ethanol as the oxygenate of choice. Finally, the
Energy Policy Act of 2005 repealed the oxygenate requirement in favor of mandated use of
renewable fuels (i.e., the RFS mandate currently in place).
Methanol remains as an EPA-approved additive to gasoline, but has not achieved significant penetration because it does not qualify under the RFS mandate. Under EPA rules,
methanol and ethanol are considered to be fuel additives, and must have a waiver to be
permitted in gasoline blends. Methanol currently has a fuel additive waiver from the EPA so
that it can be used (with a co-solvent) in up to 5% blends with gasoline. However, the
waiver for methanol cannot be used in addition to the waiver for up to 10% ethanol blends
in gasoline. Since the RFS essentially requires 10% ethanol in all gasoline, and since methanol does not qualify for RFS compliance purposes, the use of methanol in gasoline currently is effectively blocked.
Thus, the most plausible scenario for introduction of large quantities of methanol fuel into
the LDV market is through the introduction of dedicated methanol mono-fuel vehicles or
the introduction of FFVs capable of utilizing methanol-rich blends, such as M85. The idea of
the GEM fuel blend presented to symposium participants represents a novel option for
introducing methanol into existing FFVs currently on the road.
70
Are Alternative Fuels Needed in the US Transportation Fuel Market?

The discussion of various new policy measures to support the deployment of alternative fuels
raised a more fundamental question by some symposium participants, namely, whether the
United States should seek to develop an alternative fuels market for LDVs. These participants
noted that new CAFE standards, combined with further advances in LDV technology, have the
potential to bring about substantial reductions in the demand for liquid fuels. This reduction in
demand provides an alternative policy approach to addressing energy security, national security,
and environmental objectives established at the outset of the symposium.
Several participants noted that conventional vehicles are expected to experience gradual vehicle
size and weight reduction of up to 15% by 2030 as a result of changes in vehicle design and the
use of lighter materials, which will directly impact projected fuel consumption and will improve
vehicle acceleration performance (projected to be a 10% increase). Based on current estimates
where a 10% reduction in vehicle weight gives a 6% reduction in fuel consumption the 15%
weight reduction by 2030 is projected to produce an 8% reduction in fuel consumption.
One of the participants, MIT Professor John Heywood, presented some projections from his
research team on the demand for different transportation fuel and the relative shares of different
types of vehicles. Figure 34 shows the volume of each fuel for a 100km-travel for different types
of fuel. A similar trend is expected for light-duty trucks except for a 30%40% higher amount of
fuel consumption due to the particular vehicles higher weight.
Figure 34 Fuel Consumption Forecast of the Average Car for Different Powertrains over Time to 2050
Relative Fuel Consumption of Cars

1.20
1.00
0.80
0.60
Gasoline Naturally
Aspirated Spark
Ignition (NA SI)
0.40
Gasoline Turbo
0.20
Diesel
PHEV
0.00
2005
2010
2015
2020
2025
2030
2035
2040
2045
Source: Bastani, Heywood, Hope, 2012.
1926
1930
1934
1938
1942
1946
1950
1954
1958
1962
1966
1970
1974
1978
1982
1986
1990
1994
1998
2002
2006
The impact of fuel efficiency on total demand is illustrated in Figure 35. The figure shows that by
2030, fuel economy could potentially reduce oil consumption by nearly half. Most participants
50similar
50 trade-offs;
50 50 reducingNet
(MW)
agreed that all available alternative fuels had reasonably
oilLoad
consumption by improving fuel economy provided a reasonably stable interim solution in addressing the
200
300
400
500
600
700
oil security problem.
MIT 2Energy Initiative Symposium on Prospects for Bi-Fuel and Flex-Fuel Light-Duty Vehicles | April 19, 2012
71
Figure 35 Impact of Fuel-Economy Standards on Light-Duty Oil Demand
Million Barrels per Day
12
10
AEO 2007
AEO 2010
(including 20102016 standards)
3% Annual Improvements
(20172025)
6% Annual Improvements
(20172025)
2010
2015
2020
2025
2030
Source: DOE, EIA, SAFE Analysis.
1926
1930
1934
1938
1942
1946
1950
1954
1958
1962
1966
1970
1974
1978
1982
1986
1990
1994
1998
2002
2006
There was some debate on the extent of the progress that has been made as a result of CAFE
standards, noting that it was often masked by higher oil prices. Participants also discussed the
fact that reliance solely on CAFE standards as the mechanism to reduce petroleum demand had
significant limitations. Two issues the rebound effect and new source bias were identified in
the discussion. The rebound effect is the tendency for individuals to drive more due to cheaper
operating costs, as increased fuel efficiency reduces the effective price per mile. This effect has
been estimated to be anywhere between 4.5% and 31%, and can offset efficiency gains. New
source bias refers to reduced purchases of new vehicles due to the higher vehicle prices associ50
50 50 50
Net Load (MW)
ated with more stringent efficiency regulations; the net result is that older, less-efficient vehicles
200 longer. National
300
400
500Administration
600
700
tend to stay on the roads
Highway Traffic
and Safety
(NHTSA)
and EPA estimates in support of the new CAFE standards assume a 10% rebound effect and a
-1 price elasticity of demand for vehicles. Some participants noted that this analysis does not
fully account for the impact on the used vehicle market or scrappage (though the negative price
elasticity picks up some of the reduced demand for these vehicles given higher prices).
2
72
2006
2002
1998
1994
1990
1986
1982
1978
1974
1970
1966
1962
1958
1954
1950
1946
1942
1938
1934
1930
1926
Participants reviewed analysis indicating that a policy solely focused on increasing fuel efficiency
has diminishing returns. For example, replacing a sedan in the fleet with a hybrid equivalent costs
the manufacturer
around $3,000 given current hybrid technology (although exact costs are
1
unknown, the difference in price between a hybrid car and its non-hybrid counterpart is approximately this amount). On the other hand, replacing an efficient non-hybrid vehicle with an electric
0
vehicle costs anywhere between $10,000 and $30,000, mostly due to the cost of the battery
(which, based on warranty information, is projected to need to be replaced more frequently than
the battery in a hybrid due to its usage and cycles). Yet, reductions in gasoline consumption are
much larger from replacing the sedan with the hybrid than replacing the efficient vehicle with an
electric. Consider two vehicles: a 12-MPG vehicle and a 30-MPG vehicle. Increasing the 12-MPG
vehicle by 2 MPG would result in fuel savings of 1.19 gallons per 100 miles driven. On the other
hand, increasing the 30-MPG vehicle to 40 MPG would result in 0.08 gallons of fuel saved per 100
miles (assuming no change in Vehicle Miles Traveled (VMT) if the rebound effect occurs, then
the 40-MPG vehicle will be driven more than the 30-MPG vehicle, thus reducing even further the
amount of gasoline saved).
Figure 36 demonstrates graphically the downward slope of these returns. This figure shows
gasoline savings by increases in fuel efficiency (under assumed VMT of 15,000 or 7,500). For
example, replacing a 20-MPG conventional gasoline vehicle with a 40-MPG hybrid vehicle would
save 375 gallons (at 15,000 VMT) at a cost of $3,000, while replacing it with a 100-MPG electric
vehicle would save 600 gallons at an average cost of $20,000. This implies marginal costs of
$8/galreduced vs. $33/gal reduced, demonstrating that although total gallons reduced are
higher, it is less efficient (in an economic perspective) to replace a conventional gasoline vehicle
with an EV compared to replacing it with a hybrid. An even larger number of gallons could be
saved by replacing a 10-MPG vehicle with a traditional gasoline 20-MPG vehicle and without
spending thousands of dollars to do so.
Figure 36 Annual Fuel Savings with Higher CAFE Standards
1,600
750 gal/yr
375 gal/yr
Gallons per Year
1,400
250 gal/yr
125 gal/yr
125 gal/yr
63 gal/yr
751 gal/yr
38 gal/yr
50 gal/yr
25 gal/yr
36 gal/yr
18 gal/yr
1,200
1,000
800
15,0
0
600
7,50
400
0m
i/yr
0m
i/yr
200
0
10
20
30
40
50
60
70
MPG
Source: William Chernicoff, Energy & Environmental Research Group Manager, Toyota Motor North American, Inc.
EISA 2007
Possible Additional Areas for Government Intervention

50
50
50 50
Net Load (MW)
200 strongly supported

300
400objective of achieving
500
600
A number of participants
the policy
a substantial
market penetration for alternative fuels in the LDV market. Participants discussed additional
measures to complement existing policies that aim to address externalities created by oil depenU.S. Average
Fuel Prices
dency, create a controlled environment to test new technologies, support
greaterRetail
investment
in
R&D, and better inform consumers of alternative fuels and vehicle choices.
2
2006
2002
1998
1994
1990
1986
1982
1978
1974
1970
1966
1962
1958
1954
1950
1946
1942
1938
1934
1930
1926
73
700
This discussion was framed by two concerns:

The
chicken and egg problem associated with what comes first: increasing alternative fuel
production and building fuel distribution infrastructure or promoting the manufacturing of
increased numbers of bi-fuel and flex-fuel vehicles. Vehicle manufacturers argue that there
are an insufficient number of refueling stations, which deters consumers from investing in the
vehicle, while fuel suppliers argue that there is insufficient demand for fuel to justify building
refueling stations. While not fully resolved, participants believed that an Open Fuel Standard
for vehicles may be the appropriate path forward to resolve this dilemma.
The
tendency to gravitate toward low-hanging fruit policies in which only fuel and vehicle
technologies that are currently economically viable are incentivized, to the detriment of
developing advanced technologies that may prove superior. Participants generally believed
that a mix of policies was appropriate, with strong emphasis on R&D and technology
advancement.
The principal alternative policy mechanisms discussed by the participants included the following:
Open Fuel Standard
A key regulatory option discussed by participants was the proposed Open Fuel Standard. The
Open Fuel Standard is a mandate that OEMs manufacture FFVs capable of operating on a variety
of fuels and fuel mixtures without the need for aftermarket adjustments. An Open Fuel Standard
is a broad-based mandate that avoids the issue of picking winners and losers among particular
combinations of alternative fueled vehicles, but rather would allow the market to decide the
economically viable options. Requiring flex-fuel capability on OEM vehicles also facilitates consumer acceptance because consumers would be able to purchase vehicles with confidence that
they meet all applicable environmental emissions standards and certifications.
The Open Fuel Standard Act of 2011 provides one possible blueprint for an Open Fuel Standard.
As proposed, the proposed legislation would require each OEM to manufacture a minimum
proportion of vehicles meeting the standard, on a mandated schedule of:

50% qualified vehicles in model year 2014;
80% qualified vehicles in model year 2016; and
95% qualified vehicles in model year 2017 and each subsequent year.
The legislation defines a qualified vehicle broadly to include:

A vehicle that operates solely on natural gas, hydrogen, or biodiesel;
An FFV capable of operating on gasoline, E85, and M85;
A PEV; or
A
74
vehicle propelled solely by fuel cell or by a technology other than an ICE.
Participants noted that the rapid phase-in schedule in the proposed legislation would favor
vehicle options that are developed and have sufficient production and distributional infrastructure in place.
Fuel Tax
While fuel economy standards could help stabilize domestic oil demand, the presence of a
rebound effect the tendency for people to drive more when they have a more efficient vehicle
could further negate its benefits. Some participants proposed implementing a fuel tax to curb
demand and directly holding consumers responsible for reducing their fuel consumption.
However, this could reduce the benefit of using alternative fuels and create long-term issues
with replacing lost revenue from reduced fuel consumption.
Government Fleet Programs
Federal, state, and local governments can take the lead with respect to government-owned and
operated LDV fleets. A controlled fleet program can provide a useful demonstration to test the
technical performance and economic competitiveness issues regarding alternative fuels. For
example, participants discussed a new initiative starting in Oklahoma and Colorado that is supported by 13 governors to help CNG vehicles gain market traction as well as consumer acceptance.
The proposal commits the governors to transitioning their state fleet vehicles to run on CNG and
work with auto manufacturers to help drive down the vehicle price point. Proponents note that
this initiative parallels policies for EVs, and is not meant to pick a winner.
Better Informing the Public
Some participants argued that rather than help a particular technology become commercially
viable, the governments role is in educating consumers and in making consumers internalize the
externalities associated with gasoline consumption to change their behavior. In doing so, the
most efficient vehicle would emerge naturally. Participants noted that certain alternative vehicle
technologies, namely methanol and CNG-powered vehicles, were more impacted by negative
consumer and public attitudes that sometimes could be attributed to sensationalized news.
Federal R&D Support
Federally funded R&D through agencies, including the National Science Foundation and US DOE,
such as ARPA-E, is critical in discovering newer and/or more affordable technologies.
CNG Refueling Programs
One often overlooked but important aspect of using CNG-powered vehicles is the refueling
process itself. As a compressed gas, CNG would require specific refueling tools that often require
proper training or otherwise could lead to dangerous leakages. Government could provide a
useful role in setting specific standards on this process, so as to minimize potential technology
incompatibilities.
Another option discussed by the participants is the possibility of financial incentives for purchasing a CNG home-refueling device, which would help reduce the cost barrier as well as mitigate
some of the problems with fuel accessibility.
75
Vehicle and Fuel Certification Process

Fuel certification and emission standards should be developed for this new type of tri-flex-fuel by
the EPA. This might be a very complex or time-consuming process for this kind of a complex
blend fuel.
76
endnotes
1
MITEI, Electrification of the Transportation System, Symposium Report, April 2010.
2 MITEI, Electrification of the Transportation System, Symposium Report, April 2010.

3 US Energy Information Administration, Annual Energy Review 2011, published September 2012.
4 US Energy Information Administration, Annual Energy Review 2011.
5
Steven E. Koonin, The Evolving Fuels Context for Light-Duty Vehicles, MITEI Symposium White Paper, April 2012.
6 Steven Koonin, Symposium White Paper.

7 Testimony of Dr. Gal Luft, Executive Director, Institute for the Analysis of Global Security, presented before the
House Committee on Foreign Affairs, December 16, 2011.
10 www.autoalliance.org/auto-economics, and www.api.org/policy-and-issues/policy-items/jobs/energy-works
11 Ulrich Kramer and James E. Anderson, Ford Motor Company, Prospects for Flexible- and Bi-Fuel Light-Duty
Vehicles: Consumer Choice and Public Attitudes, Symposium White Paper, April 2012.
14 MITEI, The Future of Natural Gas: An MIT Interdisciplinary Study, 2011.
15 MITEI, Electrification of the Transportation System, Symposium Report, April 2010.
16 Fuels and vehicles that were not discussed in the symposium, and are not further discussed in this report, include
butanol, coal-to-liquids, hydrogen, dual-fuel vehicles, dedicated vehicles, and biodiesel/diesel.
17 Dual-fuel vehicles are another type of alternative vehicle in which there are two independent fuel systems that can
run on both fuels simultaneously or on one fuel alone.
18 Others have defined drop-in fuel as not only being compatible with vehicles, but also with current infrastructure.
19 Recently, the EPA has tested and approved E15 for use in vehicles from model years 2001 and newer. However,
GreenWire reported on Jan 29, 2013, that a study conducted by the Coordinating Research Council (CRC) found
gasoline with 15% ethanol by volume (E15) damages critical fuel components in an automobile fuel system. In their
study, researchers used several car models the 2007 Nissan Altima, 2001 Chevrolet Cavalier, 2004 Ford Focus,
2003 Nissan Maxima, and 2004 Ford Ranger. The fuel components used in this study represent 29 million vehicles
that are currently on the road. The study found that E15 could completely fail fuel systems in some cars. Based on
this result, the American Petroleum Institute argues that the testing on E15 by EPA and DOE was not correct and
that the Renewable Fuel Standard should be repealed. Biofuels groups showed their disagreement with the result
of the study. The CRC study by contrast doesnt reflect a single mile driven, but rather, car components tested in
isolation. By researchers own admission, testing also included an aggressive E15 blend that includes more water
and acid than what consumers would use in their cars, said Fuel America, a coalition of biofuels supporters.
20 Ulrich Kramer and James Anderson, Symposium White Paper.
21 David J. Ramberg and John E. Parsons, The Weak Tie Between Natural Gas and Oil Prices, The Energy Journal,
Vol. 33, No. 2, 2012.
22 Michael D. Jackson, The Case for Bi-fuel Natural Gas Vehicles, Symposium White Paper, April 2012.
23 Oak Ridge National Laboratory, Transportation Energy Data Book, 30th Edition, 2011.
25 Paul Jarod Murphy, The Role of Natural Gas as a Vehicle Transportation Fuel, Masters Thesis, Massachusetts
Institute of Technology, June 2010.
26 Natural Gas Vehicle Association, http://www.ngvc.org/gov_policy/index.html
27 Energy Policy Act of 1992 Fleet Program, NGVAMERICA, http://www.ngvc.org/gov_policy/fed_regs/fed_
EPAFleetPrg.html
28 Fact Sheet: Converting light-duty vehicles to natural gas, NGVAMERICA, http://www.ngvc.org/pdfs/FAQs_
Converting_to_NGVs.pdf
29 Fact Sheet, converting light-duty vehicles to natural gas, NGVAMERICA, http://www.ngvc.org/pdfs/FAQs_
30 US Department of Energy, Alternative Fuels Data Center.
31 Michael Jackson, Symposium White Paper.
77
36 Fact Sheet: Converting light-duty vehicles to natural gas, NGVAMERICA, http://www.ngvc.org/pdfs/FAQs_

37 Paul Jarod Murphy, The Role of Natural Gas as a Vehicle Transportation Fuel, Massachusetts Institute of
Technology, June 2012.
39 This discussion is adapted from E-85 and Flex-Fuel Technology, Blaine M. Heisner, Southern Illinois University
Carbondale, 2008.
40 From Flexible Fuel Vehicles, Alternative Fuels Data Center (AFDC), http://www.afdc.energy.gov/vehicles/flexible_
fuel.html
42 Roberta J. Nichols, The Methanol Story: A Sustainable Fuel for the Future, Journal of Scientific and Industrial
Research, Vol. 62, JanuaryFebruary 2003.
43 Roberta J. Nichols, op. cit.
45 Vehicle changes for E85 Conversion, Coleman Jones, Clean Cities Webcast.
46 Elisheba Spiller, Regulations and Incentives for Alternative Fuels and Vehicles, Symposium White Paper,
April 2012.
47 MITEI, The Future of Natural Gas: An Interdisciplinary Study, 2011.
48 L. Bromberg and W.K. Cheng, Methanol as an Alternative Transportation Fuel in the US: Options for Sustainable
and/or Energy-Secure Transportation, Sloan Automotive Laboratory, Massachusetts Institute of Technology,
November 2010.
53 Molly F. Sherlock, Energy Tax Policy: Historical Perspectives on and Current Status of Energy Tax Expenditures,
Congressional Research Service, May 2011.
56 MITEI, The Future of Natural Gas: An MIT Interdisciplinary Study, 2011.
59 US Department of Energy, Clean Cities, Vehicle Technologies Program, http://www.afdc.energy.gov/uploads/
publications/clean_cities_overview.pdf
60 http://www.afdc.energy.gov/fuels/natural_gas_cng_stations.html#fastfill
61 Bromberg and Cheng, op. cit.
62 M.A. Weiss et.al., On the Road in 2020, MIT Energy Laboratory Report MIT EL-00-003, October 2000.
63 R. Abbott et. al., Evaluation of Ultra Clear Fuels from Natural Gas, Conoco Phillips, Nexant and Pennsylvania State
University, Final Report for the US Department of Energy, 2006.
64 MIT, The Future of Natural Gas.
65 Bromberg and Cheng, op. cit.
66 Nichols R J et al., Cold Start Studies, Canadian Flexible Fuel Vehicle Program, Final Report, Ontario Ministry
of Energy, March 31, 1987.
68 S. Yeh, An Empirical Analysis on the Adoption of Alternative Fuel Vehicles: The Case of Natural Gas Vehicles,
Energy Policy, 35(11):5865-5875.
69 MITEI, The Future of Natural Gas: An Interdisciplinary MIT Study, 2011.
70 X-Prize Foundation, National Household Travel Survey Data Summary for X-Prize, http://www.progressiveautoxprize.org/files/downloads/auto/AXP_FHWA_driving_stats.pdf. March 2007.
75 China: The Leader in Methanol Transportation, Methanol Institute http://www.methanol.org/Methanol-Basics/
Resources/China-Methanol.aspx
76 Natural Gas Flex Cars in Brazil, Anna Fitzpatrick, April 12, 2011, The Rio Times.
78
Index of Figures
Figure 1. Consumption of Alternative Fuels by Fuel Type, 2010 (Excludes E10) . . . . . . . . . . . . . 9
Figure 2. The Interaction of Policy, Technical, and Economic Constraints
on Alternative Vehicle-Fuel Adoption, Deployment, and Use . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Figure 3. Relationship among the Prices of Various Alternative Fuels . . . . . . . . . . . . . . . . . . . . 16
Figure 4. Modifications for CNG-Dedicated Vehicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Figure 5. Global Distribution of Mono-Fuel and Bi-Fuel NGVs, 2010 . . . . . . . . . . . . . . . . . . . . . . 21
Figure 6. Components to Convert and Operate Conventional Vehicles with CNG . . . . . . . . . . . 23
Figure 7. Special Features of the FFV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Figure 8. Illustration of Dielectric Flex-Fuel Sensor (by Duralast) . . . . . . . . . . . . . . . . . . . . . . . . 26
Figure 9. Concept of a Two-Tank Design for the FFV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Figure 10. Combinations of GEM Fuel Blends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Figure 11. Percent Change in GHG Emissions Relative to Gasoline . . . . . . . . . . . . . . . . . . . . . . . 34
Figure 12. Comparison of Fueling Infrastructure for Various Alternative Fuels . . . . . . . . . . . . . 35
Figure 13. Comparison of US Average Retail Fuel Prices per GGE . . . . . . . . . . . . . . . . . . . . . . . 37
Figure 14. US Production, Consumption, and Trade of Fuel Ethanol . . . . . . . . . . . . . . . . . . . . . . 38
Figure 15. Historical Relationship between E85 and Gasoline Prices . . . . . . . . . . . . . . . . . . . . . 39
Figure 16. US Ethanol Production Facilities and Areas of FFVs . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Figure 17. Schematic of US Rail and Truck Ethanol Distribution System . . . . . . . . . . . . . . . . . . 41
Figure 18. Proposed Dedicated Ethanol Pipeline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Figure 19. US Ethanol Refueling Stations and Areas with FFVs . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Figure 20. US Natural Gas Pipeline Compressor Stations Illustration, 2008 . . . . . . . . . . . . . . . 44
Figure 21. US CNG Refueling Stations and Interstate Highways . . . . . . . . . . . . . . . . . . . . . . . . . 45
Figure 22. US CNG Existing and Proposed Refueling Stations
and Clean Cities Coalitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Figure 23. Illustration of a CNG Fast-Fill Fueling Station . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
79
Figure 24. Illustration of a CNG Time-Fill Fueling Station . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

Figure 25. Conversion of Natural Gas to Alternative Fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Figure 26. Normalized Costs of Liquid Fuels, E85, Gasoline at the Gas Station,
and Estimated Costs of Methanol at the Station . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Figure 27. Scenarios for Pricing of Alternative Fuels Relative to Gasoline . . . . . . . . . . . . . . . . . 52
Figure 28. Possible Price Arbitrage under Conditions of Large Volumes
of Alternative Fuel Blends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Figure 29. Natural Gas and Gasoline Prices, 20042012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Figure 30. Timeline of Federal Alternative Fuels Legislation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Figure 31. Policy Instruments Affecting All Elements
of the Alternative Fuels Supply Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Figure 32. Growth of US Ethanol Production in Response to Federal Mandates . . . . . . . . . . . . 68
Figure 33. RFS2 Mandate and Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Figure 34. Fuel Consumption Forecast of the Average Car
for Different Powertrains over Time to 2050 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Figure 35. Impact of Fuel-Economy Standards on Light-Duty Oil Demand . . . . . . . . . . . . . . . . 72
Figure 36. Annual Fuel Savings with Higher CAFE Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
80
I n d e x o f Ta b l e s
Table 1. Types of Alternative Fuel Light-Duty AFVs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Table 2. Configurations of AFVs and Fueling Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Table 3. Cost Comparison of Dedicated NGV and Conventional Gasoline-Powered Vehicle . . . 21
Table 4. Various Types of CNG Fuel Tanks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Table 5. Approved Methanol Gasoline Blends with Requirements
for Co-Solvent Alcohols and Additives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Table 6. Comparison of Fuel Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Table 7. GEM Ternary Blend Fuels Used in the FFV Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Table 8. Conversion Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Table 9. January 2012 Overall Average US Retail Fuel Prices . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Table 10. Comparison of the Number of Refueling Stations in the US . . . . . . . . . . . . . . . . . . . . 38
Table 11. Ethanol Production Costs from Various US Feedstock Materials . . . . . . . . . . . . . . . . 39
Table 12. Cost of Adding E85 Fueling Capability to Existing Gasoline Stations . . . . . . . . . . . . . 43
Table 13. Cost for CNG Refueling Station . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Table 14. US Methanol Production, 2009 (Millions of Gallons) . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Table 15. Illustrative Methanol Production Costs, Relative to Gasoline
(Excluding Taxes) at $2.30 per Gallon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Table 16. Conditions for Price Decoupling with the Vehicle-Fuel Options . . . . . . . . . . . . . . . . . 54
Table 17. Comparison of Vehicle-Fuel Options from a Consumer Perspective . . . . . . . . . . . . . . 56
Table 18. Comparison of the Number of Refueling Stations in the US . . . . . . . . . . . . . . . . . . . . 57
Table 19. The Effect of $1 Increase in the Gasoline Price . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Table 20. Alternative Fuel Experience in Other Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Table 21. Minimum Cost-Benefit Recommended
for Different Alternative Transportation Fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Table 22. Summary of Vehicle-Fuel Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
81
APPENDICES

A. Glossary of Terms
B. Abbreviations/Acronyms

C. Comprehensive Summary of Vehicle-Fuel Options
D. Symposium Agenda

E. Symposium Participants
F. White Papers
82
Glossary of Terms
Vehicle Terms
Battery electric vehicle (BEV): A vehicle that uses batteries to store the electrical energy that
powers the motor.
Bi-fuel vehicle: A vehicle that is capable of operating on and switching between two fuels
generally gasoline or diesel and an alternative fuel that are stored in separate tanks. Unlike
a flex-fuel vehicle, a bi-fuel vehicle engine runs on one fuel at a time and the fuels are
not mixed.
Conventional gasoline vehicle: A vehicle that runs on conventional gasoline fuel.
CNG-dedicated vehicle: A vehicle that runs on only compressed natural gas.
CNG/Gasoline Bi-fuel vehicle: A vehicle that is capable of operating and switching between
CNG and gasoline that are stored in separate tanks.
Dedicated or mono-fuel vehicle: Any vehicle engineered and designed to be operated using
a single fuel.
Dual-fuel vehicle: A type of a FFV in which there are two independent fuel systems that can
operate on both fuels simultaneously or on one fuel alone.
Electric vehicle (EV): An electric vehicle (EV), also referred to as an electric drive vehicle, uses
one or more electric motors <http://en.wikipedia.org/wiki/Electric_motor> or traction motors
<http://en.wikipedia.org/wiki/Traction_motor> for propulsion <http://en.wikipedia.org/wiki/
Ground_propulsion>. Three main types of electric vehicles <http://en.wikipedia.org/wiki/Vehicle>
exist, those that are directly powered from an external power station, those that are powered by
stored electricity originally from an external power source, and those that are powered by an
onboard electrical generator, such as an ICE (a hybrid electric vehicle) or a hydrogen fuel cell.
Flex-fuel vehicle (FFV): A vehicle designed to run on more than one fuel, usually gasoline
blended with ethanol or methanol. The most common FFVs in the world use ethanol as their
alternative fuel source. Unlike bi-fuel vehicles, FFVs store two fuels in the same tank.
Heavy-duty vehicle: An on-road vehicle with a gross vehicle weight rating equal to or greater
than 26,001 pounds. Transit buses and large delivery trucks fall into this category.
Hybrid electric vehicle (HEV): A vehicle powered by 1) an ICE or other propulsion source that
can be run on conventional or alternative fuel and 2) an electric motor that uses energy stored in
a battery. Hybrid electric vehicles combine the benefits of high fuel economy and low emissions
with the power and range of conventional vehicles.
Light-duty vehicle (LDV): An on-road vehicle with a gross vehicle weight rating equal to or less
than 8,500 pounds. Automobiles, motorcycles, minivans, SUVs and other small pickups fall into
this category.
83
Medium-duty vehicle: An on-road vehicle with a gross vehicle weight rating between 8,501
and 26,000 pounds. Some larger cargo vans, pickup trucks, and maintenance trucks fall into this
category.
Natural gas vehicle (NGV): A natural gas vehicle (NGV) is an alternative fuel vehicle
<http://en.wikipedia.org/wiki/Alternative_fuel_vehicle> that uses compressed natural gas (CNG)
<http://en.wikipedia.org/wiki/Compressed_natural_gas> or liquefied natural gas (LNG)
<http://en.wikipedia.org/wiki/Liquefied_natural_gas> as a cleaner alternative to other fossil fuels
<http://en.wikipedia.org/wiki/Fossil_fuel>. Natural gas vehicles should not be confused with
vehicles powered by propane <http://en.wikipedia.org/wiki/Autogas> (LPG) <http://en.wikipedia.
org/wiki/Liquefied_petroleum_gas>, which is a fuel with a fundamentally different composition.
Worldwide, there were 14.8 million natural gas vehicles by 2011.
Plug-in hybrid electric vehicle (PHEV): A plug-in hybrid electric vehicle (PHEV), plug-in
hybrid vehicle (PHV), or plug-in hybrid is a hybrid vehicle <http://en.wikipedia.org/wiki/Hybrid_
electric_vehicle> which utilizes rechargeable batteries <http://en.wikipedia.org/wiki/
Rechargeable_battery>, or another energy storage device, that can be restored to full charge by
connecting a plug to an external electric power <http://en.wikipedia.org/wiki/Electric_power>
source (usually a normal electric wall socket <http://en.wikipedia.org/wiki/AC_power_plugs_and_
sockets>). A PHEV shares the characteristics of both a conventional hybrid electric vehicle, having
an electric motor <http://en.wikipedia.org/wiki/Electric_motor> and an internal combustion
engine (ICE) <http://en.wikipedia.org/wiki/Internal_combustion_engine>; and of an all-electric
vehicle <http://en.wikipedia.org/wiki/All-electric_vehicle>, having a plug <http://en.wikipedia.org/
wiki/AC_power_plugs_and_sockets> to connect to the electrical grid <http://en.wikipedia.org/
wiki/Electrical_grid>. Most PHEVs on the road today are passenger cars, but there are also PHEV
versions of commercial vehicles and vans, utility trucks, buses, trains, motorcycles, scooters
<http://en.wikipedia.org/wiki/Scooter_(motorcycle)> , and military vehicles.
Tri-flex fuel vehicle: A vehicle that is capable of operating on a blended mixture of gasoline
and two alternative fuels in a single tank. In the symposium, primarily gasoline/ethanol/methanol
vehicles were considered.
Fuel Terms
Alternative fuel: Any fuel material that is not conventional fuel. Alternative fuels for transportation include methanol, denatured ethanol, compressed or liquefied natural gas, liquefied petroleum gas (propane), hydrogen, coal-derived liquid fuels, cellulosic biofuel, and electricity.
Biodiesel: Vegetable oil or animal fatbased diesel fuel. Biodiesel can be used alone or as a
mixture with diesel fuel in any diesel engines.
Conventional (traditional) fuel: Fuel that is petroleum-based (e.g., gasoline and diesel).
Drop-in fuel: Fuel that can currently be blended with conventional petroleum-based fuel
(gasoline, diesel) and used in conventional gasoline- or diesel-powered vehicles without requiring
vehicle modifications (e.g., E5-E15, M5, and biodiesel).
84
Drop-out fuel (non-drop-in fuel): Fuel that is not drop-in fuel. There are two types of drop-out
fuels:

lendable drop-out fuel: Fuel that can be blended with gasoline but requires vehicle and/
B
or infrastructure modifications for use (e.g., E85, M85).

Physical
drop-out fuel: Fuel that cannot be blended with conventional petroleum-based

fuel (e.g., electricity, CNG).
Ethanol blend fuel: A mixture of liquid ethanol and gasoline in various ratios. E numbers
describe the percentage of ethanol fuel in the mixture by volume. For example, E15 is 15%
anhydrous ethanol and 85% gasoline by volume.
Methanol blend fuel: A mixture of liquid methanol and gasoline in various ratios. M numbers
describe the percentage of ethanol fuel in the mixture by volume.
Tri-flex fuel: A fuel mixture of gasoline and two alternative fuels in a single tank. In the symposium,
a tri-flex fuel composed of gasoline, ethanol, and methanol was discussed.
XTL: Any alternative liquid fuel produced from conversion of a solid or gaseous feedstock. This
includes, Coal-to-Liquids (CTL), Gas-to-Liquids (GTL), and Coal/Biomass-to-Liquids (CBTL).
85
ABBREVIATIONS / ACRONYMS

AFDC
Alternative Fuels Data Center

AFV
Alternative Fuel Vehicle

AKI
Anti-Knock Index

ARPA-E
Advanced Research Projects Agency Energy

BEV
Battery Electric Vehicle

CAFE
Corporate Average Fuel Economy

CBL Coal/Biomass-to-Liquids

cf
Cubic Feet

CNG
Compressed Natural Gas

CTL Coal-to-Liquids

DME
Dimethyl Ether

DOE
Department of Energy

DOT
Department of Transportation

E10
Low-level Blend, 10% Ethanol, 90% Gasoline

E85
Ethanol Fuel Blend

EIA
Energy Information Administration

EPA
Environmental Protection Agency

EU
European Union

EV
Electric Vehicle

FFV
Flex-Fuel Vehicle

gal Gallon

GDP
Gross Domestic Product
GEM Gasoline/ Ethanol / Methanol

GGE
Gallon of Gasoline Equivalent

GHG
Greenhouse Gas

GREET
Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation

GTL Gas-to-Liquids

HEV
Hybrid Electric Vehicle

ICE
Internal Combustion Engine

LDV
Light-Duty Vehicle

LNG
Liquefied Natural Gas

LPG
Liquefied Petroleum Gas

M5
Methanol with Gasoline

M100
Pure Methanol

mcf
Thousand Cubic Feet

MON
Motor Octane Number

MPG
Miles Per Gallon
MTBE Methyl-Tertiary-Butyl-Ether

NA SI
Naturally Aspirated Spark Ignition

NEVC
New England Vehicle Council

NGV
Natural Gas Vehicle

NHTSA
National Highway Traffic Safety Administration

NREL
National Renewable Energy Laboratory

OEM
Original Equipment Manufacturer

OPEC
Organization of the Petroleum Exporting Countries

PCM
Powertrain Control Module

PHEV
Plug-in Hybrid Electric Vehicle

psi
Pounds per square inch

R,D,&D
Research, Development, and Deployment

R&D
Research and Development

RFS
Renewable Fuels Standard

RON
Research Octane Number

SUV
Sport Utility Vehicle

tcf
Trillion Cubic Feet

USDA
US Department of Agriculture

VEETC
Volumetric Ethanol Excise Tax Credit

VMT
Vehicle Miles Traveled
86
Comprehensive Summary of Vehicle-Fuel Options

This section provides a comprehensive summary of possible alternative fuels and vehicles
options. These options include all the ideas proposed by the participants from the symposium
as well as those reflected in the discussions after the symposium. The options are listed below, in
Table 22 and then discussed in more detail.
Table 22 Summary of Vehicle-Fuel Combinations
Option #
Fuel Options
Vehicle Options
Increase Vehicle Energy Efficiency
Conventional gasoline vehicle
E5E15
M5 (maximum volume allowed, from 1986)
Biodiesel up to 100%
Conventional diesel vehicle
Drop-in Fuels
Drop-out Fuels
5
Ethanol (E16-E85)
FFV
Methanol (M6-M85)
FFV for methanol
Tri-flex fuel
(Gasoline + ethanol + methanol)
Tri-flex fuel vehicle
CNG
CNG-dedicated vehicle
Blendable drop-out fuels
Physical drop-out fuels
9
10
Bi-fuel vehicle
Electricity
11
Electricity-dedicated vehicle
Hybrid electric vehicle
Source: MITEI.
Except for the biodiesel option in the Drop-in Fuels category, the other two fuel options (E5-E15
and M5) cannot achieve gasoline price decoupling from the price of corresponding alternative
fuels. This is because at E15 and M5 level, the total supply of either ethanol or methanol is too
small so that the demand and the supply curve of both ethanol and methanol intersect at points
where the price of either the two fuels is higher than the price of gasoline.
However, it is important to remember that those two options are still beneficial to the society,
since they diversify fuel options for consumers. The existence of an alternative fuel compared
with the gasoline-only situation creates 1) consumer surplus (until the point when the methanol
supply curve is smaller than the world gasoline price) and 2) fuel arbitrage.
Option 1. Increase Vehicle Energy Efficiency
Compression-ignition engine with diesel fuel operates with about 20% higher efficiency compared to non-turbocharged gasoline engines on an energy-equivalent basis and about 30% higher
efficiency on a fuel-volume basis. Efficiency of diesel engines will gradually improve but the
increase will not be as high as the increase gain in gasoline engines. Transmission efficiency of
diesel vehicles will improve by about 10% as powertrain incorporates more efficient shifting
mechanisms. Currently the United States does not extensively use diesel and on a fuel-volume
basis, diesel is more expensive than gasoline.
87
Vehicles for Drop-in Fuels

Option 2. E5-E15 Fuel + Conventional Gasoline Vehicle
Currently up to E15 (15% of ethanol by volume) can be used for conventional gasoline vehicles
without vehicle modifications. It was the general view of participants that at this maximum level
of ethanol blend, ethanol price is coupled with the gasoline price since the ethanol supply is not
large enough.
Option 3. M5 + Conventional Gasoline Vehicle
In the United States, the maximum level of methanol that can be blended with gasoline for
conventional gasoline vehicles is set at 5% by volume. Again, the amount of total methanol
supply is restricted at a certain level such that the price of methanol in the market is determined
at a point higher than the price of gasoline. Therefore, the methanol price is coupled with the
gasoline price.
Option 4. Biodiesel up to 100% + Conventional Diesel Vehicle
Biodiesel refers to a vegetable oil or animal fatbased diesel fuel and it can be used with and
without petro-diesel. In other words, biodiesel can be used from 0% to 100% by volume for
conventional diesel vehicles without vehicle modification. In this sense, biodiesel is a perfect
drop-in fuel.
For biodiesel to bring about price decoupling with diesel price, the supply curve of biodiesel
should intersect with the demand curve at a price lower than the price of diesel. For this scenario
to happen, the number of diesel vehicles needs to substantially increase in the United States.
Vehicles for Drop-out (Non-Drop-in) Fuels

According to the definition of a drop-in fuel, there exist two different types of drop-out fuels.
First, fuels that can be blended with gasoline but require vehicle modifications are classified as
drop-out fuels. This includes ethanol (E16-E85), methanol (M6-M85), and tri-flex fuel (mixture of
gasoline + ethanol + methanol). Second, fuels that are not physically blendable with gasoline are
another type of drop-out fuels. Compressed natural gas and electricity are the examples.
A. Blendable Drop-out Fuel Options
Option 5. Ethanol (E16-E85) + Current FFV
Current FFVs can run on ethanol and gasoline blend up to 85% of ethanol by volume (E85).
Ethanol price decoupling with the price of gasoline can occur when the demand curve and the
supply curve of ethanol intersect at a point where the ethanol price is lower than the price of
gasoline. This scenario requires a wider distribution of FFVs, an increase in ethanol production.
88
Option 6. Methanol (M6-M85) + FFV for Methanol

Current FFVs are designed and certified to run on ethanol blends but not on methanol blends.
Methanol fuel is corrosive to engines and fuel lines in vehicles, therefore, high levels of methanol
as fuel requires vehicle modifications. Methanol price decoupling with the price of gasoline
occurs when the demand curve and the supply curve of methanol determine the methanol price
that is lower than the gasoline price. For this to be realized, both the number of FFVs for methanol
and the production of methanol fuel need to be increased substantially.
Option 7. Tri-Flex Fuel (Gasoline + Ethanol + Methanol) + Tri-Flex Fuel Vehicle
Tri-flex vehicles are capable of operating on a fuel mixture of gasoline and two other alternative
fuels in a single tank. At the symposium Turner et al. from Lotus Engineering argued that a fuel
blend of gasoline, ethanol and methanol (called GEM blend) can be produced in a way that the
blend has the same stoichiometric property as that of E85 and as a result, the difference between
this new blend and E85 is invisible to a current FFV. They also argued that producing vehicles that
can run on the GEM blend is not highly challenging since current FFVs have been already tested
with M100. In addition, since there are three fuel sources, the opportunities for fuel arbitrage
are maximized.
B. Physical Drop-out Fuels Options
Option 8. CNG + CNG-Dedicated Vehicle
CNG-dedicated vehicles run on only CNG. Compressed natural gas requires high compression
(typically at 3,0003,600 psi) for storage, which makes it more expensive to operate refueling
stations. In 2009, the United States had 114,270 CNG vehicles and most of them were buses.
Although CNG is a physical drop-out fuel, whether its price can be easily decoupled with the
price of gasoline is less certain because of the historical link between the price of natural gas and
the price of oil in the United States. There is no fuel arbitrage opportunity since CNG is the only
fuel source.
Option 9. CNG / Gasoline + Bi-Fuel Vehicle
CNG/Gasoline bi-fuel vehicles are capable of operating and switching between CNG and gasoline
that are stored in separate tanks. Many gasoline-powered vehicles can be converted to bi-fuel
vehicles in the aftermarket. Due to the same reason mentioned in the previous case, the price
decoupling effect is not certain. Fuel arbitrage opportunities exist since there are two fuel options.
Option 10. Electricity + Electricity-Dedicated Vehicle (BEV)
Battery electric vehicles use batteries to store the electrical energy that powers the motor. Battery
electric vehicles rely solely on electricity. Since electricity is a physical drop-out fuel and the
overlap between the electricity market and the gasoline market is minimal, the price decoupling
is naturally achievable. Fuel arbitrage opportunities do not exist since electricity is the only fuel
source.
89
Option 11. Electricity/Gasoline + Hybrid Electric Vehicle

Hybrid electric vehicles are powered by 1) an ICE or other propulsion source that can be run on
conventional or alternative fuel and 2) an electric motor that uses energy stored in a battery. Due
to the same reasons mentioned in the BEV case (Option 10), price decoupling is easily achievable.
There is fuel arbitrage opportunity since there are two fuel options. However, the benefit is
limited because HEVs cannot only rely on electricity due to their small battery capacity.
90
Symposium Agenda
2012 MITEI Symposium
Prospects for Bi-Fuel and Flex-Fuel Light-Duty Vehicles
Boston Marriott Cambridge
Two Cambridge Center, 50 Broadway, Cambridge, MA
April 19, 2012
8:309:00
Breakfast
9:0010:30
Framing the Issues
Policy
White Papers/Speakers:
Overview
Dr. Steven E. Koonin, Institute for Defense Analysis

Prof. John Heywood, MIT

Discussant #1: Prof. Bruce Dale, Michigan State University

Discussant #2: Dr. Gal Luft, Institute for the Analysis of Global Security
10:3010:45 Morning Break

10:4512:45 Vehicles
Bi Fuel
White Paper/Speaker:

Mr. Michael D. Jackson, TIAX LLC

Discussant #1: Prof. William Green, MIT

Discussant #2: Mr. Richard Kolodziej, NGVAmerica

Discussant #3: Mr. Kevin Stork, US DOE
Tri-Flex Fuel White Paper/Speaker:

Dr. James Turner, Lotus Powertrain

Discussant #1: Dr. Jos Coelho Baeta, Fiat South America

Discussant #2: Mr. Norman Brinkman, General Motors

Discussant #3: Dr. Daniel Cohn, MIT
12:451:30
Lunch Break
1:302:30
Consumer Choice and Public Attitudes


Dr. Ulrich Kramer and Dr. James E. Anderson, Ford Motor Company

Discussant #1: Dr. Alicia Birky, TA Engineering, Inc.

