A Governmental View of Oxygenates for Use As
Motor Fuels and Motor Fuel Components
for the
1983 Miayear Refining Meeting of tne API
oy Charles Gray, Jr.
Office of Mobile Sources
Environmental Protection Agency
May 11, 1983

The Office of Mobile Sources within the Environmental
Protection Agency has studied and evaluated
alternative transportation fuels since its formation
in 1970. EPA's responsibilities under the Clean Air
Act also have necessitated a significant regulatory
role dealing with transportation fuels.	In
particular, Section 211 of the Clean Air Act requires
EPA to play a key role in the introduction of new
fuels and fuel additives. Perhaps most visible was
EPA"s role in the introduction of unleaaed gasoline to
permit the use of catalytic converters on 1975 and
later model year automobiles. More recently EPA nas
responded to a growing interest in the use of
oxygenates (in particular metnanol) for use in motor
vehicles and for blending with gasoline.
From a technology assessment perspective, EPA has been
motivated to investigate alternative transportation
fuels because of concerns over diesel engine
emissions. Diesel engines, which are becoming more
numerous in both passenger cars and trucks, produce
relatively high levels of particulate matter and
oxides of nitrogen. In a sernwe, fuel modification can
be considered as an emission control technique.
EPA studies have suggested that one alternative
transportation fuel stands out above all alternatives
from both environmental and economic perspectives;
that fuel is methanol. Not only is methanol
producible from a wide variety of raw materials,
including coal, but the production technology is
currently available and appears economical relative to
other synfuel processes. In addition, the use of pure
methanol in motor vehicles appears to have significant
potential to result in improved emissions and thermal
efficiency relative to alternative internal combustion
engine fuels. The results of EPA's latest technology
assessment are summarized in a recent report to
Cong ress [1] .
Tnere has also been considerable interest in using
alternative fuels in blends to extend petroleum
supplies and raise octane ratings. Based upon the
information currently available to EPA, the only
prime-component oxygenates which seem potentially
practical as motor vehicle fuel components are ethanol
and methanol. Recent interest has concentrated on the
use of methanol as a prime additive to gasoline.

Methanol as a Motor Fuel Component
Methanol, usually along with cosolvents and sometimes
certain inhibitors, is oeing seriously considered as a
motor fuel component. Most of the current attention
deals with methanol as a gasoline component, although
EPA has performed some preliminary investigations into
the use of methanol as a diesel ruel component [2] .
Methanol concentrations in gasolines of up to about 10
percent are being considered. Clearly there are
certain advantages of such blends, such as an
economical means of extending petroleum reserves ana
lower carbon monoxide emissions. However, there are
several potential problems with methanol/gasoline
blends including: 1) higher evaporative emissions, 2)
poor vehicle dnveability due to an enleanment effect,
3) material compatibility problems, 4) phase
separation potential and 5) further stress on
distillate availability.
Higher evaporative emissions may be a result of
short-term or long-term effects. Higher short-term
evaporative emissions would likely result from
methanol blends due to the increased low-end
volatility characteristics o&wnethanol blends, unless
appropriate volatility adjustments are maoe.
Long-term evaporative emission concerns are associated
with the possibility that methanol may reduce the
effectiveness of the charcoal canister in the
evaporative emission control system. Poor vehicle
driveability may be experienced with some vehicles
calibrated near to or lean of stoichiometric due to
the additional enleanment effect of methanol/gasoline
blends. While certain 1981 and later model year
vehicles have closed-loop fuel control and may avoid
lean dnveability problems, concern remains for the
otner vehicles. The materials compatibility concern
stems from methanol's deleterious effect on certain
metals and elastomers. The materials compatibility
problem can be either a short-term or a long-term
concern. The tendency of phase separation m
methanol/gasoline blends is well known, and can be
initiated by small amounts of water. In addition to
possible safety problems (i.e., vehicle stalling),
phase separation woula also worsen materials problems
for those materials in contact with the methanol
phase. Finally, using methanol in gasoline may
exace rbate a potential future problem of distillate
availability. Extending gasoline supplies alone would
reduce the amount of oil refined and thus reduce

distillate production. There is a possibility, with
trie increased use of diesel engines in transportation
and other demands on distillate fuels, that extending
gasoline supplies through the use of methanol blenas
may be councerproductive.
EPA1s involvement in considering these potential
problems is limited to the provisions of Section
211(f) (1) of the Clean Air Act which prohibit the
lntroauction of any new fuel or fuel additive for
general use in 1975 and later model year automobiles,
unless the EPA Administrator waives the restriction by
determining that the new fuel or fuel additive will
not cause or contribute to a failure of the vehicle
emission standards.	EPA has granted several
methanol/gasoline waivers and intenas to continue to
work with interested parties to help define the
constraints on the appropriate use of methanol as a
motor fuel component.
Neat Methanol as a Motor Fuel
As mentioned previously, EPA's Office of Mobile
Sources has studied and evaluated alternative
transportation fuels for several years. The remainder
of this paper: 1) presents the highlights of an
analysis which led the Agency to conclude that
methanol is the most promising candidate future
transportation fuel, 2) updates the results of certain
recently completed and ongoing EPA engine/vehicle test
programs, and 3) presents and discusses some concerns
associated with the widespread use of methanol as a
transportation fuel.
Extensive discussion would be necessary to properly
evaluate all important considerations for tne various
candidate alternative transportation fuels, but
because of the brevity required for this paper, it
must suffice to focus on the highlignts which have led
EPA technical analysis to dace to conclude that
metnanol appears to be the most promising future
surface transportation fuel. How the cransition from
the petroleum era to the methanol era can occur and
when it will or should occur are important questions
but are also beyond the scope of this paper.
This paper will begin with the technology and cost of
producing methanol relative to other synthetic fuels
from coal, our primary domestic energy resource. Only
the use of coal is considered here because it will be