Discussant #2: Prof. Stephen Ansolabere, Harvard University
2:302:45
Afternoon Break
2:454:00
Policy & Regulation


Dr. Elisheba Beia Spiller, Resources for the Future

Discussant #1: Prof. Christopher Knittel, MIT

Discussant #2: Mr. Ronald Minsk, Securing Americas Future Energy

Discussant #3: Mr. Jay Albert, Deputy Secretary of Energy, State of Oklahoma

Discussant #4: Mr. Michael Carr, Senior Counsel, Senate Committee on Energy
and Natural Resources
4:004:30

Closing Discussion
Chair:
Prof. Ernest Moniz, MIT Prof. John Deutch, MIT
91
S y m p o s i u m Pa r t i c i pa n t s
Prospects for Bi-Fuel and Flex-Fuel Light-Duty Vehicles
Boston Marriott Cambridge
Two Cambridge Center, 50 Broadway, Cambridge, MA
April 19, 2012
Jay Albert Office of the Secretary of Energy,
Discussant
State of Oklahoma
James Anderson
Ford Motor Company
Speaker
Stephen Ansolabehere
Harvard University
Discussant
Robert Armstrong
MIT
Alicia Birky
TA Engineering, Inc.
Discussant
John Bradley
EBS, LLC
Norman Brinkman
General Motors
Discussant
Leslie Bromberg
MIT
Michael Carr US Senate Committee on Energy
Discussant
and Natural Resources
Alice Chao
MIT
William Chernicoff
Toyota Motor North America
Eric Chow
MIT
Andrew Cockerill
BP
Jos Coelho Baeta
FIAT
Discussant
Daniel Cohn
MITEI
Discussant
Bruce Coventry
Nostrum Motors
Bruce Dale
Michigan State University
Discussant
Nicola De Blasio
Eni
John Deutch
MIT Closing
Comments
Carmine Difiglio
US Department of Energy
Greg Dolan
Methanol Institute
Michael Gallagher
Westport Innovations Inc.
Ahmed Ghoniem
MIT
Karen Gibson
MITEI
William Green
MIT
Discussant
Tiffany Groode
IHS
John Heywood
MIT
Joseph Hezir
MITEI
Ben Iosefa
Methanex Corporation
Michael Jackson
TIAX LLC
Speaker
Melanie Kenderdine
MITEI
Chris Knittel
MIT
Discussant
Michael Knotek Renewable and Sustainable Energy Institute
(RASEI)/CU Boulder
Richard Kolodziej
NGVAmerica
Discussant
92
Steven Koonin Institute for Defense Analyses Science

Speaker
and Technology Policy
Ulrich Kramer
Ford
Speaker
Leebong Lee
MIT
Gal Luft
US Energy Security Council
Discussant
Richard MacMillan
MIT
Rebecca Marshall-Howarth
MITEI
Ronald Minsk
Securing Americas Future Energy
Discussant
Ernest Moniz
MITEI Closing
Comments
Francis OSullivan
MITEI
Matthew Pearlson
MIT
Scott Rackey
OsComp Systems
Matt Roberts
Methanol Institute
Jess Rodriguez
Exelon
Pedro Santos
OsComp Systems
Leonardo Silva
Universidade de Brasilia
Brian Siu
Natural Resources Defense Council
Elisheba Beia Spiller
Resources for the Future
Discussant
Gregory Stephanopoulos
MIT
Kevin Stork
US Department of Energy
Discussant
Tom Stricker
Toyota Motor North America
Steve Tullos
Entergy
James Turner
Lotus Engineering
Speaker
Kaushik Vyas
Nostrum Power
John Wall
Cummins
Craig Wildman
Chevron
Stephen Zoepf
MIT ESD
93
The Evolving Fuels Context for Light-Duty Vehicles

Steven E. Koonin
April, 2012
Abstract
This paper discusses the evolving fuels context for light-duty vehicles, with an
emphasis on supply and infrastructure. Key points of techno-economic analyses of diverse
liquid hydrocarbons (crude-derived, XTL, and biofuels) and non-liquid fuels (natural gas,
hydrogen, and electricity) are compared. The challenges in effecting a transition to any
alternative fuel are discussed.1
1. Introduction
The 240 million light duty vehicles (LDVs) in the US are powered almost exclusively
by crude-derived gasoline-like fuels. These passenger cars and light trucks consume some
130 billion gallons of fuel annually and account for 8.7 million barrels per day (Mbpd) of
liquids2 consumption (45% of the US total).
Gasoline has dominated US LDVs for several reasons: its high energy density (some
30 times greater than batteries), its convenience, safety, and historical low cost, the
interdependence among fuel production/distribution/vehicles, as well as consumer
expectations and incumbent business interests.
Yet present circumstances have
drawbacks, mostly importantly that the US imports about 45% of the liquids it consumes
(Figure 1).
Figure 1: US Liquids History, 1949-2010. Source: EIA

1
Much of the material in this paper derives from the recent DOE Quadrennial Technology Review (QTR). A report
on that process can be accessed at http://energy.gov/sites/prod/files/ReportOnTheFirstQTR.pdf and the
companion Technology Assessments are expected to be released in the future.
2
Liquids include crude oil, natural gas liquids, and biofuels
Indeed, concerns about foreign oil and a quest for energy independence have garnered
the resolve of every president for the past four decades, as illustrated in Figure 2.
Figure 2: US liquids imports. Import fraction is given at the time of each presidential quote.
Todays oil problem can be parsed into several related but distinct sub-problems:
Physical security: While roughly half of US imports come from the Western
Hemisphere (most importantly Canada, Mexico and Venezuela), the other half
arrives via longer and less secure routes (Figure 3). Any extended physical
interruption of supply would have severe consequences for the Nation; the Strategic
Petroleum Reserve stores 700 M bbl (70 days of imports) against such
contingencies.
Figure 3: Origin of US petroleum imports, 2010. Source: EIA, Petroleum Supply Monthly
Trade deficit: At a price of $100/bbl, US oil imports debit the balance of trade by
almost $1B/day. In 2010, US exports were $1.8T, while imports were $2.3T, for a
$500B trade deficit, of which oil-related imports were half, the largest single debit
item.
Price volatility: Gasoline is one of the few globally-traded commodities that US
consumers purchase directly, keeping its price in the public consciousness. That
price, driven by the underlying crude price, has risen dramatically in the past
decade and remains volatile, as shown in Figure 4.
Figure 4: US retail gasoline price and light sweet crude price. Abscissa is year, left ordinate
$/gal, right ordinate $/bbl Source: CNBC, derived from EIA/DOE/WSJ/Haver data. Accessed at
http://media.cnbc.com/i/CNBC/Sections/News_And_Analysis/__Story_Inserts/graphics/__CHARTS_SPECIAL
/MISCELLANEOUS/CNBC_US_RETAIL_GASOLINE_PRICE_520.gif
Greenhouse gas emissions: LDVs account for about 1/3 of US energy-related

emissions. Although the stationary energy sector (heat and power) has larger and
more price-sensitive3 opportunities for mitigation, any serious commitment to
reduce emissions will entail changes to the LDV sector.
Demand-side measures that reduce oil use, such as price signals, regulations (e.g., the
recently strengthened CAFE standards), and behavior (e.g., shifts to public transport) offer
some of the most timely, material, and cost-effective solutions to all of the oil subproblems. But there are also supply-side actions that can be categorized as:
Increased domestic production of crude
Increased production of alternative liquid fuels (XTL4, biofuels)
Transition to a non-liquid fuel, such as natural gas, electricity, or hydrogen
Each of these measures must be judged in terms of its technical feasibility, materiality,
timeliness, economics, and the varying proportions in which it addresses the different subproblems.
For example, a carbon price of $40/t CO2 would be sufficient to induce a shift away from coal-fired power to lowGHG generation technologies. Yet it corresponds to only a $0.35 increase to the price of gasoline.
4
Thermochemical processes that convert coal, gas, or other organic feedstocks into liquid hydrocarbon fuels
After a review of relevant aspects of the oil scene, this paper considers each of the
LDV supply-side options in turn, with an emphasis on the fuels and infrastructure. The
companion white paper by John Heywood emphasizes vehicle issues.
2. Liquid fuels
A holy grail supply-side solution to all of the oil sub-problems would be a low-cost,
low-carbon, liquid fuel compatible with existing infrastructure and vehicles (drop-in)
that could be produced domestically at scale. Unfortunately, technology, economics, and
politics conspire to make such a solution currently unimaginable, so that trade-offs will be
necessary. A brief synopsis of the current oil scene is an important prelude to discussions
of specific solutions.
2.1 Todays oil scene
Crude oil5 has been produced and consumed for more than 150 years. It is a finite
resource traded on a global market with one price (varying somewhat with quality and
location) set by demand and supply, the latter modulated by the OPEC cartel.
World liquids demand, currently about 87 Mbpd, is projected to rise by almost 1
Mbpd every year, driven largely by economic development in Asia, as shown in Figure 5.
EIA projections, necessarily an imprecise art, show the OPEC cartel continuing to satisfy
about 40% of that demand for the next 25 year, and a growing wedge of unconventional
production (tar sands, tight oil).
Figure 5: Global liquids production, 1990-2035. Source: EIA

5
The terms oil and liquids are used interchangeably for convenience; a note will be made where the
distinction is important
High oil prices invariably increase the volume of an on-going discussion best
paraphrased as were running out! But an average global consumption of 100 Mbpd over
the next 25 years requires 1,000 B bbl (about what the world has consumed over the past
century), easily accommodated at reasonable cost by the supply curve shown in Figure 6.
However, the substantial investment to produce that increasingly difficult oil will require
new technology, access to resource, and favorable business cases.
Figure 6: Long-term oil-supply cost curve. Shale oil is an additional significant resource that is
not included on this four-year-old chart. Source: IEA WEO 2008, p. 220
Although there is more than enough hydrocarbon in the ground, the easy oil
reserves (those with a low cost of production) are increasingly concentrated in the hands
of a few countries, as shown in Figure 7.
Figure 7: Percentage of conventional reserves in key NOC (National Oil Company) hands.
Source: Morgan Stanley
The decisions and fates of countries like Venezuela, Saudi Arabia, Iran, and Russia are
therefore important elements in thinking about security of supply. More costly, but more
significant, opportunities are distributed differently around the globe. Of particular
interest for the US are the North American tar sands and oil shales.
Figure 8 below shows the history and projections of US liquids supply. Total
consumption is expected to be essentially flat, as improvements in vehicle efficiency are
offset by growing vehicle miles travelled6. Biofuels and Natural Gas Liquids (NGLs) are
expected to grow as imports decline modestly. Most surprising is how little change is
projected over 25 years, but EIA projections, by design, make conservative technology and
regulatory assumptions.
Figure 8: History and projection of US liquids consumption and sources. Source: EIA
2.2 Increased domestic crude production

The US is the worlds largest consumer of oil (22%), but even as the worlds third
largest producer (after Saudi Arabia and Russia), it provides only 11% of global supply.
The security and balance of trade impacts of increased domestic production are scaled by
US imports (9 Mbpd), but as oil is a freely-traded global commodity, the price impact must
be judged against the global 87 Mbpd, growing 1 Mbpd each year (white line at the bottom
of Figure 5.) There is no GHG benefit of increased domestic production, and likely a slight
drawback, as the production, refining, and use of more difficult oils generates at least 10%
more life-cycle GHGs than conventional crude.
A convenient threshold of materiality for increased US production is 1 Mbpd. Figure
1 shows that, following a more rapid run-up before 1970, changes of that magnitude have
taken about a decade. More detail can be seen in Figure 9, which shows that, even with
great effort, it has taken a decade to bring on 1 Mbpd production in either the Gulf of
6
US population is projected to increase at 1% per annum through mid-Century.
Mexico or through corn ethanol (energy equivalent). It is sobering to view these histories
against the much larger and more rapid changes in Saudi production also shown.
Figure 9: Saudi, US Gulf of Mexico, and corn ethanol production, 1993-2011. Source: EIA
A 2011 National Petroleum Council study7 attempted to quantify the potential for
increased North American production. The 2035 High Potential case shown in Figure 10
Figure 10: Limiting projections for NA liquids production. Source: NPC NARD report, Figure 15. p. 49
Prudent Development: Realizing the Potential of North Americas Abundant Natural Gas and Oil Resources,
(NARD) available at http://www.npc.org/. Given the general alignment and tight coupling of the North American
economies, it is not unreasonable to treat them as a unit when discussing security and balance of trade.
compounds favorable assumptions about technology, access, infrastructure, and regulation.

Oil sands, NGLs, and tight oil are the largest contributors to a dramatically increased
production. None of these sources are without its drawbacks, although all three are
currently growing rapidly [tight oil production has almost already hit this projection].
While US (or North American) production during 1950-1970 grew at a rate comparable to
that implied by the High Potential case (more than doubling over 25 years), it has declined
slightly at a much slower rate during the past 40 years. If realized, the High Potential case
would improve security, balance of trade, and is material enough to impact global price, but
would deteriorate progress on GHG emissions.
2.3 XTL8
Various thermochemical and biological processes can be exploited to turn any
carbonaceous material into a liquid fuel, as shown in Figure 11. As with increased oil
domestic production, the desideratum of a drop in product fully compatible with gasoline
vehicles and fueling infrastructure means that 1 Mbpd-scale deployment can improve
security and trade, but will not lower price and will not necessarily reduce GHG emissions.
This section covers thermochemical processes that convert coal, gas, and biomass to fuel,
while biological processes are covered in the following section.
Figure 11: Summary of feedstocks, pathways, and products for alternative hydrocarbon fuels
Viable domestic deployment of Coal-to-Liquids, Gas-to-Liquids, or Coal/Biomass-toLiquids (respectively CTL, GTL, and CBTL, and collectively XTL) requires a material and
economic domestic carbonaceous feedstock. Figure 12 shows annual US carbon flows in
various streams. Assuming a 50% carbon efficiency, the nominal 1 Mbpd, requires about
100 MtC/yr, which could be accommodated by modest increases in the coal, natural gas, or
biomass streams.
The GHG impacts of various XTL schemes are shown in Figure 13. Apart from
unsequestered CTL, life-cycle emissions are comparable to, or smaller than, crude-derived
8
Material in this section is reproduced or adapted from the QTR and its forthcoming Technology Assessments.
gasoline. As gasification produces relatively pure CO2 streams, the incremental costs of
Carbon Capture and Storage (CCS) in an XTL process are only those of compression,
transportation, and storage.
U.S. Annual Carbon Source (milion tonnes)
600
500
400
300
200
100
15% transportation fuels
Net CO2 (Tonnes per Barrel Gasoline-Equivalent

Product)
Figure 12: Annual U.S. carbon flows (in MtC). As shown by the height of the brown line, some 70
Mt of carbon in non-oil feedstocks would be required to replace 15% of current transport fuels if no
carbon were lost in the conversion process.
1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
Figure 13: Life Cycle Carbon Emissions for Various Transportation Fuels. The greenhouse gas
emissions from some alternative fuels are less than those from conventional fuels, while others are
higher. Source: America's Energy Future Panel on Alternative Liquid Transportation Fuels, accessed at:
http://www.nap.edu/openbook.php?record_id=12620&page=250
10
XTL processes currently produce over 370 kbpd of liquid fuels and specialty
chemicals, some 0.4% of global liquids production. The most common product is diesel
fuel via Fischer-Tropsch (F-T) synthesis; GTL facilities deployed in the last 10 years are
responsible for most of the synthetic fuel production and utilize the newest F-T
technologies. The syngas product of gasification (a mixture of CO and H2) can also be
converted to methanol, which can then be converted to gasoline via the Mobil MTG process.
CTL and GTL F-T facilities without CCS have been deployed at the 150 kbpd scale but a
CBTL facility has yet to be deployed. The largest GTL methanol plants deployed are less
than 50 kbpd.
Current studies suggest that GTL can be economically viable at crude oil prices
above $80/bbl9, while CTL w/ CCS can be viable at crude oil prices above $97/bbl. The
large capital outlays associated with CTL, CBTL, and GTL facilitiesnormally built at 50
kbpd or greater to maximize economies of scalepresent a significant risk to potential
investors, especially given oil price volatility. At capital costs of some $150k and $80k per
daily barrel for CTL/CBTL and GTL, respectively, a 150 kbpd CTL plant with CCS is a $22B
bet that crude prices will average above $97/bbl over the amortization period.
The CO2 from an XTL plant can be used for enhanced oil recovery (EOR), thereby
improving the fuel production economics. Current annual U.S. use is some 60 MT of CO2,
mostly from natural sources (e.g., natural gas separation plants). At a typical 50% carbon
efficiency, the current CO2 usage for EOR would be met by 300 kbpd of synthetic fuel
production.
While there is some potential for incremental technical improvements, the greatest
hurdles to deploying XTL are economic, not technical. Relative to other methods for
producing fuels, thermochemical conversion has both higher production and capital costs.
Industry understands the risks and is well-poised to deploy should economic conditions
warrant.
2.4 Biofuels10
Federal policies have encouraged the domestic production of biofuels through fuel
standards, blender subsidies, and import tariffs. Corn ethanol production now exceeds
more than 14 B gallons annually, amounting to more than 10% of the US gasoline pool on
volume basis (7% energy basis); more than half of the US corn crop is now devoted to
ethanol production.
Corn ethanol addresses the security and trade subproblems, but not the GHG issue;
direct life-cycle emissions from corn ethanol are only slightly less than conventional
gasoline. Brazilian cane ethanol could be imported on material scales. Doing so would
improve security by diversifying supply and reduce direct GHG emissions, but would not
improve the trade balance. If the cost of ethanol production from whatever source is less
than the cost of gasoline, as is currently the case, fuel prices could be reduced and
stabilized if penetration levels were significantly greater than todays.
9
Assumes a a 12% Internal Rate of Return on Equity, gas prices of $5-10 mmBTU, and a 10% Capital Charge Factor.
Material in this section has been reproduced or adapted from the QTR Section 7 and its companion Technology
Assessments
10
11
Unfortunately, neither the corn nor the ethanol in corn ethanol is optimal.
Resource requirements and the interactions with food and feed make crop-based biofuels
problematic and it is difficult to imagine corn production growing substantially beyond
current levels. The Energy Independence and Security Act of 2007 therefore mandates that
by 2022 the US produce annually 16 billion gallons of cellulosic biofuels (made from the
structural material of plants), together with 15 billion gallons of corn ethanol. While the
corn ethanol target will be achieved easily, there are no commercial scale cellulosic biofuels
plants in the US today and 2011 production of cellulosic biofuels was no more than 7
million gallons.
Cellulosic ethanol production is hindered by economics. Feedstocks from food-crop
residues, dedicated energy crops, forest materials, and municipal solid waste are
collectively ample to make a material impact (Figure 12). Gathering and processing costs
are the barrier. Feedstock transport costs limit the size of an economic processing facility
to some 10 kbpd, preventing economies of scale. Processing costs are currently too high by
about a factor of two, as the lignocellulose must be decomposed into lignin (generally
burned for power) and sugars (both C5 and C6) which are then fermented to ethanol.
Ethanol is not an optimal motor fuel. Its energy density is only 70% that of gasoline,
and it is hydroscopic and corrosive. Low blends (perhaps to 15 volume percent, or E15) can
be accommodated by the existing infrastructure and vehicles, but E85 requires an
upgraded fueling infrastructure and flex-fuel vehicles (the latter at only a $300 premium).
Research has been underway for some years to develop organisms that will ferment sugars
to higher alcohols (most famously butanol), which have a higher energy density and are
more compatible with the gasoline system.
High-value coproducts (i.e., chemicals for the pharmaceutical, cosmetic, and food
science markets) can augment the economics of early-stage biofuel production. But the
coproduct market will saturate as fuel production is taken to materiality.
3. Non-liquid fuels
The liquid fuel solutions discussed above directly address the security, trade, and
perhaps GHG sub-problems, but would not materially impact price except at the most
aggressive levels of deployment. Indeed, any drop-in product that is a minority11 of the US
gasoline pool will continue to sell at the global crude derived price.12 In other words,
energy independence will not guarantee price independence. That point is vividly
illustrated by noting that:
Fuel riots in the UK in 2000 protested fuel price increases driven by a rising
global crude price. Yet at the time the UK was more than energy independent,
exporting half of its 3 Mbpd production (largely from the North Sea).
After correcting for taxes and exchange rate, the price of gasoline in the US is, to
within a few percent, the same as it is in Germany, even though the US produces
200 times as much crude.
11
If the product cost were less than the crude-derived price, manufacturers would not leave money on the table,
while if it were greater, a mandate or subsidy would be required for a viable business. The minimum penetration
required for a drop-in to set the market price remains a point of discussion.
12
Short term, infrastructure constraints can cause a violation of this general rule.
12
The impact of price volatility is most directly addressed by improved vehicle efficiency. A
supply side measure toward that same end is a shift to a source of LDV energy not fungible
with gasoline (ie, a drop-out fuel rather than drop-in). We review in this section the
three major drop-out possibilities: natural gas, electricity, and hydrogen. Figure 14 shows
the impact that additional transport demand might have on each of these alternative
energy sources, while the damping of price volatility enabled by drop outs is illustrated
by Figure 15.
Figure 14: Estimated supply impacts of satisfying 50% of todays LDV demand by various drop out
fuels. Running half of todays LDVs on Compressed Natural Gas (CNG) would increase current NG demand by
about 1/3. Running those same vehicles entirely on electricity (BEVs) would increase electricity demand by
1/6 and, if all that electricity were generated from natural gas, gas demand would increase by the same
fraction. If Fuel Cell Vehicles (FCEVs) powered by hydrogen were deployed, hydrogen production would
need to increase by more than a factor of three, and either electricity demand would increase by 1/3 or
natural gas demand would increase by 1/5, depending upon how the hydrogen were produced. Finally, for
comparison, biofuel production would have to increase six-fold from todays corn ethanol volume. Source:
QTR, Section 6.
Figure 15: Relation of Fuel Prices to Crude Oil Price, 2000-2011. Data from EIA and Nebraska
Energy Office
13
The price of residential electricity is quite stable because of regulation and because the cost
of fuels is only one component of the total cost. The price of natural gas in the US has
decoupled from that of oil and dropped to extraordinarily low levels because of booming
shale gas production, as shown in Figure 16. On an energy-equivalent basis, wholesale gas
is now only 1/8 the cost of crude oil.
Figure 16: Historic and projected sources of US natural gas. Source: EIA
3.1 Natural gas13

In 2008, 150,000 of the Nations 250 million road vehicles (less than 0.1%, mostly
buses and corporate-fleet vehicles) were powered by compressed natural gas (CNG); About
20% of transit buses in use (and 20% of bus sales) are fueled by natural gas. Advantages of
CNG as a substitute for liquid fuels include a lower and more stable fuel price, GHG
reduction of up to 10% compared to gasoline, and the existing natural gas distribution
infrastructure: the U.S. has more than 210 natural gas pipeline systems totaling over
300,000 miles of transmission pipelines and 1.9 million miles of distribution lines.
However, this infrastructure is optimized to supply power plants and commercial and
residential end users, not to fuel vehicles - natural gas must be compressed (typically to
3500 psi) to meet the volume requirements of mobile applications.
The barriers to natural gas deployment for LDVs include:
13
Material in this section is reproduced or adapted from the QTR, Section 7.2 and its Technology Assessments
14
Limited availability of CNG refueling stations (the closest to the authors home is 14
miles distant, although its fuel prices are half that of gasoline on an energyequivalent basis). Fewer than 1% of U.S. fueling stations supply CNG.
High capital cost ($5,000) and long fueling times (overnight) for home refueling
Vehicle price premium of $2,000-4,000 relative to gasoline-fueled
Limited vehicle range and smaller trunk space (each about about half that of
gasoline fueled) due to lower fuel density; the density issue would be alleviated by
developing in-tank methane-absorbing materials such as metal-organic frameworks.
These barriers are well-described in a recent Consumer Reports article,

http://www.consumerreports.org/cro/2012/03/the-natural-gas-alternative/index.htm .
Globally, propane is the most widely used drop-out fuel. Outside of the United
States, propaneor liquefied petroleum gas (LPG)is commonly referred to as autogas
when used as an automotive fuel. EIA estimates that roughly 147,000 vehicles operate in
the U.S. using 130 million gasoline-equivalent gallons of propane fuel a year (a bit more
than 0.1% of gasoline use). Propane vehicles in the U.S. are primarily used in fleet or rural
(e.g., farming) applications. Propane fueling stations, though the most numerous of all
drop-outs, are unevenly distributed around the country.
Virtually all propane is produced as a co-product of natural gas processing or an oil
refinery product. US production in 2010 was about 20 billion gallons. Hence, a 4-fold
increase in production would be required to meet 50% of todays LDV demand and
continue to satisfy other uses.
US propane production could increase strongly as
economics of shale gas production drive interest toward wet gas..
3.2 Hydrogen14
Research on fuel cells has led to significant progress in recent years, helping to
reduce their cost by a factor of five and on-vehicle hydrogen storage has improved to
acceptable ranges for an LDV. But significant further improvements in key technologies
remain to be demonstrated to create a viable LDV system. Further progress could well
bring the cost of driving (vehicle plus fuel) for FCEVs within the range of other drop-out
technologies. However, those other technologies are currently economically superior and
will continue to improve rapidly.
Although the projected cost of hydrogen (dispensed to the vehicle) via some
production pathways is less than $5/gge (gasoline gallon equivalent, roughly 1 kg of
hydrogen), given the higher vehicle cost, $24/gge is required for FCEVs to be competitive.
Reforming natural gas is the most mature and lowest cost method to produce hydrogen; it
14
Most of the material in this section has been reproduced or adapted from QTR (Section 6 and the Technology
Assessments).
15
is used to produce over 90% of the 9 million tons of merchant hydrogen annually in the U.S.
(equivalent to about 7% of todays U.S. gasoline supply). R&D has advanced the state of
hydrogen production from distributed natural gas so that it could be technically possible to
achieve high volume production costs of approximately $3/gge, but industry estimates
indicate a more realistic high-volume cost of $7/gge over the near-term. Production of
hydrogen from renewable sources (biomass, algae, solar, etc.) offers lower GHG emissions,
but is less mature and more costly today; there are also serious scaling issues in some of
these sources.
3.3 Electricity15
Degrees of electrification for electric drive vehicles range from mild and strong
hybrid electric vehicles (HEVs), through plug-in hybrid electric vehicles (PHEVs), to pure
electric vehicles powered by batteries (all-electric vehicles, or AEVs). HEVs and PHEVs
offer increased fuel economy but still require some liquid hydrocarbons, while AEVs do not
require liquid hydrocarbons, and thus fully decouple transport from oil.
The vehicle industry is more than a decade into the commercial deployment of
electric powertrains in HEVs, and is generating expertise in integrating conventional and
electric powertrains. Although HEVs are currently only 3% of new LDV sales, market
penetration is increasing. General Motors, Nissan, and Toyota have undertaken massproduction of plug-in vehicles, so expertise in the next generation EV powertrains is
growing.
PHEVs and HEVs are more energy-efficient than conventional vehicles because
electric motors are four times more efficient than todays ICEs, hybridization of the
powertrain allows for use of more efficient ICEs than conventional powertrains, and
because regenerative braking allows energy to be recovered and reused.16 PHEVs further
reduce oil consumption by replacing liquid fuels with electricity. As shown in Figure 17, a
PHEV with a 40 mile all-electric range would replace at least 2/3 of gasoline consumption
with electricity. This modeling is in accord with the real-world experiences of Chevy Volt
owners.
HEVs have a price premium of $2,000-$4,000, which offers an attractively short
payback period in an era of $4 gasoline price. The premium for a PHEV40 is three two
fours times larger and such vehicles are not now an economic proposition; with declining
battery costs and/or rising gasoline price, they would become so.
15
Most of the material in this section is reproduced or adapted from the QTR, Section 6.
16
Typical efficiencies for a PHEV and AEV are 2.5 mi/kWh and 4 mi/kWh, respectively, corresponding to a fuel costs
of $0.06mi and $0.04/mi if electricity is $0.15 /kWh. At $4 gasoline, the fuel cost for a 40 mpg ICE is $0.10/mi.
16
Figure17: Impacts of plug-in hybrid electric range and charging infrastructure. Utility factor
is the fraction of vehicle miles that would be driven on electric power without recharging. Different
charging scenarios are shown. The benefit of ubiquitous charging becomes smaller as the allelectric range increases; for most applications, home charging is sufficient. From a forthcoming
EPRI report, Understanding the Effects and Infrastructure Needs of Plug-In Electric Vehicle (PEV)
Charging.
There is more than enough electrical generation capacity to power the LDV fleet. As
shown in Figure 14, only 15% more electricity would be required to fully power half of
todays LDV fleet. Since the average capacity factor of installed generation is less than 50%
due the diurnal load variation and inability to store large amounts of electrical energy,
charging at night will not be an energy issue. But it could be a power issue, as discussed in
the following section.
The GHG impact of LDV electrification will, of course, depending upon the carbon
intensity of grid power. With the current average US carbon intensity of power, they would
reduce GHGs by about 1/3, a figure that would improve as the grid decarbonizes and/or
PHEVs are fueled with low-carbon biofuels.
3.4 Effecting a transition
It is plausible that technical advances and/or rising gasoline prices will make the
driving costs per mile (vehicle + fuel) of any or all of the drop-out fuel possibilities
economically attractive. But a large-scale transition of LDVs to any dropout fuel will not be
simple, as vehicles, fuel production, and fueling infrastructure, each separate industries,
17
must shift simultaneously while maintaining a fuel anywhere capability. Further, it

seems likely that at most one drop-out fuel system would be supported to supplement the
current gasoline system. Important questions then are: Should a drop-out system be
deployed? Which one? and How would the transition be effected?
New LDV technologies will be deployed most rapidly and seamlessly if they can
integrate with the existing energy infrastructure. The fueling patterns of on-road transport
require extensive infrastructure (Figure 18). The U.S. has 55,000 miles of crude oil pipeline
feeding 150 refineries and another 95,000 miles of refined product pipelines and many
local delivery trucks supplying 160,000 gas stations.
Current Number of Fueling Stations in the U.S.
180,000
160,000
159,006
140,000
120,000
100,000
80,000
60,000
40,000
20,000
2449
0
Gasoline
2407
623
Electricity * Ethanol-85 Biodiesel-20
893
58
CNG
Hydrogen
Figure18: Current fueling stations in the United States. There are many more fueling
stations for gasoline than for other fuels. *Electricity stations are the publicly available stations
only. Not shown are the millions of existing locations for home charging. Source: DOE EERE (for
alternative fueling stations) and EIA (for gasoline stations).
Among the drop-out possibilities, electricity is most favored in these considerations

because the LDV transition to it can be gradual and graceful. The existing grid
infrastructure can accommodate significant immediate deployment of HEV and PHEV
vehicles, and could eventually support full electrification of the light-duty vehicle fleet with
some upgrades and modifications. There are 11 million miles of electrical distribution
circuits that can, today, accommodate virtually unconstrained residential 120V wall outlet
charging of plug-in hybrid electric vehicles (Level 1, 2 kW, equivalent to a hair dryer).
Ubiquitous charging does not significantly increase the utility factor of a PHEV with an
electric range greater than 40 miles (Figure). Among the drop-out possibilities, PHEVs will
therefore see fastest deployment and can have the greatest near-term impact on oil
consumption.
18
Charging time is a potential barrier to further electrification; ten hours are required
to fully charge a PHEV with a 40 mile electric range from a 120V charger. Vehicles with
longer electric ranges will require faster charging, which would eventually require grid
upgrades. While the household circuits necessary for 240V Level 2 chargers (>3 kW) are
commonly used for appliances, obtaining vehicle access to those circuits may require
specialized wiring and could affect grid distribution circuits if deployed in clusters. Fewer
than 2% of U.S. fueling stations currently offer 240V charging for EVs (Figure 18). Direct
current fast charging (Level 3, 480V DC, 50kW) would stress todays grid and would
require special infrastructure and power management for widespread deployment.
As the market progresses from HEVs to PHEVs of various ranges to AEVs, the
demands on the electric charging infrastructure will gradually increase. These increases
can be accommodated as they occur, allowing for a smooth path toward greater
electrification.
In contrast, the U.S. hydrogen fueling infrastructure is extremely limited. Fewer
than 0.05% of U.S. fueling stations supply hydrogen. Hydrogen can be centrally generated
and distributed in the U.S. by truck or through the 1,200 miles of pipelines mostly in
Illinois, California, and along the Gulf Coast. Mass-market FCEVs would therefore require
vastly expanded hydrogen generation, distribution, and fueling infrastructure, which will
hinder, if not limit, their impact in the transport sector.
Infrastructure requirements vary across application. Vehicle fleets with their own
fueling infrastructure could benefit from specialized fuels. Examples include overhead
electrification for designated public transportation routes and hydrogen or CNG fueling at
fleet depots. However, these are specialized applications, and technology pathways that
leverage existing infrastructure are more likely to succeed in mass markets. Because of
their infrastructure requirements, AEVs and FCEVs are most easily introduced into vehicle
fleets with a captive fueling infrastructure.
_______________________________________________________________________________________
Steven E. Koonin was the second Under Secretary for Science at the US Department of Energy, serving
from May 2009 thru November 2011. In that capacity, he was oversaw technical activities across the
Departments science, energy, and security activities and led the Departments first Quadrennial
Technology for energy, from which some of the material in this paper is drawn. Prior to joining the
government, Koonin spent 5 years as Chief Scientist for BP, plc. where he played a central role in
establishing the Energy Biosciences Institute. Koonin was a professor of theoretical physics at Caltech
from 1975-2006 and was the Institutes Provost for almost a decade. He is a member of the U.S.
National Academy of Sciences and the JASON advisory group. Koonin holds a BS in Physics from Caltech
and a PhD in Theoretical Physics from MIT (1975) and is currently an adjunct staff member at the
Institute for Defense Analyses. He will take up an academic position in 2012.
Light-Duty Vehicles: A Discussion of the Evolving Engine, Vehicle and Fuel

Requirements Context

John B. Heywood

Sun Jae Professor of Mechanical Engineering Emeritus
Sloan Automotive Laboratory
M.I.T.

Abstract

This paper discusses the evolving vehicle context within which assessments and
choices of alternative fuels for transportation should take place. The focus is on anticipated
engine, powertrain, and other propulsion system improvements (especially their average
efficiency) in the context of steadily reducing vehicle driving resistances, and especially
weight. The fuel requirements of spark-ignition, diesel, and hybrid engines are discussed
and connected to the supply, distribution, and refueling requirements of alternative fuels.
An illustrative scenario of alternative fuels is described and used to indicate likely future
fuel demand.

1. Background

As the price of oil continues to rise, and the potential for growth of petroleum-based
transportation fuel in the longer-term is uncertain, the question of other energy sources
and fuels is obviously important. An awareness of the vehicle context, especially the
engines or other propulsion system changes in progress, is a necessary preliminary to
developing effective alternative fuels strategies. This paper addresses the question of how
this engine and vehicle context is likely to evolve over the next twenty years especially in
relation to fuel requirements. The focus is on the current state of the in-use light-duty
vehicle fleet (cars and light-trucks) in the United States and how, as new and more fuel
efficient technology and propulsion systems enter and leave the vehicle fleet or parc
through sales and scrappage, the fuel demand and greenhouse gas emissions of this critical
component of the total U.S. energy sector will change.

The current U.S. transportation fuels situation from an energy perspective is as
follows. Ethanol, made from corn grain is approaching 10 percent of the gasoline market.
It is largely blended with and sold as gasoline. There are some 2500 refueling stations
where ethanol fuel (as E85) can be purchased: there are close to 120,000 re-fueling
stations nationwide. A modest amount (or order 1 percent) of biodiesel fuel is blended
with regular diesel fuel. In the total in-use fleet of 260 million vehicles there are about 10
million flexible-fuel vehicles in use that can satisfactorily use gasoline, E85, or any mixture
of the two. Vehicle models that can use electricity directly (plug-in hybrids, PHEVs, battery
electric vehicles, BEVs) are entering the market but as yet sales volumes are very small.
There is also modest use of natural gas in vehicles in the U.S.in buses, and in a small
number of dual-fuel and dedicated NG vehicles.

In terms of mainstream technology the dominant propulsion systems are gasoline-

fueled spark-ignition engines and diesel engines. While the light-duty vehicle market in
Europe is about half gasoline/petrol and half diesel, in the U.S. gasoline engines dominate.
The performance and fuel consumptions of these engines has and continues to steadily
improve. Table 1 lists the primary nearer-term opportunities for improving the efficiency
of gasoline-fueled engine vehicles. They are, in order of importance:

Turbocharging naturally-aspirated gasoline engines which constitute some 90% of
the market, with direct injection of fuel into the cylinder, and significant engine
downsizing since the torque per unit of engine displaced volume is substantially
increased.

Improving the base engine efficiency through variable valve timing/control,
increasing engine compression ratio, and reductions in powertrain friction, trends
that are already underway.

Introducing engine shut down at idle: so-called engine stop/start. In urban driving
this reduces average fuel consumption by up to 5%.

Steadily reducing the weight of vehicles through substitution of lighter materials,
vehicle redesign, and shifting the vehicle size distribution downwards.

Beyond 2016, the auto manufacturers are likely to use these technologies more widely
in their sales mix to meet the 2025 CAFE mpg targets now being finalized, step up the pace
of ongoing incremental improvements, take additional weight out of vehicles beyond the 5
10% reduction expected by 2016, increase the percentage of hybrids, and introduce some
electrified vehicles (PHEVs and BEVs) to gain the miles-per-gallon credits these alternative
energy vehicles are being awarded.

Table 1: Opportunities for Improving Powertrain Efficiency

Technology
Compared with
% Gain in MPG

Diesel engine
Gasoline engine
25 30%
Gasoline direct injection +
Multiport fuel injection
Up to 12%
Turbo
Dual-clutch transmission
Automatic transmission
Up to 10%
Cylinder deactivation
Using all cylinders
6%
Cont. variable valve timing
Fixed valve settings
5%
Stop-start system
Idling
5%
5 or 6-speed transmission
4-speed
3 5%

Source: Automotive News, May 25, 2009

2. Evolving Vehicle Context for Alternative Fuels

To assess the roles that alternative transportation fuels might play in the U.S., we
need to extend our planning timeframe form the above nearer-term where impacts will
largely come from improvement that can be made in mainstream spark-ignition and diesel
engines, and expanding hybrid use, over the next 20 or more years. This evolving
powertrain, vehicle, and fuels-requirement context involves both the quantitative
performance and efficiency gains that improved mainstream and new technology could
provide (at the vehicle level), and the deployment potential over time of these technology
changes so their impacts on fleet fuel consumption (and other issues such as greenhouse
gas (GHG) emissions) can be assessed.

As described above, in the powertrain arena, we expect that spark-ignition engines
(largely gasoline fueled) will improve their efficiency through many advances in their
component technologies, and through better optimization. By 2030 it is anticipated that a
significant fraction (some 50% ore more) of these spark-ignition engines will be direct-
injection, turbocharged, and downsized engines. This improves engine part-load efficiency
significantly, largely by reducing the relative (and negative) impact of engine friction. Also,
through turbocharging, automakers can continue to offer both enhanced (increasing)
vehicle performance and higher fuel economy at the same time. There is a negative trade-
off between performance and fuel consumption (1): thus this increased vehicle
performance does decrease the fuel economy benefit by some 10-15%.

In terms of engine fuel requirements, this turbocharging trend, and the increasing
efficiency trend it enables, is constrained by the abnormal engine combustion phenomenon
of knockthe rapid spontaneous release of a portion of the fuel-air mixtures chemical
energy inside the engine cylinder towards the end of the normally continuous mixture
burning process. This knock constraint depends on the fuels spontaneous ignition
characteristics and is quantified by the fuels octane number. Use of direct fuel-injection in
these turbocharged engines helps with this knock constraint because as the in-cylinder fuel
spray evaporates, it cools the unburned mixture and the temperature of that mixture which
is a critical controlling variable.

What this says is that alternative liquid fuels with high knock resistancehigh
octane numberare most desirable. Note that interest is growing in increasing the octane
rating of current gasolines, if it can be shown that the higher efficiency of the higher
compression ratio gasoline engines these better fuels would enable more than offsets the
additional energy used in the production of these higher octane gasolines.

Note also that the volatilityease of vaporizationof spark-ignition engine fuels is
a key characteristic in achieving fast and repeatable engine starting and realizing the very
low emission levels of hydrocarbons required now (and more stringent levels in the
future). This is another major constraint on the properties of alternative fuels.

Diesel engines will steadily improve also, but their efficiency will not increase as
much as gasoline engine efficiency will increase. However, diesels, already some 20%
more efficient than non-turbocharged gasoline engines on an energy consumption basis
(some 30% more efficient on a fuel volume basisgallons per 100 miles, liters per 100
km). This gap is expected to narrow over time. However, in the U.S. there is no tradition of
extensive diesel use in light-duty vehicles, fuel costs are still relatively low, and diesel (per
gallon) is more expensive than gasoline. So, the diesel sales fraction in U.S. is not expected
to exceed 5-10% of vehicle sales over the next couple of decades. Hybrid vehicles (see
below) appear to be a more attractive option. Also, transmission efficiencies will improve
by about 10% as powertrain and vehicle designers incorporate more gears and more
efficient shifting mechanisms, or continuously variable transmissions.

Alternative propulsion systems are already in production though still at modest
levels. Hybrid sales (HEVs) are expected to grow steadily from todays 3% level to 10-20%
of new vehicles in 2030, as the HEV cost premium above a conventional engine vehicle is
reduced. Use of electricity in light-duty vehicles, in PHEVs and BEVs, will grow more
slowly. Sales of such vehicles in 2030 might reach 10% if the battery cost premium falls
sufficiently, though there is a growing consensus that the necessary cost reductions will
need the successful development of new battery chemistries, a research and development
task that would take at least 15 years.

From 2020 to 2030, the use of natural gas as a vehicle fuel may well grow, but
prospects are uncertain. In the light-duty vehicle area its use is likely to be largely confined
to localized fleets. Natural gas vehicles, either as single-fueled vehicles or as dual fuel
with both natural gas and gasoline fuel systems on the vehicleare significantly more
costly and a readily and broadly available distribution and refueling system for natural gas
does not yet exist. Thus, at present the use of natural gas as a vehicle fuel is inconvenient
though natural gas per unit energy content is significantly cheaper than petroleum-based
fuels and is expected to remain so.

Fuel-cell hybrid systems, which many view as a promising longer-term option for
the larger end of the light-duty vehicle size distribution, will continue to be developed and
tested under real world conditions essentially as production prototypes. Sales volumes
in the 2020-2030 timeframe might be several percent. The deployment of a widespread
hydrogen distribution system will be a critical constraint, and at best will take time to
develop. However, the cost premium of the fuel-cell propulsion system has been
substantially reduced, and that cost reduction trend continues.

There will be vehicle changes, too. A steady reduction in vehicle weight, up to some
10-15% by 2030, from vehicle design changes and use of lighter-weight materials is
anticipated. A comparable weight reduction (by 2030) is likely to occur in parallel due to
downsizing of the U.S. new vehicle size distribution. This downsizing is starting as
gasoline prices steadily rise, and due to consumers caution in the current economy. A
10% reduction in vehicle weight results in about a 6% reduction in fuel consumption.

An important less obvious trend is the ongoing negative impact of increasing

vehicle performance on fuel consumption. An average vehicle acceleration performance
escalation of some 10% where it eventually levels out by about 2030 is anticipated for an
average car value of about 9 seconds today to about 8 seconds). Some 15% degradation of
the anticipated potential 2030 fuel consumption benefit will result as a consequence. Note
that smaller lighter vehicles with less powerful engines may compromise vehicle
drivability, performance up a grade, vehicle towing capability, load-carrying capacity, etc.,
especially in certain important categories of vehicles where one or more of these
capabilities is critical.

What will be the fuels requirements of these evolving future vehicles? Multi-
component hydrocarbon fuels closely comparable to what we have today would be the
best solution. Such fuels would have a high octane rating and appropriate wide volatility
range: liquid fuels are obviously the most desirable. It will be challenging to replace
petroleum-based fuels as their supply, though pulled by increasing demand, over time will
become limiting. Oil sands and heavy oil are a growing source of liquid fuels: currently
supplying some 10% of demand in the U.S. Global supply projections of petroleum and oil
sands/heavy oil based fuels, out to 2030, suggest modest growth of order 1% (at best) with
a gradual leveling-off and then decline thereafter.

Ethanol, which at present is an easier end-product to produce than many of the
alternatives, is about 7% (on an energy basis) of U.S. transportation fuel supply. 10% corn-
based ethanol is a likely upper bound in the U.S. Other biomass sources and fuel
production approaches are moving toward pilot production but are not yet at that stage.
Currently, plausible best fuel choices from biomass are unclear.

Methanol, from energy density (half that of gasoline) and toxicity perspectives, is
not an ideal end product, and the broader motivation in the U.S. for methanol, beyond its
potential for lower cost, is unclear.

While we are steadily learning about the environmental and other potential impacts
of large-scale use of biomass for transportation, the likely magnitude of these impacts is
often uncertain. The issues are: competition with food in agricultural land use, the
ecological impacts of large scale biomass production for fuel, the greenhouse gas emissions
impacts of increasing and changing land use patterns to produce fuels from biomass, water
impactsthe amounts required for this agricultural expansion and the amounts used in
biomass processing, and the environmental consequences of increased use of fertilizer, etc.

Increasingly, many professionals in this area, are recognizing the real challenges
involved in setting up new fuel distribution and vehicle refueling infrastructures for
alternative fuels such as ethanol and methanol, for natural gas, and for hydrogen, and in
parallel implementing at scale the propulsion system and on-vehicle fuel storage
technology needed to use these fuels. As a consequence, attention at least for the nearer-
term, is shifting to whether drop-in fuels that are compatible with existing fuels
gasoline and dieseland thus do not require any major changes on the fuel side and the
vehicle side, are a realistic alternative. Given this difficult choice with substantial
uncertainty, it is not surprising that the major growing non-petroleum based fuel supply is
gasoline and diesel from Canadian tar/oil sands.

3. Illustrative Demand Scenarios

A second part of this evolving context is the anticipated U.S. demand for
transportation fuel and how that will change over time. This has been an important focus
of my MIT teams research. Here, I will summarize one of our recent assessments (2).
Many others, of course, are active in this area. While different groups make different
assumptions, especially about the rates at which we progress to lower fuel-consuming
vehicles technology, the general trends in these studies are similar.

Figure 1 shows our recent projections of vehicle fuel consumption (in liters/100
km) of different powertrain technology vehicles into the future. Light-trucks show similar
trends but with fuel consumptions some 30-40% higher due primarily to their higher
weight. Figure 2 shows the assumed market shares of the various powertrains as a
percentage of new vehicle sales in each year, out to 2050. Note that in this study (2),
assumed values for the some 40 input parameters required for each simulation were
specified by a minimum and maximum value, and a modal value for each triangular
distribution. These figures show mean values. A steady transition over time to more
efficient propulsion systems (and increasingly lighter vehicles) is assumed. (We update
these assumptions periodically. With the 2025 CAFE targets now part of a NHTSA/EPA
rulemaking, we are assuming a higher proportion of gasoline engines is likely to be
turbocharged (increasing that percentage from 20 or so to approaching 50%. The net
impact of this change on fleet fuel consumption and GHG emissions is modest.) Many other
assumptions related to the in-use fleet size and turnover, vehicle kilometers traveled,
sources of alternative fuels to petroleum-based gasoline and diesel, any electricity and
hydrogen used, extent of vehicle performance escalation, are required: see reference (2).
Also, a Monte Carlo probabilistic methodology is used to generate a distribution of outputs
form the input distributions specified as assumptions (3).

The results for the LDV U.S. in-use fleets fuel consumption (in billion liters of
gasoline equivalent per year), is shown in Fig. 3. We see that the mean projected fuel
consumption changes little over the next decade, and then decreases at some 1 to 1.5
percent per year. By 2040 fleet fuel consumption would be down by about 20% from its
2010 to 2020 value. The U.S. in-use fleets GHG emissions decrease from 2010-2020 levels
also by about 20% (note these are life-cycle emissions, and several other fuel-related
factors come in). The dashed lines in Fig. 3 show the 75% and 25% probability pathways
(one standard deviation), and 95% and 5% probability pathways (two standard
deviations) in this calculation, which embodies uncertainty. These scenario analyses give
us a useful sense of what future demand for transportation fuels is likely to be.

Figure 1. Relative fuel consumption of the average car for the different powertrains,
assumed scenario input, over time to 2050. Hybrids and plug-in have the same fuel
consumption for liquid-fuel driven miles. (2)

Figure 2. Powertrain new vehicle market share, mean input values 2010-2050. (2)

Figure 3. U.S. light-duty fleet fuel use (billion liters gasoline equivalent/year) over time
out to 2050. (2)

4. Alternative Fuels: Overall Objectives

Our overall objectives are to displace a significant fraction of the petroleum and oil-
sands based fuels we are using at roughly equivalent cost, and do this in ways that also
reduce the LDV fleets greenhouse gas emissions. Both these objectives are furthered if the
powertrain that use these alternative fuels are of equal or higher fuel efficiency than the
steadily improving gasoline engine. It is clearly beneficial if vehicle engines can operate
satisfactorily on both the alternative fuel and gasoline. Higher compression ratios, higher
turbocharger boosting levels and thus greater engine downsizing, all improve powertrain-
in-vehicle efficiency. Thus alternative liquid fuels should match or exceed the anti-knock
rating (octane number) of gasoline.

Light-duty vehicles must, of course, meet current and future vehicle air-pollutant
requirements. Future standards will be lighter than todays requirements and, the most
demanding requirement, the HC emission standards (emissions must be less than 1/10,000
of the vehicles fuel usage), are expected to be further reduced. Thus the
volatility/evaporation characteristics of alternative fuels will need to be comparable to
those of gasoline (which also need to be tightly controlled as well), to ensure very clean
engine start-ups. Note that deployment of engine start/stop technology makes this even
more important.

Obviously, a high specific energy density (per unit mass and per unit volume) is
important in fuel production, distribution, storage, and refueling at the service station, and
for fuel storage on the vehicle. Here the alcohols, ethanol and methanol, are at a
disadvantage because they are partly oxidized already (they have specific chemical energy
densities of 0.7 and 0.5 relative to gasoline, respectively).

The above summary indicates that drop-in fuelshydrocarbons with properties
little different from petroleum-based gasoline, maybe with higher octane ratings, are an
attractive option if the availability of primary energy sources of such fuels and their
processing technology indicates their potential for large-scale production at marketable
prices. If alternative fuels can be produced that are fully miscible with gasoline or diesel,
and could even enhance the characteristics of these petroleum-based fuels (for example
through higher octane) they would have a significant advantage.

Alternative fuels that would need a separate (and therefore new) supply, and
distribution, and refueling system, and which would need vehicle modifications to use
these fuels would be disadvantaged.

The anticipated future prices of these different alternative fuels relative to the
prices of petroleum-based gasoline and diesel, is clearly a major factor in choosing among
the alternatives.

5. Vehicle and Fuel Options

Here I list and briefly describe our vehicle and fuel options, with the next 20 years as
the timescale. Table 2 summarizes the several alternatives.

We can blend these new fuels with conventional fuels. We are already doing this with
ethanol as E10 and we may move to E15. With ethanol and methanol, which is not fully
miscible, there is an upper bound on the amount that can be absorbed by blending.

Thermochemical conversion of biomass and other sources to gasoline-and diesel-like
fuels has the potential for producing drop-in fuelsend products fully miscible with
gasoline and diesel. As discussed previously, should the cost of producing these fuels prove
to be competitive, their development and use would be an especially attractive option
because propulsion system and vehicle technology changes, and fuel distribution and
refueling infrastructure changes would be minimum.

Flex-fuel vehicles that can operate with any mixture of gasoline and ethanol have
been brought into the vehicle fleet over the past decade or so. The auto manufacturers
incentives were the governments CAFE credit incentive that this approach (with its low
costssome $100 per vehicle) could open the LDV fuel market to growing ethanol use, and
There are currently about 10 million flex-fuel vehicles in use in the U.S., about 4% of the in-
use fleet. E85 refueling stations have spread and there are now about 2500 such stations,
about 2% of the 120,000 U.S. refueling stations. Only about 500,000 of the 10-million flex-
fuel vehicles regularly use E85. Barriers to increased ethanol use are the fuels cost, its
availability (production and retailing are currently concentrated in the U.S. Midwest) and
limited supply (most of the available ethanol is blended).

10
_______________________________________________________________________________________________________
Table 2: Vehicle/Alternative Fuels Options

(a) Blend new fuels with existing fuels: e.g., E10, maybe E15. Upper bound on
penetration.

(b) Produce new fuels that are fully miscible with gasoline and diesel.

(c) Expand production of flex-fuel vehicles; achieve adequate distribution of alternative-
fuel refueling stations.

(d) Produce dedicated optimized alternative-fuel-vehicles: e.g., natural-gas vehicles.

(e) Dual-fuel vehicles: e.g., both gasoline and natural gas fuel tanks and fuel-injection
systems on the vehicle.

(f) More focused approaches: Separate on-board tank for anti-knock fuel (e.g., ethanol):
suppresses knock with gasoline and increases gasoline engine efficiency.
_________________________________________________________________________________________________________

An approach to expand this ethanol path under consideration is to require all vehicles
sold be made bi-flex-fuel (gasoline and ethanol) or tri-flex-fuel (gasoline, ethanol, and
methanol). Thus, over time, use of these alcohol fuelswhich do have attractive
combustion and knock-resisting characteristicscould then expand. An important
question is whether uncertainly as to the long-term potential for these two alcohol fuels
relative to other options such as producing drop-in fuels thermochemically from biomass
and other sources, makes it premature to attempt a mandate.

Development and limited production of dedicated fuel vehicles is occurring. Honda
is selling a natural-gas-fueled LDV. Also, in other parts of the world (Sweden, Brazil) E100,
ethanol-fueled vehicles, have been offered. The latter usually require a small gasoline tank
on-board to achieve adequate low-emissions engine starting.

Another option is dual-fuel vehicles such as natural gas and gasoline. These vehicles
require two on-board fuel storage systems and fuel injection systems. Dual fuel vehicles, as
with the flex-fuel vehicles, may not be able to get the optimum use out of each of the two
fuels due to their different characteristics. Each of the fuels in these two pairsnatural gas
and gasoline, or gasoline and ethanolhas different knock resistance and thus octane
rating. The engine compression ratio is fixed by the basic geometrical design of the engine,
so it has to be set (more or less) at a value determined by the lower octane rating fuel
(gasoline, compared to natural gas; gasoline, compared with ethanol). So optimum
efficiency in the absence of variable compression ratio engines is not obtained with each
fuel. Variable valve control can help here but at a loss in power. The added costs of dual-
fuel spark-ignition engines involving natural gas are substantial.

11

In addition to the broader options outlined above, there are some more specific
engine-fuels opportunities. One concept that I, with Dan Cohn and Leslie Bromberg here at
MIT are exploring uses direct-injection of ethanol (or methanol) into the cylinders of a
gasoline engine when that engine (with gasoline) is about to knock. Thus the major
efficiency constraint on compression ratio and high turbocharger boost pressures is
removed, the engine can be downsized substantially, and its efficiency significantly
increased (doubling the benefits that a direct-injection, turbocharged and downsized
standard gasoline engine achieves). This can be done with modest amounts (5% or less) of
ethanol but a small additional tank and fuel pump for this anti-knock fuel are required.
This approach to constraining or removing knock is also applicable to flex-fuel vehicles and
natural gas vehicles to optimize their operation and performance. It can utilize more than a
modest amount of the alternative fuel, if more is available. This concept is being explored
by some industrial groups. There are several potential refueling options: one is to
distribute the anti-knock fuel (say ethanol plus some water) in a manner analogous to how
windshield washer fluid is distributed.

6. Key Questions

This Symposium is taking place because developing a significant supply of
alternative fuels is important as the cost of petroleum-based transportation fuels rises, and
(in due course) their availability becomes a serious constraint. Our discussions today are
also important because we have yet to identify clearly the most advantageous and viable
path towards this goala substantial supply of one or more alternative fuel that is cost
effective as it is used in light-duty vehicles. We need to acknowledge that our knowledge
base for identifying the more promising fuel options (along with the vehicle propulsion
systems these fuels require) is currently insufficient.

The challenge of building-up significant supply of these fuels can usefully be
separated into two steps. First, how can we best get started on exploring the various
options in ever-greater depth and thus narrowing our many possible choices in a rational
way? Second, we need to explore how to grow the supply of the most promising of these
options, to significant scale, as we steadily become wiser. A key piece of these questions
is what the appropriate role of our Federal Government in this process should be. The
basic question is how do we break out of the chicken and egg constraint circlefuels first
or vehicles first: how can we best grow both together?

One approach to moving us forward would be to identify the (limited number of)
promising options, gaining real-world experience with the required vehicle technology,
fuel supply and distribution, in a step-by-step manner. Some of this is already happening
with limited fleet studies that are localized so that fuel supply and distribution, and actual
vehicle use, are not severely constrained. A steadily expanding set of fleet studies, which
may well need to be incentivized by Federal funding, may be a promising way to get us
started more seriously towards our broader goal. At present, our progress is limited.

12
References

1. Cheah, L.W.B., A.P., Bodek, K.M., Kasseris, E.P., Heywood, J.B. 2009. The Trade-off
between Automobile Acceleration Performance, Weight, and Fuel Consumption, SAE
paper 2008-01-1524, SAE International Journal of Fuels and Lubricants, 1, 771-777.

2. Bastani, P., Heywood, J.B., Hope, C., 2012. The Effect of Uncertainty on the U.S.
Transport-Related GHG Emissions and Fuel Consumption Out to 2050. Transportation
Research Part A: Policy and Practice, 46, 517-548.

3. Bastani, P., Heywood, J.B., Hope, C., 2012. A Forward-Looking Stochastic Fleet
Assessment Model for Analyzing the Impact of Uncertainty on Light-Duty Vehicles Fuel
Use and Emissions, SAE paper 2012-01-0647, SAE World Congress, April 24-26, 2012,
Detroit, MI.

Acknowledgement

My students and my work in these areas has been sponsored by several
organizations: Chevron, CONCAWE, DOE, Eni, MIT-Portugal Program, Shell.

The Case for Bi-Fuel Natural Gas Vehicles

Michael D. Jackson
TIAX LLC
208132 Stevens Creek Blvd, Ste 250
Cupertino, California 95014
April 14, 2012
A Position Paper for the 2012 MITEI Symposium
Prospects for Flexible- and Bi-Fuel Light Duty Vehicles
Summary
The purpose of this paper is to examine the feasibility of a bi-fuel natural gas vehicle for
the U.S. light duty retail market. Although both dedicated and bi-fuel natural gas
vehicles (NGVs) have been marketed in the U.S., the vehicles have been designed to
maximize compressed natural gas storage within constraints of costs and volume.
Throughout the world bi-fuel NGVs are the predominate design. In developing countries
most bi-fuel vehicles are converted from existing gasoline vehicles. Costs of conversion
kits including storage tanks (mostly Type I steel tanks) are relative low as are installation
costs. Coupled with high gasoline prices and low CNG prices, these bi-fuel conversions
continue to capture market share. Fueling infrastructure is being built to meet customers
demand for the cheaper natural gas fuel. The bi-fuel concept allows for some leeway in
station build-out, since gasoline can still be used if needed in these vehicles.
Automakers marketing vehicles in Europe have further evolved the bi-fuel concept to
take advantage of the relatively high gasoline and low CNG prices. CNG prices in
Europe are 30 to 50 percent of gasoline prices. Automakers are providing bi-fuel
vehicles with underfloor CNG storage so as to not compromise vehicle functionally.
They are also optimizing fuel consumption and performance. Automakers are offering
enough storage to provide good range on CNG and have maintained a somewhat smaller
gasoline storage tank. This strategy is aimed at mostly CNG use and therefore requires
the investment and built-out of CNG stations. Bi-fuel vehicles are more expensive than
gasoline counterparts, but the price of CNG makes reasonable paybacks possible.
In the U.S. the NGV strategy for light-duty vehicles has been to maximize CNG range of
either dedicated or bi-fuel NGVs. Typically, however, dedicated vehicles have reduced
range compared to their gasoline counterpart due to limited vehicle space available for
CNG storage tanks. Similarly, for bi-fuel options vehicle space is further limited by the
retained gasoline fuel tank. In either design, the CNG tanks are often mounted in the
vehicles trunk or pickup bed, thus reducing storage space or payload. The current
designs really address the commercial light duty market for those users that can justify
1
the higher upfront costs based on their duty cycle. With U.S. fuel prices, this usually
means that NGVs are only economical for fleets that use a lot of fuel/drive a lot of miles.
This design philosophy effectively excludes the light duty retail market where the
average annual mileage is 12,000 and the average annual fuel use is less than 500 gallons.
At these low utilization rates, it is hard to payback the higher upfront costs of NGVs fast
enough to interest consumers. However, if storage could be reduced, costs could be
lowered enough to possibly interest consumers. Two thirds of all drivers travel 40 miles
or less per day which means depending on vehicle that CNG storage of 1 or 2 gallons
gasoline equivalent could be sufficient. Home refueling would most likely also be
needed to eliminate daily CNG refueling trips. A simple analysis indicates that the
combination of reduced vehicle costs and additional costs for home fueling appliance
may be attractive to consumers.
Additionally, with small natural gas storage volumes it may be possible to further reduce
system complexity and costs by lowering the storage pressure. This would need to be
investigated further.
Introduction
In the world today, there are over 12.7 million natural gas vehicles (NGVs) operating.1
Most of these vehicles are light-duty (passenger and light commercial) vehicles and are
largely converted from gasoline to natural gas. Table 1 shows the worlds distribution of
light, medium and heavy duty natural gas vehicles in 2010. Almost all the light duty
conversions retain the gasoline fueling system and are then capable of operating on
natural gas and gasolineso called bi-fuel vehicles. The majority of heavy duty
applications using natural gas are buses due to the emissions benefits of natural gas
compared to uncontrolled or minimally controlled diesel technologies in developing
countries. Except for buses, there is little penetration of natural gas technologies in the
heavy duty sector; the U.S. is the exception and this discussed further below.
Table 1. World NGV Population by Vehicle Class
Total NGV population Cars, Buses and Trucks
LD+MD
+HD
Vehicles
11,931,328
LD Cars and
Commercial
Vehicles
MD+HD
Buses
MD+HD
Trucks
Others
11,236,843
400,370
206,789
87,326
Source: Gas Vehicles Report, October 2010

1
http://en.wikipedia.org/wiki/Natural_gas_vehicle
Figure 1 shows the distribution of NGVs by country. The majority of NGVs are
concentrated in Latin America and Asia Pacific regions. Most of these vehicles in these
regions are converted gasoline vehicles. Conversion costs in these countries are low due
to low cost conversion kits (less sophisticated gasoline technologies), low cost CNG
cylinders (steel), and low cost labor. These regions also have reasonably high gasoline
prices and natural gas costs are often 30% to 50% cheaper. Low conversion costs
coupled with fuel savingsand often government incentivesresults in quick payback
periods.
Source: J. Seisler, Clean Fuels Consulting working paper to TIAX on International Perspective NGV Market
Analysis: Light- and Medium-Duty Vehicle ownership and Production, April 2011.
Figure 1. Regional Distribution of NGVs throughout the World

Key to the penetration of NGVs worldwide has been the installation of CNG fueling
stations to meet vehicle fueling demands. Even thought the vehicles are capable of
gasoline or natural gas operation, the low cost of natural gas in comparison to gasoline
has lead to the demand for natural gas and the build-out of natural gas fueling
infrastructure. As shown in Table 2, the number of vehicles per station varies from 112
for the U.S. to 1,890 for India. The U.S. number is biased by more heavy duty
applications that use more fuel per vehicle. The average of this data set is 840 vehicles
per station. This is consistent with station costs and a reasonable rate of return based on
economic analysis performed by TIAX.
The purpose of this paper is to examine the viability of light-duty NGVs for the U.S.
retail market. As indicated most of the NGVs operating in the world are bi-fuel vehicles
that have been converted to natural gas. European automakers have been introducing
many new bi-fuel models into the market place and the next section reviews this
experience. How the world and European experience are related to U.S. conditions is
then discussed. Finally, the paper ends with proposed bi-fuel approaches for the U.S.
light-duty retail market.
Table 2. 2010 Fueling Infrastructure for Selected Countries

Country
India
Iran
Egypt
Argentina
Kyrgyzstan
Brazil
Italy
Bolivia
Pakistan
Peru
Bulgaria
Ukraine
Myanmar
Colombia
Thailand
Uzbekistan
China
United States
Totals
NGVs
Fueling
Stations
1,080,000
1,954,925
122,271
1,901,116
6,000
1,664,847
730,000
140,400
2,740,000
103,712
60,270
200,000
22,821
340,000
218,459
47,000
450,000
112,000
11,893,821
571
1,574
119
1,878
6
1,725
790
156
3,285
137
81
285
38
614
459
133
1,350
1,000
14,201
Vehicles
per fuel
station
1,891
1,242
1,027
1,012
1,000
965
924
900
834
757
744
702
601
554
476
353
333
112
838
Source: http://www.iangv.org/tools-resources/statistics.html
European Development of NGVs

Like the developing countries today, Europe NGV development started with conversions
or retrofits of existing gasoline vehicles. Italy, for example, started with conversions of
gasoline vehicles to NGVs in the 1930s. NGV technology has vastly improved since
these early conversions mostly due to the substantial improvements in gasoline
technologies over the last 20 years. Early conversions were performed on carbureted
gasoline vehicles without emissions controls. As emission controls were phased in
carburetors gave way to closed loop carbureted technologies which then gave way to
close loop fuel injection systems. Aftertreatment catalyst also became much more
efficient driving gasoline emissions to extremely low levels. Natural gas technologies
kept up with the advances in gasoline technology with multi point sequential injection,
engine controls, and exhaust aftertreatment.
Gaseous storage technology also evolved. Natural gas has a low energy density
compared to petroleum fuels and can be improved by compression. This however
requires high pressure storage containers which are more costly than gasoline or diesel
fuel tanks. Four types of storage cylinders are now manufactured:
Type I all steel cylinder
Type II fiberglass, hoop wound aluminum cylinder
Type III fully wrapped metal liner cylinder
Type IV 100% full composite cylinderliner and wrapping
Type I steel cylinders are the least expensive but weigh the most. Type IV cylinders are
the most expensivedue mostly to the cost of carbon fiberand weigh the least. For
light duty vehicles every 3 percent increase in weigh reduces fuel consumption by 0.6 to
0.9 percent.2 Pressure has also changed over the years from 2400 psi in some of the
earlier applications to 3000 psi used in Europe today to 3600 psi used in the U.S. Higher
pressure allows for more storage of natural gas and longer vehicle range.
European vehicle manufacturers are now offering a variety of natural gas bi-fuel models
for the retail light-duty market. According to NGVA, auto manufacturers including Fiat,
Mercedes Benz, Opel, Seat, Scoda, VW, Audi, Volvo, and Saab now offer 22 passenger
car models. Table 2 provides examples of OEM offerings for small and medium size
NGVs. All these models meet Euro V emission standards. A distinction is now be made
between NGVs with smaller gasoline tanks (<15 L) and those with larger gasoline tanks.
The former are referred to as mono-fuel and the latter as bi-fuel.
Table 2. Example of OEM Vehicles Available in Europe
OEM
Model
Fiat
Fiat
Fiat
Fiat
Mercedes
Benz
Mercedes
Benz
Opel
Panda 1.2 8V
Punto Evo 1.4 8V
Qubo 1.4 8V
Fiorino 1.4 8V
B 180 NGT
69
70
70
70
CNG
storage
kg
15
15
15
15
E 200 NGT
163
19.5
Power (hp)
Gasoline
storage (L)1
Range
(km)
CO2
g/km2
30
45
45
45
800
1000
950
960
107
115
114
119
54
1070
149
Zafira Tourer 1.6 CNG Turbo

150
25
14
680
129
ecoFlex
Opel
Comobal 1.4 CNG Turbo
120
16-22
22
750
134
ecoFLEX
VW
Up!CNG
68
11
79-86
VW
Passat 1.4 TSI EcoFuel
150
21
31
940
117
VW
Touran 1.4 TSI EcoFuel
150
18-21
11
650
128
VW
Caddy 2.0 EcoFuel
109
37
13
760
156
Source: www.ngvaeurope.eu/cars
1. According to Directive 2007/46/EC concerning type approval of vehicles, all petrol/gas vehicles
having a petrol tank not exceeding 15 liters should be classified as mono-fuel, beyond 15 liters
petrol tank size, the classification would be bi-fuel.
2. Regulation (EC) No 443/2009 says that "in the case of bi-fuelled vehicles (petrol/gas) the
certificates of conformity of which bear specific CO2 emission figures for both types of fuel,
Member States shall use only the figure measured for gas".
Assessment of Fuel Economy Technologies for Medium- and Heavy-Duty Vehicles, with Matthew A.
Kromer and Wendy W. Bockholt, Final Report prepared for the National Academy of Sciences,
Washington, DC, November 3, 2009.
European auto manufacturers have also introduced a number of vans that have
applications from commercial to private use. These manufactures and models include
Iveco Daily, Fiat Ducato, Mercedes Benz Sprinter, Fiat Dobolo cargo, Fiat Fiorino, and
VW T5.3 Clearly, either in the passenger car segment or the van segment European auto
manufacturers are now providing products instead of the previous retrofits/conversion
companies. Some conversions are still being done, but primarily through a qualified
vehicle manufacturing (QVM) program like Volvos V70 CNG bi-fuel vehicle. This is a
result of the more complicated emissions and engine/powertrain controls and
aftertreatment on modern gasoline vehicles. Integrating natural gas technologies to these
very complex gasoline technologies requires close interaction with the automakers.
An example of the packaging of natural gas components in these European NGVs is
shown in Figure 2. In this bi-fuel example, Fiat has located the CNG tanks underfloor
along with the gasoline tank. Fueling for gasoline or CNG is in a common vehicle
location as shown in the figure. Locating the fuel tanks underfloor does not compromise
the space in the vehicle as has been the practice in most NGVs to date.
Keeping the gasoline fuel system has several advantages. Key advantage is that the
vehicle is not solely dependent on CNG fueling infrastructure. Gasoline can be used in
cases where CNG fueling is not available or to extend the driving range. Operating on
gasoline can also be used to help to meet the very tight emission standards. For example,
auto makers are adopting the strategy of starting up on gasoline and then switching to
Figure 2. Example of European NGV Packing
http://www.ngvaeurope.eu/vans
natural gas to minimize methane emissions during cold starts with natural gas. Of course,
there are also disadvantages of bi-fuel operation since the engine can not necessarily be
optimized for natural gas operation. Higher compression ratio associated with 130 octane
rating of natural gas is not possible with todays engine technology without affecting
gasoline performance. Valves and valve seats have to be hardened for natural gas
operation. Ignition systems also need to be evaluated including spark plug durability.
Also, aftertreatment systems need to be optimized for both gasoline and natural gas and
methane emissions in natural gas operation need to be managed. European automakers
are adding a catalyst to reduce methane emissions from bi-fuel vehicles. Off setting
some of these issues, is that gasoline technology today is much more flexible than the
mechanical systems of the past. Nevertheless, natural gas technology will have to keep
up with the advancing improvements in gasoline technology aimed at improving fuel
consumption and CO2 emissions.
European gasoline prices are quite high compared to U.S. prices. Some example prices
for February 21, 20124 were:
Italy
1.80 /L (8.98 $/gal assuming 1.32 /$)
Germany
1.68 /L (8.38 $/gal)
Sweden
1.63 /L (8.13 $/gal)
France
1.60 /L (7.98 $/gal)
CNG prices in these same countries range from 0.80 /kg to 1.15 /kg or on an equivalent
gasoline energy basis (liter gasoline equivalent, Lge) 0.53 /Lge to 0.77 /Lge.5 CNG is
therefore 30% to 50% cheaper than gasoline. This fuel savings can be use to offset the
higher costs of the CNG equipped vehicles. Figure 3 shows a simple payback analysis
for the fuel prices in Italy. Average annual vehicle kilometers travel in the EU15 is
10,450.6 Fuel consumption was assumed at 7.8 L/100km. With these assumptions nearly
all incremental costs are within an acceptable 3 year payback. Lower fuel consumption
increases the payback period.
A more specific analysis was also performed for the recently announced Opel Zafira
Tourer. The CNG version of this vehicle has a best in class 530 km natural gas range
with 25 kg CNG capacity and a 14 L auxiliary gasoline tank.7 This vehicle is a multi
passenger vehicle (MPV) with seating up to 7. Figure 4 shows a schematic of the vehicle
with CNG tanks located underfloor. The CNG version of this vehicle which includes start
stop technology is priced at 27,950 (recommended price in Germany including VAT).
A comparably equipped gasoline version of this vehicle (1.4 L turbo rated at 103
kW/140hp) retails for 24,150 with fuel consumption of 6.3 L/100km slightly better than
the CNG version.
http://www.drive-alive.co.uk/fuel_prices_europe.html
http://www.cngprices.com/station_map.php accessed April 13, 2012
6
J. Seisler, Clean Fuels Consulting working paper to TIAX on International Perspective NGV Market
Analysis: Light- and Medium-Duty Vehicle ownership and Production, April 2011.
7
http://media.opel.com/content/media/intl/en/opel/news.detail.print.html/content/Pages/news/intl/en/2011/O
PEL/12_08_opel_zafira_tourer_cng
5
14.00
Incremental vehicle costs
12.00
CNG .81 e uros/kg

Petrol 1.80 e uros/L
European Car
10450 VKT
7.8 L/100km
Simple Payback (yrs)
4500 e uros
10.00
3400 e uros
8.00
6.00
2090 e uros
4.00
2.00
0.00
0
Price of Petrol (euros/L)
Figure 3. Payback in years for incremental vehicle costs decreases with

higher gasoline prices
Source: Opel
Figure 4. Opel CNG Zafira Tourer 1.6 L Turbo ecoFLEX with 110 kW/150 hp.
Natural fuel consumption 4.7 kg/100km
A simple payback analysis was performed for this vehicle using the gasoline fuel prices
in Germany and a range of CNG prices. These results are shown in Figure 5. For
gasoline at 1.68/L and CNG prices ranging from 0.80/kg to 1.15/kg, paybacks range
from 4.4 to 5.75 years. Although outside the 3 year payback target, little changes in
either CNG or gasoline pricing would potentially make a difference on consumer
acceptance.
8.00
7.00
Opel Zafira Tourer Manual

10450 VKT
CNG 7.05 Lge/100km
Petrol 6.3 L/100km
1.15 e uros/kg
Simple Payback (yrs)
6.00
0.80 e uros/kg
5.00
Zafira vehicle costs with delivery

CNG 28,610 euros
Petrol 25,510 euros
4.00
3.00
Petrol 1.68 e uros/L

2.00
1.00
0.00
Price of Petrol (euros/L)
Figure 5. Simple payback analysis for recently introduced Opel CNG Zafira
Tourer
European automakers are leading the world in the development and sales of CNG bi-fuel
(and mono-fuel) vehicles. CNG bi-fuel vehicles sold in Europe to retail customers have
integrated the CNG and gasoline storage tanks so as to not affect vehicle functionally.
They have also designed these vehicles to have comparable attributes on vehicle range
and performance. Retail customers are not sacrificing vehicle attributes with these
offerings and provided the customer has convenient access to CNG fueling
acceptable savings are possible if natural gas is used.
United States Development of NGVs

U.S. experience also paralleled Europe with the first NGVs converted gasoline vehicles
using technology developed in Italy and the Netherlands. This was followed by
companies offering conversion systems and subsequently by the U.S. automakers
developing dedicated and bi-fuel technology in the late 1980s and early 1990s. Although
conversions were favored initially, the development of sophisticated light-duty emission
controls in the 1990s made it difficult for the conversions to meet emission levels
achieved by the gasoline vehicles without substantially more integration with the OEM
(original equipment manufacturer) engine and emissions systems. In fact some retrofit
and conversion systems actually increased tailpipe emissions, leading the U.S. EPA and
California ARB to require conversion and retrofit suppliers to emission certify their
systems. Ultimately, this increased the cost of the retrofit systems since the certification
costs are amortized over a small number of vehicles.
Energy legislation in the U.S. required government and fuel provider fleets to purchase
light duty alternative fuel vehicles (EPAct 1992)8 which help to develop the demand for
NGVs in the late 1990s early 2000s. Alcohol flexible fuel vehicles were also introduced
into the market in the mid to late 1990s. Automakers received CAFE (corporate average
fuel economy) credits for manufacturing these alternative fuel vehicles.
Fuel use was not required by EPAct and many of the bi-fuel or FFVs used only gasoline.
This was a result of very sparse or non-existing fueling facilities for alcohol fuels (first
methanol and then ethanol) and compressed natural gas. One exception was vehicles
placed in many of the utilities around the U.S. (gas and/or electricity suppliers). Here the
utilities built the infrastructure to supply high pressure natural gas (CNG) to their
dedicated and bi-fuel light duty vehicles purchased to meet EPAct requirements. These
stations were also used by other fleets to fuel their NGVs.
The second factor that hurt the penetration of alternative fuel vehicles and use of
alternative fuels was the drop in oil prices after the first Gulf war (1992) and the relative
stability of prices throughout the 1990s. Low oil prices drove down the price of gasoline
and the lower the price differential between gasoline and natural gas. This made it
particularly hard for natural gas to complete with gasoline, since fuel savings were
insufficient to reasonably payback the higher upfront vehicle costs.
Unlike Europe and other regions in the world, U.S. gasoline prices are much lower due to
higher taxing of gasoline in these other regions. As shown previously, these higher
gasoline prices coupled with low CNG prices makes it possible to amortize the higher
CNG vehicle costs over reasonable payback periods. For the U.S. market with low
gasoline or diesel fuel prices, the primary factor affecting payback periods is the amount
of fuel used. Figure 6 shows estimated average fuel economy and fuel use for various
U.S. vehicle segments. The light duty segment fuel economy assumes full adoption of
the recent fuel economy rule making (average fuel economy for MY 2016).9 In this
example, the high fuel use fleets are mostly heavy duty, but not illustrated are light-duty
fleet applications like taxi cabs that can annually travel 70,000 miles or more.
Energy Policy Act of 1992. See for an overview of the alternative fuel requirements:
http://www.afdc.energy.gov/afdc/laws/key_legislation
9
NHTSA, Corporate Average Fuel Economy for MY 2011 Passenger Cars and Light Trucks, Final
Regulatory Impact Analysis, March 2009
10
Source: TIAX estimates made in 2010
Figure 6. Fuel Use and Average Fuel Economy of U.S. Vehicle Segments
Figure 7 illustrates how effective fuel use is in paying back the higher upfront costs of
NGVs. Shown in this figure are the incremental costs that can be amortized in 3 years (3
year payback) for two fuel price differentials. The discount rate in this analysis was 8
percent. Heavy duty vehicles using upward of 20,000 gallons per year (line haul tractor
trailer) can afford incremental costs ranging from $50,000 to $110,000 depending on the
price spread between natural gas and diesel. Conversely, with lower incremental costs
the payback period would be reduced. Transit buses use 13,000 gallons of diesel fuel per
year and can afford increased costs ranging from $34,000 to $69,000.
Transit buses and refuse applications have been very successful at converting from diesel
to natural gas. This success has depended on a variety of factors:
1. reasonable vehicle payback periods or user economics
2. high enough fuel demand at return to base facilities to justify fueling
infrastructure
3. economics of scale and reasonable fueling station costs to provide high fuel price
differentials
4. little or no vehicle attribute differences between natural gas and diesel vehicles
5. lower local emissions (at least up until 2010 diesel technologies)10
Conversely, the penetration of NGVs into the tractor trailer truck segment has been much
slower as a result of little or no fueling infrastructure and limited product from the engine
and truck manufacturers. This is currently changing especially with the growing price
differentials between diesel and natural gas.
10
Natural gas still has lower overall emissions of criteria or local emissions due lower upstream fuel cycle
emissions.
11
Incremental Vehicle Costs for 3 yr Payback
$120,000
8% discoutn rate
$100,000
$80,000
Fuel Price Differential $2
$60,000
$40,000
Fuel Price Differential $1
$20,000
$0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
20,000
Annual Fuel Use (gallons of gasoline or diesel)
Figure 7. Annual fuel use required to amortize incremental vehicle costs

over 3 years
The heavy-duty sector has been relatively successful for NGVs provided the applications
duty cycle uses enough fuel and a reasonable business case can be made for building
fueling infrastructure. If the fleet is large enough or if the demand from several fleets can
be aggregated, then a good business case is possible for both the end user and fuel
supplier.
The situation is not so clear for high duty vehicles especially for the retail customer
where annual gasoline use is 500 gallons or less.11 It is very difficult to offset the higher
NGV costs with low annual use. Figure 8 compares payback periods for several
advanced light-duty vehicles including both natural gas and gasoline hybrids. The
advanced technologies are compared to equivalent gasoline models and the assumptions
on vehicle price (MSRP) and fuel economy are shown in the figure. At a gasoline price
of $3/gal only the Toyota Prius has a payback less than 5 yrs. Only at near gasoline price
highs (near $4/gal) does the Honda GX dedicated NGV start to have reasonable payback.
11
Much of the following discussion was taken from work performed by theCarLab as a subcontractor to
TIAX.
12
16.0
Annual Mileage 10,000
14.0
CNG Price $2.00/gge
Years to Payback
12.0
10.0
Prius-Camry
8.0
Camry Hybrid-Camry
Civic Hybrid-Civic LX
6.0
Civic GX-Civic LX
4.0
2.0
0.0
2
5
6
7
Gasoline Price ($/gallon)
Assumptions:
Vehicle Make and Model
MSRP
MPG
Honda Civic GX
$25,280
28
Honda Civic Hybrid
$23,800
42
Honda Civic LX
$18,360
29
Toyota Camry
$19,720
26
Toyota Camry Hybrid
$26,150
34
Toyota Prius
$22,800
50
(8% discount rate is included in this payback analysis)
Source: theCarLab
Figure 8. Years to payback higher initial costs of natural gas and

gasoline hybrid technologies.
The Honda GX is a dedicate NGV and therefore dependents on a convenient fueling

infrastructure. Honda did for a period market a home CNG appliance called Phill
through Fuelmaker. The appliance is now being market by an Italian Company BRC Gas
Equipment. Phill costs were about $4,500 and depending on the customers residential
gas rate, installation costs, operating and maintenance costs, the resulting cost of CNG
could be in the $3 to $5 per gasoline gallon equivalent.12 Anecdotal comments on
Hondas experience of selling both the GX and Phill in Southern California indicated that
once the customer understood the existing fueling infrastructure this was deemed
acceptable and not worth the additional investment for home refueling. Of course, not all
regions of California let alone the U.S. have as much CNG infrastructure as Southern
California. Other regions are therefore dependent on the need to aggregate demand as
12
Whyatt, GA, Issues Affecting Adoption of Natural Gas Fuel in Light- and Heavy-Duty Vehicles,
Pacific Northwest National Laboratories, PNNL-19745, September 2010.
13
fast as possible in order to not strand CNG station investments. A bi-fuel vehicle could
help in this regard provided station investments are made.
The relative success of hybrid electric vehicles (HEVs) for personal consumer use
strongly illustrates the advantage of leveraging gasolines extensive and familiar
distribution system to lower consumer objections to alternative powertrains. The recent
sales volume of FlexFuel gasoline/ethanol vehicles (FFVs) also illustrates this point,
even if few buyers actually use the intended ethanol capability. Both vehicle types have
outsold other alternative fuels/powertrains precisely because consumers are not asked to
change their behavior. In such cases, only economic resistance is then left to overcome.
In the case of FFVs, the case to adopt these vehicles is strengthened by the fact that
incremental vehicle costs to the buyer are essentially zero. Fundamentally, FFVs,
hybrids, and bi-fuel NGVs are essentially equivalent, differing only in the form of
alternative energy storageethanol, battery, and natural gas, respectively (Figure 9).
Figure 9. Fundamentally, FFVs, hybrids, and bi-fuel NGVs are

essentially equivalent, differing only in the form of alternative
energy storageethanol, battery, and natural gas,
respectively.
Ethanol flex-fuel capability was essentially provided without incremental cost to
consumers because the costs to OEMs were minor to allow the vehicles to run on alcohol.
Millions of such vehicles are and were sold, as consumers effectively faced no real tradeoff relative to the gasoline-only version. HEVs, however, have significant cost (for
batteries, controllers, motors, and other components) that ultimately must be recovered by
OEMs from consumers. In comparison to dedicated EVs, though, HEVs have a cost
advantage principally because the volume of battery they must contain is far lower than
required for dedicated EVs, and these batteries are by far the highest cost item in the
vehicles build. Hybrids are therefore much less expensive to build and buy than full
EVs, which, when combined with their easy use of existing fueling infrastructure, make
14
them much more rational options for consumers than dedicated EVs. Therefore, while
HEVs have outsold pure battery electric vehicles (BEVs), their cost premium continues
to be a constraint on their success over gasoline vehicles.
In contrast, the cost of a bi-fuel NGV is nearly the same as that for a dedicated NGV,
which means actual cost or purchase price does not affect a comparison of the two.
Instead, the relative advantages of each must be compared from the perspective of the
end user. Here again, bi-fuel has the clear advantage precisely because the buyer is not
forced to change behavior, especially in cases where range or resultant drive routes might
be impacted. Instead, drivers of such vehicles can selectively take advantage of the lower
operating cost and greener footprint of natural gas, knowing that there is no walk home
factor that threatens their convenience or safety should travel take them beyond natural
gas pumps. Drivers of such vehicles simply have more choice when the fuel range and
availability issues that plague EVs, hydrogen vehicles, and dedicated NGVs are removed.
As HEVs (the equivalent of bi-fuel EVs) are to dedicated EVs, bi-fuel NGVs are
potential fatal competitors to dedicated NGVs at least in the light-duty retail market. As it
is with HEVs, the cost premium of bi-fuel NGVs is a natural constraint on their success
over gasoline vehicles.
The lower natural gas energy density compared to gasoline means that sufficient tank
volumes cannot be achieved for almost any dedicated NGV light car or truck without
degrading the effective size of the remaining vehicle as shown in Figure 10 for two
vehicle examples. Compromises, for any reason, to occupant package (logically limited
to second or third row seating volume) have historically been detrimental in terms of
buyer appeal and subsequent market share. This has led many OEMs and small volume
manufacturers (SVMs) to move natural gas tanks into cargo areas, a fact aptly
demonstrated by NGV conversions of cars such as the Ford Crown Victoria for taxi use.
While regulated livery fleets are often forced to accept such impositions on usability
even despite still extant luggage capacity demands private consumers are so far not
inclined to do so.
Vehicle performance is more challenging for NGVs, as private users will accept little
compromises in the long term. Natural gas offers slightly lower performance relative to
gasoline in unmodified gasoline engines, and this presents both a planning and
engineering challenge. Here, NGV creators must resist the temptation to apply natural gas
to the lowest specification gasoline engines offered in particular models in an effort
maximize fuel economy. Rather, conversions and bi-fuel NGV installations are better
applied to mid and upper trim level powertrains to meet or exceed customer expectations,
especially as natural gas is in the nascent stages of broad market exposure. Looking much
farther forward, it is obvious that dedicated NGVs designed from the ground up should
have engines optimized for natural gas, especially in terms of usable compression ratio.
15
Figure 10. Recently introduced 2012 Dodge Ram 2500 Pickup Truck and
Ford Crown Vic taxi cab application. Both pictures show how
CNG storage tanks were integrated in vehicle.
As a variety of Battery Electric Vehicles (BEVs) and HEVs enter the North American
market, consumers are becoming conditioned to calculate payback period when
considering any AFV. Volume hybrids, for instance, owe much of their success to
relatively favorable payback scenarios, with the Prius having the best payback of all
contemporary hybrids (as was shown in Figure 8). Current annualized fuel costs for light
vehicles are relatively insignificant when compared to the cost premiums for acquisition
of most AFVs. This issue is critical if NGVs are going to be accepted by the retail
customer.
Bi-Fuel Vehicles for the U.S. Retail Market

The U.S. retail market for light-duty vehicles is highly competitive with automakers
providing a variety of gasoline and alternative fueled vehicles. Hybrid electric vehicles
have been quite successful at least compared to other alternatives due to favorable user
economics. Since U.S. energy pricing for petroleum based fuels and natural gas is less
favorable than other regions, a different vehicle approach maybe needed for NGVs to
compete against gasoline. Unlike European NGV designs, the fuel price differential is
insufficient to support the higher costs of their bi-fuel vehicles. Instead, U.S. bi-fuel
vehicles will require lower costs than the European market.
The dominant costs of bi-fuel NGVs are the CNG storage tanks. If bi-fuel NGVs are
going to be successful in the U.S. market, storage costs need to be reduced. Currently,
just the opposite approaches seem to be happening both in Europe and the U.S. The
prevailing experience seems to be to maximum CNG storage and therefore range
achievable on natural gas. In other words, design a bi-fuel vehicle that maximizes CNG
range but also can operate on gasoline in order to extend range beyond that of the volume
and costs constraints of CNG. Another way to approach the design of bi-fuel NGVs is to
design the CNG storage system to meet only the demands of most every day drivers and
to have a gasoline capacity that would match the range requirements of consumers. This
approach is similar to the philosophy of HEVs and to the plug-in HEVs (PHEVs). Like
16
batteries natural gas storage is both expensive and takes up lots of space. Minimizing
CNG storage would reduce costs and make it easier to package on vehicles.
Cumulative Percentage of All