the key to any syntnetic fuels program which intends
to significantly decrease our dependence on foreign
petroleum. From a broader raw material availability
perspective, methanol is also attractive. Methanol is
proaucea by reacting carbon monoxide and hydrogen over
a catalyst, ana processes have been in commercial use
for many years. The question of whether a given
material can be converted into metnanol is actually
one of whether it can be converted into carbon
monoxide and hydrogen. Because partial oxidation of
almost any carbonaceous material in the presence of
steam can be made to produce carbon monoxide and
hydrogen, there are a great many raw materials which
can be converted into methanol. These induce wood,
peat, municipal solid waste, natural gas, and all
grades of coal including lignite. This range of
possiole raw materials may offer the long-term
advantage of a geogr apmcally aiverse methanol fuel
industry, since various feedstocks are spread across
tne country. However, the cost of methanol and other
indirect liquefaction products from domestic resources
should be significantly less from all grades of coal
than from the other raw materials. Thus, m the
near-term, coal would likely be the raw material of
choice. While direct liqi^faction processes can
process subbituminous (Western) or bituminous
(Eastern) coal, the economics are such that bituminous
coal would likely be used. A major advantage of
methanol over direct liquefaction products is that its
economics appears to be less dependent on coal type.
Production and Distnoution
The chemical industry already produces methanol from
natural gas and residual oil, so the technology to
produce methanol from synthesis gas is fully
commercial. Methanol is produced from coal via
indirect liquefaction, a two-step process consisting
of first gasifying coal into carbon monoxide ano
nydrogen and then synthesizing this gas into
methanol. The second step of the process, methanol
synthesis, is tne same regardless of the feeastock
used to produce the synthesis gas. Coal gasification
is also commercially proven, as first-generation
gasifiers have been in operation tor thirty years.
However, more efficient second-generation gasifiers
have been under development for over twenty years and
a number of these gasifiers now appear to be ready for

commercialization. Thus, the production of methanol
from coal appears to be achievable today and only
waiting for the proper economic conditions.
Other indirect liquefaction processes also are
commercial or near commercial. The Fischer-Tropsch
process is Definitely commercially proven, as
full-scale planes are currently operating in South
Africa using first-generation coal gasifiers. The
other indirect process, the t4obil MTG process,
converts methanol into gasoline. Up to the methanol
conversion step, the technical feasibility of this
process is the same as that for a methanol-f rom-coal
facility, which was already described above. The
final methanol-to-gasoline step is not as commercially
ready, however, having oeen only demonstrated in small
pilot plant units.
Direct liquefaction processes, on the other hana, are
a numoer of years away from commercialization. Large
pilot plant work is currently underway and progress is
being made. However, significant technical problems
still remain to be overcome. In addition, many of the
key processing steps have not yet been integratea, but
have only been tested individually. Thus, most plans
call for a large-scale demonstration plant to be built
to develop confidence in the entire process before any
commercial plants would be planned. Overall, the
direct liquefaction processes are not in the same
state of commercial readiness as are the indirect
liquefaction processes.
A variety of individual coal gasifier types are
available but there are several properties of coal
which are important in the selection of an individual
gasifier. Without going into the aetails of the
various gasifiers and coal properties and tneir
interactions, four processes were chosen from the many
considered for detailed economic analysis. These were
the Texaco, the modified WinKler, the modified Lurgi,
and che Koppers-Totzek processes. Synthetic gasoline
was studied from methanol via the Mobil MTG process.
And the Fisctier-Tropsch process was usea as a final
means for producing transportation fuel via indirect
1ique faction.
Like indirect liquefaction, numerous direct
liquefaction processes could be evaluated; however,
the three direct liquefaction processes chosen for
detailed economic analysis were the Exxon Donor

Solvent (EDS) , the H Coal, and the Solvent Refined
Coal-II (SRC-II) processes.
Various results of the EPA study [1] will oe sum-
marized in this paper. First, trie thermodynamic
conversion efficiencies and product mixes for several
coal conversion technologies are presented in Taole
1. An effort has been made to put all of the
efficiencies and other information presented in these
comparisons on the same, and most comparable, basis.
Further, the product slates were set up to maximize
the amount of useable transportation fuel. The two
efficiencies shown for methanol in Table 1 represent
the range of efficiencies expected for the nine
different gasifler/synthesis combinations considered
in the EPA study. It is worth noting that the
efficiency of producing methanol can be significantly
improved by coproaucing substitute natural gas (SNG),
however, as mentioned earlier, a major criterion was
to maximize the production of transportation fuels.
Of course, further improvements in all the processes
are possible as optimization work continues. Table 1
shows that direct liquefaction processes are projected
to be somewhat more efficient than processes which
would produce methanol. however, the methanol
production processes yield 100 percent methanol which
is considered a premium transportation fuel whereas
portions of the direct liquefaction product slates are
not suitable transportation fuels.
Recalling that diesel-fueled engines are more
efficient than gasoline-fueled engines and that the
current transportation fuel demana is shifting
progressively toward diesel fuel, a synthetic gasoline
product may not be most desirable. It is also
important to note that the distillate produced by
direct liquefaction will probably not be available for
use as diesel fuel. Coal liquefaction distillates
need to oe severely hyorotreateo to reduce the
aromatic content to a sufficient degree (less tnan 25
percent aromatics) so that a cetane numoer of 36-39
can oe obtained. These cetane numbers are still lower
than the minimum ASTM specification for diesel fuel of
40, and well below the current national average of
46. Therefore, additives to boost the cetane number
would likely be required even before severely
hydrotreated coal distillate could be used as diesel
fuel. This would be more severe hydr ot reating than
indicated in Table 1, and would therefore lower the
indicated efficiencies and yields.