Drivers (%)
The PHEV analogy is interesting from two perspectives. First, PHEV battery energy
storage is been designed to meet the average daily mileage of most retail users of 40
miles or less per day. As shown in Figure 11, 2/3 of all drivers in the U.S. drive fewer
than 40 miles per day and over five days per week 52 weeks per year this results in
10,400 miles per year close to the nominal 12,000 miles annually driven in the U.S.13
This being the case, then natural gas storage should be reduced to provide this range.
Depending on the vehicle and model year this is equivalent to 1 or 2 gge or 14 to 28 liters
(water volume at 250 bar). This is almost a 10 times reduction from current CNG
vehicles. The Honda GX has a 8.3 gge (113 liter) storage tank and the Opel Zafira
Tourer has a 9.8 gge (173 liter at 200 bar) storage tank.
100
80
60
40
20
0
0
20
40
60
80
100
Daily Miles Driven
Figure 11. 67 percent of all drivers in the U.S. drive fewer than
40 miles daily, a consideration when designing
alternative fuel tank capacities.14
Secondly, the PHEV analogy requires home refueling. Most consumers would be
unwilling to refuel their vehicle every day unless it was convenient. Cars and light trucks
today are designed with a refueling range of around 350 miles. At average annual
mileage, this works out to 40 fueling events per year or nominally once per week.
Asking consumers to fuel once per day is unacceptable. The savings from reducing the
storage costs could offset the costs of a home refueler. However, this home appliance
would not have to be designed to the same characteristics as the Phill unit. Phill was
designed to provide 0.42 gge/hr at 3600 psi. For a 2 gge storage tank, this rate could be
halved and with some storage this rate could be further reduced.
13
There is distribution of daily mileages that will limit the penetration of vehicles designed for a daily
range of 40 miles.
14
XPrize Foundation. National Household Travel Survey Data Summary for XPrize.
http://www.progressiveautoxprize.org/files/downloads/auto/AXP_FHWA_driving_stats.pdf. March 2007.
17
Figure 12 shows how the economics of this concept might play out for a bi-fuel Honda
Civic GX. Here it was assumed that the incremental vehicle costs could be reduced from
$6,920 to $4,080 by reducing CNG storage from 8.3 gge to 1.5 gge or enough for 40 +
miles on CNG. Secondly, it was assumed that a simpler home CNG appliance could be
manufactured and installed for $3,000. A discount rate of 8 percent was also assumed.
As shown, with gasoline prices at $4/gal and CNG prices at $1.8/gge it would take over 7
years to payback both the vehicle and fuel appliance costs. This is probably too high but
some of these costs would be offset by not having as frequent visits to gasoline stations
(assuming this is a benefit consumers are willing to value).
40
35
Bi-Fuel GX Concept
1.5 gge CNG Storage
Incemental V ehicle Costs $4,080
Home Appliance $3,000
80% CNG 20% G asoline
12,000 miles/yr
Years to Payback
30
25
20
15
10
5
0
0
10
Gasoline Price ($/gal)
Figure 12. Estimated payback for 40 mile CNG range bi-fuel vehicle
It is possible the vehicle costs could be further reduced in volume production especially
since the tank volumes and presumably costs have been substantially reduced. A more
sophisticated analysis would be needed to investigate this. Similarly, a simplified design
and cost analysis of a home fueling appliance is also needed.
Another interesting option with reduced storage is to reduce storage pressure. Much of
the costs of the tanks and compression are related to the systems operating pressure. If
vehicles only need several gge of natural gas, it may be possible to simplify storage by
reducing the pressure. Pressure and volume are related so decreasing the pressure would
increase the volume, but perhaps more conformable shapes could be used to help vehicle
packaging. Reduced pressure would also reduce stages of compression possibly
simplifying compressor design and function. It is possible that at low pressures total
vehicle and home appliance costs could be further reduced.
18
An obvious drawback of reducing storage pressure is that the existing CNG fueling
stations would not be usable unless the pressure was regulated down. Even in this
situation the potential safety issues may out weigh the advantages of lower natural gas
storage pressures. Other disadvantages of bi-fuel compared to dedicated operation are
compromises in:
vehicle performance (power and torque)
fuel consumption
emissions
Some of these disadvantages can be overcome with todays technologies and some will
require more advanced engine and powertrain technologies. For example, the emission
performance of todays vehicles is extremely low whether for a gasoline or natural gas
vehicle. Integrating gasoline and natural gas together and meeting the most stringent
emissions will require effort, but this should not be insurmountable.
Policies at the federal and state levels may also need to be changed so that bi-fuel NGVs
have the same incentives as other alternative fuels. In recent TIAX work, NGVs both
dedicated and bi-fuel were compared on a full fuel cycle analysis to electric vehicles (for
different generation mixes), and ethanol vehicles (with different feedstocks). Societal
costs were estimated for criteria (or local) pollutants, greenhouse gases, and petroleum
dependency. Surprisingly, the societal benefits were about the same for each alterative
averaging about $3,000 over the vehicles lifetime.
Acknowledgement
Sources and references for this work were taken from recent TIAX work on light- and
heavy-duty natural gas vehicles; TIAX staff that contributed to this work included Karen
Law, Jeffrey Rosenfeld, Michael Chan, and Jon Leonard. TIAXs efforts were also
supported by efforts from Jeffrey Seisler, Clean Fuels Consulting, and theCarLab. The
concept of lower CNG storage volumes was advanced by theCarLab.
19
Evolution of alcohol fuel blends towards a sustainable

transport energy economy
J.W.G. Turner and R.J. Pearson
Lotus Engineering
E. Dekker
BioMCN
B. Iosefa
Methanex Corporation
G.A. Dolan
Methanol Institute
K. Johansson and K. ac Bergstrm
Saab Automobile Powertrain AB
ABSTRACT
Work undertaken by Lotus Engineering and partners has shown that the miscibility of
the low-carbon-number alcohols with gasoline provides a powerful tool to enable the
introduction of methanol as a transport fuel in a wholly evolutionary manner. This is
primarily facilitated by the fact that the CAF regulations in the US (together with
other incentives in other countries, such as Sweden) have created a situation in which
there are 7-8 million E85/gasoline flex-fuel vehicles on the roads there, for which
insufficient E85 can be provided at an affordable price. This paper will show that the
introduction of methanol (which can be made extremely simply and cheaply from
natural gas) into gasoline-ethanol mixtures, can be used to create drop-in fuels
equivalent to E85 and can bring the price of an alcohol-based fuel for spark-ignition
engines down to less than that of gasoline (on a per-unit-energy basis, before tax is
applied). It can thus more than compete with that fuel. This opens up the possibility
of using the USs reserves of naturally gas (be it conventional or unconventional types
such as shale) immediately to manufacture methanol to displace gasoline, as a bridge
to a broader energy economy based on higher concentrations of methanol made from
renewable sources.
The vehicle work conducted has shown the possibility of realizing this state of affairs,
and related laboratory tests on some of the potential fuel blends have similarly
demonstrated that they possess some of the necessary characteristics to be truly dropin alternatives to E85. These necessary characteristics are considered to be equal
volumetric energy content (to enable compliance with on-board diagnostics
requirements), equal octane numbers and latent heat (to provide invisibility to the
combustion and air handling systems), and inherent miscibility with gasoline (to avoid
the requirement to change the fundamental nature of the vehicle fuel system). A
further requirement in practice is that the vehicles still be of adequate performance
with regard to tailpipe emissions standards using the existing exhaust after treatment
systems fitted to them. In the present work this is demonstrated by reporting data for
oxides of nitrogen emissions (NOx) taken from a standard production Saab 93
certified to the EU5 emissions standard using ternary blends (such NOx emissions
being especially important from a human health point of view in built-up areas),
which shows that generally such emissions are significantly lower for all of the
alcohol blends than for gasoline. All results are found to be well within the EU5
limits, with the gasoline results showing that the after treatment system was indeed
functioning correctly.
INTRODUCTION
Around the world, concerns with climate change and energy security have prompted
the investigation and introduction of renewable fuels in order to reduce usage of fossil
oil. In the US, the Energy Independence and Security Act of 2007 (and related
Renewable Fuel Standard 2) has mandated that a total of 36 billion US gallons of
ethanol be used in the fuel pool by 2022 [1], and in the European Union (EU) the
Renewable Energy Directive (RED) seeks to establish a minimum proportion of
renewable energy in the fuel pool of 10% by 2020 [2].
The conversion of fossil hydrocarbons to carbon dioxide (CO2) causes atmospheric
levels of greenhouse gas to increase, which, due to the fact that much of the worlds
oil supply comes from areas outside of those of the main consumer regions, gives rise
to a further concern with respect to security of energy supply.
The European situation is complicated by the facts that diesel penetration in the
vehicle pool is high (at approximately 50% in the light-duty sector) and the volume of
bio components in diesel which it is practical to include in the fuel is limited to
approximately 7% by volume if future emissions standards are to be met. Together
these imply that the proportion of ethanol blended into gasoline in Europe will have to
be approximately 13% by energy, which equates to ~20% by volume as a result of the
lower volumetric lower heating value (LHV) of ethanol versus gasoline. Although
most current vehicles fitted with spark-ignition (SI) engines can accept 10% by
volume ethanol as standard (a situation essentially initiated by the presence of 10%
ethanol in gasoline in wide areas of the US), 20% is beyond their capability.
Realization of this fact, coupled to the impending fines on the fuel suppliers if they do
not meet their legal obligations under the RED, has prompted calls by one oil major to
give vehicle OEMs credits in terms of tailpipe CO2 for any E851/gasoline flex-fuel
vehicles that they manufacture, in order to produce a larger market for high-blendconcentration ethanol fuels [3].
This situation is desirable since selling large volumes of ethanol is probably the most
pragmatic way for the fuel suppliers to comply with the requirements of the RED and
the related Fuel Quality Directive (FQD) (which defines minimum standards before a
fuel can be considered a biofuel). In actual fact, the EU vehicle tailpipe CO2 fines
system does presently allow a 5% reduction in tailpipe CO2 to be claimed for any
flex-fuel vehicle that an OEM sells, provided one-third of the fuel forecourts in the
country in which it is sold has at least one E85 pump [4]. It could be said, therefore,
that the potential remedy to the RED impasse for the fuel suppliers is in fact in their
own hands. Furthermore, for a theoretical vehicle at the 2011 EU average of 145.1
1
Throughout this paper, the use of E followed by a number refers to the proportion by volume of
ethanol in a blend. The same applies for M (methanol) and G (gasoline). Thus E85 is nominally 85%
ethanol in bulk gasoline (a high blend rate), and E10 is 10% ethanol in bulk gasoline (a low blend rate).
gCO2/km, and at the highest proposed fine rate in 2015 of 95/gCO2, this represents a
saving to the OEM of 689 per car, which the authors contend is significantly greater
than the costs of modifying a standard gasoline-fuelled vehicle to be flex-fuel with
E85 in the first place. Thus all of the notional prerequisites are in place for ethanol to
become a major transport fuel, which begs the question as to why this should not
already be so.
Ethanol as a minor blend component in gasoline has two main benefits: firstly, there
is the renewability and energy security factor, and secondly it is an excellent octane
enhancer, in part because of its high heat of vaporization [5]. This latter fact means
that even at low blend rates of 5-10% it can provide a significant uplift in octane
number, which concomitantly means that a fuel supplier can reduce the volume of
other octane enhancers in the bulk gasoline [6], reducing its net price and increasing
profits. Unfortunately this low blend effect means that the price of ethanol is kept
high and is closely tied to the price of gasoline. Thus, when it is used at high blend
rates to make E85, there is little decontenting possible in the gasoline comprising the
remaining 15% of the fuel; in fact, in commercial E85, the bulk gasoline often has to
have its composition altered to facilitate cold starting, ethanol being a difficult fuel in
this respect2.
Hence, any mechanism to offset the high price of ethanol while still permitting its use
in large volumes across the fuel pool will be of benefit to the fuel suppliers and, if
they are encouraged to put pumps with the necessary capability on sufficient fuel
station forecourts it would also be of benefit to OEMs selling in the EU, providing
fuel renewability factors as mandated by the RED and FQD are adhered to. If the
resulting fuel was cheaper than gasoline to use in terms of operating cost the
consumer would readily move to its use. Approached in terms of taxation per unit
energy, migration to this situation could be achieved without a reduction in tax take
for governments, together with no requirement for direct subsidies, which are
necessary in the case of electrification of the vehicle fleet. Hence all stakeholders
could benefit if a suitable introduction mechanism could be found.
At the same time, the biomass limit for ethanol production has been used by some as a
reason not to pursue alcohol fuels for transport, since only about 27% of the energy
required can be gathered within it (this figure varies country by country)3. The
biomass limit only applies to fuels made using biological processes (such as
bioethanol and biodiesel). In fact, using thermochemical processes, it is possible to
manufacture liquid fuels from anything containing carbon and hydrogen via FischerTropsch chemistry or a syngas-to-methanol-to-gasoline (or similar) process.
Thermochemical routes therefore open up the possibility of using more waste as a
carbon feed stock, meaning that the amount of renewable fuel which could be
manufactured moves beyond the biomass limit and prevents more conventional
biofuels from being regarded as a strategic dead end. As an end game, in order to
cover the full amounts of energy necessary for transport, atmospheric CO2 and
2
Note that commercial E85 is often not configured with 85% ethanol; US limits are 51-83% by
volume. Generally, in winter months ethanol concentration is often reduced to 70% to aid cold
starting, and even in summer months the ethanol component may only comprise 77%.
3
It is interesting to contrast this with the fact that even in optimistic scenarios electric vehicles are not
expected to penetrate to more than 10% in the short term, yet that is not seen as a reason not to pursue
them vigorously.
molecular hydrogen could be used as the physical ingredients to carry renewable

energy either in liquid or gaseous form [7-9].
Thus there exist various possibilities to increase renewable fuel supply as a result of
the miscibility between gasoline and the alcohols, the fact that a liquid fuel
infrastructure already exists, that the necessary vehicles also exist and are cheap to
manufacture, and that the feed stocks required are not limited if the necessary
technologies can be developed. The only thing missing is to construct a route to
enable this scenario to play out, with the necessary fuel and vehicle specification
changes linked to it. A first step along this road to energy security and sustaianability
would be to employ ternary (three-component) blends of gasoline, ethanol and
methanol.
TERNARY BLENDS OF GASOLINE, ETHANOL AND METHANOL
Gasoline, ethanol and methanol are all miscible together and ternary (i.e. threecomponent) blends can be configured to have the same target stoichiometric air-fuel
ratios (AFRs) as any binary gasoline-ethanol blend. In the present paper we
concentrate on such GEM ternary blends with a target stoichiometric AFR of 9.7,
i.e. that of E85, but equally ternary blends targeted at E10, E22, etc. could be
arranged. For a fixed stoichiometric AFR the relationship between them is defined by
linear volumetric relationships and this has been discussed in detail in earlier
publications [10,11]. Furthermore, when configured in an iso-stoichiometric manner,
all such blends have near-identical volumetric lower heating value (LHV) and
practically the same octane numbers when configured to constant stoichiometric
AFRs; they also have extremely close enthalpies of vaporization (to +/- 2%). For the
case of a stoichiometric AFR of 9.7:1 the volume relationship between the
components is shown in Figure 1, as determined using the Lotus Fuel Mixture
Database [12]. In this figure one can see that as the volume percentage of ethanol is
reduced, so the rate of increase of the methanol proportion is faster than that of the
gasoline proportion. This is because as one volume unit of ethanol is removed, a
volume unit of the binary gasoline-methanol mixture with the same stoichiometric
AFR as ethanol (i.e. 9:1) has to be used to replace it, and the necessary volume ratio
of gasoline:methanol is 32.7:67.3, as discussed in detail in [10].
Several points of interest arise from Figure 1: firstly (and most obviously) is that E85
contains no methanol; secondly, that the binary equivalent of E85 for a gasoline and
methanol mixture occurs at volume percentages of 44 and 56, respectively (i.e. the left
hand limit, where no ethanol is present); and thirdly, the ratio where the proportion of
gasoline and methanol are equal occurs at approximately 42.5 volume percent
ethanol. Aspects related to this will be returned to later.
Initial experimental results using four such GEM blends in a production vehicle
showed that, provided a certain minimum level of cosolvent was present, the blends
were invisible to the vehicles on-board diagnostics (OBD) system [10]. Ethanol
performs the cosolvency function in gasoline-methanol mixtures, and the minimum
level of ethanol concentration in a GEM blend was further investigated in a car
certified to a higher emissions standard and using a different alcohol concentration
sensor technology (actually a physical sensor as opposed to the virtual sensor the first
vehicle used). Here, no such minimum requirement for ethanol was identified,
despite repeated cold soaks to -20C and cold start tests [11]. From these pieces of
work it is presumed that a minimum ethanol concentration is needed to ensure
satisfactory operation of all of the vehicles in the fleet, since they do not all use the
same alcohol sensing technology.
Proportion of Gasoline or Methanol in Ternary

Blend / [% vol.]
60
55
50
45
40
35
30
25
20
15
10
5
0
0
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
Proportion of Ethanol in Ternary Blend / [% vol.]

Gasoline
Methanol
Fig. 1: Relationship between blend proportions of gasoline, ethanol and methanol in

iso-stoichiometric ternary blends configured with a stoichiometric AFR of 9.7. Blend
ratios determined using the Lotus Fuel Mixture Database [12]
Note that it is possible to produce ternary blends of other alcohols with gasoline,
should their use be beneficial with regard to the utilization of all available feed stocks,
or even quaternary (or higher number) blends; examples of these may be mixtures of
gasoline, methanol and butanol with or without ethanol respectively. It is intended to
investigate blend ratios of these in a later publication [13].
TEST FUELS, VEHICLE AND EXPERIMENTAL RESULTS
Test Fuels
The fuel blends used were as described in Table 1. Note that the terminology used to
describe the blends from this point hence in this paper is as follows: G, E and M refer
to gasoline, ethanol and methanol, and the percentage proportion by volume is given
after each letter (i.e., GEM component ratios). Hence E85 would be G15 E85 M0 and
from Figure 1 the binary equivalent using gasoline and methanol only would be G44
E0 M56. The blends were given the names shown in the table.
Several points of interest arise from the choice of these fuel blends. Blend C takes the
same amount of ethanol as was used to make one volume unit of Blend A (the
commercial E85 surrogate) and spreads it across four times the volume of fuel.
Similarly, Blend D4 takes the same volume of ethanol and spreads it across 8.5 times
the volume of fuel. Thus, if the amount of ethanol that can be supplied is constrained
for any reason by feed stock supply, a desire to avoid interference with the food
chain, or concern over indirect land use change (ILUC), for example one can extend
how far the limited amount of ethanol can reach into the fuel pool by introducing
methanol in a ternary blend with it. The situation is improved if the methanol used is
better, from an energy security or carbon intensity perspective, than gasoline. It
should be pointed out that this is effectively the situation in the US, if one considers
that the Energy Independence and Security Act mandates the production of a specified
amount of ethanol. This can be coupled to the recent shale gas finds and the ease with
which methane can be turned into methanol, and is synergistic with the fact that there
exist many more vehicles which can take high-alcohol blend fuels than currently use
them. The subjects of gasoline displacement and cost will be returned to in the
Discussion.
Table 1: GEM ternary blend fuels used in the vehicle tests described. Properties
calculated using Lotus Fuel Mixture Database [12] or measured to the relevant
ASTM standards where applicable
Original Blends
Fuel
GEM Component Ratios
Stoichiometric AFR
Density (kg/l)
Gravimetric LHV (MJ/kg)
Volumetric LHV (MJ/l)
Carbon Intensity (gCO2/l)
Carbon Intensity (gCO2/MJ)
RON (to ASTM D2699)
MON (to ASTM D2700)
Sensitivity
Blend A
G15 E85 M0
9.69
0.781
29.09
22.71
1627.9
71.69
107.4
89.7
17.7
Blend C
G37 E21 M42
9.71
0.769
29.56
22.71
1623.9
71.49
106.4
89.3
17.1
Blend D4
G40 E10 M50
9.65
0.767
29.46
22.60
1613.9
71.42
105.6
89.0
16.6
Blend D
G44 E0 M56
9.69
0.765
29.66
22.69
1620.2
71.41
106.1
89.0
16.2
Test Vehicle and Facility

A production flex-fuel Saab 93 BioPower station wagon was used for these tests. This
vehicle was fitted with an automatic gearbox and was certified to Euro 5 emissions
standards. It was tested on the rolling road dynamometer at Lotus Engineering and
was operating with the standard production flex-fuel calibration and OBD system.
The drive cycle used was the New European Drive Cycle (NEDC). Two gasoline
baseline results were taken before and after the tests, and the order in which the fuels
were tested was gasoline, Blends A, C, D, D4 and then the repeat of the gasoline tests.
The same procedure was followed for each test fuel, in line with the requirements for
testing vehicles on the NEDC. Each fuel was tested twice on sequential days, and on
each day a hot NEDC test was conducted after the cold test. Only the cold test results
will be reported here, since this is what is used to determine the emissions compliance
of a vehicle. The Euro 5 oxides of nitrogen (NOx) emissions limit is 0.06 g/km [14]4.
Note that the NOx limit for Euro 5 regulations stated also applies at Euro 6; the major difference for
spark-ignition engines at Euro 6 level is that there are additional particulate number limits. Euro 5
came into effect in September 2009 and Euro 6 will come into effect in September 2014.
Experimental Results
Figures 2 and 3 show the fuel consumption (in miles per US gallon5) and tailpipe CO2
emissions (in terms of gCO2/km, which is the parameter used to establish a
manufacturers total tailpipe CO2 emissions for the purposes of establishing any fiscal
penalties in Europe, weighted by sales volume [4]), respectively. Figure 4 shows the
energy utilization of the vehicle, calculated using the data in Table 1.
Drive Cycle Fuel Consumption / [mpg US]
25
20
15
10
0
Gasoline (Start)
Gasoline (End)
D4
Blend Designation
Day 1 Cold
Day 2 Cold
Fig. 2: Production flex-fuel vehicle fuel consumption (in terms of miles per US gallon)
when operated on four GEM blends and gasoline on the New European Drive Cycle.
Vehicle certified to Euro 5 emissions level
From the data in Figure 4 one can see that the vehicle was energetically more efficient
when operated on the alcohol blends than it was when operated on gasoline. The
result for the second cold test on Blend A (G15 E85 M0) is considered a slight outlier,
but nevertheless (and disregarding the Blend A result from the second day) the
improvement in energy utilization across all of the alcohols was 2.8-4.9% for the first
day and 2.0-3.4% for the second day [15]. This improvement in energy utilization
was echoed in a higher result when the vehicle was hot in earlier work with a car with
a different alcohol sensing system and certified to an earlier emissions level (Euro 4),
where 3-5% improvement was seen when the vehicle was warm [10]. The
implications are that there would be a reduction in energy consumption from a fleet of
vehicles using such alcohol blends versus gasoline, with obvious advantages if those
fuels were to have to be synthesized in the future from another feed stock, e.g. from
shale gas.
In order to convert miles per US gallon to miles per Imperial gallon, divide the data in Figure 2 by
0.833.
Drive Cycle Tailpipe CO2 / [gCO2/km]
250
240
230
220
210
200
190
Gasoline (Start)
Gasoline (End)
D4
Blend Designation
Day 1 Cold
Day 2 Cold
Fig. 3: Production flex-fuel vehicle tailpipe CO2 emissions (in terms of gCO2/km)
Vehicle certified to Euro 5 emissions level
Drive Cycle Energy Utilization / [MJ/km]
3.5
3.4
3.3
3.2
3.1
3.0
2.9
2.8
Gasoline (Start)
Gasoline (End)
D4
Blend Designation
Day 1 Cold
Day 2 Cold
Fig. 4: Production flex-fuel vehicle drive cycle energy utilization (in terms of MJ/km)
Vehicle certified to Euro 5 emissions level. Data calculated from tailpipe CO2
emissions shown in Figure 3 using the Lotus Fuels Mixture Database [12]
Results for NOx emissions are shown in Figures 5(a) and 5(b), in absolute terms and
as an average percentage of the regulated maximum of 0.06 g/km, respectively.
NOx Emissions on NEDC / [g/km]
0.020
0.016
0.012
0.008
0.004
0.000
Gasoline (Start)
Gasoline (End)
D4
Blend Designation
Day 1
Day 2
Fig. 5(a): Production flex-fuel vehicle tailpipe NOx emissions in g/km when operated
on four GEM blends and gasoline on the New European Drive Cycle. Vehicle
certified to Euro 5 emissions level
Proportion of Legislated NOx Emissions Limit
30%
25%
20%
15%
10%
5%
0%
Gasoline (Start)
Gasoline (End)
D4
Blend Designation
Fig. 5(b): Production flex-fuel vehicle averaged tailpipe NOx emissions as a

proportion of the maximum permitted value when operated on four GEM blends and
gasoline on the New European Drive Cycle. Vehicle certified to Euro 5 emissions
level
From the results of Figure 5 it can be seen that the vehicle has no problem delivering
legal NOx emissions when operated on the four ternary GEM fuels. The average of
the four alcohol blend fuels is 0.0075 g/km, and is over 50% less than the average of
the two gasoline tests (0.0151 g/km). Additionally, the results of all the fuels are less
than 30% of the legislated maximum for NOx of 0.06 g/km, which is significantly
lower than the normal engineering target of 50% to ensure compliance of the whole
fleet over the lifetime of the vehicles. (Note that after catalyst light-off there will be
virtually zero emissions anyway due to the conversion efficiency of three-waycatalysts.) Therefore, from these results there is likely to be little concern with regard
to NOx emissions when existing flex-fuel vehicles are operated on any of the GEM
ternary blends. Results for hydrocarbon and carbon monoxide emissions will be
reported in a later publication.
Finally, the vehicle exhibited no driveability problems when using any of the ternary
blends, and the on-board diagnostics were not upset, as shown by the fact that there
was no malfunction indicator lamp (MIL) activity on the dashboard, regardless of fuel
blend used. Approximately 1500 km were covered on a wide range of the GEM fuels
(both as specific blends and as general tankfuls of one blend following another) and it
always started well and has done so ever since as far as the authors are aware. For
more details of this, plus the cold-temperature operation testing that was carried out,
see [11].
DISCUSSION
The results presented here suggest that ternary blends can be true drop-in alternatives
to E85, and that the NOx exhaust emissions important to human health will be lower
than those for gasoline. This is important when considering how they can help with
energy security in countries where they can be manufactured from indigenous feed
stocks, such as is the case with the recent shale gas finds in the US [16], which create
an opportunity for it to become more energy independent. The scale of the
opportunity was illustrated by Moniz et al. [16], who estimated that with recent finds
the total US reserves of natural gas equal 92 times the current annual consumption,
and thus these resources can provide a bridge to a low-carbon future. From this work,
the issue of how to apply this opportunity to increase energy independence to
transport (which is especially reliant on imported oil) is one that can be addressed by
two routes in terms of making liquid energy carriers: full Fischer-Tropsch (FT)
synthesis of liquid hydrocarbon fuels, or conventional methanol synthesis from
natural (shale) gas.
While full FT synthesis produces drop-in fuels for all vehicles (including ships and
aircraft), direct methanol synthesis is a more efficient means of converting methane to
a liquid fuel, and furthermore, requires less investment in plant and is economical on a
smaller scale. An extension of this could see small, economical methanol plants
feeding the fuel pool with their products directly (via ternary blending) or providing
methanol as a feed stock for larger methanol-to-synfuels (MtSynFuels) plants. This
might help to open up some more of the stranded shale gas fields because of the
relative ease of transporting energy dense liquids over distance.
If the methanol produced in this manner is introduced in the near-term via the ternary
blending approach discussed above, one can extend the available ethanol significantly
and displace more gasoline. For illustrative purposes, there follows an assessment of
how much methanol fuel could be used. Of the 36 billion US gallons of ethanol
which the US Energy Independence and Security Act mandates for 2022, some can be
blended into gasoline. Currently the permitted level is 10%, although EPA is moving
towards 15% in the future for 2001 and newer light-duty motor vehicles (subject to
certain conditions) [17]. Assuming that 140 billion US gallons of gasoline are used
for light-duty vehicles from 2016 onwards6, and that ~12% of it by volume is ethanol
(most in E10 but some in E15), let us assume that there will be 19 billion US gallons
available for flex-fuel vehicles, which, at an E85 blend rate of 85% (disregarding the
fact that less ethanol is typically used in commercial E85 in the winter months),
implies that 22.4 billion gallons of E85 could be supplied.
These 22.4 billion gallons of E85 are equivalent in energy terms to 16.1 billion
gallons of gasoline, although they do contain 3.4 billion gallons of gasoline
themselves (the 15% gasoline in E85). Effectively, 19 billion gallons of ethanol is
equivalent to 12.7 billion gallons of gasoline (i.e. the ratios of the volumetric LHVs of
gasoline and ethanol, 31.6 MJ/l for and 21.2 MJ/l respectively) Thus, 140 billion
gallons are reduced to 140-12.7 = 127.3 billion gallons of gasoline, and there is a
reduction in gasoline usage of 9.1%.
Consider now that the 19 billion gallons of ethanol instead be used to manufacture a
ternary blend such as Blend C (G37 E21 M42). As mentioned earlier, it is possible to
show that the methanol displaces gasoline if the total ethanol volume in the fuel pool
is held constant. Figure 6 shows this relationship; on the left-hand side of the figure
one supplies four units of energy as three units of gasoline and one unit of E85, and
on the right-hand side all four units are supplied as Blend C instead. Note that there is
effectively the same volume of ethanol on both sides of the figure which is the case
when ethanol supply is constrained. Summing the gasoline volume on both sides one
arrives at 231 volume units on the left (i.e. the traditional approach) and 148 on the
right, i.e. 35.9% extra gasoline has been displaced over and above that already
supplied by the ethanol. Put another way, 168 volume units of methanol have
displaced 83 units of gasoline.
Because of the blend proportions then for ternary Blend C one would require twice as
much methanol i.e. 38 billion gallons (from a total of 90.5 billion gallons of Blend
C that can be made from 19 billion gallons of ethanol). The situation compared to the
traditional E85 approach is that the 38 billion gallons of methanol have been used to
displace 18.8 billion gallons of gasoline (i.e. 38 x 83/168, which again is the ratio of
the volumetric LHVs of gasoline and methanol, 31.6 MJ/l for and 15.7 MJ/l
respectively).
Now one can see that in addition to the 12.7 billion gallons of gasoline displaced by
the ethanol, there is an additional 18.8 billion gallons displaced by the methanol, and
the gearing on the ethanol is considerable. Effectively, instead of the 140 billion
gallons of gasoline needed, the new volume required is 140-12.7-18.8 = 108.5 billion
gallons, or a reduction of 22.5% by volume of gasoline in the entire fuel pool with the
same volume of ethanol being supplied.
Based on the actual 2007 consumption of 134.8 US gallons, with an assumption that vehicle fuel
economy improves on the one hand and that there are more vehicles on the road on the other.
Equivalent Energy on Each Side

Volumes for Equal Energy / [Volume Units]
100
90
80
42
42
42
42
21
21
21
21
37
37
37
37
70
60
85
50
40
72
72
72
30
20
10
15
0
Gasoline
Gasoline
Gasoline
Blend Designation
Gasoline
Ethanol
Methanol
Fig. 6: Example of enhanced gasoline displacement by the introduction of methanol

into ternary Blend C equivalent to E85 (for explanation, see text)
Note that the above argument is not extended to the 17 billion gallons of ethanol
going into the remaining gasoline. In fact an iterative approach must be taken as the
displacement of more gasoline (containing ethanol) implies more of it being available
for a ternary blend. Furthermore, a ternary blending approach can be adopted at any
blend rate, so assuming that methanol can be introduced via this method into E10 or
E15 equivalents, a significant further proportion of gasoline could perhaps be
replaced. This mechanism and overall system of displacement is perhaps worthy of
further investigation and modelling.
Such an approach is academic if the vehicles do not exist to take the fuel (which can
be easily rectified since the cost of making a flex-fuel vehicle is very low, and the
CAF regulations in the US are forcing this anyway) or if the blends are too
expensive for the vehicle owner to use. Fortunately the low price of methanol means
that the cost of ternary blend fuels can be lower than gasoline, on a per-unit-energy
basis. As previously mentioned E85 is more expensive than gasoline in energy terms
because the twin benefits of ethanol being renewable and a significant octane
enhancer at low blend ratios drive its price up. It is interesting to illustrate the
potential in cost reduction in ternary blends due to the introduction of the methanol
blend component using assumed prices for gasoline, ethanol and methanol. We shall
take these to be 3.21, 2.30 and 1.11 dollars per US gallon respectively, which was the
case in September 2011 (before tax). Figure 7 shows how the price per unit energy
relative to gasoline changes as the proportion of methanol in the ternary blend is
increased. Only 24% methanol is required for the blend to be on a par with gasoline;
at this point the user would see a reduction in operating costs anyway because of the
reported higher efficiency with an alcohol blend fuel. Blends A to D are shown on
Figure 7; specifically Blend D4 (G40 E10 M50), considered a practical fuel in terms
of low-temperature phase separation, would be approximately 9.3% cheaper than

gasoline on an energy basis.
Blend A
Energy Cost - Increase over Gasoline / [%]
10
0
0
10
20
30
40
50
60
Blend B
-5
Blend C
-10
Blend D
-15
Proportion of Methanol in Ternary Blend / [%]

1
Fig. 7: Variation in energetic cost of GEM ternary blends versus that of gasoline as a
function of the methanol concentration. Assumed costs per US gallon: gasoline
$3.21, ethanol $2.30 and methanol $1.11. Ternary blends equivalent to E85
It should be remembered that a proportion of the gasoline required can be also made
by either the FT or a MtSynFuels process (using methanol synthesis as an
intermediate step). This will help with the gradual balancing of the two fuel products
against the introduction of the necessary E85/gasoline flex-fuel vehicles. Given that
the necessary fuel energy can be supplied in this manner, and that eventually a
practical limit will be reached in terms of utilization in the existing vehicle technology
(and that heavy-duty vehicles will otherwise continue to need diesel-type fuels, with
the attendant energy losses from their onward synthesis from methanol) the remainder
of this paper will discuss a pathway from ternary blends to the supply of fuels in full
amounts to the light- and heavy-duty markets.
Having shown that the ternary blend approach produces functionally invisible drop-in
blends suitable for E85/gasoline flex-fuel vehicles, further work will investigate the
effect of such blends on fuel systems materials. The production flex-fuel vehicle used
for these tests exhibited no problems in this regard, and has not done so ever since as
far as the authors are aware. It is hoped that since many flex-fuel fuel system
components are (it is believed) tested with methanol as a default that there will be no
danger to existing vehicles through moving to an E85-equivalent blend containing
methanol as well; even so, any potential issues can be mitigated by a phased
introduction, which will be discussed in the following section.
Potential Rollout of the GEM Fuels and Possiblel Future Scenarios

From the work conducted to date it is entirely possible that methanol can be
introduced into the fuel pool for existing flex-fuel vehicles (blended at E85-equivalent
stoichiometry of 9.7:1) or for normal vehicles at blends equivalent to E10 or E15 in
the near future. It is suggested that initial rollout be for E85/gasoline flex-fuel
vehicles, since their smaller number automatically keeps the number of cars using the
fuels down. Obviously some form of fleet test and further validation in-vehicle needs
to be carried out before any rollout can be fully imagined, and it is hoped to do this
with a small number of vehicles. Following successful conclusion of fleet trials, the
release of the blends can be carried out in a manner controlled by both geography and
blend ratio (obviously a blend containing much less methanol than Blend B can be
created such a blend is discussed later). This will allow the evolutionary change of
the fuel and vehicles to gradually-increasing amounts of methanol in a steady and
controllable manner. It is imagined that the ramp-up in plants converting shale
methane to methanol would effectively mirror this, making the whole process
complimentary.
Given that the existing light-duty fleet can start to use methanol by its incorporation
as a blend component in a GEM fuel compatible with E85/gasoline vehicles, at some
point the number of suitable vehicles able to take the fuel will become a limiting
factor. It is suggested that early on in the process of GEM fuel introduction, given its
successful implementation, government would enact legislation to encourage the
wider production of the number of flex-fuel cars necessary so that the demand side is
not a limiting factor. It is suggested that the approach of shale gas to methanol and
use in the fuel pool would, at this point, have been considered a success, and that
more far-reaching strategies could be created at that point. This section will discuss
some such options.
With minimal impact on the vehicle manufacturers, it could be made mandatory that
all spark-ignition vehicles should be made E85/gasoline flex-fuel. Some extension of
the CAF regulations would help to offset any costs incurred by the vehicle OEMs in
doing this. However, considering the longer term, it would perhaps be advantageous
to encourage the engineering of vehicles for sale which could take more than the
maximum proportion of the methanol blend component in a blend equivalent to E85.
This limit is Blend D (G44 E0 M56), although the actual maximum methanol
proportion may need to be lower than this due to the need to have a cosolvent; it is
suggested that Blend D4 (G40 E10 M50) would represent some upper limit in
vehicles with E85/gasoline flex-fuel technology, due to the desire to be sure of
avoiding low-temperature phase separation [11].
Since M85 was very successfully used before in the California methanol trial [18], it
might be considered to make sense to move to that blend rate of methanol and
gasoline as the next step, but it may equally be considered desirable to move straight
to M100, while maintaining flex-fuel capability. This has been shown to be
straightforward in previous work by the authors [19] and has also been called for in
[16], together with more support of the US Open Fuel Standard (OFS). The vehicle
engineering costs are likely to be similar to just providing M85 flex-fuel capability,
since new technologies to aid starting in the form of direct injection are becoming
commonplace now, and the efficacy of such technologies in cold starting pure
alcohols having been known for some time [20]. A significant secondary advantage
of this larger step is that the ensuing demand for pure methanol would then permit the
use and adoption of either direct methanol fuel cells (DMFCs), proton exchange
membrane (PEM) fuel cells with a simple reformer or optimized solid oxide fuel cells
(SOFCs). In separate work, Bromberg and Cohn have suggested that heavy-duty
trucks could move to M100 with the fuel being supplied by the smaller infrastructure
necessary for such vehicles, which would limit the expenditure necessary [21]. This
infrastructure would also play its part in the gradual evolution towards a full alcoholbased energy economy, since the necessary modifications to the heavy duty
infrastructure could lead those in the light-duty infrastructure.
That emissions compliance is possible to achieve with current technology even at very
high methanol concentrations was demonstrated at Euro 4 emissions level in [19]. In
line with the above comments regarding NOx emissions for the ternary blends tests
described above, Figure 8 reproduces the NOx results from [19], with the approximate
blend ratios in the tank for each different test shown on the bars. Note that in the
work reported in [19], constant stoichiometry was not aimed for in the fuel blends
tested; rather the mixtures tested in that work were arbitrary since it was aimed at
showing that any blend of gasoline, ethanol and methanol in a single vehicle fuel tank
could be automatically compensated for by a modern engine management system.
The modified Lotus vehicle used for this work was fitted with the standardspecification gasoline catalyst and was certified to Euro 4 emissions level, for which
the NOx limit was 0.08 g/km. Figure 8 shows that the working engineering limit of
approximately 50% when operating on gasoline was achieved for NOx. However,
from the changes in the alcohol concentrations it is clear to see that in general the
higher the proportion of ethanol or methanol (or both) the lower the tailpipe NOx
emissions. Test 3 in Figure 8 uses G12 E0 M88 which is close to the notional M85
blend used in [18], and represents a reduction in NOx of nearly 70% versus gasoline;
furthermore, the calibration was being refined as the test numbers increased, so the
final value for M100 could be expected to be lower (for more details of the other
emissions and how these interact, together with potential trade-offs enabled by the
extremely low NOx output, see [19]).
In parallel with the above, Cohn and co-workers have proposed using the direct
injection (DI) of ethanol or methanol in SI engines employing port-fuel injection
(PFI) of gasoline as way of increasing the knock limit due to the chemical octane of
the fuel coupled to the physical octane effects due to the high latent heat [22]. This
they proposed under the banner of Ethanol Boosting Systems and their work was
continued by Stein et al. [23]. The gearing on gasoline displacement was found to be
significant since the direct injection of low-carbon-number alcohols helps to offset
enrichment fuelling and to permit higher boost pressures, and thus greater degrees of
downsizing.
Importantly with regard to this approach of PFI gasoline with DI of alcohol, the
ternary GEM blends equivalent to E85 discussed earlier in this paper could be used
instead of E85. This is because, when calculated on basis the of their mass ratios, the
latent heat of all such ternary blends is the same from Blend A to Blend D to within
+/- 2% (see Appendix I of [11]). Functionally this would not be expected to
adversely impact the EBS concept, and it also acts as another means of introducing
methanol into the fuel pool, should any such concept be commercialized. It
represents another aspect of the invisibility of the blends to E85-optimized

combustion systems.
0.060
0.040
G58 E11 M31
G36 E43 M21
G46 E25 M29
G30 E70 M0
G47 E0 M53
0.010
G12 E0 M88
G30 E0 M70
0.020
G72 E0 M28
0.030
Gasoline
NOx Emissions on NEDC / [g/km]
0.050
0.000
1
Test Number
Fig. 8: Prototype gasoline-ethanol-methanol tri-flex-fuel vehicle tailpipe NOx

emissions in g/km when operated on various blends of alcohol and gasoline on the
New European Drive Cycle. Vehicle certified to Euro 4 emissions level. Calibration
being developed from one test to the next; for more information see [19]. Note: fuel
blend rates are not configured to a fixed, target stoichiometric AFR value (see text);
limit = 0.08 g/km
Bromberg and Cohn discuss the use of DI of methanol in heavy-duty engines in
general in [25] and Brusstar and co-workers have investigated alcohol fuels in veryhigh-compression ratio SI engines, showing that higher peak thermal efficiencies can
be achieved with such concepts than the diesel engines on which they are based
[26,27].
Thus high-blend alcohol fuels can offer the prospect of a future energy economy with
increased energy security due to the high energy conversion efficiencies possible with
these energy carriers. Furthermore, because of the miscibility with gasoline, flex-fuel
approaches can ensure that the driver will not be left without a fuel to operate the
vehicle on (although, as alcohol fuels become more commonplace, the engines may
be biased towards operation on alcohols, and their may be a concomitant reduction in
performance and range on gasoline; modern engine management systems will still
permit safe operation despite the high compression ratios which such engines may
adopt due to the superior characteristics of such high-blend alcohol fuels). A
suggested time line for this process, showing how the fuels and their manufacturing
processes interact with the vehicles, is shown in Figure 9, where it is proposed that the
first ternary blend introduced contains only a relatively small percentage of methanol,
and that this is ramped up in proportion and/or geographical area over time until all of
the fuel is at Blend D4, which would represent saturation point for the vehicles was it
not for mandating that all SI vehicles be capable of operating on E85 or similar (in
Figure 9 see the arrow moving from a potential introductory blend which we call
Blend B1 (G20 E70 M10) to Blend D4 (G40 E10 M50)).
Heavy
Duty
Introduce methanolconsuming fuel cells
Now
Mandate E85
flex-fuel
compatibility
2015
Conduct
ternary blend
fleet trials
Fuels
Introduction of methanolconsuming trucks
Light
Duty
Vehicles
Either DI or PFI - mandate M100 and E100

full flex-fuel compatibility from the outset
Mandate M100 and

E100 full tri-flex-fuel
compatibility (OFS)
2020
Commission first shalegas-to-liquid plants
Begin phase-out of
non-domestic gasoline
Introduce ternary blends and

increase methanol component
Blend B1
(G20 E70 M10)
Gradual methanol blend

rate phase-in by region
and proportion
2030
2025
Introduce
M100
Blend D4
(G40 E10 M50)
Fig. 9: Roadmap for introduction of increasing amounts of methanol into the US fuel
pool via GEM ternary blends, eventually leading to M100.
OFS = Open Fuel Standard
If the gasoline price does not increase further then the energy in Blend B1 (G20 E70
M10) would cost about 4.8% more than gasoline, which would likely be offset by the
higher efficiency of the vehicles, so this blend could be expected to be cost neutral.
However, it is not unreasonable to assume that the gasoline price will increase, and an
increase of 10% would make Blend B1 2.3% cheaper (Blend D4: 12.7% cheaper).
Thus Blend B1 would appear to be a practical target introduction blend; furthermore,
since there would now only be 70% ethanol and both gasoline and methanol cold start
more easily, it may be possible to stay with this blend ratio year-round (see [11] for
the effect of the introduction of methanol on the cold startability of ternary blends).
Eventually, there will be supply side limitations even with methanol made from shale
gas, and it must also be remembered that this is a finite resource. Many researchers
have proposed that methanol (and higher hydrocarbons, albeit at an efficiency
penalty) can be made using CO2 extracted from the atmosphere, electrolytic hydrogen
and renewable energy [7,28-32]. This has the potential to provide liquid transport
fuels in full amounts, which fuels using biomass as a feedstock cannot do due to the
biomass limit. It can be seen how the gradual introduction of such fuels would be
facilitated by the vehicles and infrastructure having already moved in that direction.
The high value of transport fuel will ensure that the investment necessary can be
supported, and the volume used will help to bypass the issues faced by renewable
energy in general, i.e. that the ability of the electricity grid to absorb renewable
electricity is limited by the base load condition (which cannot be circumvented), and
the fact that electricity cannot easily be stored. When the wind blows and the
renewable energy output is above what the electricity grid can absorb, conversion to a
hydrocarbon energy carrier is an excellent means of buffering such renewable energy
[7,9].
Taking all of the foregoing into account, alcohol fuels therefore represent a pragmatic
solution to future transport energy requirements for all stakeholders, since a
continuous process of gradual evolution to a practical end game can be followed, with
no quantum investment necessary at any stage by governments, OEMs, fuel suppliers
or customers in either infrastructure or vehicles. This is because the alcohols are
miscible with the gasoline that we use now, many flex-fuel vehicles already exist to
use it, and it is feasible to make all future vehicles alcohol-compatible at minimal
extra cost as the fuels become available in larger amounts.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Energy Independence and Security Act of 2007, Public Law 110-140, 110th
Congress, DOCID: f:publ140.110, 2007.
On the promotion of the use of energy from renewable sources and amending
and subsequently repealing Directives 2001/77/EC and 2003/30/EC, Directive
2009/28/EC of the European Parliament and of the Council, 23rd April 2009.
Cooper, J., Future fuels for transport: Regulatory and supply drivers, Internal
Combustion Engines: Improving Performance, Fuel Economy, and Emissions,
Institution of Mechanical Engineers Conference, London, UK, 29th-20th
November, 2011.
Setting emission performance standards for new passenger cars as part of the
Communitys integrated approach to reduce CO2 emissions from light-duty
vehicles, Regulation (EC) 443/2009 of the European Parliament and of the
Council, 23rd April, 2009.
Milpied, J., Jeuland, N., Plassat, G., Guichaous, S., Dioc, N., Marchal, A. and
Schmerlzle, P., Impact of Fuel Properties on the Performances and Knock
Behaviour of a Downsized Turbocharged DI SI Engine - Focus on Octane
Numbers and Latent Heat of Vaporization, SAE paper number 2009-01-0324,
SAE 2009 World Congress, Detroit, Michigan, USA, 20th-23rd April, 2009.
Anderson, J.E., Kramer, U., Mueller, S.A., and Wallington, T.J., Octane
numbers of ethanol- and methanol-gasoline blends estimated from molar
concentrations, Energy Fuels, Vol. 24, pp. 6576-6585, 2010.
Pearson, R.J., Eisaman, M.D., Turner, J.W.G., Edwards, P.P., Jiang, Z.,
Kuznetsov, V.L., Littau, K.A., di Marco, L., and Taylor, S.R.G., Energy
storage via carbon-neutral fuels made from CO2, water, and renewable energy,
Special Issue of Proc. IEEE: Addressing the intermittency challenge: Massive
energy storage in a sustainable future. To be published Feb. 2012.
Olah, G. A., Goeppert, A. and Prakash, G.K.S., Beyond Oil and Gas: The
Methanol Economy, Second Edition, Wiley-VCH Verlag GmbH & Co. KGaA,
Weinheim, 2009. ISBN 98-3-527-32422-4.
Sterner, M., Bioenergy and renewable power methane in integrated 100%
renewable energy systems, Dr.-Ing Thesis, University of Kassel, September
2009, ISBN 978-3-89958-798-2.
Turner, J.W.G., Pearson, R.J., Purvis, R., Dekker, E., Johansson, K. and
Bergstrm, K. ac, GEM Ternary Blends: Removing the Biomass Limit by
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
using Iso-Stoichiometric mixtures of Gasoline, Ethanol and Methanol, SAE

paper number 2011-24-0113, The 10th International Conference on Engines and
Vehicles, Capri, Naples, Italy, 11th-16th September, 2011.
Turner, J.W.G., Pearson, R.J., McGregor, M.A., Ramsay, J.M., Dekker, E.,
Iosefa, B., Dolan, G.A., Johansson, K. and Bergstrm, K. ac, GEM Ternary
Blends: Testing Iso-Stoichiometric mixtures of Gasoline, Ethanol and Methanol
in a Production Flex-Fuel Vehicle Fitted with a Physical Alcohol Sensor, SAE
paper number 2012-01-0113, SAE 2012 World Congress, Detroit, 24th-26th
April, 2012.
Lotus Fuels Mixture Database, proprietary software of Lotus Engineering, 2008.
Pearson, R.J., Turner, J.W.G., Davy, M. and Stone, R., Iso-stoichiometric
ternary blends: Characterization of some physico-chemical properties of
potential future transitional fuels, in preparation for submission to Proc. Instn
Mech. Engrs Journal of Automotive Engineering, Part D.
Worldwide Emissions Standards Passenger Cars and Light Duty Vehicles
2011/2012, Delphi Corporation, Troy, Michigan, USA.; available on line at
www.delphi.com/emissions-pc, last accessed 28th March, 2012.
Turner, J.W.G., Pearson, R.J., Dekker, E., Iosefa, B., Dolan, G.A., Johansson,
K. and Bergstrm, K. ac, Extending the role of alcohols as transport fuels using
iso-stoichiometric ternary blends of gasoline, ethanol and methanol, submitted
to Journal of Applied Energy, 2012.
Moniz, E.J., Jacoby, H.D., Meggs, A.J.M., Armstrong, R.C., Cohn, D.R.,
Connors, S.R., Deutch, J.M., Ejaz, Q.J., Kaufman, G.M., Kenderdine, M.A.,
O'Sullivan, F., Paltsev, S., Parsons, J.E., Perez-Arriaga, I., Reilly, J.M., Seto, C.,
Yang, Y., McRae, G.S., Ruppel, C., Fletcher, S., Hagerty, J.M., Kragha, O.,
Leung, T., Murphy, P., Rachokonda, A., Tapia-Ahumuda, K., Toukan, I., Ucok,
D. and Yao, Y., The Future of Natural Gas - An Interdisciplinary MIT Study,
Massachusetts
Institute
of
Technology,
available
at:
http://web.mit.edu/mitei/research/studies/documents/natural-gas2011/NaturalGas_Report.pdf, 2011, last accessed 28th March, 2012.
E15
(a
blend
of
gasoline
and
ethanol),
available
at:
http://www.epa.gov/otaq/regs/fuels/additive/e15/, last accessed 28th March,
2012.
Ward, P.F. and Teague, J.M., Fifteen years of fuel methanol distribution, 11th
International Symposium on Alcohol Fuels, Sun City, South Africa, 14th-17th
April, 1996.
Pearson, R.J., Turner, J.W.G. and Peck, A.J., Gasoline-ethanol-methanol trifuel vehicle development and its role in expediting sustainable organic fuels for
transport, 2009 I.Mech.E. Low Carbon Vehicles Conference, London, UK,
20th-21st May, 2009, pp. 89-110.
Siewert, R.M. and Groff, E.G., Unassisted Cold Starts to -29C and SteadyState Tests of a Direct-Injection Stratified-Charge (DISC) Engine Operated on
Neat Alcohols, SAE paper number 872066, 1987, doi:10.4271/872066.
Bromberg, L. and Cohn, D., Alcohol Fueled Heavy Duty Vehicles Using
Clean, High Efficiency Engines
Cohn, D.R., Bromberg, L. and Heywood, J.B., Direct Injection Ethanol
Boosted Gasoline Engines: Biofuel Leveraging for Cost Effective Reduction of
Oil Dependence and CO2 Emissions, MIT report PSFC/JA-06-16, MIT Plasma
Science and Fusion Centre, 20th April, 2005, available at:
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
http://www.psfc.mit.edu/library1/catalog/reports/2000/06ja/06ja016/06ja016_ful
l.pdf, last accessed 28th March, 2012.
Stein, R.A., House, C.J. and Leone, T.G., Optimal Use of E85 in a
Turbocharged Direct Injection Engine, SAE paper number 2009-01-1490, SAE
Int. J. Fuels Lubr. 2(1):670-682, 2009, doi:10.4271/2009-01-1490.
Blumberg, P.N., Bromberg, L., Kang, H. and Tai, C., Simulation of High
Efficiency Heavy Duty SI Engines Using Direct Injection of Alcohol for Knock
Avoidance, SAE paper number 2008-01-2447 and SAE Int. J. Engines
1(1):1186-1195, 2009, doi:10.4271/2008-01-2447.
Bromberg, L. and Cohn, D., Alcohol Fueled Heavy Duty Vehicles Using
Clean, High Efficiency Engines, SAE paper number 2010-01-2199
doi:10.4271/2010-01-2199.
Brusstar, M., Stuhldreher, M., Swain, D. and Pidgeon, W., High Efficiency and
Low Emissions from a Port-Injected Engine with Neat Alcohol Fuels, SAE
paper number 2002-01-2743, doi:10.4271/2002-01-2743.
Brusstar, M.J. and Gray, C.L., High Efficiency with Future Alcohol Fuels in a
Stoichiometric Medium Duty Spark Ignition Engine, SAE paper number 200701-3993, doi:10.4271/2007-01-3993.
Steinberg, M., Production of synthetic methanol from air and water using
controlled thermonuclear reactor power I. Technology and energy
requirement, Energy Conversion, vol. 17, pp. 97-112, 1977.
Bandi, A., Specht, M., Weimer, T. and Schaber K., CO2 recycling for hydrogen
storage and transportation electrochemical CO2 removal and fixation, Energy
Conversion and Management, vol. 36(6-9), pp. 899-902, 1995.
Stucki, S., Schuler, A. and Constantinescu, M,. Coupled CO2 recovery from the
atmosphere and water electrolysis: feasibility of a new process for hydrogen
storage , Int. J. Hydrogen Energy, vol. 20, 8, pp. 653-663, 1995.
Weimer, T., Schaber, K., Specht, M. and Bandi, A., Methanol from
atmospheric carbon dioxide: a liquid zero emission fuel for the future, Energy
Conversion and Management, vol. 37(6-8), pp. 1351-6, 1996.
Specht, M., Staiss., F., Bandi, A. and Weimer, T., Comparison of the
renewable transport fuels, liquid hydrogen and methanol, with gasoline
energetic and economic aspects, Int. J. Hydrogen Energy, vol. 23, 5., pp. 387396, 1998.
ACKNOWLEDGEMENTS
The authors from Lotus would like to thank the directors of Lotus Cars for permission
to publish this paper and for encouragement to conduct the research. Mark McGregor
and John Ramsay were heavily involved in the sequencing and conduct of the tests,
and their rigour and enthusiasm is much appreciated.
The authors would also like to acknowledge the insights gained from discussions
with, among others, the following people: Tony Marmont, Peter Harrison and Bob
Jennings of Air Fuel Synthesis, Matt Eisaman of Brookhaven National Laboratory,
Paul Wuebben of Carbon Recycling International, Steve Brueckner of Chrysler, Leon
di Marco of FSK Technologies, Gordon Taylor of GT-Systems, Tim Fox and Philippa
Oldham of the Institution of Mechanical Engineers, Martin Davy and Colin Garner of
Loughborough University, Leslie Bromberg and Daniel Cohn of Massachusetts
Institute of Technology, Frank Zeman of New York Institute of Technology, Arthur
Bell, Gareth Floweday and Andre Swarts of Sasol Technology, Karl Littau of
Stanford University, Matt Brusstar of the United States Environmental Protection
Agency, Chris Brace of the University of Bath, Peter Edwards and Richard Stone of
the University of Oxford, and George Olah, Surya Prakash, and Alain Goeppert of the
University of Southern California.
ABBREVIATIONS
AFR
CAF
CO2
DI
DMFC
EBS
EPA
EU
FQD
GEM
ILUC
LHV
MIL
MtSynFuels
OBD
OEM
OFS
PEM
PFI
NEDC
NOx
RED
SI
SOFC
Air-fuel ratio
Corporate Average Fuel Economy
Carbon dioxide
Direct injection
Direct methanol fuel cell
Ethanol Boosting Systems
Environmental Protection Agency
European Union
Fuel Quality Directive
Gasoline, ethanol and methanol
Indirect land use change
Lower heating value
Malfunction indicator lamp
Methanol-to-synfuels
On-board diagnostics
Original equipment manufacturers (i.e., vehicle manufactuers)
Open Fuel Standard
Proton exchange membrane
Port-fuel injection
New European Drive Cycle
Oxides of nitrogen
Renewable Energy Directive
Spark-ignition
Solid oxide fuel cell
2012 MIT Energy Initiative Symposium
Prospects for Flexible- and Bi-Fuel Light Duty Vehicles:

Consumer Choice and Public Attitudes
Ulrich Kramer*, James E. Anderson**
Ford Motor Company, Research & Advanced Engineering, Europe* & North America**
ABSTRACT
Based on an analysis of several case studies of alternative fuel
introductions [ethanol, biodiesel, liquefied petroleum gas
(LPG), compressed natural gas (CNG)], requirements for
alternative fuels, vehicles, and the fueling infrastructure are
postulated that are necessary for successful market
implementation. Affordable vehicle technology and costcompetitive fuel were identified as the most critical factors.
Payback periods for additional vehicle costs associated with
different alternative fuels are discussed. Fuel costs need to be
consistently competitive in both the near-term and the longterm as demand for the fuel rises.
For the vehicles, other considerations include backwardscompatibility or capability for two fuels, retrofit kits
controlled by original equipment manufacturers (OEMs), and
emissions compliance. For the fuel distribution infrastructure,
affordable development and initially sufficient filling station
numbers are required. For the fuel, important factors include
energy density and adequate fill time, as well as the need for
incentives and sufficient natural resource availability for
sustainable fuels.
For the long-term sustainability of an alternative future fuel,
there should be a future source that is non-fossil (low CO2
emissions), renewable, and cost-competitive even when
required in large volumes. Also considered are two possible
future sustainable fuel scenarios involving ethanol and
renewable methane. Ethanol in E85 can be used in todays
flex-fuel vehicles (FFVs) to overcome backwards
compatibility limits of the existing fleet, allowing time for a
compatible fleet to be deployed. Renewable methane (biomethane, e-methane) could be used at any blend level in
todays compressed natural gas vehicles (CNGVs). Near-term
fuel flexibility from FFVs and bi-fuel or mono-fuel CNGVs is
a key enabler for both scenarios.
1. INTRODUCTION
Rising energy costs (particularly oil price), energy security,
and greenhouse gas (GHG) emissions are the main drivers of
the active, ongoing discussion of alternative fuels in the
transportation sector. Several alternative fuels have been
proposed and brought into different markets in recent years,
Page 1 of 21
including natural gas, liquefied propane and butane gas,

biodiesel and ethanol as both neat fuels and blend components
in diesel and gasoline, respectively, and electricity. Many
introductions have failed or have only led to niche
applications, whereas others have been truly successful in
local markets. Since only a few alternative fuels are
compatible with conventional vehicle technology, several
fuels require additional vehicle actions to ensure
compatibility. Depending on the technology used, different
types of alternative fuel vehicles have been developed or
proposed. Given the variety of possible configurations,
consistent terminology and definitions would be desirable for
the various industries and regulatory bodies involved. The
definitions presented in Table 1 were developed [1] based on a
review of various national regulations and industry standards.
Table 1 Types of Alternative Fuel Vehicles [1]
Dedicated- Any vehicle engineered and designed to be
fuel vehicle operated using a single fuel.
Any vehicle engineered and designed to be
operated using a single fuel, but with a petrol
Mono-fuel
system for emergency purposes or starting
vehicle
only, with petrol tank capacity of no more than
15 liters.
Bi-fuel
operated on two or more different fuels using
vehicle
two independent fuel systems, but not on a
mixture of the fuels.
Flex-fuel
operated on the original fuel(s), alternative
vehicle
fuel(s), or a mixture of two or more fuels that
(FFV)
are combusted together.
Vehicle with two independent fuel systems
Dual-fuel
that can run on both fuels simultaneously. It
vehicle
also may run on one fuel alone.
In this paper several case studies of alternative fuel
introduction are considered and reasons leading to success or
failure in each case are identified. A general set of lessons
learned for successful market introduction is outlined. In
addition, basic requirements of the alternative fuel, vehicles,
and fueling infrastructure are postulated that are necessary for
successful market implementation.
2. ALTERNATIVE FUEL MARKET

EXAMPLES
A wide variety of alternative fuels are in use in markets
globally. Ethanol has been used as an alternative to gasoline as
both a neat fuel and as blend component in various
concentrations. Other important alternative fuels currently in
use include biodiesel (fatty acid methyl esters, FAME) and
hydrogenated vegetable oil in diesel blended at various
concentrations and in a neat form, as well as liquefied
petroleum gas (LPG, a mixture of propane and butane) and
compressed natural gas (CNG) in gaseous fuel applications.
Electricity as a vehicle energy source is seeing increasing
development, in both dedicated-fuel vehicles (battery electric
vehicles, BEVs) and dual-fuel vehicles (plug-in hybrid
vehicles, PHEVs), but is not discussed here as BEV and
PHEV market development has only recently started.
Likewise, hydrogen is not discussed here as there is no
example yet of a large-scale market introduction, as is also the
case for PHEVs and BEVs.
Examination of examples of actual market introduction of
these fuels allows some conclusions to be drawn that can be
generalized for future fuel introduction scenarios.
2.1 ETHANOL BRAZIL

Brazil has the most fully developed market for ethanol used in
light duty vehicles (LDVs). Today there is no gasoline-type
fuel available in Brazil that does not contain ethanol. Gasoline
in Brazil contains 1825% ethanol by volume and is sold as a
fuel called gasohol. In addition, hydrous ethanol (E100) is
sold, consisting of at least 94.5% v/v ethanol, with the balance
being water and allowed minor components such as
hydrocarbons and other alcohols [2].
Table 2 E100 History in Brazil [3,4]
1974
1979
1985
1990s
2002
2003
2008
Brazilian government issued national alcohol

program to develop a renewable fuel for vehicle
purposes from sugar cane.
First E100 dedicated-fuel vehicle built: Fiat
model 147
Dedicated-fuel E100 vehicles comprise more
than 80% of LDV production.
Challenges for dedicated-fuel E100 vehicle
market: low oil prices and high sugar prices
=> Ethanol shortage in internal market
=> Consumers stop purchasing dedicated-fuel
E100 vehicles.
First flexible-fuel vehicle (FFV) demonstrated:
Ford Fiesta
OEMs begin offering FFVs in the market
FFVs comprise 87% of new LDV registrations.
In 2003-2004, Volkswagen, Fiat, GM, and Ford brought their

first Flexible-Fuel Vehicles (FFVs) to the Brazilian market.
Unlike their dedicated-fuel predecessors, FFVs could be
operated with gasohol or E100 (or any mixture of the two),
and thus allowed a choice between the two fuels at each fill.
By 2008, FFVs made up 87% of registrations of new
passenger cars and light commercial vehicles. With this
flexibility, consumers began purchasing E100 in increasing
amounts after 2006 as oil prices increased once again. Based
on the chronology, it appears that FFVs and high oil prices
have been key factors in the revival of the Brazilian E100 fuel
and vehicle market [5].
Ethanol was introduced in scale in Brazil in the 1970s when

oil prices rose rapidly and the Brazilian government initiated
the Pro-lcool program to develop a renewable fuel for
vehicle purposes from sugar cane [3,4,5]. A timeline of the
development of Brazils ethanol market is provided in Table 2.
In 1979 the first E100-vehicle was built, a Fiat model 147,
which was a dedicated-fuel application. As can be seen in
Figure 1, the E100 dedicated-fuel vehicle market first grew
significantly and successfully. By 1985, E100 dedicated-fuel
vehicles comprised more than 80% of LDV production.
Subsequently, the ethanol market struggled and production of
E100 dedicated-fuel vehicles declined [3]. High sugar prices
led to an ethanol shortage and higher ethanol prices.
Meanwhile, petroleum prices dropped and ethanol became
more expensive than gasoline. As a result, demand for E100
dedicated-fuel vehicles rapidly declined. As seen in Figure 2,
while gasoline and diesel fuel demand increased from 1986 to
2006, ethanol consumption was unchanged.
Page 2 of 21
Figure 1 Registrations of new light-duty vehicles in Brazil

by vehicle type, 1975 to 2009 [6].
Another reason for the market success of E100 and FFVs
appears to be the fuel price benefit of E100 versus gasohol, in
part due to higher taxes on gasoline. For the period 2003 to
2010, de Freitas [5] showed that the price differential between
the national-average prices for E100 and gasohol (after
correcting for energy content differences) were within 15% (in
either direction). Price differentials are likely to vary more by

region than the national average. In general, the availability of
both gasohol and E100 in the fuel marketplace, and the
availability of FFVs to allow free consumer choice of either
fuel based on price, provides a mechanism to dampen the
effects of price fluctuations in the oil, gasoline, sugar, and
ethanol markets.
Sufficient feedstock supply and production capacity is

required.
Flexible fuel demand (enabled by cost-efficient FFVs)
should help stabilize the alternative fuel price relative
to the competing fuel.
Long-term, consistent governmental policies that can be
relied upon by industry and consumers contribute greatly
to successful implementation.
2.2 ETHANOL UNITED STATES

The oil crisis of the 1970s was a key motivation for
development of alternatives fuels and FFVs in the US, as was
the case in Brazil. Methanol was initially identified as the
preferred alternative fuel due to low production costs and
abundant feedstock (coal, natural gas) and the air quality
benefits relative to gasoline. As such, FFVs were first
designed to operate on 85% methanol (M85) or gasoline [8].
Figure 2 Automotive fuel consumption in Brazil, 1970 to

2010 [7]
A third reason for the success of FFVs is the relatively simple
and cost-effective vehicle technology that is required to
upgrade a conventional vehicle to a FFV. The flex-fuel
technology for Brazil mainly consists of hardened valve seats
and valves, a dedicated flex-fuel controls system with two data
maps, and a separate cold start system with a separate small
fuel tank containing gasohol for cold engine starts. The next
generation of cold start system being introduced uses a heated
fuel system to replace the secondary fuels. Given the high
percentage of FFVs in new vehicle sales, any changes in
vehicle costs (net of additional vehicle cost [on-cost] and
lower vehicle taxes for FFVs) have obviously been acceptable
to consumers, with the benefit that it allows them to
participate in the fuel price benefits of E100 while being
insulated from E100 shortages or price spikes relative to
gasohol.
The following conclusions can be drawn from the Brazilian
market case.
Lessons learned:
Flex-fuel capability significantly supports the successful
introduction and market penetration of a new alternative
fuel.
Low vehicle on-cost for flex-fuel capability supports the
alternative fuel market development.
Market fuel prices need to remain competitive even if fuel
demand rises (no steep fuel price increase when fuel
demand exceeds feedstock or production capacity).
Page 3 of 21
The first FFVs were sold in the US retail market in the early
1990s and were designed for M85 capability. A few years
later, in part due to greater emphasis on addressing global
climate change, FFVs were instead being designed for ethanol
(E85). Ethanol production in the US, primarily from starch
obtained from corn, received considerable support from the
agricultural industry. It also was understood to address the
initial objective of reducing petroleum consumption and the
new objective of reducing GHG emissions.
Policy mechanisms stimulating production of FFVs by
automakers began with the Alternative Motor Fuel Act of
1988, which contained incentives in the form of credits that
could be applied to corporate fuel economy targets within the
Corporate Average Fuel Economy (CAFE) program. The next
year, the federal governmental committed to major purchases
of alternative fuel vehicles for federal fleets [8]. The Energy
Policy Act of 1992 mandated the purchase of alternative fuel
vehicles by certain federal and state government fleets. The
Energy Policy Act of 2005 provided additional mechanisms to
further promote alternative fuel vehicle acquisition (including
FFVs), develop alternative fuel supply infrastructure
(including E85), and mandate alternative fuel usage [9].
The FFVs and fuels in the US are different from those in
Brazil. In Brazil, FFVs use either anhydrous gasohol (E18
E25) or hydrous E100. In the US and Europe, FFVs are
designed to be fueled with anhydrous E0 (or E5 or E10),
anhydrous E85 (85% v/v ethanol), or any mixture of these.
The vehicle technologies are very similar, except for a
different cold start system [10,11]. Due to the lack of a volatile
gasoline fraction in hydrous E100, Brazilian FFVs use E22
fuel from a secondary tank (or a heated fuel system) for cold
starts below approximately 15C. FFVs in the US have no
secondary fuel tank and can usually start on E85 down to
approximately -15C (5F) without any auxiliaries. For cold
start at lower temperatures, an engine block heater can be
included (e.g., in Europe and the northern US). In these cold
climates, the E85 itself is sold with lower ethanol content (as
low as 70% v/v in Sweden [12] and Germany [13] and now as
low as 51% v/v in the US [14]). Cold starting below -15C
(5F) is possible with these lower ethanol content forms of
E85 without utilizing auxiliary devices. The technology used
in typical US and European FFVs is shown in Figure 3.
Despite the increasing numbers of FFVs on the road, E85 use

has not been as significant. Although the use of fuel ethanol in
the US grew steadily through the year 2000 as shown in
Figure 5, nearly all of it was used in low level blends in
gasoline (up to E10) rather than in E85 [17]. At that time,
ethanol and methyl tertiary butyl ether (MTBE) were being
used as oxygenates for tailpipe emissions reductions and
octane rating value in the fuel [15]. Ethanol use accelerated
after 2000 as the use of MTBE was phased out due to
groundwater contamination issues.
Figure 3 US and European FFV Technology [10,11 ]
Figure 5 Ethanol and gasoline consumption in US road

transportation, 1980 to 2010 [15].
From the outset, FFVs in the US were generally sold without a

price premium relative to comparable gasoline versions [8],
despite higher production costs (engineering, tooling,
materials, and controls).
FFV production began with a few thousand produced each
year from 1993 to 1997, and then increased to several hundred
thousand per year (Figure 4). In 2006, the three major USbased automakers (GM, Ford, Chrysler) announced their
intention to double production of FFVs by 2010 and that FFVs
would comprise half of new LDV offerings by 2012 if the
appropriate supporting fuel infrastructure existed. As of late
2011, over 9 million FFVs were registered in the US
(approximately 4% of all LDVs) [15], with their numbers
steadily increasing at over 1 million FFVs per year.
Figure 4 FFV percentage of new vehicle registrations in

the US, 1993 to 2011, estimated from FFV and total vehicle
registrations by vehicle model year as of March 2012 [16].
Page 4 of 21
The first Renewable Fuel Standard (RFS1) was created from

the Energy Policy Act of 2005 and mandated alternative fuel
use of 4 billion US gallons (15 billion liters) in 2006
increasing to 7.5 million gallons (28 billion liters) in 2012
(Figure 6). Most of the mandate was expected to be fulfilled
by ethanol, but it did not differentiate between ethanol used in
low-level blends and E85. Starting in 2003, several states
began mandating minimum concentrations of ethanol in
gasoline, generally E10.
Figure 6 Renewable fuel targets for the US mandated by

RFS1 (2006 to 2011) and RFS2 (2010 to 2022). RFS2
includes specific requirements for conventional (corn) biofuel
and advanced biofuels, the latter including specific
requirements for cellulosic biofuel, biomass-based diesel, and
other advanced biofuel. Actual historical ethanol and
biodiesel use is also shown (2000 to 2011) [15,18].
In 2007, shortly after the RFS1 schedule was finalized, the

Energy Independence and Security Act became law. This act
called for a new Renewable Fuels Standard (RFS2) that
accelerated and extended the mandated volumes of renewable
fuel, starting at 11.1 billion gallons (42 billion liters) in 2009
increasing to 36 billion gallons (136 billion liters) in 2022
(Figure 6). Within these total mandated volumes, there are
mandates for specific types of fuels, including advanced
biofuels (defined as having at least 50% GHG reduction
relative to gasoline), cellulosic ethanol, and biomass-based
diesel. At the time, most of this renewable fuel was expected
to be supplied as ethanol.
Although fuel ethanol consumption in the US has grown
rapidly in the last decade (Figure 5), only 1.0-1.5% of this was
used in E85 [17] while the balance was blended into gasoline
at levels up to E10. Use of E85 in FFVs has grown steadily,
but the volumes have been limited. In 2009, E85 use was
approximately 0.05% that of total highway gasoline use on an
energy-equivalent basis [17]. Reasons for the low E85 use
include the limited availability of E85 at service stations,
higher E85 prices than gasoline on an energy-equivalent basis,
perceived lower fuel economy (volumetric basis) and lower
travel range compared to gasoline, and ethanols greater value
for fuel suppliers in low-level blends [15].
As shown in Figure 7, the number of filling stations supplying
E85 in the US has continuously grown since 2004, aided by
various government incentives. As of early 2012,
approximately 2500 stations sold E85, with nearly half located
in six states (MN, IL, IA, IN, WI, and MI) [19] that have a
significant agricultural base and are major corn producers. The
current number of E85 stations represents less than 2% of the
total number of filling stations in the US, and is only half the
corresponding percentage of FFVs in the LDV fleet (4%).
Thus, at present, the E85 fueling infrastructure can be seen as
lagging the E85 vehicle fleet.
contained approximately 74% v/v ethanol on an average basis

[22]. Gasoline contained 1 to 10% v/v ethanol over the last
decade based on Figure 5. As such, for energy equivalent
pricing, E85 should have been priced (on a per-gallon basis)
at a discount of 2328% relative to E0 gasoline or 2126%
relative to E10.
Average retail prices of E85 and gasoline in the US (both
adjusted to E0 energy-equivalent price) are shown in Figure 8,
as well as the infrequent cost benefit for E85. With the
exception of a brief period in early 2009 after a rapid drop in
fuel prices, the pricing of E85 has generally provided less
consumer value than gasoline on an energy-equivalent basis.
Thus, the insufficient price discount for E85 has probably
contributed to the low E85 use observed to date in the US.
Figure 8 E85 and gasoline prices (energy equivalent) in US

and calculated E85 cost benefit, 2000 to 2011. Retail fuel
prices from [23], and assuming average of 74% v/v ethanol in
E85 and gasoline ethanol content implied from Figure 5.
Going forward, RFS2 requires continuing increases in the
amount of renewable fuel in road transportation. In the near
term, most of this is expected to be supplied as ethanol. In
2012, the gasoline pool will become effectively saturated with
E10 and other ethanol outlets will be needed. This E10 blend
wall issue is amplified by the fact that total LDV energy
demand is expected to decrease in the future as a result of
more stringent fuel economy requirements [15,22].
Figure 7 Filling stations offering M85 and E85 in the US,

1992 to 2011 [20].
As discussed earlier, due to the energy content difference, E85
prices need to be lower than gasoline on a volumetric basis
($/gallon) to be competitive on an energy basis. Until 2011,
the US industry specification for E85 has required between
68% and 83% v/v ethanol, depending on climate [21], and has
Page 5 of 21
Although there have been regulatory efforts to increase the

ethanol content in regular gasoline, exemplified by the recent
US EPA waiver allowing E15 for MY2001+ vehicles, there
are several administrative, technical, and marketing hurdles
for E15, and it is not yet present in the marketplace.
Alternatively, the use of high-level ethanol fuel blends (up to
E85) in FFVs is an immediately available outlet. High-level
ethanol fuel blends have been commercially identified as E85
(containing 7085% v/v denatured ethanol), but a recent
change now allows for a wider range (5183% v/v) of ethanol
content [14]. For E85 to see greater use, it will need to be
priced more attractively for consumers.
More cost-competitive pricing for E85 could be facilitated in

the near future by the Renewable Identification Number (RIN)
system within RFS1 and RFS2. As part of the Energy Policy
Act of 2005, the RIN system was initiated to allow more
efficient compliance by the fuel supply industry. The RINs are
generated when renewable fuel is produced or imported and
are transferred as it is blended into motor vehicle fuel for the
marketplace. Fuel blenders must acquire a certain number of
RINs for every gallon of fuel prepared, either by selling the
biofuel blend or by buying RINs from others who have done
so. (The RIN requirement is determined annually by US EPA
to ensure that the national RFS2 mandates are met) [24]. If
insufficient amounts of biofuel are being blended into fuel,
then insufficient numbers of RINs are being generated and
RINs will be in greater demand and command a higher price.
This mechanism should provide an incentive to sell E85 at
lower cost relative to gasoline based on the value of the
additional RINs that would be generated. (This mechanism has
recently come into play in the US biodiesel market [24].)
Now that the US gasoline market is nearly saturated with E10,
the RIN mechanism should encourage fuel suppliers to price
the higher ethanol content fuels (E85 and possibly E15) more
competitively as their renewable fuel obligation continues to
increase. If the RFS2 total renewable fuel mandates are
retained (and not downgraded as has recently been the case for
the cellulosic ethanol mandate [24]), then E85 (and by
extension FFVs) should become more attractive to consumers.
This situation should also provide additional motivation for
the installation of E85 pumps at filling stations. Thus,
although E85 consumption has been somewhat limited thus
far, the growing presence of FFVs in the vehicle fleet may be
a critical enabler that allows the RFS2 mandate to be met in
the future.
2.3 ETHANOL EUROPE

In Europe, Sweden was the first country to introduce FFVs
and has developed a strong market. FFVs have also been
introduced in other European countries, but with far less
success. The FFVs in Europe use the same technology as those
in the US (Figure 3).
2.3.1 ETHANOL SWEDEN

In the late 1990s, the cities of Stockholm and Gothenburg
started a purchasing consortium of communities and private
companies committed to buy several thousand ethanol cars for
municipal fleets and public transport if a company could
supply them [25]. Ford accepted the challenge and developed
a FFV specifically for the Swedish market. The vehicle, the
Ford Focus Flex-Fuel, was launched in 2001. Saab followed in
2003, Volvo in 2006, and other original equipment
manufacturers (OEMs) followed later. As shown in Figure 9,
FFV sales in Sweden started to increase rapidly in 2004 and
continued through 2008 when 22% of all new cars sold were
FFVs.
This early success was enabled by several measures taken by
the Swedish government, including a lower sales tax rate for
E85 than gasoline, incentives for FFV purchases, as well as
local incentives for FFVs (e.g. exemption from congestion
charges, free city parking) [26,27]. Similar incentives were
provided for other alternative fuels. Some of these incentives
were linked to the inclusion of FFVs and other alternative-fuel
vehicles in the federal and local clean vehicle programs.
Lessons learned:
Alternative fuels need to be priced competitively (on at
least an energy equivalent basis) for consumers to choose
to purchase them in meaningful quantities.
Vehicles designed and built with compatibility for an
alternative fuel need to enter the marketplace and
accumulate in the on-road fleet before the alternative fuel
is made available; otherwise there is no viable outlet for
the fuel.
Without a consumer pull for the alternative fuel (attractive
energy equivalent price), incentives are needed to induce
automakers to produce vehicles compatible with that fuel.
Without charging the vehicle on-cost for an FFV to the
consumer, the FFV fleet size can grow significantly.
However, competitive fuel pricing is needed to ensure that
the alternative fuel (here E85) will be used to a similar
extent.
Incentives to install alternative fuel tanks and pumps at
filling stations are helpful, but not sufficient to ensure
consumption of that fuel, particularly if the fuel cannot be
(or is not) priced competitively.
Page 6 of 21
Figure 9 Fraction of new vehicle sales in Sweden by type,

2001 to 2010 [28]
The clean vehicle standard, introduced in 2005 defines a
clean vehicle as one that is driven primarily with renewable
fuels or electricity or one that is conventionally-fuelled with
less than 120 g/km CO2 emissions. Government fleets are
required to purchase clean vehicles. The vast majority of
vehicles meeting the clean vehicle standard have been FFVs
[27].
However, as shown in Figure 9, sales of FFVs declined after

2008. At the same time, diesel car sales rose significantly. The
reasons for this change likely included the rapid drop in oil
and diesel prices after the economic crisis in 2008 and the
greater availability of diesel cars with CO2 emissions below
120 g/km that meet the clean vehicle standard (including
meeting EU4+ emissions limits). Such vehicles are attractive
due to vehicle purchase incentives and annual vehicle taxes
that are linked to CO2 emissions [29].
In terms of fuel infrastructure, a law was passed in 2006 that
required all fuel stations above a certain size to offer at least
one alternative fuel. Stations selling E85 numbered less than
100 in 2003 (2% of all stations) but steadily increased to
nearly 1700 in 2011 (59% of all stations) [30]. Stations chose
to install E85 pumps (SEK 350,000400,000; 40,000
45,000; US $50,000$55,000) rather than biogas pumps due to
a ten-fold lower installation cost [27].
Retail fuel pricing has provided an inconsistent benefit for E85
relative to the prevailing E5 gasoline [30]. As shown in
Figure 10, after adjusting for energy content differences in the
two fuels, the E85 price benefit has generally varied between
+10% and -10% that of gasoline, with mostly positive pricing
prior to late 2008. However, in late 2008 a drop in oil and
gasoline prices and an increase in E85 price resulted in E85
having up to a 30% cost penalty relative to gasoline. Since that
time, oil prices have steadily risen again, gasoline has become
more expensive, and E85 retail pricing has been more
competitive.
Furthermore, the volume of ethanol blended into E85 was

approximately equal to that blended into E5 gasoline in 2011.
Figure 11 E85 and E5 gasoline consumption in Sweden,

2005 to 2011 [30]
The Swedish example demonstrates that the actual fuel price
benefit (based on energy content) of an alternative fuel is an
extremely important factor for its success in the market. This
is particularly true for cases in which the fuels are in direct
competition with little to no performance difference. This
situation exists with gasoline and E85 in markets with
significant penetration of FFVs.
The availability of E85 pumps is another important factor that
determines the utilization rate of E85. The Swedish policy of
requiring alternative fuels at filling stations has undoubtedly
been important. Consistently low E85 prices would also offer
an attractive investment climate that would help the
development of a comprehensive fuel pump network.
Lessons learned:
Consumer fuel price benefits (versus gasoline and diesel)
are very important. An energy-based price benefit of 5
10% for E85 seems to be sufficiently attractive. The
benefit need not be continuous, but should occur with
sufficient frequency that consumers see a benefit for
considering the alternative fuel and vehicle.
Sufficient availability of E85 pumps is obviously an
important requirement for enabling significant E85 usage.
Figure 10 E85 and E5 gasoline prices (energy equivalent)

in Sweden and E85 cost benefit, 2005 to 2012. Retail fuel
prices from [30] and assuming 85% v/v ethanol in E85.
Despite the decline in FFV sales after 2008, E85 has been sold
in steadily increasing volumes through 2011, as shown in
Figure 11. The exception was 2009 when there was a 19%
year-over-year decline, likely due to uncompetitive E85 prices
(Figure 10). In 2011, sales of E85 were approximately 4% that
of E5 gasoline [30] after adjusting for energy content.
Page 7 of 21
Governmental actions are likely to be necessary to

accelerate development of the fuel supply network for the
alternative fuel, particularly if its price is not consistently
attractive relative to competing fuels.
2.3.2 ETHANOL GERMANY

While E85 and FFVs have seen some success in Sweden,
ethanol has not been nearly as successful in the rest of Europe.
The following sections describe the German experiences with
E85 and E10 as examples.
2.3.2.1 E85 GERMANY

The first FFVs entered the German market in 2005 with the
same technical content as in Sweden and the US. Unlike
Sweden, significant incentives have not been provided for
FFVs in Germany. In 2010, FFVs represented 0.05% of light
duty vehicle sales (1409 out of 2.9 million vehicles) [31].
E85 prices in Germany tend to be more attractive than typical
95-RON gasoline, even after adjusting for energy content. As
shown in Figure 12 for the period from January 2007 to
November 2011, the energy-equivalent E85 price was usually
510% less than gasoline, but also exceeded it by up to 18%
when gasoline prices were low. In general, the E85 price
advantage has been small and the benefit inconsistent.
No incentives have been granted for E85, thus it has remained
a niche product in Germany thus far. (Only 1% of the ethanol
sold in Germany in 2010 was as E85; the rest was blended into
gasoline as ethanol or ethyl tertiary butyl ether [ETBE] [32]).
Thus, the modest but inconsistent cost benefit of E85 in
Germany (-18% to +10% in 2006 to 2011) seems to be
insufficient to attract many consumers to FFV technology, at
least under the given competitive conditions with other
alternative fuels (CNG and LPG) and a strong diesel market,
as well as a poorly developed E85 infrastructure.
Gasoline 95 RON
15.0
160.00
10.0
150.00
5.0
140.00
0.0
130.00
-5.0
120.00
-10.0
110.00
-15.0
Consumer fuel price benefits (versus gasoline and diesel)

need to be positive and stable over time. Although an
(energy-based) consumer fuel price benefit of 515% may
be sufficient, market development may be hindered if that
benefit is inconsistent.
Without governmental actions to accelerate infrastructure
development for an alternative fuel, and with a modest but
inconsistent fuel price benefit, the fuel distribution system
may grow slowly.
As with cost-efficient vehicle technology, a new fuel
introduction is likely to fail if the reduction in cost of
ownership is small and inconsistent.
2.3.2.2 E10 GERMANY
100.00
-20.0
Jun- Dec- Jun- Dec- Jun- Nov- May- Nov- May- Nov- May- Nov- May06
06
07
07
08
08
09
09
10
10
11
11
12
Figure 12 Unstable E85 Cost Benefit vs. Gasoline in

Germany, 2007 to 2011 [33,34,35]
As shown in Table 5 (in Section 2.6 below), E85 FFVs have
had the lowest vehicle on-cost, and partly as a result, have the
shortest payback period of all reasonably available alternative
fuels in Germany, shorter than diesel, LPG and CNG. But, as
discussed in Sections 2.5 and 2.6, LPG is the clearly preferred
alternative fuel in Germany, even though it has a longer
payback period than E85. Consumers probably prefer LPG
because the long-term cost savings are much greater than with
E85. Furthermore, E85 fuel stations are very limited in
number as compared to LPG. As of March 2012 in Germany,
there were approximately 6500 LPG stations and only 311
E85 stations [36]; out of the total 14,700 filling stations [37],
this represents 44% with LPG and 2% with E85. One reason is
that the installation of E85 stations has not been supported by
Page 8 of 21
Lessons learned:
E85 Cost Benefit
170.00
E85 Cost Benefit / %
Retail Price / ct/Liter gasoline equivalent
E85 corrected by energy content
incentives. Furthermore, legal hurdles for installing an E85

station have been very high and in some parts of Germany it
has been generally forbidden to build E85 stations. In most
parts of Germany, E85 pumps have only received interim
exceptions to operate, not permanent legal approval. As a
result, long-term planning is not possible and many
bureaucratic hurdles have to be overcome. Thus far the
sizeable investment to install an E85 tank and pump
(approximately 20,000 [38]) has been a questionable
investment and has been avoided.
The recent attempt to increase the ethanol content of gasoline

in Germany from 5% v/v to 10% v/v (E10) has not gone
smoothly. Concerns had been identified for materials
capability with E10, particularly for the high pressure fuel
systems of the first-generation direct injection engines (not
sold in the US). After extensive discussions and capability
reviews by the OEMs, E10 fuel was introduced in 2011 [39].
Because 7% of the German vehicle stock was identified as not
being E10-capable [40], vehicle compatibility lists were issued
(a very challenging process in itself) and a protection grade
fuel (E5) had to be kept in the market. This approach caused
considerable consumer confusion. Consumers have expressed
uncertainty about the correct fuel for their vehicle and have
mostly chosen to avoid E10, resulting in much less E10
consumption than expected [40].
Lessons learned:
Widespread blending of the standard market fuel with a
new fuel is ideally accomplished with complete
backwards-compatibility with the existing vehicle stock. If
the fleet is only partially compatible, then vehicle
compatibility lists must be issued, detracting from
consumer confidence in the new fuel.
Depending on the fuel, the maximum blend rates can be
very limited without introducing compatibility issues.
2.4 BIODIESEL GERMANY

In the 1990s, biodiesel in a neat form (B100) was introduced
in Germany. Rapeseed for biodiesel production was cultivated
on fallow land in the 1990s. Because the German government
did not impose any taxes on biodiesel during the introduction
phase, B100 could be sold with a consumer price benefit
relative to fossil diesel [41]. For example, from 2004 to 2006,
the production cost of B100 (approx. 0.76 /liter [41]) was
much greater than the pre-tax cost of diesel (0.350.53 /liter
[33]). However, due to taxes on the diesel fuel the retail price
for B100 was approximately 515% lower than fossil diesel
on an energy-equivalent basis.
In addition, some OEMs announced that their light duty
vehicles in the existing market fleet were compatible with
B100. This very important step led to B100 becoming rapidly
accepted and purchased by consumers within a short time.
As can be seen in Figure 13, biodiesel sales rose from 1990 to
2006, mostly as B100. By 2006 there were approximately
1900 biodiesel stations in Germany [41]. Then, deficits in fuel
tax income, caused by the increasing B100 sales, were
recognized by the German government. As a consequence,
B100 as neat fuel was taxed as of 2008. Compensatory
regulations were enacted that supported greater blending of
biodiesel into fossil diesel, such that the maximum allowed
biodiesel blend limit was extended from 5% v/v (B5) to 7%
v/v (B7).
At the same time vehicle incapability issues increased because
many modern diesel vehicles equipped with common-rail
fuel injection systems, particulate filters and post-injection
strategies for purging these filters were not B100-capable.
In 2008, B100 sales declined by 0.74 Mt or 41% from the
prior year. Despite the regulations supporting greater biodiesel
blending in fossil diesel, the increase of the blended biodiesel
was only 0.19 Mt such that total biodiesel sales declined by
17% [41].
and the existing European diesel fuel standard that limits the
maximum blend rate to B7 [44]. The original German
government proposal in 2008 was to increase the biodiesel
blend limit to 10% v/v (B10). But because of OEM concerns
about incompatibility with B10, with risks such as oil dilution,
oil degradation, deposit formation, and materials
compatibility, the limit was set to the current B7 and the
complete vehicle stock was declared to be capable. This
approach largely created widespread consumer acceptance,
unlike the recent transition from E5 to E10 in which the entire
fleet was not declared to be compatible.
Lessons learned:
Backwards vehicle fleet capability is critical to the success
of an alternative fuel introduction.
When a sufficient number of capable vehicles are available
in the market, a fuel cost benefit of 515% versus the
established fuel (in this case diesel) seems to be sufficient
to generate consumer demand.
Incentives (here sales tax reduction) can spur the growth of
an alternative fuel market and can make an alternative fuel
successful if a sufficient fraction of the existing vehicle
fleet is declared compatible with the fuel.
Governments will be motivated to withdraw incentives if
they become too costly (paradoxically due to successful
growth of the alternative fuel market), which can rapidly
reverse the market success.
The greater the cost-competitiveness of the alternative fuel
in the long-term (without subsidies), the more likely the
fuel will be able to avoid a market collapse as incentives
are reduced. However, the long-term cost-competitiveness
of an alternative fuel may be influenced by differing policy
treatment to account for differences in perceived external
costs.
2.5 LPG EUROPE

LPG consists primarily of a mixture of propane and butane.
Under typical storage pressure (812 bar under normal
ambient conditions [49]), the fuel is in a liquid form. Even
though LPG is not renewable, it can deliver an approximately
10% tank-to-wheel (TTW) CO2 emissions reduction versus
gasoline, due to its greater hydrogen-to-carbon ratio [45].
Figure 13 Biodiesel Sales in Germany, 1990 to 2011

[42,43]
At present, biodiesel is primarily used in Germany as a blend
component in fossil diesel in concentrations up to B7, because
of backwards compatibility limits of the existing vehicle fleet
Page 9 of 21
LPG is currently the most used alternative fuel in Europe and

has a 3% European market share [46]. For comparison,
biofuels represented approximately 2% of road transport
energy used in the EU in 2007 [47]. European LPG vehicle
(LPGV) registrations have increased greatly in the last decade.
Most LPGVs are retrofit systems. More than 450,000 LPGVs
were registered in Germany in late 2011 (at the same time
about 75,000 natural gas vehicles were registered) [48]. The
total number of new OEM bi-fuel LPGVs registered from
2006 to 2010 was only 43,000 [31,48]. Prior to this, there were
essentially no OEM LPGV registrations. Assuming that all of
these OEM LPGVs are still on the German market (average
age of German cars is 8.5 years [48]), their share of the total
LPGV fleet was 9.4% in late 2011. This implies that
approximately 90% of German LPGVs are retrofits. The share
of retrofitted LPGVs in other European countries is even
greater [46].
The majority of retrofitted LPG systems use a gaseous LPG
port fuel injection system [49,50]. The additional LPG tank,
with typical capacity of about 40 liters (10 US gallons)
enabling a 400500 km (250300 mile) range, is usually
mounted in the spare wheel well. The additional system
weight, including tank, is about 60 kg (130 lb). Usually LPG
is conveyed by the vapor pressure of the fuel in the fuel tank.
The LPG first flows to the evaporator where it is vaporized.
The gaseous fuel is injected through separate fuel injectors
into the intake manifold. To start the engine at low
temperatures, these retrofitted LPGVs need the additional
gasoline capability (bi-fuel) from the existing fuel tank, since
the evaporator and fuel supply do not work properly at low
temperatures. At very low temperatures the LPG has a very
low vapor pressure and fuel does not flow to the injectors. In
that case the system is automatically switched to gasoline
operation (bi-fuel capability required). Retrofitters also offer
systems with liquid LPG injection into the manifold, which
require an additional fuel pump but eliminate the need for an
evaporator. There are also systems offered (typically not
approved by OEMs) where LPG is directly injected into a
modified gasoline high-pressure direct injection system.
Typical retrofit kits utilize a slave control unit to operate the
LPG injectors, which is placed between the injection signal
output of the engine control unit (ECU) and the LPG injectors.
When the driver selects LPG operation, the gasoline injectors
are switched off. Most OEM bi-fuel vehicles also utilize this
kind of control system.
The LPG infrastructure has been growing in response to
vehicle registrations and consumer demand. As shown in
Figure 14, LPG stations in Germany have seen significant
growth since 2005 and are now widely available, considerably
more than CNG stations [51]. In early 2012, LPG and CNG
were available at approximately 44% and 6%, respectively, of
German filling stations. Meanwhile, there is a very sufficient
LPG infrastructure in many parts of Europe, with Turkey,
Poland, and Italy being the most developed. The main reasons
for the success of LPG are incentives and tax reductions in
many European countries.
Figure 14 LPG and CNG filling stations in Germany, 2002

to 2012 [51].
Another important reason for LPG success is the availability
of aftermarket retrofit conversion kits for used gasoline
vehicles (bi-fuel vehicles). The conversion of used cars is
particularly attractive in cost-sensitive markets, as the
strongest LPG markets tend to be, because the main reason for
consumers to change from a well-established fuel (gasoline) to
an alternative fuel such as LPG is the lower operational cost.
LPG aftermarket conversion kits are available for only
moderate on-costs (German OEM consumer on-cost
approximately 20002500 [$ 26003300)] per vehicle
[52,53]; Eastern Europe retrofit, non-OEM on-cost starting at
approx. 700 per vehicle). The less expensive, usually OEM
uncontrolled retrofit kits are typically of lower quality and
durability. For example, OEMs typically upgrade the valves
and valve seat inserts of their LPG engines, whereas retrofit
conversions usually do not use these relatively expensive
replacement parts. Therefore the engines of many aftermarket
converted vehicles will have considerably reduced durability,
since the poor lubricity of LPG fuel leads to increased valve
seat wear [54,55].
Lessons learned:
Consistently low consumer fuel prices with readily
available vehicle bi-fuel capability are strong driving
forces for the market penetration of an alternative fuel.
Cost-efficient retrofit conversion possibilities for existing
vehicles can significantly help to develop the fuel market.
However, typically OEM uncontrolled retrofit kits
reduce engine durability. Therefore, reliance on
aftermarket vehicle conversion through retrofit kits should
be considered very carefully in any introduction strategy
for an alternative fuel.
2.6 CNG ASIA AND EUROPE