There have been many studies undertaken in the last
few years to project the cost of producing synthetic
fuels from coal. However, the conclusions from these
studies vary widely and make generalized conclusions
about the relative economics of producing one
synthetic fuel versus another very difficult. This
variability appears to be due to a number of reasons.
One, the economic bases usea for the various studies
often differ, affecting cost by as much as 100
percent. Two, each study uses the best information
available at the time of that study. Since the
product mixes, efficiencies, and costs or many of
these processes, especially the direct liquefaction
processes, change frequently as more is understood
about each process, studies performed even two or
three years ago cannot be compared directly to the
latest studies. This is especially true in cases
where information was originally nonexistent ano
assumptions had to be made.
The EPA study attempted to focus on the original
engineering studies to assess the reasonableness of
the various cost estimates. Tne available designs of
each process were compared to ascertain which ones
were outdated or based on ^inaccurate assumptions.
After performing this "technical screening", placing
everything on the same economic basis, and adjusting
for plant size, surprisingly good agreement within
each process was found. Where there are still
differences, these differences can usually be
explained oy differences in the process technologies
For the presentation of results, product fuel costs
were normalized to two sets of economic and financial
parameters. One set uses an overall annual capital
charge rate (CCR) of 11.5 percent, representing
utility financing, and the other an overall annual CCR
of 30 percent to represent a private financing
option. Sacn plant was also scaled up or down to a
production level or 50,000 fuel-ol1-equlvalent barrels
per day, all results were expressed in 1981 dollars,
and all common costs were normalized (for coal,
utilities, etc.). As an important caveat, the final
answer will, of course, remain unclear until
commercialization of these processes, and the
conclusions presented here must be considered in this

Determining the economics of production of useable
synthetic liquid fuels is prooably the most difficult
of tne three economic areas; the other two being the
distribution economics and the economics of use.
However, the differences and uncertainties surrounding
the economics of production for the various synthetic
liquid fuels appear to be minor compared to the
economic differences associated with the end-use of
each fuel. (This will become clear later when the
engine efficiencies and costs for the various
alternative transportation fuels are considered.) It
is most important to consider the overall
efficiencies, economics and environmental impacts from
the base energy form/location through final end-use
(for transportation, passenger-miles or ton-miles per
energy unit) when considering choices among various
alternative transportation fuels. By computing the
cost of producing and distributing each synfuel on a
per energy basis (dollars per million Btu), the
combined cost at vehicle delivery will appropriately
reflect all costs up to that point. Then the cost ana
efficiency of utilization will allow the final
comparison of overall economics.
The costs of producing finisi»d products from several
of the synthetic fuel processes studied are presented
in Table 2. First, the capital cost for each process
was determined and then the product cost based on the
two different capital charge rates was determined.
Finally, the cost of refining and upgrading the
synthetic crudes is added, and the final product costs
are shown in Table 2 along with the resultant product
mix. The overall costs were allocated among the
various products in a aianner that would simulate their
market demand. The conclusion from the available data
is that, in general, it appears that the indirect coal
liquefaction processes can produce useable
transportation fuels at a lower cost than direct
liquefaction technologies.
Since distribution systems already exist for yasolme,
tne short-term distribution economics would, of
course, favor the continued use of this fuel over the
introduction of methanol. Gasoline also has about
twice the energy density of methanol. Thus, because
transportation cost depends primarily on volume,
gasoline would also be less expensive to transport per
Btu on a long-term basis.

Analysis of fuel distribution costs were diviaed into
three parts: 1) cost of distribution from refinery or
plant gate (if no refining is required) to the
regional distributor, 2) cost of distribution from the
regional distributor to the retailer, ana 3) cost of
distribution by the retailer (i.e., at the service
station) . Our estimates of the total cost for
distributing methanol are $1.69 to $1.88 per million
Btus, and for gasoline $1.34 per million Btus.
Gasoline has a significant advantage over methanol in
terms of percentage, but the absolute difference is
only $0.35-0.54 per million Btus.
All of the methanol and synthetic gasoline production
processes will affect the environment to some extent,
as will the distribution and storage of the fuels
themselves. Coal contains many elements and compounds
in addition to hydrogen and carbon, such as organic
nitrogen-containing compounds, organic and inorganic
sulfur, trace metals, etc. The conversion of coal to
other fuels offers a number of opportunities for these
pollutants to reach the environment in harmful ways,
regardless of the particular conversion process used.
Focusing primarily on the differences between indirect
and direct liquefaction, one potential advantage of
gasification over direct liquefaction is the fact that
most of the organic nitrogen and sulfur is broken down
into simple compounds like ammonia and hydrogen
sulfide. Tnese are relatively easy to separate from
the carbon monoxide and hydrogen which make up the
ma]or parts of the synthesis gas. Also, since the
carbon monoxide and hydrogen entering the methanol
synthesis unit must be essentially free of sulfur to
prevent rapid catalyst deactivation, there is a
built-in economic motivation for its removal.
Direct coal liquefaction, on the other hand,
inherently leaves some of the sulfur and nitrogen in
the liquid phase, found with the organics. Tne most
effective cecnnique to remove tnese elements is
hyorogenation, which also is used to upgrade the
fuel. However, nydrogenation is expensive, because of
the large amounts of hyorogen consumed, and will
likely be limited to only the degree that is necessary
to market the fuel. If tne fuel is upgraded to
gasoline or high quality fuel oil, most of the sulfur
and nitrogen will be removed and cnere should not be
any significant sulfur or nitrogen emission problems.
However, that portion of the synthetic crude which may