Natural gas (NG) is mostly methane (CH4) with a wide range
of other components (e.g. N2, CO2, C3H8, H2, etc.) depending
on the source and how it is processed. Liquid fuels are
generally preferred to gaseous fuels in vehicle applications
because of their ease of handling and storage. However, NG is
an attractive automotive fuel because it is available
Page 10 of 21
worldwide, found in abundant supply, and much of it can be

developed at relatively low cost [56]. The current mean
projection of the recoverable NG resource is 16,200 trillion
cubic feet (460 trillion m3), or 150 times current annual global
NG consumption [57]. Furthermore, because of its greater
hydrogen-to-carbon ratio, it provides a 20-25% TTW CO2
emissions reduction versus gasoline [45].
In the mid-1990s, compressed natural gas (CNG) became a
significant automotive fuel for mass-production passenger
vehicles in the European market when several auto
manufacturers introduced CNG-capable vehicles [58]. For
automotive application, NG is compressed up to 250 bar
pressure and the CNG remains gaseous under all typical
temperature conditions [59].
Practically all CNG vehicles (CNGVs) use a retrofitted port
fuel injection system [49,60]. Because CNG remains a gas at
typical storage temperature and pressure, the additional CNG
containers require much more storage volume than LPG tanks
and as such cannot be mounted in the spare wheel well. CNG
containers are typically placed on the trunk floor (decreasing
trunk capacity) or as an under-floor system when the vehicle
platform is supporting this (OEM solution only). OEMs also
typically upgrade the valves and valve seat inserts due to the
poor lubricity of CNG [55].
CNG flows using the pressure within the CNG storage
container. A pressure regulator reduces the CNG pressure to
approximately 210 bar (system-dependent) before it is
injected through separate CNG injectors into the intake
manifold. Typical retrofit kits and many OEM bi-fuel vehicles
utilize a slave control unit to operate the CNG injectors,
placed between the injection signal output of the ECU and the
CNG injectors. When the driver selects CNG operation, the
gasoline injectors are switched off.
The typical amount of CNG stored on the vehicle is about 12
20 kg (2644 lb) depending on vehicle size, which usually
enables a cruising range of 250450 km (160280 miles). The
additional weight of the complete CNG system is
approximately 150 kg (330 lb), while the package space
occupied by the fuel tanks is approximately 100 liters (3.5 ft3)
[49].
Worldwide in 2010, CNGVs totalled 12.7 million and their
numbers grew by 12% from the prior year. More than 18,000
CNG filling stations were available worldwide in 2010, with
year-over-year growth of 10% [61]. In Table 3, the 20 largest
CNGV markets are listed in rank order of CNGVs.
Asia is the largest CNG vehicle market and also accounts for
the worlds largest growth in CNGVs (42% since 2001) [61].
Pakistan is the worldwide leader with 2.7 million CNGVs or
61% of all passenger vehicles. India has 1.1 million CNGVs
but these comprise only 1% of passenger vehicles. China and
Thailand are also among the top ten CNGV markets.
Page 11 of 21
Table 3 Worldwide CNGV Markets [61]
Country
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Pakistan
Iran
Argentina
Brazil
India
Italy
China
Colombia
Thailand
Ukraine
Bangladesh
Bolivia
Egypt
United States
Peru
Armenia
Russia
Germany
Bulgaria
Uzbekistan
No. of
CNGVs
2,740,000
1,954,925
1,901,116
1,664,847
1,080,000
730,000
450,000
340,000
218,459
200,000
193,521
140,400
122,271
112,000
103,712
101,352
100,000
91,500
60,270
47,000
No. of
CNG
Stations
Year
3,285
1,574
1,878
1,725
571
790
1,350
614
426
285
546
156
119
1,000
137
297
244
900
81
133
2010
2010
2010
2010
2010
2010
2009
2010
2010
2006
2010
2010
2009
2010
2010
2010
2009
2010
2009
2010
CNG development in Pakistan started in the 1980s in an effort

to reduce dependency on petroleum. The national Oil and Gas
Regulatory Authority has regulated all CNG activities since
1992. The use of incentives enabled Pakistan to become the
largest user of CNG in the world by 2008. In addition to the
duty-free import of CNG kits and other CNG equipment,
policy incentives have included a consumer price advantage
versus gasoline of approximately 60% ensured by national
fuel price controls. Almost 70% of the CNGVs are aftermarket conversions. Approximately 3,300 CNG stations are
operational and more than 10,000 CNG conversion facilities
are supporting the installation and maintenance of CNG kits
and vehicles [62].
In India, several government mandates led to the development
of the CNGV market. In many areas, public transportation
vehicles (buses, taxis, and three-wheelers) are obligated to use
CNG. The price of CNG in India is approximately 40% less
than diesel fuel and more than 50% less than gasoline.
(Because NG is a gas, its price is often provided in terms of a
gasoline gallon equivalent.) Lower sales tax on CNG
contributes to the favorable pricing (an average of 11%
compared to 12% to 33% for gasoline and diesel).
Concessions on import taxes for CNG kit components are also
provided [62].
Most other regions of the world have also seen a significant
growth in CNGV markets. After Asia, the region with the
second largest CNGV market growth (20012010) is Latin
America with 18%, followed by Africa (15%) and Europe
(14%). The only region without growth was North America
(-0.1%) [61]. The more cost-sensitive markets in developing

countries appear to have preferentially adopted CNG relative
to more developed countries.
CNG was 58% lower cost than gasoline and 50% lower than
diesel, but only 29% less than LPG, while LPG itself had a
cost benefit of 40% versus gasoline and 30% versus diesel.
Besides the two successful CNG markets in Asia already

discussed, it is worthwhile to consider Europe with its
relatively fragmented market situation. Italy is the largest
CNGV market in Europe (No. 6 worldwide) and was the first
European country that attempted to develop a CNGV market
(in the 1970s). All Italian CNGVs were retrofits until the mid1990s when OEM CNGVs became available [63]. Various
incentives for CNG fuel stations, retrofit conversions, and
favorable CNG tax treatment have aided the market [64].
The number of CNGVs has reached this high level with the
help of a very active retrofit conversion industry. There has
also been a strong economic reason behind this growth. Fuel
costs with CNG have been about 60% less than gasoline and
33% less than diesel. The availability of more new car models
with OEM CNG options is pushing the growth of the CNGV
market further [58].
In Germany, the second largest European market, CNGVs
have been significantly less successful than in Italy. The
market development started in the mid-1990s with the
introduction of OEM CNGVs. In a few years, more than
90,000 CNGVs were present, appearing to be a good start.
This development has been pushed by two main strategic
initiatives and policies from the German NG industry and
government. The NG industry committed to a rapid
development of the public CNG filling station network, now
more than 900 locations (the most of any European country).
In 1994, the German government committed to a reduced tax
rate for NG as a vehicle fuel until 2009. In 2002 the German
government acknowledged CNGs potential and extended the
tax benefits until 2020, but did not extend the tax benefit of
LPG (a competing alternative fuel) past 2009 [65].
However, in 2006 a new German government revised the
expiry dates for the tax reduction to 2018 for both CNG and
LPG [65]. As can be seen in Figure 15, this policy change
directly impacted new CNGV registrations after 2006. New
CNGV registrations (bi-fuel and mono-fuel) peaked in 2006
2008 at approximately 11,000 vehicles per year. In the next
two years, annual CNGV sales decreased by more than 60% to
4500 vehicles. In the same time (20062010), registrations of
LPGVs doubled from 4000 to 8000 vehicles per year, despite
the fact that LPG has been more expensive than CNG on an
energy basis [66]. In 2011, there were approximately 96,000
CNGVs in Germany, accounting for only 0.2 % of the German
vehicle stock [66].
In December 2011, the average cost of CNG in Germany was
0.74 /m, or 1.03 /kg, while gasoline cost 1.60 /liter, diesel
1.49 /liter, and LPG 0.73 /liter [66]. Fuels are priced based
on different units of measure and have different energy
content, thus Table 4 summarizes these fuel prices corrected to
an equal energy basis of 1 liter of gasoline. On this basis,
Page 12 of 21
Figure 15 New vehicle registrations of alternative fuel

vehicles in Germany (excluding retrofits), 2002 to 2010 [31]
Thus LPG replaced CNG in Germany as the most-used
alternative fuel despite a CNG price advantage of nearly 30%
versus LPG. There are several likely reasons for this. First,
CNGVs are more expensive than LPGVs due to a more
expensive fuel system. For vehicles purchased from OEMs,
the consumer on-cost for a CNGV is approximately 3400
($ 4500) relative to a gasoline version [67], whereas the
on-cost for a LPGV is approximately 20002500 ($ 2600
3300) per vehicle [52]. While retrofitting for CNG is difficult
and is in the same cost range as OEM solutions [68], lowerquality retrofit kits for LPG can be bought and installed
starting at 700.
Table 4 German Fuel Prices and Taxes per Energy
Content of 1 Liter of Gasoline Equivalent, December 2011,
adapted from [66].
Fuel
Gasoline
Diesel
CNG
LPG
Fuel quantity
with energy
content of liter
of gasoline
1 liter
0.92 liter
0.66 kg
1.31 liter
Fuel Price,
per liter
gasoline
equivalent
1.60
1.37
0.68
0.96
Fuel Tax,
per liter
gasoline
equivalent
0.665
0.432
0.120
0.120
As shown in Table 5 (a scenario comparing the 2008-MY

Ford Focus with CNG, LPG, diesel and gasoline powertrains
[69], assuming German fuel prices as of December 2010), the
payback time for the CNGV is significantly higher than that of
the LPG or diesel version. This calculation neglects the
increased service and system inspection requirements
(enforced by law) for the gaseous fuel systems, and
differences in insurance and automobile taxes.
Table 5 Payback time for CNGV, LPG, E85 and diesel

vehicle on-cost in Germany based on fuel prices in Dec. 2010
(excluding costs for maintenance, inspections, insurance, and
automobile taxes).
Ford Focus
CNG
Ford Focus
Ford Focus
Ford Focus
FFV (E85)
Gasoline
Diesel
Hyphotetical*
2.0l NA 126 hp 2.0l NA 141 hp 2.0l NA 145 hp 2.0l NA 145 hp 2.0l TC 136 hp
Fuel Price /
liter or kg (CNG)
Fuel Consumption
(NEDC) / l/100km
or kg / 100km (CNG)
Vehicle On-Cost vs.
Gasoline /
Yearly Mileage / km
15,000
30,000
45,000
Yearly Mileage / km
15,000
30,000
45,000
Yearly Mileage / km
15,000
30,000
45,000
Ford Focus
LPG
1.03
0.73
1.06
1.60
1.49
5.7
9.5
9.8
7.2
5.8
2,500
250
0
Yearly fuel costs /
1,040.25
1,558.20
1,728.00
2,080.50
3,116.40
3,456.00
3,120.75
4,674.60
5,184.00
Yearly fuel cost benefit versus gasoline /
687.75
169.80
basis
1,375.50
339.60
basis
2,063.25
509.40
basis
Pay Back Period / years
3.6
1.5
basis
1.8
0.7
basis
1.2
0.5
basis
2,000
3,400
880.65
1,761.30
2,641.95
847.35
1,694.70
2,542.05
4.0
2.0
1.3
1,296.30
2,592.60
3,888.90
431.70
863.40
1,295.10
4.6
2.3
1.5
* European E85 FFV only existing as 1.8l version (fuel consumption calculated)
For the assumptions shown, and considering the German

average mileage per vehicle of less than 15,000 km/year
(13,200 km/year in 2002 [70]), the payback period is
estimated to be 4.0 years for the CNGV, 3.6 years for the
LPGV, and 1.5 years for the E85 FFV. Diesel vehicles have a
longer (4.6-year) payback, but are attractive to German drivers
due to their superior drivability vs. conventional natural
aspirated spark-ignited powertrains, a fully-developed fueling
infrastructure and decent long-term fuel cost savings. For
those drivers who focus primarily on cost, LPG is preferred to
CNG, diesel, and E85 (despite the shorter payback period for
E85). Consumers probably prefer LPG because few E85
stations are available and the long-term cost savings are much
higher than with E85. (E85 pays back early because of the low
vehicle on-cost, but is inferior in the long-term because of
lesser fuel cost savings.) With aftermarket retrofit kits, used
cars can be converted to LPGVs at even lower cost and would
have an even shorter payback period.
In order for CNGVs to gain the same payback time as an
OEM-built LPGV in the cost scenario above, the CNG price
would need to decrease from 1.03 /liter to 0.92 /liter, which
would result in an energy-corrected CNG cost of 0.61 /liter
gasoline-equivalent, or a fuel cost benefit of 62% versus
gasoline. A second possibility would be to reduce the CNGV
on-cost ( 3400) by approximately 10% (to 3000) to
compete with OEM-built LPGVs, or by approximately 60%
(to 1300) to compete with aftermarket retrofit LPG
conversions, which is unlikely to be feasible.
Another important factor for the relatively low CNGV
penetration in Germany is the greater initial cost for filling
stations to provide CNG versus LPG and the resulting effect
on infrastructure development. The equipment costs for CNG
stations are reported to be 1015 times greater than for LPG
stations (approx. 200,000350,000 for CNG [71] versus
Page 13 of 21
20,00025,000 for LPG). The lower investment cost for

LPG stations together with the shorter consumer payback
period for LPGVs has led to a considerable increase in the
number of LPG stations since 2005, while the number of CNG
stations has only increased slightly (Figure 14). While in 2004
there were approximately the same number of CNG stations
and LPG stations in Germany (about 500 each), by March
2012 there were approximately seven times as many LPG
stations (6500) as CNG stations (900). Of the total number of
German filling stations in March 2012, 44% sold LPG
whereas only 6% sold CNG.
Finally, given the greater costs associated with vehicle and
filling station conversions relative to LPG, CNG can only
become a significant automotive fuel if it is priced more
favorably with respect to LPG. As the German market has
demonstrated, it is difficult to develop multiple alternative fuel
markets at the same time.
Lessons learned:
Due to the high conversion costs of CNGVs and the slowly
growing fuel station infrastructure (due to high
infrastructure investment costs), CNG as a fuel appears to
require an energy-based cost benefit versus gasoline on the
order of 5070 % to develop a significantly growing CNG
market (examples: Pakistan, India, Germany).
Cost-efficient retrofit conversion possibilities for existing
vehicles as well as cost-efficient OEM-built CNGVs may
help to develop the market (examples: Pakistan, India,
Italy) but cannot guarantee it.
The LDV fuel market is very cost sensitive. Consumers
will choose the fuel that offers the lowest total cost over
the first few years of vehicle ownership.
Fuel cost benefits and operational cost benefits must be
reliable in the long term; otherwise, consumers are likely to
avoid the alternative fuel (e.g., Germany).
Simultaneous development of multiple alternative fuel
markets can be difficult. An effective future fuel strategy
needs to consider the impact of competition between
alternative fuels (e.g., Germany) as well as with the
existing conventional fuels.
2.7 SUMMARY OF LESSONS

LEARNEDALTERNATIVE FUELS
AND INTRODUCTION STRATEGY
From the observations in the preceding sections, some basic
requirements can be derived for future alternative fuels that
allow for development of a significant market:
1. Affordability and cost competitiveness is the most
important factor to attract potential consumers to an
alternative fuel and vehicle technology. The vehicle and
fuel costs need to be competitive with alternatives. The

consumer must have a reasonable chance to recover the
vehicle on-cost and to considerably save on operational
costs over the first few years.
For FFVs, with low consumer on-cost and moderate E85
infrastructure costs, an energy-based fuel cost benefit of at
least 5 % seems to be required.
For OEM-built LPGVs, with a greater consumer on-cost
(approx. 20002500 [US $ 26003300]) and moderate
LPG infrastructure costs, a reliable fuel cost benefit of at
least 40 % versus gasoline has been sufficient.
For CNGVs, with a greater consumer on-cost (approx.
3500 [US $ 4500]) and high CNG infrastructure costs, a
reliable fuel cost benefit of 5070 % has been sufficient to
develop its market (e.g., Pakistan, India).
For any alternative fuel, it is important that the cost benefit
be reliable and stable in the long-term (e.g., FFVs in
Germany and Brazil). Fuel prices need to remain
competitive even if fuel demand rises with the successful
development of that market (e.g., FFVs in Brazil).
Sufficient feedstock and fuel production capacity is
required.
Governmental actions are typically necessary to facilitate
development of the alternative fuel market, particularly if
its price is not consistently attractive relative to competing
fuels. When an artificially-stimulated and tax-incentived
fuel market grows too much, the lost tax revenue may spur
the government to reduce the incentives, which can lead to
reversal of the market success (e.g., B100 in Germany).
Thus, ideally for a large-scale introduction of an alternative
fuel, the fuel should be cost-efficient in the long-term and
its price resilient to relaxation of subsidies.
2. Backwards-compatibility of vehicles is very beneficial
for the successful development of an alternative fuel
market (e.g., B100 in Germany). If the vehicle fleet is
backwards-compatible, a fuel cost benefit in the range of
515 % versus the established fuel seems to be sufficient
(e.g., B100 Germany).
Alternative fuel can also be blended into existing fossil
fuels if the existing vehicle fleet is compatible. Even if
only a small percentage of the existing vehicle stock is not
compatible with the alternative fuel blend, then vehicle
compatibility lists must be issued and protection grade
fuels must be retained for incompatible vehicles, which can
be a politically very complicated process (e.g., E10 in
Germany).
3. Affordable distribution infrastructure is another
important factor for the development of an alternative fuel
market (e.g., CNGV and LPG in Europe). High
infrastructure investment costs will lead to slow growth in
fuel station numbers. A sufficient supply network is
Page 14 of 21
required for consumer acceptance and significant use of

that fuel (e.g., E85 in Sweden). If the supply network is
insufficient, then cost-effective bi-fuel or flex-fuel
capability in vehicles can compensate.
4. Vehicle capability for two fuels (bi-fuel vehicle, monofuel vehicle, or FFV) overcomes issues associated with an
alternative fuel that is not backwards-compatible to the
vehicle fleet and/or supported by a sufficient supply
infrastructure (e.g., B100 in Germany, LPG/CNG in
Europe). Capability for two fuels provides consumers with
confidence in the ability to refuel even if the alternative
fuel is unavailable and with certainty for avoiding
uncompetitive fuel pricing if the fuel cost benefits are
inconsistent. For example, in the Brazilian E100
introduction, dedicated-fuel vehicles eventually failed
while FFVs have become successful. Lower vehicle oncost for the second-fuel capability provides earlier payback
for the additional fuel system installation and improves
market success. Capability for two fuels does not
necessarily mean that the powertrains are still optimized
for gasoline and do not exploit the potential of some
alternative fuels (e.g. high octane of ethanol or methane).
A bi-fuel vehicle can be optimized to the alternative fuel
and in certain situations would provide degraded
functionality in the gasoline operation mode. Some monofuel applications on the market are already optimized for
the alternative fuel [72].
5. Retrofit kits can significantly help to develop an
alternative fuel market (e.g., LPG in Europe; CNG in Italy,
Pakistan, India), because they allow the conversion of used
vehicles already in the fleet. The conversion of used
vehicles is particularly attractive in cost-sensitive markets,
as most alternative fuels markets are, since operational cost
is the main reason for these conversions. Retrofit kits are
often available at much lower cost than OEM solutions
(e.g., LPG in Europe), but these are usually OEM
uncontrolled and of lower quality and with fewer
upgraded components than the OEM systems. For
example, OEMs typically upgrade the valves and valve
seat inserts of their engines for CNGV, LPGV, and FFV
applications, whereas retrofit kits usually do not contain
these relatively expensive upgrades. Therefore, engines
with these retrofit conversions can have reduced durability,
since the poorer lubricity of gaseous and alcohol fuels can
lead to increased valve seat wear. Therefore the supporting
effect of aftermarket vehicle conversion with retrofit kits
should be carefully considered by OEMs in any
introduction strategy for an alternative fuel.
6. Sufficient fuel energy density is important, since the fuel
storage capacity in passenger cars is limited. Long travel
range of a vehicle is a consumer demand and is not readily
compromised. Also, many consumers are aware of their
volumetric fuel economy and are dissatisfied when it is
reduced by the fuel, especially when the fuel price has not
been discounted appropriately. Most alternative fuels
reduce the vehicle range (or reduce the available interior
space with greater fuel storage) in comparison to gasoline

and diesel. This is particularly an issue for gaseous fuels,
but even more for battery electric vehicles (BEVs).
7. Acceptable fuel fill time is also important, as consumers
have become accustomed to filling their vehicles with
gasoline or diesel within a few minutes. Increasing the
filling time to hours (as for BEVs) is a strong source of
dissatisfaction. Therefore a compromise in filling time
needs to be adequately balanced by other positive
attributes.
8. Sustainability attributes of the fuel and vehicle should be
maximized from a WTW perspective, considering the
production and use phases and end-of-life disposition. In
addition to GHG emissions, all other tailpipe emissions
should be minimized and need to meet applicable
regulatory limits. Fuels that enable reduced tailpipe
emissions are more likely to be supported. Other factors
that should be considered include land use, energy use,
water use, strategic materials, and social issues. Future fuel
strategies should ideally have an endpoint that includes a
sustainable, non-fossil, renewable fuel.
9. Incentives for sustainable alternative fuels are initially
required if they have higher production and/or distribution
costs than gasoline/diesel (after tax) in order to be
affordable and cost-competitive. Although continuation of
increasing oil prices in the future would make alternative
fuels more attractive, some stimulus is usually required for
development of the distribution infrastructure. Truly
sustainable (non-fossil, renewable, and GHG-reducing)
fuels are typically more expensive than gasoline/diesel
fuels [71]. Therefore incentives will most likely be
required to bring these fuels into the market. Governments
may also support their long-term cost-competitiveness
through differing policy treatment that accounts for
differences in perceived external costs (e.g., climate
impact). Governments pursuing climate-friendly policies
are unlikely to support unsustainable fuels in the long term.
10. Scale: For any fuel, there must be enough feedstock
available (at a cost allowing a competitive fuel price) to
develop and sustain the market in the long term. If not, the
eventual feedstock scarcity will result in demand
exceeding supply, and will cause fuel price increases. If the
market relies on dedicated-fuel vehicles, this can lead to
collapse of the market for that vehicle (e.g., E100 vehicles
in Brazil). If there is fuel flexibility, then the sales of the
fuel will drop until demand is in balance with supply (e.g.,
B100 in Germany). The future fuel supply needs to be
scalable with future demand when the market develops
successfully.
Page 15 of 21
3. POSSIBLE FUTURE SCENARIOS

Considering the above requirements for alternative future
fuels, two requirements are essential for long-term
sustainability. There should be a future source of the fuel that
is non-fossil and renewable and that is eventually costcompetitive even in large volumes without compromising
basic societal resource needs (food, water, etc.).
Future fuel prices are difficult to project, however some trends
are evident. In 2011, the International Energy Agency (IEA)
Renewable Energy Division provided scenarios of future
prices of fossil and renewable fuel under different fuel price
assumptions [73]. In the long term (2050), two fuels were
projected to be less expensive (pre-tax) than gasoline in both
scenarios: methane and ethanol. Methane can be used as
automotive fuel in CNGVs in any blend up to 100%. Ethanol
can be used as a blend component with gasoline in fairly
limited concentrations in conventional vehicles, but at higher
concentrations in FFVs. Possible development scenarios for
these two alternative fuels are discussed in the following
section.
3.1 ETHANOL SCENARIO

In the future, higher ethanol volumes are expected in the US
fuel pool, driven in the short term by RFS2 [15,22]. In the EU,
the Renewable Energy Directive and policies by some
member states are expected to result in greater ethanol use
than at present. In the long term, greater ethanol volumes may
be available due to advances and cost reductions in cellulosic
ethanol production, gains in feedstock yields and overall
production, policy mechanisms, and/or high oil prices.
An attractive opportunity provided by increasing ethanol
availability is to increase the octane rating of regular-grade
gasoline [15]. The octane rating of the fuel that will be used in
an engine determines the limits to which the engine can be
designed and operated to extract useful work out of the fuel,
without leading to engine knock. The major automakers
design around the most heavily used fuel, typically the regular
grade (i.e., lowest octane rating). If minimum octane ratings
for regular gasoline were increased, engines in future vehicles
would be designed for greater efficiency through higher
compression ratios and/or turbocharging and downsizing.
Vehicles already on the road would benefit through a reduced
need for spark retard and enrichment [15].
Ethanol could enable such a change. Ethanol has higher octane
ratings than typical petroleum refinery streams used to make
gasoline. Despite this, the E10 on the US market today just
meets the minimum octane ratings for regular gasoline
because the refining and blending industry adds the ethanol to
a gasoline blendstock for oxygenate blending (BOB) that has
considerably lower octane ratings [15], thereby reducing
production costs. An alternative scenario, as shown in
Figure 16, would be to increase octane ratings in concert with
ethanol addition, particularly for ethanol above the current

E10 level. The anti-knock quality of such blends is also
effectively increased due to the greater heat of vaporization of
ethanol, particularly when used in direct injection engines.
Both of these properties would enable the engine design
modifications and engine operation adjustments described
above, leading to greater efficiency across the entire vehicle
fleet.
in the future. FFVs would also be compatible with possible

future intermediate ethanol-content blends. FFVs could
become particularly desirable to consumers if higher ethanol
blend fuels are attractively priced.
As a long-term strategy, use of E85 in FFVs would provide a
lesser benefit than blending ethanol across the fuel pool and
increasing its minimum octane rating. There are a few reasons
for this. First, as shown in Figure 16, ethanol provides a
greater incremental octane enhancement when blended at low
concentrations (in all gasoline) than at high concentrations (in
a proportionally smaller E85 volume). Second, although E85
has a high octane rating, FFVs as designed today would not
greatly benefit from it. If new FFVs were optimized for E85,
they would not provide competitive performance when using
lower-octane E0-E10 gasoline (e.g., considerable decreases in
power and torque) and thus would be inferior to nonoptimized vehicles when driven with the more widelyavailable E10 fuel. Therefore such vehicles would not be
purchased by sufficient numbers of consumers, particularly if
E85 is not reliably cost-competitive and widely available.
3.2 METHANE SCENARIO
Figure 16 Estimated Research Octane Number (RON)

values for ethanol-gasoline blends with contour lines of
constant blendstock RON. Based on approach in [74].
Such a change would of course involve challenges [15],
including how to accomplish the transition of the gasoline
supply from E10 as the primary blend level to one with higher
ethanol content. Todays US filling stations and vehicles
(designed around E10 as the maximum ethanol content) would
need to be gradually replaced with a significant number that
were compatible with the higher blend. Likewise, time would
be required for the ethanol supply chain to grow to where it
has proven that it could reliably produce the ethanol volumes
needed to supply the higher blend level. Meanwhile,
specifications for the new fuel (most critically the ethanol
content range and minimum octane rating) would need to be
agreed upon to ensure that fuel suppliers and vehicle
manufacturers are working around a common set of fuel
properties. Once those first transition requirements were met,
then roll-out of the new fuel could begin. After sufficient
amounts of the new fuel were available in the marketplace,
automakers could start to provide advanced vehicles that were
optimized (dedicated) for the new fuel.
FFVs could play an important role in this scenario by
providing a means to bridge the fuel supply transition (i.e.,
while the ethanol industry was ramping up production, but
before the ethanol blend level could be increased in the
general gasoline pool.) The additional ethanol could be
consumed as high-level blends (e.g., E85) by the many FFVs
already present in the US and new FFVs that will be produced
Page 16 of 21
Although NG is very abundant and attractively priced at

present, it originates from fossil sources and is thus not
renewable or sustainable in the long-term. However, NG could
be replaced in the future by sustainable sources of methane
with minimal, if any, changes to the existing NG distribution
system or users. Bio-methane or biogas is produced from the
anaerobic microbial digestion of biomass, including municipal
waste, sewage, animal manure, plant residues, and also crops
(e.g., corn). Methane can also be produced directly from CO2
and H2 by methanation, using CO2 (e.g., extracted from flue
gas) and H2 obtained from water via electrolysis using
renewable electricity such as solar or wind power. Methane
from this potential production pathway is sometimes called
Wind-Methane or E-Gas [75,76,77]. Here, the term
e-methane will be used to describe methane derived using
renewable electricity sources.
CNG vehicles and the NG distribution system are fully
compatible with e-methane and post-processed bio-methane
(dried and cleaned) to meet appropriate requirements (e.g.,
German DIN 51624 standard [78]). Unlike ethanol blending in
gasoline, properly treated bio-methane and e-methane can be
blended at any concentration into NG. Therefore, CNG
vehicles could instead be considered compressed methane
vehicles and all methane production pathways can be
considered.
NG itself is considered an attractive alternative transportation
fuel capable of providing long-term energy security, as it is
available worldwide [56] with reserves that are 150 times
current annual global consumption [57] and is available at low
cost. Bio-methane and e-methane can be considered long-term
alternatives that could provide both energy security and
sustainability. For example, in the UK the main feedstocks for

bio-methane are agricultural manure and food wastes. The UK
generates 30 million dry tonnes of these waste materials
annually, capable of producing 6.3 million tonnes of oil
equivalent as methane gas, or equivalent to 16% of UK
transport fuel demand [79].
Sustainability includes many other factors, including cost,
GHG emissions, land use, water use, competition with food
supply, etc. Bio-methane is particularly attractive because
much of it today is produced from a low- or zero-value
feedstock that requires disposal (sewage, manure, plant
waste). Scaling up bio-methane production is possible but will
be more challenging because it would require more valuable
feedstocks (energy crops) that have other potential uses,
including other biofuel options. E-methane is an intriguing
possible source of renewable methane, though its production
at scale and economic viability are yet to be demonstrated. If
successful, there would also be demand for renewable
methane in other sectors that currently use NG including
electricity generation, industrial uses, and heating.
NG presently offers a 2030% TTW CO2 reduction potential
due to its favorable hydrogen-to-carbon ratio and efficiency
potential because of its high knock resistance [45,80].
According to a joint CONCAWE-EUCAR-JRC study [81],
fossil CNG used in CNGVs today offer a 24% well-to-wheel
(WTW) CO2 emissions reduction versus gasoline. With biomethane or e-methane close to 100% fossil CO2 reduction
would be possible.
As already discussed, NG is consistently less expensive than
gasoline and diesel (on an energy basis) in many countries.
NG is likely able to pay back vehicle on-costs in a reasonable
time. With an expanding infrastructure and rising vehicle
numbers, the positive effects of scale will drive costs down.
NG could therefore support the development of a methane
infrastructure and CNGV fleet. According to recent IEA fuel
price scenarios [73], bio-methane is assumed to become cost
competitive with oil in the next few decades however the
potential scale of future production and the fraction that would
be used in the transportation sector are unclear.
Renewable wind and solar power are likely to grow
significantly in some countries (e.g., in Germany where
nuclear power generation has come into disfavor). The
intermittency of wind and solar power has been cited as a
limitation, and suggests the need for large-scale energy storage
to fully utilize their potential. One proposal is to convert
excess solar and wind power to e-methane with storage in
vast, already existing, underground cavities [75]. With
growing renewable power sources, e-methane could become
available in scale in the long-term.
As of 2010, there were 12.7 million CNGVs in the world and
more than 18,000 CNG filling stations, both growing at a rate
of 10% or more annually. Dedicated-fuel CNGVs with better
efficiency (NG-optimized engines and less vehicle weight),
reduced package restrictions (elimination of gasoline fuel
Page 17 of 21
system), and lower on-cost would be enabled with the

availability of a mature infrastructure. The factors above
suggest that methane may become a more important
automotive fuel in the future.
A successful introduction strategy would initially require a
stable, reliable, predictable and sufficient fuel cost benefit for
NG harmonized ideally over continents. In the long-term more
sustainable methane (bio-methane, e-methane, or other) may
become available and could be blended with fossil methane.
4. SUMMARY/CONCLUSIONS
Based on the analysis of several case studies of alternative fuel
introductions, the basic requirements for alternative fuels,
vehicles, and the fueling infrastructure are postulated that are
necessary for successful market implementation.
To successfully introduce a new fuel into the market it is
critical that both the fuel and vehicle technology are
affordable and cost-competitive. The consumer must have a
very good chance of recovering the vehicle on-cost and to
realize considerable operational cost savings over the first few
years. The fuel cost benefit must be reliable and stable in the
long term, even if fuel demand rises with the successful
development of that market. Thus sufficient feedstock and fuel
production capacity must be available for large-scale
introduction.
A precondition for a successful new fuel introduction is the
existence of backwards-compatible vehicles in the market,
which can use a cost-efficient fuel without any vehicle
changes. If backwards compatibility is not possible, alternative
fuels can also be blended into existing fossil fuels at a blend
content that is compatible with the existing vehicle fleet
which, depending on the fuel, can be very limited. If even a
small fraction of the existing vehicle stock is not compatible
with the alternative fuel blend, then vehicle compatibility lists
must be issued and protection grade fuels must be retained for
incompatible vehicles, which can be a politically very
challenging process.
An affordable distribution infrastructure is another
important factor for the development of an alternative fuel
market. High infrastructure investment costs will lead to slow
growth in fuel station numbers. A sufficient supply network is
required for consumer acceptance of the fuel; otherwise the
use will be limited to captive vehicle fleets at best. If the
supply network is insufficient, then vehicle bi-fuel or flex-fuel
capability can compensate. In that case, the vehicle on-cost
needs to be relatively low because of the limited refuelling
opportunities that provide the offsetting lower operational
(fuel) costs.
The vehicle capability for two fuels (bi-fuel vehicle, monofuel vehicle, or FFV) is critical to the success of any
alternative fuel that is not backwards-compatible to the vehicle
fleet and/or supported by a sufficient supply infrastructure.

This capability also provides consumers with confidence in
buying the alternative fuel vehicle if the fuel cost benefits are
inconsistent. To bring bi-fuel capable vehicles into the market,
OEM-controlled retrofit kits could be of significantly help,
because they allow the conversion of used vehicles already in
the fleet (particularly attractive in cost-sensitive markets).
However, the use of retrofit kits should be carefully
considered in terms of their impact on vehicle durability and
vehicle quality.
Incentives for sustainable alternative fuels are initially
required for fuels with higher production and distribution costs
than gasoline/diesel to develop the market. Long-term
alternative fuel strategies should have an endpoint that
includes a non-fossil, sustainable, and cost-effective
renewable fuel, recognizing that the cost competitiveness of
the alternative fuel could be influenced by differing policy
treatment to account for differences in perceived external
costs.
In the long term (2050), two sustainable gasoline substitutes
have been suggested as becoming less expensive (pre-tax)
than gasoline: renewable methane and ethanol. While ethanol
can be used as a blend component with gasoline in nearly any
concentration in FFVs, any kind of methane (NG, biomethane, e-methane) can be used in CNGVs.
For the development of future ethanol markets, the availability
of FFVs in the vehicle fleet (already significant in the US) is
important. To achieve the maximum benefit from greater
future ethanol availability, the ethanol could eventually be
blended across the entire gasoline pool and would enable
increases in minimum octane ratings. If not, considerable
efficiency and fuel savings potential may be lost. The
existence of a large FFV fleet provides time to transition the
rest of the fleet to compatibility with those future blends.
For the development of future compressed methane
vehicles, inexpensive CNG can be used to stimulate
infrastructure development and market penetration with bifuel and mono-fuel CNGVs. Once the infrastructure is built up
and a sufficient number of CNGVs are in the fleet, future
availability of cost-competitive bio-methane and e-methane
could be readily transitioned into the fuel market.
6. ABBREVIATIONS
BEV
CNG
CNGV
ECU
ETBE
FAME
FFV
GHG
LPG
LPGV
MTBE
NG
OEM
TTW
WTT
WTW
Battery Electric Vehicle

Compressed Natural Gas
CNG Vehicle
Engine Control Unit
Ethyl Tertiary Butyl Ether
Fatty Acid Methyl Esters (Biodiesel)
Flexible Fuel Vehicle
Greenhouse Gas
LPG Vehicle
Methyl Tertiary Butyl Ether
Natural Gas
Original Equipment Manufacturer
Tank to Wheel
Well to Tank
Well to Wheel
7. CONTACT INFORMATION
Ulrich Kramer
Ford Motor Company
Research and Advanced Engineering Europe
Powertrain Research & Advanced
Spessart Strasse, D-ME/5-B8
D-50725 Cologne, Germany
James E. Anderson
Ford Motor Company
Research and Advanced Engineering
Systems Analytics and Environmental Sciences Department,
PO Box 2053, Mail Drop RIC-2122
Dearborn, MI 48121
8. REFERENCES
1
5. ACKNOWLEDGMENTS
The authors gratefully acknowledge the assistance of our
colleagues in preparing this paper: Leandro Benvenutti, Tim
Wallington, and Wulf-Peter-Schmidt.
Kauling, D.C. US Light Vehicle Development Gaps

Current and Future Technology, Natural Gas Vehicle
Technology Forum, San Francisco, California, October 25-26,
2011.
2
Agncia Nacional do Petrleo, Gs Natural e

Biocombustveis (ANP, Brazil National Agency of
Petroleum). Resoluo ANP No. 7, 2011.
3
Nass, L.L., Pereira, P.A.A. Ellis, D. (2007) Biofuels in

Brazil: An Overview, Crop Science 47: 2228-2237.
Page 18 of 21
Soccol, C.R., Vandenberghe, L.P.S., Costa, B.,

Woiciechowski, A.L., de Carvalho, J.C., Medeiros, A.B.P.,
Francisco, A.M., Bonomi, L.J. (2005) Brazilian biofuel
program: An overview, Journal of Scientific & Industrial
Research 64: 897-904.
18
Energy Information Agency (EIA). Monthly Energy

Review, US DOE, DOE/EIA-0035(2012/02), February 2012.
19
de Freitas, L.C., Kaneko, S. (2011) Ethanol demand in

Brazil: Regional approach, Energy Policy 39(5): 2289-2298.
Alternative Fuels and Advanced Vehicles Data Center.

Alternative Fueling Station Total Counts by State and Fuel
Type, US DOE,
http://www.afdc.energy.gov/afdc/fuels/stations_counts.html,
accessed March 16, 2012.
20
Assoiao Nacional dos Fabricantes de Veiculos

Automotores (ANFAVEA) (2010) Brazilian Automotive
Industry Yearbook 2010.
7
Empresa de Pesquisa Energtica (EPE) (2011) Captulo 3

(Consumo de Energia por Setor) 1970-2010, Ministrio de
Minas e Energia,
https://ben.epe.gov.br/BENSeriesCompletas.aspx, accessed
April 4, 2012.
Alternative Fuels and Advanced Vehicles Data Center.

U.S. Alternative Fueling Station Count, US DOE,
http://www.afdc.energy.gov/afdc/data/docs/alt_fueling_station
s_fuel.xls, accessed March 16, 2012.
21
ASTM International, Standard Specification for Fuel

Ethanol (Ed70-Ed85) for Automotive Spark-Ignition
Engines, ASTM D 5798-10a, Rev. 2010.
22
Nichols, R.J. (2003) The Methanol Story: A Sustainable

Fuel for the Future, Journal of Scientific & Industrial
Research 62: 97-105.
Anderson, J.E., Baker, R.E., Wallington, T.J., Ginder, J.M.

(2009) Implications of the 2007 Energy Independence and
Security Act for the US Light-Duty Vehicle Fleet, SAE
Technical Paper 2009-01-2770.
Alternative Fuels and Advanced Vehicles Data Center. Key

Federal Legislation, US Department of Energy (DOE),
http://www.afdc.energy.gov/afdc/laws/key_legislation#top,
23
Alternative Fuels & Advanced Vehicles Data Center.

Clean Cities Fuel Survey Report, May 2000January 2012
reports, US DOE,
http://www.afdc.energy.gov/afdc/price_report.html
10
24
Kramer, U., Ford Werke GmbH, Ford FlexifuelFahrzeuge, CTI-Conference Bio Fuels, Munich, Germany,
June 2627, 2007.
11
Kramer, U., Ford Werke GmbH, Ford Flexi-Fuel-Fahrzeug

konomische CO2-Reduzierung, VDI-GVC- Specialist
Committee Meeting High Temperature Technology and
Bio Fuels, Wrzburg, Germany, March 2006.
12
Standardiserings-Kommissionen i Sverige (SS),

Automotive fuels Ethanol E85 Requirements and Test
Methods, SS 155480, July 2006.
13
Deutsches Institut fr Normung e.V. (DIN), Automotive

fuels Ethanol Fuel Requirements and Test Methods,
DIN 51625, Aug. 2008.
14
ASTM International, Standard Specification for Ethanol

Fuel Blends for Flexible-Fuel Automotive Spark-Ignition
Engines, ASTM D 5798-11, Rev. 2011.
15
Anderson, J.E., DiCicco, D.M., Ginder, J.M., Kramer, U.,

Leone, T., Raney-Pablo, H.E., Wallington, T.J. (2012) High
Octane Number Ethanol-Gasoline Blends: Quantifying the
Potential Benefits in the United States, Fuel in press,
doi:10.1016/j.fuel.2012.03.017.
16
R. L. Polk & Co. database, accessed March 25, 2012.
17
Davis, S.C., Diegel, S.W., Boundy, R.G. Transportation

Energy Data Book: Edition 30, Center for Transportation
Analysis, Oak Ridge National Laboratory, Report ORNL6986, June 2011.
Page 19 of 21
McPhail, L., Westcott, P., Lutman, H. The Renewable

Identification Number System and U.S. Biofuel Mandates,
Economic Research Service, US Department of Agriculture,
Report BIO-03, November 2011.
25
Birath, K. Joint Procurement of Clean Vehicles, Policy

Notes, NICHES Consortium, undated, http://www.nichestransport.org/fileadmin/archive/Deliverables/D4.3b_5.8_b_Pol
icyNotes/14686_pn11_joint_procurement_ok_low.pdf
26
Kroh, E. (2008) FFVs flourish in Sweden, Ethanol

Producer Magazine, August 2008.
27
BioEthanol for Sustainable Transport (BEST), Promoting
Clean Cars: Case Study of Stockholm and Sweden, BEST
Deliverable No 5.12, Stockholm, February 2009.
28
Campestrini, M., Mock, P. European Vehicle Market

Statistics, 2011 Edition, International Council on Clean
Transportation, Washington, DC, 2011.
29
European Automobile Manufacturers Association (ACEA).

Overview of CO2 based motor vehicle taxes in the EU,
April 28, 2009,
http://www.acea.be/images/uploads/files/20090428_CO2_taxa
tion_overview.pdf.
30
Svenska Petroleum och Biodrivmedel Institutet. Statistik,

http://spbi.se/statistik, accessed April 4, 2012.
31
Thordsen, F., Buckle, C. Fachartikel: Emissionen und

Kraftstoffe, Kraftfahrt-Bundesamt, Germany, March 15,
2011.
http://www.kbashop.de/wcsstore/KBA/pdf/Fachartikel_Emissi
onen_Kraftstoffe.pdf
32
Fachagentur Nachwachsende Rohstoffe e.V. (FNR).

Bioethanol Glzow, Germany, http://www.biokraftstoffe.info/kraftstoffe/bioethanol/, accessed 25.March
2012
33
Minerallwirtschaftsverband e.V. (Association of the

German Petroleum Industry), Statistiken Preise, Berlin,
http://www.mwv.de/index.php/daten/statistikenpreise/,
34
Fachagentur Nachwachsende Rohstoffe e.V. (FNR), Daten

und Fakten ber Biokraftstoffe, Glzow, Germany,
http://www.bio-kraftstoffe.info/daten-und-fakten/, accessed
March 25, 2012.
http://www.bmu.de/files/na/application/pdf/fluessiggas_kohle
nstoffdixydemissionen.pdf
46
International Fuel Quality Center (IFQC), Special Report

EU: LPG and CNG Developments, February 7, 2011,
www.ifqc.org
47
Organization of Economic Co-Operation and Development

(OECD), Economic Assessment of Biofuel Support
Policies, Directorate for Trade and Agriculture, 2008.
48
Kraftfahrt Bundesamt (KBA), Statistiken; Fahrzeuge,

http://www.kba.de/cln_030/nn_124584/DE/Statistik/Fahrzeug
e/fahrzeuge__node.html, accessed April 1, 2012.
49
35
Bohn, K., CNG-Technik GmbH (Ford Werke GmbH),
36
CNG und/oder LPG?, Conference Gas Poweres Vehicles,

FKFS, Stuttgart, October 13, 2009.
Bayerisch Landwirtschaftliches Wochenblatt , April 23,

2009, www.wochenblatt-dlv.de
Ethanol-tanken.com, Tankstellen, http://www.ethanoltanken.com, accessed March 25, 2012.
37
Aral Aktiengesellschaft, Grafiken,

http://www.aral.de/aral/browsebytheme.do?categoryId=40011
41&contentId=51281, accessed April 9, 2012.
38
Baumeister, S., Ford Werke GmbH, Herausforderung beim

Aufbau einer Ethanol-Tankstelleninfrastruktur in
Deutschland, VDI Fachausschuss Sitzung Biokraftstoffe,
Wrzburg, Germany, March 2007.
50
Prins Alternative Fuel Systems, VSI LPG Vapour

Sequential Injection, product information sheet, Eindhoven,
Netherlands, 2010.
51
LPG & CNG Tankstellenverzeichnis, www.gastankstellen.de, accessed March 2012.

52
Ford Motor Company, Ford Focus Preisliste A, Germany,

February 21, 2012, http://www.ford.de/PkwModelle/FordFocus/Broschueren_Preise_Daten
39

Fuels Requirements and Test Methods Part 1: Petrol E10,
E DIN 51626-1, April 2009.
53
40
54
Verband der Automobilindustrie (VDA, Association of

German Automobile Manufacturers), personal
communication, 2011.
41
Bockey, D. (2006) Biodiesel und pflanzliche le als

Kraftstoffe aus der Nische in den Kraftstoffmarkt; Stand und
Entwicklungsperspektiven, Technikfolgenabschtzung Theorie und Praxis, 15(1):10-15.
42
Bockey, D., von Schenk, W. Status Report Biodiesel:

Biodiesel Production and Marketing in Germany 2005,
Union zur Frderung von Oel- und Proteinpflanzen e. V.
(UFOP), June 2005.
43
Verband der Deutsche Biokraftstoffindustrie (VDB),

Informationen: Biokrafstoffe in Deutschland, January 2012,
http://www.energiewende-sta.de/wpcontent/uploads/2012/02/Factsheet_Biokraftstoffe_VDB.pdf
44
Comite Europeen de Normalisation, Automotive Fuels

Diesel Requirements and Test Methods, EN 590, February
2010.
45
Wiedemann, H. K. Carbondioxyd emissions of passenger

cars powered by gas with national type approval or EC Whole
Vehicle Type approval, TV Saarland (Germany), March
2005,
Page 20 of 21
Ford Motor Company, Ford Fiesta Preisliste A, Germany,

February 21, 2012, http://www.ford.de/PkwModelle/AlternativeAntriebe/Autogas_LPG
Mantey, C.A., Mezzarano, A., Kolkemo, A. (2009)
Exhaust Valve & Valve Seat Insert Development for an
Industrial LPG Application, SAE Technical Paper 2009-011602.
55
Sato, K., Midorikawa, T., Takahashi, T., Oshige, H. (2000)

Development of Valve Seat Material for Gas-Fueled
Engines, SAE Technical Paper 2000-01-0911.
56
International Energy Agency (IEA), World Energy

Outlook 2011 Special Report: Are We Entering a Golden Age
of Gas?, Paris, France, 2011.
57
Massachusetts Institute of Technology Energy Initiative,

The Future of Natural Gas: An Interdisciplinary MIT Study,
Interim Report, 2010.
58
Natural Gas Vehicle Association (NGVA) Europe, NGV

Market Growth in Europe (1995-2010),
http://www.ngvaeurope.eu/ngv-market-growth-in-europe1995-2010, accessed March 2012.
59

Fuels Compressed Natural Gas Requirements and Test
Methods, DIN 51624, February 2008.
60
Prins Alternative Fuel Systems, VSI CNG Vapour

Sequential Injection, product information sheet, Eindhoven,
Netherlands, 2010.
61
International Association for Natural Gas Vehicles (NGV

Global), Natural Gas Vehicle Statistics,
http://www.iangv.org/tools-resources/statistics.html, accessed
March 2012.
75
62
76
Special Report: Asia Pacific: Regional Overview of

NGVs; IFQC; www.ifqc.org, January 7, 2012
63
Natural & bio Gas Vehicle Association (NGVA Europe),

NGV Market Growth in Italy (1973-2008),
http://www.ngvaeurope.eu/ngv-market-growth-in-italy-19732008; accessed March 2012.
64
Natural & bio Gas Vehicle Association (NGVA Europe),

Italy, http://www.ngvaeurope.eu/italy, accessed March
2012.
65
Vest Marketing GmbH, Besteuerung von

Erdgasfahrzeugen, Osnabrck, Germany,
http://www.gas24.de/cms/26-0-steuern-erdgas.html, accessed
March 16, 2012.
66
Natural Gas Vehicle Association (NGVA) Europe,

European NGV Statistics,
http://www.ngvaeurope.eu/european-ngv-statistics, accessed
April 4, 2012.
67
Ford Werke GmbH, Germany, Price List Ford C-MAX

CNG, August 4, 2008.
68
Vest Marketing GmbH, Kosten einer Nachrstung,

Osnabrck, Germany, http://www.gas24.de/cms/91-0-kosteneiner-nachruestung.html, accessed March 16, 2012.
69
Auto Test Magazine, Mit neuer Energie in die Zukunft,

August 2010.
70
Hautzinger, H., Stock, W., Mayer, K., Schmidt, J.,

Heidemann, D. (2005) Fahrleistungserhebung 2002 Begleitung und Auswertung, Band 1: Inlnderfahrleistung
2002; BASt Report V 120; Institut fr angewandte Verkehrsund Tourismusforschung (IVT), Heilbronn/Mannheim,
Germany.
71
German Society for Sustainable Biogas and Bioenergy

Utilization (GERBIO), Nationaler Bericht ber die Situation
des Erdgastankstellennetzes Deutschland, Report No.
WP3.1, Kirchberg, Germany, undated.
72
Jauss, A., GM Europe Engineering, Exkurs CNG &

Biomethane - State of the art and outlook; 2nd CTIConference Bio-Fuels for Combustion Engines, Mannheim,
Germany, 2008.
73
International Energy Agency (IEA), Technology Roadmap:

Biofuels for Transport, Paris, France, 2011.
74
Anderson, J.E., Kramer, U., Mueller, S.A., Wallington, T.J.

(2010) Octane Numbers of Ethanol- and Methanol-Gasoline
Blends Estimated from Molar Concentrations, Energy &
Fuels, 24:6576-6585.
Page 21 of 21
Rieke, S., SolarFuelGmbH, The storage of renewable

electricity in gas tanks as fuela new approach for sustainable
use of NGV technology, 5th Conference on Gas Powered
Vehicles, Berlin, October 1, 2010.
Specht, M., Zuberbhler, U., ZSWCenter for Solar Energy
and Hydrogen Research, Storing Renewable Energy in the
Natural Gas GridMethane via Power-to-Gas (P2G): A
Renewable Fuel for Mobility, 6th Conference on Gas
Powered Vehicles, Stuttgart, Germany, October 26-27, 2011.
77
Otten, R., Audi, Audi e-gas-Projekt; 6th Conference on

Gas Powered Vehicles, Stuttgart, Germany, October 26-27,
2011.
78

Fuels Compressed Natural Gas Requirements and Test
Methods, DIN 51624, February 2008.
79
National Society for Clean Air and Environmental

Protection (NSCA), Biogas as a road transport fuel An
assessment of the potential role of biogas as a renewable
transport fuel, England, June 2006.
80
Douailler, B., Ravet, F., Delpech, V., Soleri, D., Reveille,

B., Kumar, R. (2011) Direct Injection of CNG on High
Compression Ratio Spark Ignition Engine: Numerical and
Experimental Investigation, SAE Technical Paper
2011-01-0923.
81
CONCAWE, European Council for Automotive R&D

(EUCAR), Joint Research Centre (JRC), Well-to-Wheels
Analysis of Future Automotive Fuels and Powertrains in the
European Context: WELL-to-WHEELS Report, Version 2c,
March 2007.
Regulations and Incentives for

Alternative Fuels and Vehicles
Elisheba Spiller
Contents
1. Introduction ......................................................................................................................... 2
2. Alternative Fuels ................................................................................................................. 4
2.1 Ethanol and Cellulosic Biomass ................................................................................... 4
2.1.a US Regulatory History of Ethanol ........................................................................4
2.1.c Brazil .....................................................................................................................7
2.1.d Analysis of Environmental and Economic Impacts of Biofuels ...........................8
2.2 Alternative Refueling Stations ...................................................................................... 9
2.2.a US Legislation .....................................................................................................10
2.2.b Natural Gas and Blue Corridors ..........................................................................12
3. Alternative Vehicles .......................................................................................................... 12
3.1 Corporate Average Fuel Economy and Greenhouse Gas Standards ........................... 14
3.1.a Massachusetts vs. EPA........................................................................................15
3.1.b CAs Attempt to Regulate GHG Emissions from Mobile Sources.....................15
3.1.c The National Program .........................................................................................16
3.1.d Light-Duty Vehicles............................................................................................16
3.1.e New CAFE Standards and Compliance ..............................................................17
3.1.f Heavy-Duty Vehicles ..........................................................................................19
3.2 Market Stimulation ..................................................................................................... 20
3.2.a Mandated Adoption of Alternative Vehicles ......................................................20
3.2.b California ZEV Mandates ...................................................................................21
3.2.c Consumer Incentives ...........................................................................................22
3.2.d Demonstrations and Voluntary Programs ...........................................................24
3.3 Technological Advancements and R&D .................................................................... 25
4. Conclusion ......................................................................................................................... 26
Regulations and Incentives for

Alternative Fuels and Vehicles
Elisheba Spiller
1. Introduction
Transportation accounts for approximately one third of all US CO2 emissions and raises serious energy
security issues. Increased demand for transportation fuels and low penetration of alternative fuel vehicles
has exacerbated these policy concerns and their costs.
In addition, transportation and gasoline consumption have externalities -- spillover effects that are not
incorporated in fuel and vehicles costs and prices. These include, for example, lost productivity associated
with traffic congestion, accidents, and increased smog and pollution levels. The externalities associated
with private transportation, while appreciated for decades, have never been successfully incorporated into
the costs of petroleum based transportation fuels and vehicles.
The most efficient way to address the externalities associated with driving and gasoline consumption is
through economic instruments. Unfortunately, no single policy, tax or incentive offers an individually
optimal solution that would simultaneously diminish driving, decrease gasoline consumption, and
enhance energy security generally agreed upon policy goals. For example, a gasoline tax may change
driving patterns or it may cause individuals to simply shift to more fuel-efficient vehicles without the
alleviation of congestion. On the other hand, a congestion fee could decrease peak-time usage of
highways, though it would not necessarily affect overall gasoline consumption or the choice of fuel
efficiency. Therefore, grouping taxes or incentives (such as a gasoline tax combined with congestion
taxes) could help to alleviate multiple externalities associated with driving and gasoline consumption.
Furthermore, to maximize intended outcomes, a tax would need to vary over time and space depending on
congestion levels.
Security, equity and administrative concerns tend to impose political constraints on policy makers,
limiting their willingness to price externalities through taxes. Congestion taxes would, for example,
negatively impact low income drivers, while emissions or driving taxes are infeasible absent dashboard
technologies to measure emissions or driving patterns. Economists estimate an optimal gasoline tax (one
that addresses the multiple externalities associated with driving) to be approximately $1/gallon, more than
double the current average federal and state gasoline tax on gasoline (see Parry and Small [2005] 1
Williams [2006] 2, and West and Williams [2007] 3). Yet, discussions of any tax sufficient to alter driving
behavior are purely academic given the current anti-tax rhetoric and reticence of many policy makers in
Washington.
Policymakers have looked to alternative fueled vehicles that run on electricity, biofuels or natural gas as a
means of meeting these policy objectives, and as an alternative to taxes. Yet because the costs of these
alternatives including financial, infrastructure, adjustments and performance tradeoffs remain high,
local and federal government agencies have taken a series of steps to reach key policy goals through the
Post-Doctoral Researcher at Resources for the Future; Contact: spiller@rff.org, 202-328-5147.
implementation of a range of regulations and incentives for alternative transportation fuels (ATFs),
alternative fueled vehicles (AFVs), and electric vehicles (EVs).
Many of these incentives and regulations focus on research and development (R&D), demonstration and
deployment, and infrastructure development, and most (though not all) are designed to reduce the costs of
the alternatives. However, recent presidential administrations have implemented policies that target
certain technologies over others. For example, the Bush Administrations policies focused on promoting
fuel cell and ethanol (E85) vehicles, while the Obama Administration has placed a strong emphasis on
electric vehicles. This pattern of promoting specific technologies has been problematic as it does not
allow the more commercially and technologically viable alternative vehicles to emerge as a market
choice. Critics of picking winners argue that policies like a carbon or gasoline tax would be more
technology neutral and allow for the most efficient technologies to emerge. 4,5
In spite of the investment of billions of federal dollars over the past 30 years, market penetration of
alternative fuel vehicles remains quite low; even with the enormous federal expenditures in the promotion
of these vehicles, they accounted for less than 6% of the overall vehicle stock in 2011.6 As such, it seems
appropriate to take a step back and analyze our decision to incentivize the adoption of alternative vehicles
and fuels.
First, have the mandates, incentives and regulations focused on alternative fuels and vehicles
been successful in diminishing GHGs and other pollutants, decreasing gasoline consumption, or
increasing energy security? While it is important to analyze the effectiveness of these policies in
mainstreaming AFVs, their ability to do so matters only in so much as the underlying objectives
(such as improving environmental outcomes) are achieved. Can we, for example, be energy
independent by mandating alternative fuels and vehicles without also encouraging conservation
and decreased vehicle miles travelled?
Second, are these regulations cost-effective, and what is the cost to the government of
implementing these incentives and regulations? Given that the current government deficit exceeds
$15 trillion, implementing costly incentives and regulations may crowd out other policies and
cause significant welfare impacts.
Third, do these mandates and incentives for the adoption and utilization of expensive AFVs
impose less of a social and economic cost on society than externality taxes?
Fourth, have these policies resulted in substantial technological development?
Finally, are these policies able to be implemented given the current technological constraints we
face? Although these policies may have the goal of improving technology, in the short run the
objectives detailed in legislation might not be met, thus it may be necessary to adjust our
expectations of what government intervention can quickly achieve.
This paper discusses the incentives and regulations for alternative fuels and vehicles in more detail.
2. Alternative Fuels
2.1 Ethanol and Cellulosic Biomass
Ethanol is the most widely adopted alternative fuel in the US. Since 2005, most gasoline is mixed with up
to 10% ethanol, as a result of the federal Renewable Fuel Standard (RFS) and gasoline content
regulations. Given its relatively low cost of production (compared to other renewable fuels), the most
commonly utilized biofuel in the US is corn based ethanol. This and other incentives have boosted
national biofuel production. Cellulosic biomass (CB) and sugar-based ethanols, due to technology and
other limitations, have not been utilized to the same extent as corn ethanol in spite of concerns over it
GHG balance, associated land use and fuel for food issues associated with its production. Furthermore,
the US government has established strong barriers to the importation of sugar ethanol in spite of its
superior environmental performance. CB also has better environmental performance than corn ethanol,
yet it is technologically immature and needs additional research to improve its affordability. 7
2.1.a US Regulatory History of Ethanol

Ethanol has been used as a fuel in the US since before the Civil War. In 1862, however, the US
government placed a $2.08/gallon alcohol tax to help fund the war (equivalent to $35/gallon in 2007
dollars) and no exception was made for ethanol. The result of this tax was the replacement of ethanol by
kerosene as the fuel of choice. Fifty years later, the tax was lifted, improving market opportunities for
ethanol. In fact, the first flexible fuel vehicle (FFV) to run on ethanol, gasoline and kerosene was
produced by Ford in 1908.
Over the years, production and utilization of ethanol increased substantially, until the end of WWII, when
the increased energy demand for war materials was no longer needed. Ethanol was not viably produced
again until the late 1970s, when concerns for energy security and oil dependence after the Arab oil
embargos spurred the interest in alternative transportation fuels and the passage of the Energy Tax Act of
1978.8 This statute helped boost production of ethanol by providing a tax credit for the portion of gasoline
that was blended with at least 10% ethanol. At the time, the excise tax for gasoline was 4 cents/gallon,
which amounted to a 40 cents/gallon tax credit for every gallon of ethanol blended into gasoline. This
subsidy was increased over time: in 1983, the Surface Transportation Assistance Act increased the
subsidy to 50 cents/gallon; and in 1984, the Tax Reform Act increased the subsidy to 60 cents/gallon.
However, in 1990, the Omnibus Budget Reconciliation Act decreased the subsidy to 54 cents/gallon, and
in 1998 the Transportation Equity Act for the 21st Century phased the subsidy down to 51 cents/gallon by
2005. 9,10 The subsidy was further decreased to 45 cents/gallon in 2004, with the American Jobs Creation
Act, which also changed the recipient of the credit from the producer to the blender and was due to phase
out by December 2011. 11
By 1980, it was apparent that this incentive was largely going to ethanol importers. In fact, the
government surveyed ethanol production in the US and found only 10 producing ethanol facilities.12 The
response was the Energy Security Act of 1980, which imposed an import tariff on ethanol of 40
cents/gallon, effectively offsetting the excise tax credit for those importing ethanol. The Omnibus
Reconciliation Act of 1980 also applied a 2.5% ad valorem tariff to ethanol imports from most
countries.13 These import tariffs helped to price Brazilian sugarcane imports out of the US market and
helped boost domestic production of ethanol.
Other incentives for the production of ethanol included 10 cents/gallon to small producers (those with
production capacities below 60 million gallons per year), established in 1990 and reissued in 2004
through the Volumetric Ethanol Excise Tax Credit (VEETC part of the American Jobs Creation Act),
which also provided a $1.01/gallon credit to producers of cellulosic ethanol.14 The cost of producing
cellulosic ethanol has decreased over the years, but current costs are around $2-$3/gallon, as seen in
Figure 1. Though these costs are hard to estimate (and may vary significantly across producers), the trend
has been of decreasing costs over the last decade. The VEETC expired at the end of 2011 and cost the
government billions of dollars in subsidies over the past 30 years. 15
Figure 1. Cellulosic Ethanol Production Costs over Time
Cellulosic ethanol
production costs
Source: Novozymes, 2010. Taken from Green Car Congress New Novozymes Enzymes for Cellulosic Ethanol
Enable Production Cost Below US$2 Per Gallon
http://www.greencarcongress.com/2010/02/cellicctec2-20100216.html
Table 1. Ethanol Tax Policies

Legislation
1862 Internal Revenue Act
1906 Free Alcohol bill
1978 Energy Tax Act
1980 Energy Security Act
1980 Omnibus Budget Reconciliation Act
1983 Surface Transportation Assistance Act
1984 Tax Reform Act
1990 Omnibus Budget Reconciliation Act
Tax/Credit
$2.08/gallon tax
Repealed 1862 tax
40cents/gallon ethanol excise tax credit
Import tariff on ethanol of 40c/g
2.5% ad valorem tariff to ethanol imports
Increased ethanol excise tax credit to 50c/g
Increased credit to 60cents/gallon
Decreased credit to 54c/g, phased down to 51 in
2005; 10c/g credit to small producers
2004 American Jobs Creation Act/ Volumetric Decreased credit to blenders to 45c/g; Added
$1.01/gallon credit to producers of cellulosic
Ethanol Excise Tax Credit
ethanol
-Expired in Dec., 2011.
2.1.b Renewable Fuel Standard (2005 RFS1 and 2007 RFS2)

The Energy Policy Act of 2005 included a Renewable Fuel Standard (RFS1), which mandated a
minimum amount of alternate fuels to be blended into gasoline beginning in 2006. The mandate required
that refiners purchase a certain amount of renewable fuels (RFs) to be blended into gasoline prior to
distribution. The amount of RFs mandated by RFS1 increased yearly from 4 billion gallons in 2006 to 7.5
billion gallons by 2012.16 However, in 2007 the Energy Independence and Security Act of 2007 (EISA)
revised RFS1 and greatly expanded the mandate to 36 billion gallons of RFs by 2022 (now known as
RFS2).17 As seen in Figure 2, total gasoline consumed in the US is approximately 9 million barrels per
day,18 thus the mandate in RFS1 was equivalent to about 5% of total gasoline consumed in 2012, while
RFS2 effectively increased that percentage to 11%.
RFS1 also created an incentive for refiners to utilize cellulosic biomass, providing them with 2.5
renewable fuel credits for CB fuel, 19 meaning one gallon of CB fuel would count as 2.5 gallons of
renewable fuel. RFS1 also set a floor on the quantity of CB fuel that needed to be included in the overall
renewable fuel goal, though the actual minimum volume would depend on the amount of CB that is
projected to be available in the coming year and the projected sales of gasoline. The implementation of
the law required EPA to estimate the amount of gasoline projected to be sold and CB available, yet the
standard could not drop below the floor (as detailed in Table 2).20 While RFS1 set the floor at 250 million
gallons in 2013, RFS2 drastically increased that amount, mandating that by 2022, half of all eligible
renewable fuels must be CB.
As Figure 3 demonstrates, RFS1 and RFS2 had significant impacts on total ethanol produced. The USDA
indicates that the number of ethanol plants went from 50 in 1998 to 204 in 2010. 21
Year
2006
2007
2008
2009
2010
2011
2012
2013
2015
2020
2022
Table 2. RFS Standards

RFS1 Standard
RFS2 Standard
(Billions of Gallons)
(Billions of Gallons)
Renewable Fuels
Cellulosic
Renewable Fuels
Cellulosic
Minimum
Biomass Min.
Minimum
Biomass Min.
4.0
4.0
4.7
4.7
5.4
9.0
6.1
11.1
6.8
12.95
0.1
7.4
13.95
0.25
7.5
15.2
0.5
0.25
16.55
1.0
0.25
20.5
3.0
0.25
30.0
10.5
0.25
36.0
16.0
Figure 2. U.S. motor gasoline and diesel fuel consumption, 2000-2035 (million barrels per day)
Sources: 2010-2035 from EIA, "Annual Energy Outlook 2011," Energy Information Administration, DOE/EIA0383(2011), U.S. Department of Energy, Washington, DC, 2011.
Figure 3. Historic US Ethanol Production
Graph taken from US Department of Agriculture (2011). U.S. on Track to become Worlds Largest Ethanol
Exporter in 2011. http://www.fas.usda.gov/info/IATR/072011_Ethanol_IATR.pdf
2.1.c Brazil
Brazil is the worlds primary exporter of ethanol, and the second largest producer behind the US. Unlike
the US, however, it uses sugar and soy as feedstocks. Sugar ethanol is desirable from a GHG perspective,
decreasing GHG emissions by 78% compared to gasoline, though production techniques such as openfield burning of sugarcane leaves can dramatically diminish this benefit.22 Furthermore, land use
associated with sugarcane production in Brazil can be problematic, as it can reduce carbon capture from
trees and biodiversity when forest areas are converted to sugarcane. Walter et.al. (2011)23 take into
account sugar ethanols lifecycle and demonstrate that while the GHG benefits relative to gasoline are on
average positive, they can be (slightly) negative depending on where the deforestation takes place.
In contrast to the United States, the economy of Brazil is bio-fuels centric. In 1975, Brazil passed the
National Alcohol Program (Pro-lcool) with the goal of phasing out all gasoline based transportation
fuels.24 This Program mandated that all Brazilian vehicles run on a blend of gasoline and ethanol, with a
minimum of 10% in 1976 and increasing to 22% by 1993. In 2007, the mandate increased the minimum
percentage of ethanol to 25%, though it was dropped to 18% in 2011 given ethanol supply shortages.25
Some vehicles had to make minor adjustments to their engines to comply with the mandate at first,
though FFVs soon penetrated the market and by 2003 they comprised 90% of all vehicles purchased in
Brazil.26
Given the high level of ethanol adoption in Brazil, the question arises of whether the carbon savings from
gasoline displacement offsets the emissions due to deforestation or land use change for sugar cane
production. Lapola et.al. (2010) 27 find that the deforestation due to sugarcane and soybeans from Brazils
increased biofuels mandate would actually create by 2020 a carbon debt that would take 250 years to
pay off with gasoline. In short, the CO2 benefits from Brazils biofuel mandate are recovered only after
250 years. In sum, Brazils goal of being gasoline-free will not necessarily result in a better overall
climate outcome in the long run, given its high demand for ethanol and utilization of soy for ethanol
production. It does however satisfy Brazils energy security objectives and enables the export of a large
percentage of Brazils oil, creating higher value for its domestic energy resources.
2.1.d Analysis of Environmental and Economic Impacts of Biofuels

Concerns have been expressed about the myriad of unintended and environmental consequences
associated with the production and distribution of corn ethanol. One of these concerns is the possible
increase in food prices associated with competition for land use. However, this issue is generally
overstated. In fact, the National Academy of Sciences (NAS) estimates that although this displacement
causes food commodities, primarily corn and soy, to increase in price, a 20-40% price increase in
commodities only increases the prices of food containing these products by 1-2%.28
On the other hand, biofuel production has unintended environmental consequences both in its production
and its utilization. Depending on how the biofuels are produced, and what land-use and land cover
changes occur due to their production, the GHG benefits from gasoline displacement may be completely
offset.
Furthermore, biofuels can increase water and air pollution. Water quality can be affected by corn
production, through eutrophication or hypoxia, due to fertilizer use, decreased soil quality, and other
factors. The amount of water used to produce biofuels is also orders of magnitude greater than used in the
production of petroleum products. 29 Furthermore, biofuels emit higher quantities of other air pollutants
(such as particulate matter, ozone, and sulfur oxides) than gasoline.30 For example, 10% ethanol-gasoline
mix emits more of all air pollutants (except for carbon monoxide) than gasoline alone, while E85 emits
more acetaldehyde and formaldehyde than gasoline alone (though emits lower levels of NOx and other air
pollutants).31 Yang et.al. (2012) 32 find that gasoline blended with 85% corn ethanol results in a 6-108%
greater environmental impact than gasoline alone (on average 23%), when taking into account full
lifecycle impacts on GHG emissions, water quality, and 10 other environmental factors.
In addition to the environmental effects of biofuels production and use, the policies regarding biofuels,
such as RFS, come with significant fiscal impacts and social costs. The Congressional Budget Office
found in 2010 that the costs of the standard ranged from $1.78/gallon of corn ethanol to $3/gallon of CB,
and that the implicit cost per ton of CO2 reduced is $750/metric ton for ethanol and $275/ton for CB. 33
A major motivation for the renewable fuel standard was that it would create a technology pull sufficient
to incentivize the development of the basic technologies for CB, as well as attract the venture capital
needed for additional development and commercialization. Unfortunately, there has been little R&D in
this area, and the technological advances have not been sufficient. The National Academy of Science
conducted a report in 2011 to analyze the impact of RFS2 on the economy and environment. The outlook
was dismal: the analysis suggests that the goals set by RFS2 of 16 billion gallons of cellulosic biomass
cannot be met absent some major technological innovation and policy changes. NAS also concluded that
there is insufficient commercially viable refinery capacity required to produce the mandated amount of
CB biofuel.34
This is unfortunate: Federal yearly support of $100-$400 million in grants from 2006 to 2008 to help CB
producers build production facilities35 appear to have had minimal impact on the progress of technology
development and construction of production facilities. A perverse effect of a mandate that cannot
currently be achieved may be reliance on foreign sources for biofuels, thus diminishing the energy
security benefits that could have been achieved under the RFS. Also, biofuels are not cost competitive
with gasoline, even at todays high gasoline prices. NAS analysis indicates that biofuels will only be a
cost-effective alternative to gasoline under extreme technological innovation in refineries and oil prices of
at least $191/barrel.36
On the other hand, the RFS may help overcome the possible increase in gasoline prices due to the
removal of the ethanol subsidies. While many have speculated that the elimination of ethanol subsidies
would lead to higher pump prices, the EPA indicates that the RFS alone could result in a decrease in
gasoline prices of approximately 2.5 cents/gallon.37 It is likely that a portion of the subsidy was passed on
to the consumer. Absent any major changes in production, removal of the tax credit, while retaining the
RFS mandate, would likely result in an increase of 2 cents/gallon at the pump. On the other hand, if
producers were not passing along the subsidies, then there will be no negative impact on the consumer
from removal of the tax credit. In any case, the amount of ethanol mixed into gasoline will not change due
to the removal of the credit (given the RFS mandate), and a 2 cent/gallon increase would only result in a
$10/year increase for the average consumer (under 10,000 yearly VMT and a 20mpg vehicle). Thus, the
removal of the credit will most likely have more of an impact on the blenders profits than total gasoline
demanded, and may also make it more costly for the producers to meet the RFS standard. However, the
RFS could help to alleviate the negative economic impacts from the removal of the ethanol subsidies.
2.2 Alternative Refueling Stations

One of the main barriers to adoption of alternate fueled vehicles is the current lack of refueling stations.38
Consumers will avoid vehicles if they are not able to refuel or recharge them easily. In the classic chicken
and egg conundrum faced by alternative fuels and vehicles, producers will also be less likely to install
fueling stations if there is insufficient demand for these alternative fuels. Thus, the federal government
has promoted policies to incentivize the construction of alternative refueling stations and associated
infrastructure in order to break this unfortunate cycle.
2.2.a US Legislation
The Energy Policy Act of 2005, besides creating RFS1, also sought to promote the installation of
alternative fuel refueling stations. This statute provided a tax credit which would cover 30% of the cost of
installing an alternative refueling station (of which 85% of the volume has to consist of ethanol, CNG,
LNG, liquefied petroleum gas or hydrogen; or 20% of biodiesel; electric charging stations were not
included), up to $30,000.39 This credit was only given to commercial refueling stations - no refueling
station on personal property could receive this credit. The 2009 stimulus bill increased the amount of the
credit up to 50% of the cost, with a maximum of $50,000.40 In 2010, the Tax Relief Act extended the
alternative fuel vehicle refueling property credit through the end of 2011, but decreased the percentage
and total amount of the credit back to EPAct 2005 levels.41
EISA also took a (albeit small) step to help in the creation of refueling stations by requiring that the head
of each federal agency install at least one alternative fuel pump for service to their vehicle fleet by
January 1, 2010.42 This mandate, while not strictly enforced, required Federal agencies to report through
an online reporting tool (Federal Automotive Statistical Tool- FAST) information about the fueling center
including amounts of fuel dispensed by type. This information is then compiled by the DOE in order to
determine how many alternate fuel pumps are available and how many refueling stations are noncompliant. In June 2011, DOE reported that the percentage of agencies complying with the mandate had
increased to 66%, up from 34% in 2010. Enforcement of the mandate could have increased compliance,
suggesting that the success of such policies could be improved by incorporating incentives that
complement and help support or enforce the mandates- either positive or negative incentives such as tax
credits or penalties. The federal fleet mandate for AFVs is discussed in more detail later in this document.
Alternative refueling stations are still limited as can be seen in Figures 4 and 5, and account for less than
8% of all stations in the country (given 2010 levels of US gasoline stations).43 This suggests that the
incentives for building new stations were inadequate, and unfortunately, the lack of refueling stations
continues to present a major barrier to adopting alternative fueled vehicles.
10
Figure 4. Total Refueling Stations by Fuel Type in 2012

8000
7000
B20
6000
CNG
5000
E85
4000
ELEC
3000
HY
2000
LNG
1000
LPG
0
Total Refueling Stations by Fuel Type
Data from US DOE, Alternative Fuels and Advanced Vehicles Data Center: Alternative Fueling Station Total
Counts by State and Fuel Type
**The large amount of electric fueling stations is due primarily to an abundance of these in CA
Figure 5. US Refueling Stations 2009
Gasoline Stations
Alternative Fueled Stations
Source: Energy Information Administration. Figures taken from GAO (2000) Report to Congressional Requesters
Energy Policy Act of 1992- Limited Progress in Acquiring Alternative Fuel Vehicles and Reaching Fuel Goals.
**Each dot represents 10 refueling stations in the state (rounded up to the next 10), and the dots do not correspond
to specific locations in the state.
11
2.2.b Natural Gas and Blue Corridors

In Argentina, Brazil and Italy where natural gas is widely used in vehicles, refueling stations are very
common. These countries do however, confront inter-country transportation issues; this is especially
important for the transportation of goods in heavy duty trucks but presents issues for ease of passenger
vehicle travel as well.
This concern has led to a push for blue corridors in South America and Europe -- inter-country
pathways that connect countries with natural gas refueling stations. The development of these corridors is
facilitated by economic and political arrangements between the countries in question. In South America,
this arrangement is MERCOSUR (Mercado Comun del Sur, or The Common Southern Market),
comprised of Argentina, Brazil, Paraguay, and Uruguay. In Europe, the European Commission serves this
function. Both entities are pursuing blue corridors to promote reductions in gasoline consumption and to
help stimulate adoption of natural gas heavy duty vehicles in member countries with lower adoption rates.
These corridors allow for a sharing of the investment cost by the two connected countries.44,45 Both the
blue corridor projects in South America and Europe have begun recently (or are still in pilot programs),
and as such it is difficult to measure the impact on natural gas adoption or GHG emissions. Nevertheless,
these corridors may provide a template for an integrated approach to refueling issues in the US. Most of
the refueling stations in the US are centered in large MSAs, which makes it prohibitive to travel very far
with AFVs. This is especially problematic for long distance trucking in the US, where the lack of
refueling stations has all but prevented alternative fuel adoption by smaller, independent carriers that lack
access to or funding for private, company specific refueling infrastructures for alternative fuels. Heavy
duty trucks account for 20% of mobile GHG emissions; boosting the adoption of AFVs in this sector
could help to significantly mitigate emissions. Blue corridors in the US could similarly help enable
interstate travel, especially since incentives and policies vary widely from state to state.
3. Alternative Vehicles
Government efforts to enhance energy security and mitigate mobile emissions have also focused on
alternate fueled vehicles. During the Bush administration, policies tended to focus on incentives for AFVs
(and later fuel celled vehicles), though policies during the Obama administration have focused more on
less fuel-dependent (or independent) vehicles such as electrics, plug-in hybrids, and fuel celled vehicles.
However, while these advanced technology vehicles have become more popular over the past few years,
they still remain a relatively small portion of the vehicles on the road. Figure 6 shows sales of AFVs and
hybrids from 2005-2009. In 2009, sales totaled at less than 1.2 million vehicles, accounting for
approximately 11% of all vehicle sales, of which E85 vehicles comprised the largest portion of sales.
12
Figure 6. Sales of Alternative Vehicles 2005-2009

Number of Alternative Vehicles Sold
1,200,000
Hybrids
1,000,000
(LPG)
800,000
Liquefied Natural Gas (LNG)

600,000
Hydrogen (HYD)
400,000
Ethanol, 85 percent (E85)
200,000
Electricity (EVC)
0
2005
2006
2007
2008
2009
Data from the US Department of Energy, Alternative Fuels and Advanced Vehicles Data Center:
HEV Sales by Model, AFVs in Use.
Over the past twenty years, numerous regulations and incentives designed to stimulate the market for
alternative fuel and advanced technology vehicles have been implemented. These have included
incentives for consumer purchases under the assumption that assistance in developing markets would
lower vehicle costs and stimulate additional research. Cost reduction remains a serious issue for these
relatively immature vehicles (such as EVs and FCs, although FFVs are very similar in price relative to
their internal combustion engine vehicle counterparts).
Range limitations and battery costs are still serious concerns for electric vehicles. The current
technological leader in batteries is the nickel-metal hydride (Ni-MH) battery, though the lithium-ion (Liion) battery (such as the one used in the Nissan Leaf) has recently emerged as a promising competitor: it
is lighter, can be charged more rapidly and doesnt need to be completely discharged prior to recharging.
However, Ni-MH batteries are less expensive and are currently the battery of choice for electric vehicles.
Table 3 shows the differences in terms of energy and price between Ni-MH and Li-ion batteries compared
to conventional internal combustion engine lead acid batteries. Though these numbers may vary by
producer, they represent the general trends of power and price across different types of batteries. Energy
density is the amount of electricity that can be stored per weight; power density is the proportion of
dischargeable energy to chargeable energy; and cycle life is the number of times the battery can be
discharged and recharged.
The remainder of this section looks at some of the major regulations and incentives that have been
proposed throughout the US with regard to alternative vehicles, both electric and non-electric.
13
Table 3. Comparison of Battery Types46
Battery Type
Energy density (Wh/Kg)
Power Density (W/kg)
Cycle life
Cost ($/kWh)
Lead Acid (conv. car battery)
Ni-MH
Lithium-ion
35
180
4,500
269
60
250-1000
2,000
500-1,000
120
1,800
3,500
1,000-2,000
Sources: Deutsche Bank, 2009; METI, 2009a; Nishino, 2010; The Institute of Applied Energy, 2008; Woodbank
Communications Ltd, 2005. Table taken from Lowe et.al. (2010)
3.1 Corporate Average Fuel Economy and Greenhouse Gas Standards

The Corporate Average Fuel Economy Standards (CAFE) were established in the Energy Policy and
Conservation Act (EPCA); in 2007, EISA amended EPCA and proposed standards for MY 2011-2020
vehicles. The new EISA standards were transformative: they included a 35 MPG mandate by 2020 for
light duty vehicles.47 EISA also required, for the first time, the setting of efficiency standards for medium
and heavy-duty trucks, though did not specify a MPG mandate.48
Importantly, these new standards decreased the level of credit manufacturers would receive for selling
flex-fuel vehicles (FFVs). Previous CAFE standards incentivized the production of FFVs by allowing
manufacturers to increase the overall fleet MPG average through the sale of these vehicles regardless of
the actual fuel used (as described in more detail in the next section). This was problematic, however: as
GAO established in 200049 (confirmed later by DOE in 200850 and the EPA in 201051), FFVs were
primarily being fueled with gasoline in spite of the alternative fuel option. The EISA amendments of
EPCA directed NHTSA to phase out this incentive for FFV production completely by 2019. In short,
EISA set the highest MPG CAFE standards to date, started the process for regulating the fuel efficiency
of medium and heavy-duty trucks, and phased out the credits for AFVs.
However, there are many unintended consequences related to increased fuel efficiency standards,
including, for example, the rebound effect and new source bias. The rebound effect is the tendency for
individuals to drive more due to cheaper operating costs, as increased fuel efficiency reduces the effective
price per mile. This effect has been estimated to be anywhere between 4.5% and 31%,52 and can offset
efficiency gains. New source bias refers to reduced purchases of new vehicles due to the higher vehicle
prices associated with more stringent efficiency regulations; the net result is that older, less efficient
vehicles tend to stay on the roads longer. NHTSA/EPA estimate that its new National Program designed
to harmonize vehicle regulations, and described in more detail in Section 3.1.c -- will result in $142-182
billion in fuel savings,53 assuming a 10% rebound effect54 and a -1 price elasticity of demand for
vehicles.55 This does not, however, take into account the impact on the used vehicle market or scrappage,
thus partially ignoring new source bias (though the negative price elasticity picks up some of the reduced
demand for these vehicles given higher prices).56 Even with penalties for manufacturers, concerns remain
that the overall costs of improving efficiency could be high enough to encourage non-compliance,
reducing the overall benefits of the program.
14
The story of how current efficiency and emissions standards for mobile sources were set is quite intricate,
encompassing actions at both the State and federal level, up to and including Supreme Court decisions.
The policy process that led to these standards is described next.
3.1.a Massachusetts vs. EPA

Until 2007, the Clean Air Act (CAA) required the US EPA to regulate air pollutants from mobile sources
to protect the public health and welfare. Greenhouses gases had not been determined to be an air pollutant
under the CAA, limiting EPAs authority to regulate tailpipe GHGs.57 In 2003, several organizations
petitioned the EPA to regulate GHG emissions from mobile sources, yet EPA denied the petition, saying
it lacked authority under the CAA to do so. In response, 12 states, 2 cities, and a number of organizations
sued the EPA, to force the agency to regulate GHG emissions from mobile sources, asserting that it did
indeed have such authorities under the CAA. In 2007 the US Supreme Court in Massachusetts vs. EPA,
sided with the petitioners.58 While the Supreme Court decision was clear, the EPA remained concerned
that the costs associated with implementing national standards and regulations for GHG management
would be high and ineffective, given the global nature of the problem.
3.1.b CAs Attempt to Regulate GHG Emissions from Mobile Sources

In California, the CA Air Resources Board (CARB) had been regulating non-GHG mobile emissions
within the state for decades. The mobile source provisions in the CAA (Title II) intended for emissions
standards to be set at a national level to maintain uniformity of regulation across all states, an approach
that was beneficial to vehicle manufacturers. Title II however allowed for an exception in Section 209 (b):
any state that by March, 30, 1996 had adopted standards that were at least as stringent as the federal
standards could receive a waiver to these provisions in the CAA.59 CA was the only state that adopted
regulations prior to 1996, and thereby was the only state eligible for the waiver. However, the waiver
could be denied if it was found that: (A) the protectiveness determination of the State is arbitrary and
capricious; (B) the State does not need such State standards to meet compelling and extraordinary
conditions; or (C) such State standards and accompanying enforcement procedures are not consistent with
section 202(a) of the Act (Federal Register, 3/6/08, p. 12158).
While CA was successful in obtaining a waiver for other state-based regulations, it took the state several
years to acquire a waiver for GHG emissions regulations. In 2002, CA enacted AB 1493, placing CARB
in charge of regulating GHG emissions from mobile sources. Under this authority, CARB enacted GHG
emissions standards in 2004 requiring a gradual decline in emissions over time for each manufacturer.
This spurred Californias initial waiver request to the EPA, seeking authority to regulate GHG as air
pollutants.60 Given the lawsuit the EPA was facing at the time, it chose to defer action on CAs waiver
petition until EPAs authority to regulate GHG had been either confirmed or denied by the courts. Vehicle
manufacturers opposed CAs waiver petition, claiming its standards were excessively stringent relative to
the rest of the country. EPA initially denied CAs waiver in 2008, noting that California does not need
its greenhouse gas standards for new motor vehicles to meet compelling and extraordinary conditions
(Federal Register, 3/6/08, p.12156). The EPA claimed that since CA faced the same threat of climate
change as the rest of the country, the compelling or extraordinary conditions test in article (B) (as
detailed above) did not apply. The failure to meet one test was sufficient for the denial of the waiver.
15
However, since this initial denial, 13 additional states have adopted CARBs proposed GHG standards,
and under CAs receipt of the waiver, these states would also be allowed to implement the standards.61
3.1.c The National Program

In light of the finding on Massachusetts vs. EPA, President Obama announced in May 2009 a plan for the
EPA to impose GHG emissions standards for light duty vehicles.62 The President directed NHTSA, in
charge of setting CAFE standards, to work with the EPA to establish the National Program, setting limits
of efficiency and GHG standards for motor vehicles and ensuring that regulatory approaches were
harmonized. The result of this coordinated approach was a national standard set at the same levels of the
CA standard; thus addressing the concerns of automakers about having to satisfy different standards at the
national and state levels. Furthermore, having two different sets of regulations could actually result in
higher overall emissions. Goulder et.al. (2012)63 find that the effort by the 14 states to adopt CAs GHG
standards had negative environmental impacts: adopting states tended to have lower emissions while
emissions in non-adopting states actually increased. Given the high standards in the adopting states,
manufacturers were able to reach the federal standard faster, allowing them to sell more vehicles with
high emissions in the non-adopting states. Thus, overall emissions worsened due to the difference
between federal and state regulations. Fortunately, CA accepted the standards for the National Program as
consistent with its own.64 This agreement, along with increasing external pressure from both the President
and environmental groups, led to a revision of the waiver request, and subsequently, the EPA granted CA
the waiver in June 2009.
The National Program has enabled a uniform national standard for GHG emissions from mobile sources
across the country. It also decreased allowances and credits the manufacturer could receive for AFVs and
other vehicle technologies (such as air conditioning), though it maintains credits for electric and fuel
celled vehicles. I discuss the details of these allowances in more detail below.
3.1.d Light-Duty Vehicles

Although CA was not allowed to regulate GHG emissions from mobile sources until 2009, the CA mobile
emissions regulations were used as the basis for current standards issued by the National Program for MY
2012-2016 light duty vehicles. The National Program mandates a fleet average of 34.1 MPG and GHG
emissions of 250 grams/mile for MY 2016. It was necessary to lower the MPG requirement from 35MPG
in order to solve the problem that the GHG standards provided more allowances to manufacturers than did
the CAFE standards. On the one hand, EPCA did not allow CAFE standards to be affected by air
conditioner (A/C) credits, while the GHG standards utilized the A/C credits as an allowance to help
manufacturers reach the standard. These A/C credits were allowances given to manufacturers who made
improvements in the air conditioning system in the vehicle: since A/C is one of the most energy intensive
parts of a vehicle, improving the efficiency of these systems was considered by EPA as GHG emission
improvements. Furthermore, the 250 g/mile GHG standard would correspond to a 35MPG standard if the
manufacturers met the GHG standard through efficiency improvements alone. Thus, in order to
coordinate across these two standards and to allow for the fact that the A/C allowances differed across
these, the MPG requirement was set slightly lower than proposed in EISA.65
16
As mentioned earlier, FFVs are commonly run on gasoline alone, and these new standards dealt with this
issue directly. In previous CAFE standards, FFVs were assumed by EPCA to run 50% of the time on
gasoline, and 50% of the time on the ATF. Furthermore, the manner in which emissions for the ATF were
calculated was based on a multiplier: each gallon of ATF was counted as 0.15 gallons of gasoline. These
two assumptions jointly implied that, for example, an FFV that emits 330 g/mile of CO 2 while utilizing
ethanol and 350 g/mile of CO2 while utilizing gasoline would be estimated as having the following total
average emissions:
CO2
(330 * 0.15 350)