be burnea with little or no refinement could contain
relatively high levels of these elements and
represents more of an environmental concern than
gasification products.
Another aifference in the potential environmental
effects is in the exposure to the fuels themselves.
While coal liquids are for the most part hydrocarbons
and similar to petroleum, they have a higher aromatic
content and some may contain significant quantities of
polycyclic and heterocyclic organic compounds. Some
of these compounds are definitely mutagenic in
bioassays and many have produced tumors in animals.
Thus, while the noncarcinogenic nealth effects of
these materials would be similar to those of crude
petroleum, they would definitely have the potential to
be more carcinogenic. There is also some evidence
that much of this bioactivity can be removed by
moderate to severe levels of hydrogenation which would
occur if high grade products were produced. Thus,
again the potential hazard is dependent upon the
degree of hydrogenation given the products. Indirect
liquefaction products, on the other hand, do not
appear to exhibit mutagenicity or carcinogenicity.
Another point which deserves mentioning is the
difference between tne effect of an oil spill and a
methanol spill. The effects of oil spills are well
known. The effects of a metnanol spill are expected
to be quite different, primarily because methanol is
soluble in water. While high levels of methanol are
toxic to fisn and fauna, a methanol spill would
quickly disperse to nontoxic concentrations and,
particularly in water, leave little trace of its
presence afterwards. Water life should De able to
migrate back quickly and plant life should begin to
grow back quickly, though complete renewal would take
the tune necessary for new plants to grow.
Tnere are various concerns with the handling and use
of methanol, and these will be discussed in a later
section of the paper. While more work clearly needs
to be performed in this area, the production of
synthetic fuels from coal via indirect liquefaction
processes appears to offer environmental advantages
over the direct: liquefaction processes.

Use of Methanol in Motor Vehicles
First, the final closure on the economics of using
methanol in motor vehicles versus synthetic gasoline
will be presented. Second, the latest information on
emissions and performance (both energy efficiency and
specific power) of methanol-fueled engines as compared
to gasoline and diesel-fueled engines will be
presented, focusing primarily upon EPA's own test
The two primary effects on the economics of using a
given fuel in an engine are the efficiency of the
engine and its cost, including its emission control
system cost. In keeping with the conservative
approach used in this paper when comparing methanol to
gasoline from direct coal liquefaction, a fuel
efficiency advantage from methanol engines over their
gasoline counterparts of only 20 percent is used. As
will be discussed later, the methanol engine may well
have a fuel efficiency equal to (and perhaps
exceeding) that of a diesel, which is 25 to 30 percent
better than a gasoline engine. A final conservative
assumption that will be used is that there will be no
difference in the cost of a methanol engine as
compared to a gasoline engine. As will also be
covered in part later, such factors as significantly
higher specific power, reduced cooling system
requirements, and lower emission control system cost
of the methanol engine would likely give it a
significant cost advantage as compared to the gasoline
engine. Using these assumptions, a fuel economy of 30
mpg for the average gasoline-fueled vehicle and an
average usage of 12,000 miles per year, the annual
fuel savings relative to synthetic gasoline produced
via indirect liquefaction (Mobil tMTG process) were
determined. Table 3 presents the final vehicle use
economics. It should also be notea that the
information presented in Table 3 assumes that all
fuels were derived from bituminous coal. If a lowest
feedstock-cost approach had been taken, then the
economic benefits of methanol (and indirect gasoline)
as compared to direct liquefaction gasoline would have
been even greater than shown m Table 3.
Perhaps the most significant conclusion of this
analysis is that the efficiency and cost of the
optimum engine for each of the various alternative
transportation fuels may have an even greater
influence on the overall economics than significantly

different production efficiencies and delivered fuel
prices. Based on the information currently available,
methanol appears significantly superior to direct coal
gasoline from both the delivered fuel price and the
overall economics perspectives. Finally, it snould
again be stated that no comparison was made between
methanol and diesel fuel since none of the coal
conversion processes examined produces diesel fuel of
sufficient quality for today's diesel engines.
Methanol can be burned in a wide variety of engines.
EPA has both reviewed the results of other testing
programs [3,4] as well as begun its own engine
evaluation program [5,6,7]. There are several
features of methanol that make its use in internal
combustion engines especially attractive, and these
characteristics are discussed in previous EPA
reports. This paper will focus on the results of some
of the most recent EPA programs to characterize the
emissions and efficiency of methanol-fueled engines
and vehicles.
Methanol-fueled engines promise improved emission
characteristics over gasoline and diesel engines in a
number of areas. Especially important are low
emissions of nitrogen oxides (NOx) and an absence of
emissions of particulate matter, heavy organics and
sulfur-bearing compounds. These emissions advantages
may become even more significant in the future, as the
quality of crude oil, and diesel fuel in particular,
declines. This is especially important for diesel
fuel, since a reduction in quality will likely result
in an increase in emissions, especially particulate
matter .
Another potential benefit of methanol is the
possibility that the exhaust catalyst could be of the
oase metal variety, such as copper, chromium, or
nickel, and not require noble metals, sucn as platinum
and palladium. Unlike gasoline, methanol does noc
contain any sulfur or trace lead, which degrade base
metal catalysts. This could significantly reduce the
cost of the catalytic converter system. This change
could also improve the country's balance of payments,
since all noble metals must currently be imported.
While some base metals are also imported, their value
would be significantly less ana still produce a net
decrease in imports. Of course, it may also oe
possible that a noble metal catalyst with
significantly less noble metal loading could still be

effective. In either case, a significant potential
exists to use a lower cost catalyst with
methanol-fueleci vehicles.
Methanol engines, however, generally emit more
aldehyde emissions than gasoline or diesel engines,
including formaldehyde, a suspected carcinogen.
Uncertainties exist as to whether the increases are
significant, since the current level of formaldehyde
emission from gasoline and diesel engines is not now
thought to present a problem. Fortunately, research
testing of catalytic converters has shown that they
are able to remove up to and perhaps more than 90% of
tne aldehydes. Reducing engme-out formaldehyde
emissions and optimizing catalytic reductions will
undoubtedly be high priorities for automotive
Tables 4 and 5 present summary results from recent EPA
cnaracterization programs that have evaluated
methanol-fueled engines developed by others [5,6,7].
Although these are early prototypes and are basically
straightforward conversions of their gasoline/diesel
counterparts, they do provide EPA with the opportunity
to perform a preliminary assessment of the exhaust
composition of methanol-fueled vehicles. Exhaust from
each of the light-duty vehicles and heavy-duty engines
tested in these programs was analyzed for each of the
currently regulated pollutants: hydrocarbons, carbon
monoxide, oxides of nitrogen, and particulate. Tests
were performed for a range of other organic compounds
such as metnanol, formaldehyde, other aldehydes and
ketones, and some individual hydrocarbons. In
addition, several other unregulated pollutants
(nitrosamines and total organic amines, ammonia,
cyanide, and methyl nitrite) were measured during
tests on one of the light-duty vehicles. Phenols,
smoke, sulfate, ana benzo(a)pyrene were also analyzed
from the heavy-duty engines. The basic intent was to
identify any possible emission surprises as early as
possible witn respect to methanol-fueled engines. The
results of tne test programs compare well with what
would have been expected.
Since "hydrocarbons" are controlled because of their
role as oxidant precursors in the urban atmosphere,
the relative masses of organic emissions in gasoline
and methanol exhausts are not as important as the
relative photochemical reactivities of the organic
species. Since the exhausts from various alternative