199.8 g / mi
2
This provided a major incentive for manufacturers to sell these vehicles, as this allowance facilitated
compliance with both the emissions standards and the efficiency requirement. Given the 0.15 conversion
factor, this increased the MPG of a FFV by a factor of 6.67: for example, a 15 MPG AFV would be rated
at 100 MPG.66 This diminished greatly the environmental effectiveness of these standards, as it did not
take into account the actual fuel utilized by these vehicles. In fact, the National Programs Final Rule cites
the Regulatory Impact Analysis of RFS2 as claiming that Data show that, on average, FFVs operate on
gasoline over 99 percent of the time, and on E85 fuel less than 1 percent of the time (Federal Register
5/7/2010, p.25437).
The new standards pursuant to EISA changed this allowance structure for FFVs, providing the two
allowances (the 50/50 assumption and the conversion factor) only for MY 2012-2015 vehicles. For MY
2016 and later, these two allowances were phased out. Also, starting with MY 2016, EPA will assume
that the utilization of ATFs is negligible; the manufacturer must provide evidence demonstrating the use
of ATFs for vehicles sold or request an alternate weighting value to be determined by the EPA. This
leaves the burden of proof on the manufacturer and diminishes the incentive to produce FFVs.
Furthermore, the conversion factor no longer holds after 2016 and a vehicles actual emissions are tested
while using an ATF. Finally, the National Program limits the amount of allowances in the average fleet
MPG calculation: a manufacturer can only accrue up to 1.2MPG in allowances due to sales of FFVs.67
These standards, though they decreased the incentive to produce FFVs, created some flexibility for
electric and fuel celled vehicles. EVs, PHEVs, and FCVs are assumed to have emissions of zero g/mile,
effectively ignoring upstream and life-cycle emissions. While this benefit is phased out after the
manufacturer has sold a certain amount of these vehicles, the regulators did not consider it likely that the
limit would be reached within the time frame of the regulation.68 This was intended to provide incentives
to manufacture EVs, to be phased out when economies of scale were achieved.
3.1.e New CAFE Standards and Compliance

In May 2010, President Obama set goals for the next round of CAFE standards, affecting MY 2017-2025
light duty vehicles. The proposed standard is much more stringent, calling on the manufacturers to
increase average efficiency to 54.5 MPG by 2025.69 Because this standard is so difficult for manufacturers
to achieve, the proposed standards would actually increase allowances for AFVs and EVs relative to the
2016 rule: one EV (or FCV) would count as 2 EVs/FCVs, and one PHEV would count as 1.6 (though this
multiplier is proposed to phase down by MY 2021).70 This sort of multiplier had been previously ruled
17
out by EPA, as the agency claimed that the multiplier, in combination with the zero grams/mile
compliance value, would be excessive (Federal Register, 5/7/2010, p. 25401).
This regulation has not yet been finalized and may change after the comment period and subsequent
revisions. As it currently stands however, it creates strong incentives for automakers to manufacture EVs,
PHEVs and FCVs. The proposed regulation also brings back the zero g/mile allowance for a certain
amount of vehicles sold, further incentivizing the production of these vehicles in the beginning of the
program. Unfortunately, creating allowances that target very high levels of fuel efficiencies allows
manufacturers to sell more vehicles that do not meet the CAFE standards, thus increasing overall fuel
consumption.
The regulation may need to implement these allowances for very high efficient vehicles not only because
of the stringent standard, but also because of the way in which the CAFE standard is calculated. These
standards set a limit on the efficiency that each manufacturer needs to reach, calculated through a salesweighted harmonic average. What this implies is that the vehicles within a fleet are averaged given the
fuel economy over a set of miles, instead of over a set of gallons, as the arithmetic mean would imply. For
example, if a fleet has three regular cars and one electric vehicle, with relative efficiencies of 10, 15, 20,
and 100MPG, the harmonic average is calculated as:
4
17.6MPG
1 / 10 1 / 15 1 / 20 1 / 100
whereas the arithmetic average would be calculated as:
10 15 20 100
36.25MPG .
4
This in essence decreases the ability of a manufacturer to reach a specific goal through efficiency
improvements, as the harmonic average mitigates the impact of outliers, and thus decreases the
importance of very high fuel efficient vehicles. For example, improving the MPG of the electric vehicle in
the above equation could never bring the harmonic average to surpass 18.5MPG: even if its MPG was
infinite, the harmonic average would stay at 18.5. Therefore, utilizing harmonic rather than arithmetic
averages means that CAFE standards diminish the incentive to produce very highly efficient vehicles
(though it does incentivize increasing the MPG of low and middle efficiency vehicles, which can have
larger impacts on fuel savings). As such, if the policy wants to incentivize the production of electric
vehicles, then allowances in the standard for electric vehicles are essential.
However, increasing fuel efficiency has diminishing returns. For example, replacing a sedan in the fleet
with a hybrid equivalent costs the manufacturer around $3,00071 given current hybrid technology (though
exact costs are unknown, the difference in price between a hybrid car and its non-hybrid counterpart is
approximately this amount).72 On the other hand, replacing an efficient non-hybrid vehicle with an
electric vehicle costs anywhere between $10,000 and $30,000,73 mostly due to the cost of the battery
(which based on warranty information is projected to need to be replaced more frequently than the battery
in a hybrid due to its usage and cycles). Yet, reductions in gasoline consumption are much larger from
replacing the sedan with the hybrid than replacing the efficient vehicle with an electric. Consider two
vehicles: a 12MPG vehicle and a 30MPG vehicle. Increasing the 12MPG vehicle by 2 miles per gallon
would result in fuel savings of 1.19 gallons per 100 miles driven. On the other hand, increasing the
18
30MPG vehicle to 40MPG would result in 0.08 gallons of fuel saved per 100 miles (assuming no change
in VMT- if the rebound effect occurs, then the 40MPG vehicle will be driven more than the 30MPG
vehicle, thus reducing even further the amount of gasoline saved).
Figure 7 demonstrates graphically the downward slope of these returns. This figure shows gasoline
savings by increases in fuel efficiency (under assumed VMT of 15,000 or 7,500). For example, replacing
a 20MPG traditional gasoline vehicle with a 40MPG hybrid vehicle would save 375 gallons (at 15,000
VMT) at a cost of $3,000, while replacing it with a 100MPG electric vehicle would save 600 gallons at an
average cost of $20,000. This implies marginal costs of $8/gallon reduced vs. $33/gallon reduced,
demonstrating that although total gallons reduced are higher, it is less efficient (in an economic
perspective) to replace a traditional vehicle with an EV compared to replacing it with a hybrid. An even
larger number of gallons could be saved by replacing a 10MPG vehicle with a traditional gasoline
20MPG vehicle and without spending thousands of dollars to do so.
Figure 7. Fuel Savings by MPG, VMT
Source: William Chernicoff, Energy & Environmental Research Group Manager, Toyota Motor North America, Inc.
3.1.f Heavy-Duty Vehicles

For the first time, the EPA and NHTSA have created a comprehensive Heavy-Duty National Program of
emissions and efficiency. This legislation establishes standards separately for three different types of
vehicles;1) Class 7 and 8 combination tractor-trailers; 2) Vocational vehicles; and 3) Medium-duty pickup
trucks and vans. These regulations also incentivize the production of alternate fueled and electric vehicles,
although less so than the proposed light duty regulations.
Emissions from AFVs are calculated through CO2 tailpipe emissions testing, while EVs are assumed to
have zero emissions. FFVs are treated similarly to current regulations for light duty vehicles a 50/50
weighting assumption through MY 2015, after which the weighting factor depends on the demonstrated
ATF usage by each manufacturer. However, no conversion factor is assumed.74 Thus, the heavy-duty
19
legislation does less to promote production of AFVs than the light-duty legislation, which is unfortunate
given the extremely low level of adoption of these new technologies by the heavy-duty vehicle market.
EPA and NHTSA claim that these improvements to fuel efficiency will save the industry $50 billion in
fuel, yet this claim is troubling as the industry has not made these improvements on its own.75 This underinvestment in energy saving technologies that pay back over time has been referred to as an energy
paradox (see Harrington and Krupnick [2012]76). The trucking industry and other analysts dispute the
availability of fuel saving technologies, and attribute under-investment to hidden costs, such as
engineering tradeoffs between fuel efficiencies and torque, or safety tradeoffs.
For the regulations in the heavy-duty sector to be effective in decreasing petroleum consumption, more
incentives for alternate fuels and vehicles are necessary than what has been developed in these first
efficiency and emissions standards. In fact, merely increasing the fuel efficiency of new vehicles will not
necessarily result in a diminished use of gasoline, given rebound effects. Furthermore, hidden costs and
technological limitations are present in this industry, causing an under-utilization of efficient
technologies. Policy makers need to understand the basis behind the energy paradox and take into account
the rebound effect in order to craft policies that are effective in diminishing gasoline consumption.
3.2 Market Stimulation

The government has taken a series of steps to stimulate the market in ways that would encourage wider
use of alternative vehicles. These fall into two categories: incentivizing and mandating the adoption of
these vehicles by government agencies and consumers; and creating voluntary initiatives and
demonstrations to increase awareness and understanding. This section discusses these policies and their
impacts on the market for alternative vehicles.
3.2.a Mandated Adoption of Alternative Vehicles

EPAct 1992 presented an important set of incentives and supporting regulations for alternative fuels and
vehicles (in addition to ethanol). For example, it established goals for federal agencies to adopt alternative
fueled and electric vehicles, under the direction of the US Department of Energy (DOE). This legislation
mandated (a soft mandate) that new government fleet purchases contain a minimum of 25% AFVs in
1996, increasing to 75% in 1999.77 The control of AFV ownership by federal fleets was conducted
through voluntary provision of information.78 DOE claims that since 2003, almost 100% of the federal
fleets have complied with or exceeded this mandate, and those that fell short reached agreements with the
federal government.79
This mandate however had an unfortunate unintended consequence. For much of the period after the
passage of EPAct 1992, oil and therefore gasoline-- prices were very low. It was far less expensive for
government AFVs to run on petroleum based fuels than alternative fuels. Consequently, the mandate was
largely met but petroleum based fuel consumption did not markedly decrease.80 DOE recently reported
that even if all agencies had complied fully with the mandate, the vehicles would only have replaced less
than 1% of gasoline consumed in 2010,81 illustrating a large net economic cost the adoption of
thousands of higher-priced AFVs with little benefit on the fuel side. Though the price difference between
FFVs and traditional vehicles is negligible these days (between $50 and $10082), in 1992 it was not. The
federal fleet mandate in EPAct 1992 resulted in an extra 200,000 AFV purchases, with little to no GHG
20
emissions benefits.83 Furthermore, 200,000 vehicles is less than 1% of all vehicle purchases, so it is
unlikely that this mandate contributed to the decrease in FFV prices. CAFE standards, which included
many credits for the manufacture and use of AFVs, was likely a much more important factor in closing
this price gap than the federal fleet AFV mandate.84
The fact that these vehicles were primarily used with gasoline led to the adoption of fuel usage mandates
in future regulations. EPAct 2005 amended EPAct 1992 to mandate that agencies purchasing FFVs
operate them exclusively on alternate fuels. Those agencies not able to use alternative fuels due to lack of
fueling stations or other hardships were able to receive a waiver.85 These waivers were then utilized by
the federal government to identify areas where ATFs were not readily accessible. Using the information
gathered from these waivers, EISA 2007 mandated a 10% increase in usage of alternate fuels by federal
fleets, simultaneously with the refueling station requirement discussed in Section 2.86
In order to expand the number of first adopters outside the government, the mandates in EPAct 1992 also
affected alternative fuel providers. Alternate fuel producers and refiners were required to purchase a
minimum percentage of AFVs per year- increasing to 90% in 1999.87 In order to enforce these
regulations, penalties of $5,000-$50,000 were implemented for non-compliance.88 These regulations were
intended to create an example for the public of AFV usage, as well as to help spur the market. In 2001,
the DOE reported a 91% compliance rate amongst for fleets covered by the statute.89 Regardless of these
high compliance rates, the petroleum replacement goals set out by EPAct 1992 of 10% by 2000 and 30%
by 2010 were not met.90 Historical experience suggests that mandating AFVs at the federal fleet level may
have more value as a demonstration than as overall decline in petroleum-based fuel consumption.
3.2.b California ZEV Mandates

The state of California has implemented several mandates for the use of zero emission vehicles (ZEVs).
Given CAs major problems of transportation pollution in places like Los Angeles, there has been a push
over the decades to implement more stringent efficiency and emissions standards. CA has been able to
affect standards through the EPA waivers (as discussed in the CAFE standards section of this paper), and
has become the earliest promoter of advanced vehicle technologies. In fact, in 1990 California utilized
one of these waivers to pass a ZEV mandate, which directed that by 2003 10% of all sales by the large
manufacturers must be ZEVs.91 Unfortunately, the ZEV mandate, while popular among constituents,
became a contentious matter and was fought, amended, and changed over the decade-long policy process.
The first set of changes occurred in 1996. As the mandate ramped up ZEV sales from 2% of total vehicle
sales in 1998 to 10% in 2003, concerns of manufacturers over meeting intermediate goals resulted in the
removal of all mandates prior to 2003, while leaving intact the 10% mandate for 2003. Manufacturers
concerns about meeting the 2003 deadline intensified in 2001, causing the mandate to be changed furtherthis time, it allowed the manufacturers to meet the ZEV mandate through the production of non-ZEV
vehicles, such as Partial Zero Emissions Vehicle (PZEVs) and Advanced Technology PZEVs (AT
PZEVs). Soon after, regulators faced a lawsuit preventing them from enforcing the mandate; leading to
more regulatory changes and further expansion of the types of vehicles that met the requirements for the
mandate. In 2008, the mandate was once again changed to allow the manufacturers to produce a greater
number of PHEVs to meet the mandate, but only if they also produced a minimum number of pure
ZEVs.92
21
Even though the regulators faced massive opposition to these mandates, in 2009, they increased the
requirement for pure ZEVs from 11% in 2009 to 16% in 2018, although the manufacturers were allowed
to use a portion of PZEVs and AT PZEVs sold to meet the mandate.93 The new Advanced Clean Car
Rules of 2012 (which set the newest CA ZEV mandates) were even more stringent: credits for non-zero
emission vehicles were phased out after 2018.94
Lessons Learned from CAs ZEV Mandate History
Both the benefit and drawback of a ZEV mandate is that it places the burden on the manufacturers,
tasking them with advance technology and pricing to promote the purchase of ZEVs. While this reduces
the burden on the government and consumers, manufacturers feel pressured and will fight to avoid
compliance. Indeed, this was the case with the CA ZEV mandates: it was met at every point with lawsuits
and opposition from the manufacturers.
Would this have been different had the CA regulators produced significant financial incentives on the
consumer demand side? Though the ZEV mandate was never coupled with tax credits or rebates for
vehicle purchases, there were a number of federal and state incentives in place after 2007, including tax
credits, rebates, and HOV stickers for the purchase of ZEVs. These rebates and tax credits (which are
discussed in the next section in more detail) may have simultaneously enabled a more stringent approach
and engendered less opposition to the ZEV mandate.
Electric vehicle sales remained relatively flat between 2005 and 2009, averaging approximately 2,000
new vehicles per year with a total of 57,000 by 2009.95 CA has the majority of these vehicles: in 2009,
there were approximately 31,500 electric vehicles in use in the state, which amounted to 55% of the
national electric vehicle stock.96 This demonstrates that though CA is still the leader in ZEV ownership, it
was unable to successfully reach any of its goals in terms of percentage of ZEVs purchased.
The history of ZEV regulation in California suggests that mandates, while appealing to governments for
both budgetary and policy reasons, are not optimized without complementary incentives. Sticks without
carrots can reduce compliance and increase opposition to both mandates and ZEVs in general. Also,
mandates that are phased in over time tend to encourage lawsuits and the weakening of the original
targets or requirements.
3.2.c Consumer Incentives

In addition to incentives and mandates on vehicle manufacturers, the federal and state governments have
provided incentives to consumers for purchasing AFVs.
A major goal of EPAct 2005 was to stimulate the production and utilization of alternative vehicles:
hybrids, fuel cell vehicles, AFVs, and FFVs. To help market these vehicles, the statute established
consumer based incentives in the form of tax credits for their purchase. These tax credits ranged from
$2,500-$8,000 for light duty fuel celled vehicles, and up to $40,000 for heavy duty FCVs. The size of the
tax credit increased with the vehicles efficiency, creating an even larger incentive to purchase high fuel
efficiency vehicles. For hybrids, the credit ranged from $1,500-$3,000 for light duty vehicles and between
$3,000 and $12,000 for heavy duty vehicles, depending on the efficiency increase relative to a non-hybrid
equivalent.97
22
These credits were phased down for each manufacturer during a 15 month period, or until its first 60,000
vehicles were purchased. This helped boost purchases of these vehicles, though the economic benefits
depended on the popularity of each vehicle. For example, the Toyota Prius tax credit ended within a few
months of the regulation, while it was still possible to receive a credit for other vehicles several years
later, such as the Chevrolet Malibu Hybrid.
Consumers were already purchasing the Prius in greater quantities than other hybrids, thus the added tax
credit may have been more of a windfall than an incentive to these consumers. In fact, in 2004, Toyota
had sold over a million Prius.98 Providing consumers a credit for purchasing one of the first 600,000
Prius sold in 2005 therefore did not necessarily increase sales.
Incentivizing early adoption of alternative vehicles can help manufacturers achieve economies of scale
and help level the playing field for late adopters. On the other hand, certain manufacturers (such as
Toyota) who have already commercialized AFVs or ZEVs do not need these types of incentives.
Vehicle
(Example)
Honda Accord
Hybrid
Honda Insight
CVT
Honda Civic
Hybrid
Ford Escape
4WD
Toyota Prius
Table 4. Initial Tax Credits for Different 2005 Hybrid Models

2005 MSRP
MPG (nonMPG (city/
(non-hybrid
hybrid
Credit
2005 MSRP
highway)
version*)
version*)
Yearly
Fuel Cost
Savings**
$650
$30,655
25/33
$22,715
21/31
$266.19
$1,450
$19,845
45/49
--
--
--
$1,700
$20,415
39/43
$13,775
25/34
$560.05
$1,950
$27,445
30/28
$23,150
19/23
$740.66
$3,150
$21,815
48/45
$15,365
28/37
$543.20
MSRP data: cnet.com; MPG data: Edmunds.com

*non-hybrid versions: Honda Accord, Honda Civic, Ford Escape, and Toyota Corolla. Honda Insight has no gasoline equivalent vehicle.
**Cost savings based on 15,000 VMT, $3.5 gasoline price, and EPA 45/55 definition of average efficiency (relative to non-hybrid version)
The American Recovery and Reinvestment Act of 2009 (ARRA, the stimulus bill) provided further
incentives to consumers for the purchase of ATVs. These included a consumer tax credit of $2,500 to
$7,500 (depending on the size of the battery) for the purchase of electric vehicles. This credit ended after
the manufacturer sold 200,000 vehicles.99 ARRA also provided a consumer tax credit for electric vehicle
conversion for 10% of the cost of conversion up to $40,000.100
The recession has caused an overall decrease in vehicle purchases, making these tax credits an important
stimulus to the economy. During the recession, the government also provided a Cash for Clunkers
program to incentivize the purchase of more efficient vehicles. Cash for Clunkers provided consumers
with up to $4,500 rebates for turning in an older, low fuel efficient vehicle (a clunker) to used towards
the purchase of a more fuel efficient vehicle.101 Hybrids or EVs (as long as they cost less than $45,000)
were eligible for such purchases. Li et.al. (2011)102 and Mian and Sufi (2010)103 find that Cash for
Clunkers had small impacts on emissions and fuel efficiency; the program did however stimulate sales in
23
the (very) short term, largely by displacing future sales. This suggests that this program was more of an
economic stimulus tool than a tool to stimulate advanced technology vehicle markets.
State governments have also provided non-monetary incentives for the purchase of ATVs. In many states,
state and local governments provide stickers granting access to High Occupancy Vehicle (HOV) lanes for
hybrid, electric and partial zero emission vehicles (regardless of vehicle occupancy). The 1990 Clean Air
Act Amendments first implemented these allowances for Inherently Low Emission Vehicles (ILEVs),
vehicles categorized by the EPA as having low emissions, and which generally ran on alternative fuels.
The 1998 Transportation Equity Act for the 21st Century helped states to extend this allowance to
individual owners of ILEVs.104 This allowance was provided in many states- in fact, 9 of 20 states with
HOV lanes granted HOV stickers to ILEVs. Several of these states, including California, added hybrids to
the list of HOV allowed vehicles.105 CA currently implements HOV allowances for all electric vehicles
sold and for the first 40,000 AT PZEV until January 1, 2015. Between 2005 and 2008, California also
issued 85,000 stickers to 3 hybrid vehicle models: the Toyota Prius, the Honda Civic Hybrid, and the
Honda Insight.
The HOV allowances for hybrids in California were only available until July 2011. These stickers were
valued highly by hybrid vehicle buyers; some research suggests that that individuals were willing to pay
as much as $3,200 more for a vehicle that came with such a sticker.106 However, Diamond (2009)107 finds
that the HOV sticker did not significantly affect hybrid purchases in CA, implying that the HOV
allowance was more of a windfall benefit to hybrid owners than a hybrid vehicle market booster.
3.2.d Demonstrations and Voluntary Programs

The federal government has also established several demonstration and voluntary programs designed to
encourage adoption of AFVs largely by increasing awareness of and research into their development.
EPAct 1992 provided several financial incentives including loan guarantees and grants for trial programs
designed to boost AFV demand through demonstration of their use. The Clean Cities Program was such a
program, established to help reduce petroleum consumption, and focused on activities at the local level.108
Clean Cities has been implemented in 100 cities across the nation; it establishes partnerships of local
public and private stakeholders, to help in the adoption of advanced vehicle technologies, reduction of
gasoline consumption, and fuel economy improvements. The program also provides assistance to fleets
for employing alternative vehicles and disseminating information to educate consumers about the benefits
of AFVs. Clean Cities claims to have displaced approximately 3 billion gallons of gasoline with alternate
fuels, with the largest contributions coming from natural gas and ethanol.109
EPAct 2005 also included several voluntary programs to help generate widespread adoption of alternative
vehicles, such as fuel cell school buses.110 In general, the success of these programs is fairly limited and
some were never implemented; this was the case with the FC school bus project due to the extremely high
costs of capital (a FC school bus costs around $2-3 million compared to $350,000 regular school bus) and
hydrogen (2-4 times more expensive than diesel).111
EISA 2007 also provided incentives for the creation of voluntary programs to encourage the use of
electric vehicles, authorizing $90 million per year for this purpose.112 The Plug-in Electric Drive Vehicle
Program provided grants to local and state government agencies and private/non-profit entities to create
projects that would encourage the use of EVs and other advanced vehicle technologies. These grants were
24
intended to stimulate R&D of technology in the early stages of adoption. EISA also created the NearTerm Transportation Sector Electrification Program, which authorized $95 million a year for grants to
large scale electric transportation projects, including an electric vehicle competition. It also created an
electric vehicle education program to encourage the study of EVs in schools and Universities.113
Though these sorts of demonstrations and voluntary programs can help to increase understanding and
acceptance of alternative vehicles, their impact appears to be limited. Barriers to their adoption are still
significant, and until these are overcome, there will likely be fewer alternative vehicles than is socially
optimal.
3.3 Technological Advancements and R&D

The federal government has also invested in research to reduce the high cost of AFVs and batteries, so as
to help eliminate the barriers to adoption.
In 2007, the National Academy of Sciences recommended that the Department of Energy establish a
program to invest in the development of advanced technologies, called Advanced Research Projects
Agency-Energy (ARPA-E). This program was intended to support cutting edge technological research by
independent entities, such as Universities, firms and others, to address long-term energy issues.114
ARPA-E was authorized in the America COMPETES Act, though it did not receive any funding until
October 2009, when ARRA provided the program with $400 million. These funds are used to support 37
different energy projects focusing on renewable energy research, energy storage and fuel-independent
vehicles.115 Together, two rounds of ARPA-E projects have provided funding to advance technology
development for advanced vehicles, including EVs, AFVs, and fuel cell vehicles, as well as funding for
battery development.
EISA 2007 also implemented a grant program to help develop plug-in hybrid electric vehicles (PHEVs)
and EVs, and a $25 billion loan program to aid in the development of infrastructure that produces
alternative vehicles and their components.116 Furthermore, EISA guaranteed loans for the production of
EVs and for the production of advanced batteries.117
Federal R&D support may be the most important and impactful way to advance the adoption of
alternative vehicles. Though cost reduction and performance remain key focus areas for research, the
price of Li-ion batteries has decreased over the years, while their energy density has increased, as can be
seen in Figure 8. R&D investments can therefore help accelerate cost reductions and performance
improvements.
25
Figure 8. Historical Cost Reductions for Li-ion Batteries, 1991-2005
Source: Farrell, John (2011). Democratizing the Electricity System: A Vision for the 21st Century Grid. New
Rules Project. http://energyselfreliantstates.org/content/democratizing-electricity-system
4. Conclusion
Over the last three decades, there has been a worldwide push to adopt non-petroleum transportation fuels
and to develop vehicles that can run on these alternative fuels. The US has taken a multi-faceted approach
to encourage these developments, devising a range of incentives, mandates, tax credits, loan guarantees,
demonstration programs and voluntary programs to condition markets, require or encourage
manufacturers to produce and consumers to purchase AFVs and EVs. Some policy tools have been
complementary but success in general has been relatively limited. Opposition from interest groups, lack
of enforcement and monitoring, incoherent policies, low gasoline prices, technological stagnation, lack of
refueling stations, and other complications still present substantial barriers to broad deployment and
acceptance of AFVs and EVs. Also, these policies have arguably failed at achieving the underlying goals:
improving GHG emissions and pollution, decreasing gasoline consumption, and increasing energy
security. Questionable or limited progress towards these objectives suggests that federal dollars have not
been well spent and that we have yet to find the appropriate mix of government incentives that will enable
substantial progress towards these goals.
Providing costly incentives for highly efficient alternative vehicles, such as EVs, PHEVs, and FCVs, is
arguably a very inefficient way of reducing GHG emissions, given the high costs to government and
vehicle manufacturers and the diminishing returns to gasoline reduction from efficiency increases.
Allowances in standards for these vehicles are more costly and less effective policy instruments than
focusing on increasing the MPG of the least fuel efficient vehicles.
While it may be less economically efficient to promote electric and other high fuel efficient vehicles,
government investments in research to accelerate the advancement of the underlying technologies may be
one of the most successful ways to mainstream these vehicles. Focusing on bringing down battery costs,
for example, will result in lower vehicle prices, and increased adoption, without having to subsidize or
mandate the purchase of the vehicle. Once the costs of electric vehicles have become competitive with
internal combustion engine vehicles, adoption will occur on a much greater scale.
Clearly, many of these policies have fallen short of achieving their primary goals and are difficult to
implement, especially when the policies are structured in ways that exacerbate opposition from
26
manufacturers. The lack of penalties for non-compliance and the lack of enforcement of existing penalties
have also diminished policy effectiveness. Likewise, policies that have long waiting periods prior to their
implementation encourage lawsuits and increase uncertainty.
Given the increasing emphasis on promoting these alternative vehicles, it is time to take a step back and
ask whether this set of policy tools provides the correct avenue for reaching the goals of environmental
improvement and energy security. Stronger compliance mechanisms, coupled with regulatory certainty
and a clearer understanding of the technology constraints and market conditions is desirable and could
help achieve policy goals. However, the policies detailed in this paper do not address the other
externalities associated with driving, such as congestion and accidents, so their scope is somewhat
limited. Policies that target driving behaviors, such as a VMT or gasoline tax, for example, could arguably
do more to address all the issues mentioned above, and also fit within a much simpler administrative
framework. These types of policies have not been adopted due to the publics negative opinion of them.
Yet if we spent merely a fraction of the resources that we have placed in promoting alternative fuels and
vehicles on instead attempting to change the publics perception (such as an intensive campaign
promoting coupling taxes with lump sum rebates), we might have been able to reach a more optimal
solution.
Parry, Ian W. H. and Kenneth A. Small (2005). Does Britain or the United States Have the Right Gasoline
Tax?American Economic Review, Vol. 95, No. 4, pp. 12761289.
2
Williams, Roberton C. III (2005). An Estimate of the Second-Best Optimal Gasoline Tax Considering Both
Efficiency and Equity. Working Paper.
3
West, Sarah E. and Roberton C. Williams III (2007). Optimal taxation and cross-price effects on labor supply:
Estimates of the optimal gas tax. Journal of Public Economics, Vol. 91, No. 3-4, pp. 593617.
4
Babcock, Bruce A. (2007). High Crop Prices, Ethanol Mandates, and the Public Good: Do They Coexist? Iowa
Ag Review, Vol. 13, No. 2
5
Hahn, Robert and Caroline Cecot (2007). The Benefits and Costs of Ethanol, Working Paper 07-17, AEIBrookings Joint Center for Regulatory Studies, November 2007.
6
DOE/EIA: Annual Energy Outlook 2011, Table 58- Light-Duty Vehicle Stock by Technology Type.
7
Schnepf, Randy and Brent D. Yacobucci (2012). Renewable Fuel Standard (RFS): Overview and Issues.
Congressional Research Service Report for Congress, January 23, 2012, Page 21.
8
ICM: Ethanol, the Fuel of the Future http://www.icminc.com/timeline/1862.html
9
Cato, Greg (2010). Clearing the Air: A Policy Analysis of the Ethanol Tax Credit. Policy Perspectives, Vol. 17,
pp. 71-88.
10
TEA-21 - Transportation Equity Act for the 21st Century, 9003(b)(h)(2)
11
Yacobucci, Brent D. (2012). Biofuels Incentives: A Summary of Federal Programs. Congressional Research
Service Report for Congress, January 11, 2012.
12
Elam, Thomas E. (2007). Fuel Ethanol Subsidies: An Economic Perspective. FarmEcon.com, September 19,
2007.
13
Yacobucci, (2012).
14
Ibid.
15
US Energy Information Administration (2011). Direct Federal Financial Interventions and Subsidies in Energy in
Fiscal Year 2010. Independent Statistics and Analysis, US DOE.
16
EPAct 2005, 1501 (a)(o)(2)(B).
17
EISA, Title II, 202 (a)(2)(B)(i)(I).
18
Federal Register, Vol. 77, No. 5, 1/9/2012, page 1339.
27
19
EPAct 2005, 1501 (a)(o)(4).

Ibid, 1501 (a)(o)(2)(B)(iii)(I).
21
US Department of Agriculture (2011). U.S. on Track to become Worlds Largest Ethanol Exporter in 2011.
http://www.fas.usda.gov/info/IATR/072011_Ethanol_IATR.pdf, page 1.
22
Wang, Michael, May Wu, Hong Huo and Jiahong Liu (2008). Life-Cycle Energy Use and Greenhouse Gas
Emission Implications of Brazilian Sugarcane Ethanol Simulated with the GREET Model. International Sugar
Journal, Vol. 110, No. 1317.
23
Walter, Arnaldo et.al. (2011). Sustainability assessment of bio-ethanol production in Brazil considering land use
change, GHG emissions and socio-economic aspects. Energy Policy, Vol. 39, No. 12, pp. 5703-5716.
24
Cordonnier, Vanessa (2009). Ethanols Roots: How Brazilian Legislation Created the International Ethanol
Boom. William & Mary Environmental Law & Policy Review, Vol. 33, Issue 1, pp. 287-317.
25
Puerto Rico, Julieta Andrea (2008). "Programa de Biocombustveis no Brasil e na Colmbia: uma anlise da
implantao, resultados e perspectivas" (in Portuguese). Universidade de So Paulo. Ph.D. Dissertation Thesis.
26
Macedo Isaias, M. Lima Verde Leal and J. Azevedo Ramos da Silva (2004). Assessment of greenhouse gas
emissions in the production and use of fuel ethanol in Brazil. Secretariat of the Environment, Government of the
State of So Paulo.
27
Lapola, David M. et.al. (2010). Indirect land-use changes can overcome carbon savings from biofuels in Brazil
PNAS, Vol. 107, no. 8, pp. 33883393.
28
National Research Council (2011), page. 6.
29
Ibid, pp. 4-5.
30
Ibid.
31
National Research Council. Renewable Fuel Standard: Potential Economic and Environmental Effects of U.S.
Biofuel Policy. Washington, DC: The National Academies Press, 2011, pp. 204-206.
32
Yang, Yi, Junghan Bae, Junbeum Kim, and Sangwon Suh (2012). Replacing Gasoline with Corn Ethanol Results
in Significant Environmental Problem-Shifting. Environmental Science & Technology, doi: 10.1021/es203641p.
33
Congressional Budget Office (2010). Using Biofuel Tax Credits to Achieve Energy and Environmental Policy
Goals. July 2010, Pub. No 4044, page VIII.
34
National Research Council (2011), pp. 2-3.
35
EPAct 2005, 1512
36
National Research Council (2011), page 4.
37
US EPA (2010). Renewable Fuel Standard Program (RFS2) Regulatory Impact Analysis. EPA-420-R-10-006,
page 5.
38
Krupnick, Alan (2010). Energy, Greenhouse Gas, and Economic Implications of Natural Gas Trucks. Resources
for the Future Background Paper.
39
EPAct 2005, 1342.
40
ARRA 2009, 1123.
41
Tax Relief Act 2010, 711.
42
EISA 2007, 246(a).
43
US DOE, Alternative Fuels Data Center, Public Retail Gasoline Stations by State and Year
44
Papageorgiou, Andreas (2005). Natural Gas in Argentina. PREMIA.
45
United Nations Working Party on Gas Inland Transport Committee (2003). Blue Corridor Project: on the use of
natural gas as a motor fuel in international freight and passenger traffic.
46
Lowe, Marcy, Saori Tokuoka and Tali Trigg (2010). Lithium-ion Batteries for Electric Vehicles: The US Value
Chain. Center on Globalization, Governance & Competitiveness, Duke University. October 5, 2010.
47
EISA 2007, 102(b)(2)(A).
48
Ibid, 102(k)(1)(B).
49
United States General Accounting Office (2000). Energy Policy Act of 1992: Limited Progress in Acquiring
Alternative Fuel Vehicles and Reaching Fuel Goals. Report to Congressional Requesters, page 17.
50
Department of Energy (2008). Review of Energy Policy Act of 1992 Programs and the Alternative Fuel Provider
Fleet Mandate. Report to Congress, Prepared in Compliance with Sections 704 and 1831 of the Energy Policy Act
of 2005 and Section 501 of the Energy Policy Act of 1992.
51
EPA (2010), Renewable Fuel Standard Program (RFS2) Regulatory Impact Analysis, page 252.
52
For a review of estimated rebound effects see: Small, Kenneth and Kurt Van Dender (2007). Fuel Efficiency and
Motor Vehicle Travel: The Declining Rebound Effect. Energy Journal, Vol. 28, No. 1, pp. 25-51.
53
20
28
54
Ibid, page 25379.

Ibid, page 25517.
56
Ibid, page 25519.
57
US EPA (2008). Advance Notice of Proposed Rulemaking: Regulating Greenhouse Gas Emissions under the
Clean Air Act. Docket number EPA-HQ-OAR-2008-0318, June 2008.
58
McCarthy, James (2007). CRS Report for Congress: Californias Waiver Request to Control Greenhouse Gases
Under the Clean Air Act. Congressional Research Service, August 20, 2007, pp. CRS 3-4.
59
Clean Air Act, Title II, 209(b).
60
McCarthy (2007), pp. CRS1-3
61
Ibid, pp. CRS 4.
62
63
Goulder, Lawrence H., Mark R. Jacobsen, and Arthur A. van Benthem (2012). Unintended consequences from
nested state and federal regulations: The case of the Pavley greenhouse-gas-per-mile limits. Journal of
Environmental Economics and Management, Vol. 63, No. 2, pp. 187-207.
64
65
Ibid, page 25330.
66
Ibid, page 25432.
67
Ibid, page 25433.
68
Ibid, pp. 25435-25436.
69
70
Ibid, page 48761.
71
US DOE, Alternative Fuels and Advanced Vehicles Data Center Benefits of Hybrid and Plug-in Electric
Vehicles. http://www.afdc.energy.gov/afdc/vehicles/electric_benefits.html
72
Interview with William Chernicoff, Energy & Environmental Research Group Manager, Toyota Motor North
America, Inc. 3/19/2012.
73
Lee, Henry and Grant Lovellette (2011). Will Electric Cars Transform the US Vehicle Market? An analysis of
the key determinants. Energy Technology Innovation Policy Discussion Paper Series, Discussion Paper #2011-08,
page 10.
74
75
Ibid, page 57125.
76
Harrington, Winston and Alan Krupnick (2012). Improving Fuel Economy in Heavy-Duty Vehicles. RFF Issue
Brief 12-01, March 2012.
77
Energy Policy Act of 1992, 303(b).
78
Ibid, 310.
79
Department of Energy (2008), page 17.
80
Ibid, page 28.
81
United States General Accounting Office (2000), page 10.
82
DOE (2011). Statement of Mr. Patrick Davis, Program Director, Vehicle Technologies Program, May 5, 2011.
http://www.international.energy.gov/sites/prod/files/ciprod/documents/EE_Final_Testimony(2).pdf
83
DOE (2008), page 13.
84
Ibid, page 2.
85
EPAct 2005, 701.
86
EISA 2007, 400FF (a)(2)
87
EPAct 1992, 501.
88
Ibid, 512.
89
Langer, Therese and Daniel Williams (2002). Greener Fleets: Fuel Economy Progress and Prospects. ACEEE,
Report Number T024.
90
DOE (2008), page 33.
91
California Environmental Protection Agency Air Resources Board. Zero-Emission Vehicle Legal and Regulatory
Activities Background, webpage: http://www.arb.ca.gov/msprog/zevprog/background.htm, Accessed 3/1/2012
92
Ibid.
93
California Environmental Protection Agency Air Resources Board (2009). California Exhaust Emissions
Standards and Test Procedures for 2009 and Subsequent Model Zero-Emission Vehicles and Hybrid Electric
Vehicles, In the Passenger Car, Light-Duty Truck and Medium-Duty Vehicle Classes. as of February 13, 2010 (last
amended December 2, 2009), incorporated by reference in Section 1962.1(h), Title 13, CCR.
55
29
94
California Environmental Protection Agency Air Resources Board (2012). Final Regulation Order: 1962.1 Zero
Emission Vehicle Standards for 2009 through 2017 Model Year Passenger Cars, Light-Duty Trucks, and MediumDuty Vehicle.
95
Information Administration (2011). Alternatives to Traditional Transportation Fuels, 2009, Table V1: Estimated
Number of Alternative Fueled Vehicles in Use in the United States, by Fuel Type, 2005 2009.
http://www.eia.gov/renewable/alternative_transport_vehicles/index.cfm
96
Ibid, Table V10: Estimated Number of Electric Vehicles in Use, by State and User Group 2009.
97
EPAct 2005, 1341.
98
The Auto Channel, January 4, 2006. Toyota Reports 2005 and December Sales.
http://www.theautochannel.com/news/2006/01/04/205039.html
99
ARRA 2009, 1141.
100
Ibid, 1143.
101
Department of Transportation/NHTSA (2009). 49 CFR Parts 512 and 599: Requirements and Procedures for
Consumer Assistance to Recycle and Save Program.
102
Li, Shanjun, Joshua Linn, and Elisheba Spiller (2011). "Evaluating Cash-for-Clunkers: Program Effect on Auto
Sales and the Environment." Working Paper, Available at SSRN: http://ssrn.com/abstract=1594924
103
Mian, Atif R. and Sufi, Amir (2010). The Effects of Fiscal Stimulus: Evidence from the 2009 Cash for
Clunkers Program. Working Paper, Available at SSRN: http://ssrn.com/abstract=1670759
104
The Transportation Equity Act for the 21st Century, 1216 (a)(5).
105
Turnbull, Katherine (2005). Potential Impact of Exempt Vehicles on HOV Lanes. US Department of
Transportation Federal Highway Administration.
106
Shewmake, Sharon and Lovell Jarvis (2011). Hybrid Cars and HOV Lanes. Working Paper, Available at
SSRN: http://ssrn.com/abstract=1826382
107
Diamond, David (2009). The impact of government incentives for hybrid-electric vehicles: Evidence from US
states. Energy Policy, Vol. 37, pp. 972-983.
108
EPAct 1992, Title VII.
109
US Department of Energy, Clean Cities Program: http://www1.eere.energy.gov/cleancities/about.html
110
EPAct 2005, 743.
111
US Department of Energy (2008). Fuel Cell School Buses: Report to Congress.
112
EISA 2007, 131.
113
Ibid.
114
National Academy of Sciences (2007). Rising Above the Gathering Storm: Energizing and Employing America
for a Brighter Economic Future. National Academies Press.
115
US DOE (2009). DOE Awards $151 Million in Recovery Act Funding for ARPA-E Projects. SunShot
Initiative News, http://www1.eere.energy.gov/solar/sunshot/m/news_detail.html?news_id=15576.
30
The MIT Energy Initiative

MITEI pairs the Institutes world-class research
teams with key players across the innovation
spectrum to help accomplish two important
goals: improving todays energy systems
and transforming tomorrows global energy
marketplace.
MITEI is also a resource for policy makers and the

public, providing unbiased analysis and serving
as an honest broker for industry and government.
MITEIs educational offerings combine single-discipline
depth with multidiscipline breadth, making our
campus an energy learning laboratory. Through
research, analysis, and education, MITEI is working
to find answers to reinvent our energy world.
Linking science, innovation,

and policy to transform the
worlds energy systems
Energy Initiative
Massachusetts Institute of Technology
Mailing address
77 Massachusetts Avenue, E19-307
Cambridge, MA 02139 USA
Visiting address
400 Main Street, E19-307 (3rd Floor)
Cambridge, MA 02142 USA
617.258.8891
web.mit.edu/mitei

MIT Biofuel Report

Uploaded by

Copyright:

Available Formats

MIT Biofuel Report

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

MIT Biofuel Report

Uploaded by

Copyright:

Available Formats

Prospects for Bi-Fuel and

Flex-Fuel Light-Duty Vehicles

An MIT Energy Initiative

Prospects for Bi-Fuel and

An MIT Energy Initiative

about the report

Natural Gas (CNG) LDVs

natural gas and motor gasoline vehicles

flex fuel vehicles (FFVs) in range of E-15 to E-85

fuel vehicles fueled by gasoline/ethanol/methanol (GEM) mixtures.

MITEI Associates Program/Symposium Series

About The Report / SUMMARY FOR Policy makerS

7 Section 1 OVERVIEW AND SUMMARY

C. Comprehensive Summary of Vehicle-Fuel Options

1.The Evolving Fuels Context for Light-Duty Vehicles, Steven E. Koonin

2.Light-Duty Vehicles: A Discussion of the Evolving Engine, Vehicle,

4.Evolution of Alcohol Fuel Blends towards a Sustainable Transport Energy Economy,

5.Prospects for Flexible- and Bi-Fuel Light-Duty Vehicles; Consumer Choice

6.Regulations and Incentives for Alternative Fuels and Vehicles

The MIT Energy Initiatives Symposium

Rationale for the Symposium

a greatly expanded alternative fuels market requires significant levels

Figure 1 Consumption of Alternative Fuels by Fuel Type, 2010

natural gas accounting for half

domestic natural gas resources create new market opportunities:

Framing the Scope of the Symposium

Establishing a Conceptual Framework of Alternative LDVs and Fuels

Create greater price competition between

Expand CNG and

Expand CNG and

Reduce petroleum demand and

CAFE credits, Open Fuel Standard, RFS measures, financial incentives

Table 1 Types of Alternative Fuel Light-Duty AFVs17

Dedicated mono-fuel AFV

fuel options: In the introductory presentation, it was proposed that the

Key Issues Arising from the Symposium

Figure 3 Relationship among the Prices of Various Alternative Fuels

US Average Retail Fuel Prices

Cost per GGE

Source: Clean Cities Alternative Fuel Price Reports, http://www.afdc.energy.gov/fuels/prices.html

Guide to the Report

Only one fuel

Each fuel separately

Liquid fuel mixtures

Optimized for E85

Dedicated Mono-Fuel Vehicles

Economics of Dedicated NGVs

(Miles per Gallon (MPG)

Honda Civic GX (dedicated CNG)

Honda Civic LX (dedicated gasoline)

Regional NGV Distribution 2010

All metal (aluminum or steel)

Metal liner partially reinforced by

Metal liner reinforced by composite

Plastic gas-tight liner reinforced by

Figure 6 Components to Convert and Operate Conventional

The CNG fuel delivery system is

Vehicle Technology Modifications39

Fuel system electrical connections and

Internal engine parts: Piston

C. Comprehensive Summary of Vehicle-Fuel Options

1.The Evolving Fuels Context for Light-Duty Vehicles, Steven E. Koonin

2.Light-Duty Vehicles: A Discussion of the Evolving Engine, Vehicle,

4.Evolution of Alcohol Fuel Blends towards a Sustainable Transport Energy Economy,

5.Prospects for Flexible- and Bi-Fuel Light-Duty Vehicles; Consumer Choice

6.Regulations and Incentives for Alternative Fuels and Vehicles

The rationale for this two-tank