fuels could have significantly different photocnemical
reactivities even at similar mass emission levels,
specific hydrocarbon (organic) analysis of exhaust is
important. For example, formaldehyde is known co be
very photochemically reactive, but unburned metnanol
itself is generally considered to be of low
photochemical activity. As shown in Table 4, the
light-duty methanol vehicles emitted more methanol and
formaldehyde emissions than their gasoline
counterparts, Qut the methanol exhausts contained
almost no alkenes, aromatics, or non-methane alkanes
whicn are the major reactive components of gasoline
exhaust. Thus it is not immediately clear whether
methanol exhaust would be more or less reactive than
gasoline exhaust. The organic emissions from the
heavy-duty methanol engine shown m Table 5 were
almost exclusively unburned methanol, with only trace
levels of formaldehyde and hydrocarbons, which would
probably result in lower photochemical reactivity
impacts compared to the baseline diesel engines. It
must be noced that the methanol heavy-duty engine was
equipped with an oxidation catalyst while the diesel
engines had no form of a£tertreatment.
Both the CO ana NOx result were mixed from tne
light-duty vehicle tests (Table 4), with methanol
sometimes increasing and sometimes decreasing these
emissions. However, as with all the emissions, both
the gasoline-fueled and methanol-fueled vehicles
emitted very low emission levels, as they were all
calibrated to meet the very stringent future
California emission standards. Therefore, considering
the level of control, the differences observed are
quite small in absolute terms. One significant
observation is that a trade-off seemed to be occurmg
between NOx emissions and energy efficiency. Although
optimum methanol combustion would be expected to occur
at very lean operation, both of the methanol
light-auty vehicles utilized tnree-way catalytic
converters and thus were forced to operate at very
near stoichiometric conditions so that the NOx
reduction catalyst would perform well. The
rne thanol - f uelea Escort showed no significant
difference m energy efficiency at very low NOx
levels, while tne methanol-fueled Rabbit showed a
significant increase in efficiency but at higher NOx
levels. The methanol-fueled Escort was calibrated
with more spark retard than the Raobit thus resulting
in lower NOx and relative efficiency. Emissions of
particulate and unregulated compounds such as ammonia,

cyanide, and organic amines are consistently lower
from the me t'nanol-f ueled vehicles than from similar
gasoline-fueled vehicles. No surprises appeared in
the extensive characterization program of the
methanol-fueled vehicles. however, the optimum
emissions and efficiency configurations for
methanol-fueled vehicles may very likely be different
from the vehicles tested in this program.
EPA has several contracted methanol engine
optimization programs currently underway including
projects with Ford Motor Company to optimize the Ford
PROCO engine with methanol and with Ricardo Consulting
Engineers to evaluate methanol performance with the
Ricardo HRCC (high compression ratio, compact chamber)
combustion system when fueled with methanol, as well
as an m-house program at the Ann Arbor laboratory.
While the m-house program is only a modest effort,
preliminary results from the program are encouraging.
The m-house program is first focusing on the
determination of optimum engine efficiency and
emission configurations, with the second phase to be a
final emission control system integration and vehicle
testing. A parallel methanol exhaust aftertreatment
project is also underway and-^relirainary results from
this effort will be presented later. It is also
important to point out that development/optimization
activities for methanol engines are in a very early
pnase, as compared to gasoline and diesel engines.
Therefore, it is reasonable to expect additional
significant improvements in any methanol efficiency or
emission results that are now available.
Figures 1 and 2 present thermal efficiency values at
two different power settings from one of the engines
being evaluated in the EPA m-house program (the
Nissan NAPS-Z) compared to several advanced technology
passenger car engines. (Other m-house engines show
similar results with methanol, but the NAPS-Z engine
has more versatility in parameter control and is
receiving most of our experimental focus.) Tne NaPS-Z
engine represents an advanced technology Otto-cycle
engine, having both low friction and fast burn
characteristics. Tne Volkswagen Rabbit turbocharged
diesel engine is presented to represent
state-of-the-art indirect injection passenger car
engine tecnnology. The turbocharged, direct injection
Sofim diesel engine is representative of the best
efficiency characteristics in research passenger car
diesel engines. as can be seen, the methanol NAPS-Z

is more efficient than any of the other advanced
technology engines shown. The methanol results from
the NAPS-Z engine, while very impressive, represent
only simple compression ratio changes from the
gasoline engine and parameter optimization for
methanol fuel. Several alternative configurations are
being evaluated for methanol wnich include
water/methanol blends for reduced NOx emissions and
possibly improved efficiency and specific power,
partial methanol dissociation for increased efficiency
and reduced emissions, and turbocharger optimization.
The methanol-fueled NAPS-Z engine also produces about
25 percent more power than its gasoline
Figures 3 ana 4 present preliminary results from the
EPA/Ricardo contract. These results come from a 1.5
liter HRCC engine Ricardo prepared for EPA. The
engine has a 13:1 compression ratio, with methanol
results being generated with the engine equipped with
a very simple single barrel carburetor while the
gasoline results are taken from the same engine fitted
with an improved two carrel carburetor. The part-load
conditions presented are within the range of operating
conditions commonly encountered in a normal vehicle
application. Although these results are from the same
engine, methanol operation generally produced
efficiency improvements of 4 to 6 percent over the
gasoline fuel. These gains may be due to the effects
of lower combustion temperatures when using methanol,
the greater number of moles of combustion products per
mole of fuel for methanol, and/or the faster
combustion characteristics of methanol {higher flame
speeds) . The NOx emissions from methanol are
significantly lower than with gasoline, and achieve
very low levels at lean equivalence ratios where
methanol combustion still appeared stable. In
contrast, even with the HRCC engine, gasoline
combustion appeared unstable at equivalence ratios
lean enough to get significantly lowered NOx levels.
A significant issue for methanol-fuelea engines
designed for very low NOx levels will be whether good
vehicle dnveaoility can be achieved when operating at
an equivalence ratio near 0.7.
Another interesting result of the Ricardo study is the
assessment of idle performance. Figure 5 presents
idle fuel consumption and hydrocarbon emission results
for several engines. The curves for each engine
represent different spark timing and equivalence ratio

settings. Of significance here is that the
methanol-fueled HRCC engine idle fuel consumption is
one-half tnat of the standard production gasoline
engine, and within the range for production aiesel
If methanol-fueled engines can operate satisfactorily
at lean equivalence ratios for optimum efficiency and
very low NOx emissions, then oxidation catalysts
(perhaps with some EGR) may be sufficient to meet
stringent emission standards. As was mentioned
previously, it may also be possible to use low cost
oxidation catalysts with methanol-fueled vehicles.
The m-house test program evaluating alternative
catalysts is using the me thanol-fueled Volkswagen
Rabbit, characterized in its stock configuration in
Table 4. Table 6 presents preliminary summary aata on
several alternative catalysts. Early results are
encouraging .
Concerns Regarding Methanol Use
Several studies have been performed which assess the
environmental effects of synfuel use [1,8,9].
However, wnile EPA is con-rtnumg to evaluate the
environmental effects of the various alternative
transportation fuels production processes, this
discussion will focus on the concerns with the use of
methanol as a motor vehicle fuel. The ciiscussion will
address specific exhaust emission concerns as well as
general health and safety concerns associated with the
nandling and use of methanol. The general concerns
include: 1) the ingestion, inhalation and skin
absorption of methanol, 2) the almost invisible flame
produced when methanol burns, and 3) that the methanol
vapors in the fuel tank are likely to be lgnitaole.
Exhaust concerns include: 1) the reactivity of the
methanol exnaust, 2) the inhalation of unburned
methanol in the exhaust, and 3) che formaldehyde
levels in the exnaust.
Methanol is highly toxic in heavy exposures, leading
to blindness or death. Much of its notoriety in this
area is due to people confusing it with ethanol ana
drinking it in large quantities. Hydrocarbon fuels,
while also being toxic, do not suffer from this
confusion and are not often taken internally. The
absorption of methanol through the skin is also
hazardous, more so than gasoline (though the presence
of benzene in gasoline complicates this issue). Given

the public's ratner careless use of gasoline,
widespread use of methanol would pose some concerns in
this area. EPA has recently performed a preliminary
study on the healtn effects of methanol [10], and
results from that study will be summarized nere.
Methanol was a frequent component of liniments,
perfumes, and some patent medicines well into this
century. It has seen widespread use as a solvent and
as a high volume chemical. While bioassay studies
with methanol are limited, current results suggest
that methanol is not mutagenic. Methanol can be
introduced into the body through ingestion, inhalation
and skin absorption, and the dose rather than tne mooe
of introduction seems to be the most important
parameter in expected health effects. Defining safe
exposure and dose levels is extremely difficult
because of the very large variability in individual
response. In addition, the toxic effects are
generally delayed (twelve to eighteen hours after
dose) and may be as severe as blindness and even
death. If death does occur, it is generally
attributed to respiratory failure. To illustrate the
wide variability and individual sensitivity to
methanol ingestion, one dea^ was reported in the
literature from a dose of only 6 milliliters of
methanol (about 1 teaspoon) while recovery has been
reported for a 500 milliliter dose. Receiving a
significant methanol dose through skin absorption is a
real concern as the skin absorption rate is reported
to be 0.2 milligrams per square centimeter per minute.
Inhalation studies have shown a wide range of
effects. Headaches have been reportedly caused from
inhaling methanol concentrations as low as 22
milligrams per cubic meter. The odor threshold may be
as low as 5 milligrams per cubic meter, although some
studies indicate that the odor threshold is much
higner. The OSHA work place standard tor methanol is
200 milligrams per cuoic meter maximum. The threshold
limit value (TLV) is 260 milligrams per cubic meter,
with a 310 milligrams per cubic rueter short-term
limit. Tne Soviet Union has established a maximum
allowable concentration in work place air of 5
milligrams per cuoic meter; and 1.0 milligrams per
cubic meter one-time limit and 0.5 milligrams per
cubic meter average limit in populated places.
Because of the wide variability in response, lower
limits for the ranges of concern for methanol of 0.1
to 1.0 milliliters for ingestion and 1.0 to 3.0

milligrams pec cubic meter for inhalation nave been
suggested [10J. Skin contact should also be avoided.
While some treatment is available tor methanol
ingestion, it seems prudent that every reasonable
effort should be made to avoid methanol ingestion,
skin contact and prolonged inhalation. Several
actions can be taken to address these concerns.
Public education regarding the risk of methanol
exposure would be fundamental. Use of the term methyl
alcohol should. probably be avoided; perhaps a
non-alcohol fuel name would be wise. Warning labels
on dispensers and containers would need to be reviewed
for effectiveness.
Another concern with methanol use is that the methanol
flame is essentially invisible in daylight. Although
methanol flames are cooler tnan gasoline flames and
extinguish with water, methanol/water concentrations
down to about 26 percent methanol are still
flammable. On the other hand, methanol has a higher
flashpoint and ignition temperature than gasoline,
thus in open air it is considered less of a fire
nazard than gasoline.
Due to methanol's volatilt^ characteristics, the
vapors in a methanol tank would be expected to be
flammable. However, flame arresters in the tank
filler neck or bladder tanks seem capable of
mitigating this concern. Also, the electrical
conductivity of methanol is higher than for gasoline
which reduces the problem of static discharge causing
The possibility of using additives with methanol
should be investigated to help alleviate these
concerns. Methanol additives should be sought which
would: 1) cause vomiting if ingested, 2) make
ingestion, inhalation and skin contact unattractive
(e.g., unpleasant smell, or dye), and 3) provide a
visible flame.
Concerns with the emission products from methanol
combustion center largely around the question of the
reactivity of the exhaust and formaldehyde emissions.
The reactivity of the exhaust, wnich is primarily
methanol, was discussed earlier and still needs
further study. While most researchers agree that
methanol has a low reactivity and therefore little

potential for ozone formation, additional modeling and
smog chancer work should be conducted to resolve this
EPA has also performed a preliminary formaldehyde
nealch effects study [11]. Formaldehyde is a
photochemically active organic and may play a larger
role in methanol exhaust reactivity than tne emissions
of unburned methanol. Formaldehyde also has toxic
effects with a TLV of 3.0 milligrams per cubic meter.
The bottom end of a range of concern for formaldehyde
would probably be between 0.06 milligrams per cubic
meter and 0.2 milligrams per cubic meter [11]. In
addition, formaldehyde has been shown to cause cancer
in laboratory animals. Additional work needs to be
conducted to assess the potential impact of
formaldehyde emissions from methanol-fueled vehicles.
As mentioned previously, low formaldehyde exhaust
emission levels appear available with exhaust
catalysts. To put tne formaldehyde emission concern
in perspective, Table 7 shows formaldehyde emissions
for a variety of passenger cars. It can be seen that
formaldehyde emissions from conventional catalyst
equipped methanol-fueled vehicles are equal to or even
lower than formaldehyde emi^ions from non-catalyst
equipped gasoline-fueled vehicles, but higher than
formaldehyde emissions from catalyst equipped
gasoline-tueled vehicles.

Re fe re nee s
1.	"Preliminary Perspective on Pure Methanol Fuel for
Transportation," Environmental Protection Agency,
Office of Mobile Sources, EPA-460/3-83-003,
September 1982.
2.	"Diesel Car Particulate Control Methods," Society
of Automotive Engineers Paper 830084, Charles M.
Urban, Larry C. Landman, and Robert L>. Wagner,
March 198 3.
3. "The Utilization of Alcohol in Light-Duty Diesel
Engines," Contractor Final Report, Ricardo
Consulting Engineers, EPA-460/3-81-010, May 28,
4.	"A Brief Summary of the Technical Feasibility,
Emissions, and Fuel Economy of Pure Methanol
Engines," Internal Staff Report Jeff Alson,
EPA-AA-SDSB-8 2-01, December 1981.
5.	"Emission	Characterization	of	an
Alcohol/Diesel-Pilot Fueled Compression-Ignition
Engine and its Heavy-D-trt&y Diesel Counterpart,"
Contractor Final Report, Southwest Research
Institute, EPA-460/3-81-023, August 1981.
6.	"Characterization of Exhaust Emissions from
Methanol- and Gasoline-Fueled Automobiles,"
Contractor Final Report, Soutnwest Research
Institute, EPA-4 60/3-82-004, August 1982 .
7.	"Emission Characterization of a Spark-Ignited,
Heavy-Duty, Direct-Injected Methanol Engine,"
Contractor Final Report, Southwest Research
Institute, EPA-460/3-82-003, November 1982.
8.	"Environmental Aspects of Fuel Conversion
Technology - VI, A Symposium on Coal-Based
Synfuels" (October 1981), EPA-600/9-d2-017,
September 1982.
9.	"Effects of Synfuel Use," Masood Ghassemi, Ra^an
Iyer, Robert Scofiela, ana Joe McSorley,
Environmental Sclence and Technology, August 1981.
10. "Methanol Health Effects," Contractor Final
Report,	Midwest	Research	Institute,
EPA-460/3-81-032, December 31, 1981.

11. "Formaldehyde Health Effects," Contractor Final
Report,	Midwest	Research	Institute,
EPA-460/3-81-033, December 21, 1981.

Table 1
Process Efficiencies and Proauct Mix
of Various Coal Liquefaction Processes
Proce ss
Crude Petroleum
Solvent Refmea Coal
Indirect Liquefaction
Modified Lurgi Gas.
Lurgi Synthesis
Koppers Totzek Gas.
ICI Synthesis
Mobil MTG
Fixed Bed
Fluidizea Bed
Direct Liquefaction**
Exxon Donor Solvent
61. 8%
6 3.6%
Product Mix
(Energy Basis)
50% Gasoline
33% Distillate
15% Residual
2% LPG
100% Me OH *
100% MeOH*
88% Gasoline
12% LPG
32.7% Reg. Gasoline
14.0% Prem.Gasoline
25.6% No.2 Fuel Oil
9.6% LPG
18.1% SNG
33.1% Reg. Gasoline
11.2% Prem.Gasoline
20.4% No.2 Fuel Oil
22.3% LPG
13.0% SNG
b4.7% Gasol me
12.1% LPG
23.2% SNG
1-3% water
MeOH = 95-98% methanol,
remainder higher alcohols.
These efficiencies include the effect of refining
where needed. However, the
slates are not identical for
liquefaction processes.
refinery product
each of the direct

Table 2
Product and Capital Costs of Coal
Liquefaction Processes 11581 Dollars)
Product Cost
EDS (Bit.)
H-Coal (Bit.)
SRC-II (Bit.)
Product Mix
11. 5%
of Dollars)
32 . 7%
Reg. Gas.
$2. 65
14 . 0%
Prera. Gas.
18. 67

25. 6%
Fuel Oil
8. 20
14 .18

13 . 31

18. 1%
8. 00
13 . 83

Reg. Gas.
14 .97
11. 2%
Prem. Gas.

20 . 4%
Fuel Oil
12 .27

6. 00
11. 52

13 . 0%

64 . 7%
9. 87
IS . 06
14 . 68

23 .2%
7. 90
15. 24

Texaco (Bit.)
Koppers (Bit.)
Lurgi (Subbit.)
100% MeOH*
10 0% MeOH*
47.9% MeOH*
49.7% SNG
Incremental Cost 10-15% LPG
5. 90-
7. 04
9 .44-
10 . 41
12 .42

2.4% Gasoline
7. 04
12 . 48
riooifiea Winkier
100% MeuH*
9 . 56
(Ligm te)

Lurgi Mobil MTG
41.2% Reg. Gas.
8. 01
53.3% SImG
6. 41

5.5% LPG
11. 20
Mobil MTG
85-90% Reg. Gas.
2 .87
water , and the
* MeOH = 9 5-98% methanol,
remainder higher alcohols.
** Capital costs are instantaneous costs and do not
include refinery capital coses.
1.	99-
2.	59
2. 17
2 . 95

Cost at Pump*
($ per mBtu)
Annual Savings
(Or Cost)
Relative to
MTG Gasoline,
Table 3
Vehicle Use Economics
	Coal Liquefaction
Me thanol
(Mobil MTG)
Direct Coal
Liquefact ion
* Range of plantgate cost is the lowest cost using
tne low CCR and the highest cost using the high
CCR, all using bituminous feedstocks.
** Includes conservative assumptions for methanol: 1)
only 20% improved engine efficiency over
gasoiline, ana 2) the sant® engine cost as gasoline

Table 4
Average Light-Duty Venicle
Emission And Energy Efficiency Results
Federal Test Proceoure

Escor ts
VW Rabbits
Me th
Hydrocarbons (FID), g/mi
0.42 [c]
Carbon Monoxide, g/mi
6. 03
1. 08
0 . 88
Oxides of Nitrogen, g/mi
Particulate, mg/mi
6 . 3
11. 8
4 . 7
Methanol, mg/mi
ND [d]
Total Aldehydes and

Ketones, mg/mi
33 . 6
10. 3
Formaldehyde, mg/mi
33 . 0
10. 3
Total Individual

Hydrocarbons, mg/mi
Methane, mg/mi
14 . 0
Ethylene, mg/mi
0 . 3
Ethane, mg/mi
Acetylene, mg/mi
0 . 1
Propane, mg/mi
Propylene, mg/mi
Benzene, mg/mi
Toluene, mg/mi
17. 2
Nitrosammes , mg/mi
Ammonia, mg/mi
10. 0
Total Cyanide, mg/mi
Total Organic Amines,

he thy1 Nitrite, ppm
Energy Efficiency,
2 .25
mi/lC* Btu

[a] Gasoline-fueled.
[o] Methanol-fuelea.
[c]	Hydrocarbons as measured oy the FID (flame
ionization detector) and expressed as methanol.
[d]	None detected.
[e]	Analysis not conducted.

Table 5
Average Heavy-Duty Engine Emission
Ana Energy Efficiency Results Over
The Transient Federal Test Procedure


Me tnanol


No Cat
No Cat
Hydrocarbons,[a] g/hp-hr
Carbon Monoxide, g/hp-hr
3 . 01
Oxides of Nitrogen,Lb]
8 . 34
6. 61

Particulate, g/hp-hr
0. 52
Methanol, mg/hp-hr
Total Aldehydes ana

Ketones, mg/hp-hr
Formaldehyde, mg/hp-hr
Total Phenols, mg/hp-hr
Total Individual

Hydrocarbons, mg/hp-hr
Methane, mg/hp-hr
Ethylene, mg/hp-hr
Ethane, mg/hp-hr
Acetylene, mg/hp-hr
Propylene, mg/hp-hr
Benzene, mg/hp-hr
Sulfate, mg/hp-hr
Benzo(a)pyrene, ug/hp-hr
Fuel Consumption

lb diesel equivalent/hp-hr
0. 451
0. 538
Smoke, peak percent opacity
la 1
NA = Naturally Aspirated, TC = Turbocharged, IC =
Intercooled; the brake specific fuel consumption is
directly affected by the technology employed with
improvements expected in going from NA to TC to IC.
Therefore, direct efficiency comparisons are not
Hydrocarbons as measured by the HFID (heated clame
ionization detector) and expressed as diesel-like
No NOx correction factor used.
Does not apply.
No analysis performed.

Table 6
Methanol Catalyst Preliminary Summary
No Catalyst
Wtf Stock - 0% O2
VW Scoc,< - 3% O2
A - 0%
A - 3%
B - 0%
B - 3%
C - 0%
C - 3%
D - 0%
D - 3%
E - 0%
E - 3%
0 . 15
0 . 56
0. 19
0. 73
1. 80
0. 11
0. 69
0. 12
2 .01
0. 77
0 . 76
0 . 85
2 . 07
1. 99
0 . 74
2 . 85
1 .97
Results not complete.

Table 7
Comparison of Average FTP Formaldehyde Emissions
	for Various Light-Duty Vehicles	
HCHO Emissions
Gasoline-Fueled Vehicles
1981 3-way-catalyst equipped cars
1978, 1979 3-way-catalyst
equipped cars
1978 oxiaatlon-catalyst equipped cars
1977 non-catalyst cars
1970 non-catalyst cars
Methanol-Fuelea Vehicles
1981 3-way-catalyst equipped,
caroureted car (Escort)	33
1981 3-way-catalyst equipped,
fuel injected car (Rabbit)	10

Figure 1
M = NflPS-2
N . A.
2 . OL

N = 8

N . A .
2. OL

N = 8

5 = 5 OF I H
T/C 01
2 . SL

N = 7

V = vw
1 . 5L

N = 9

500	1 500	2500 3500	4500
1 000 2000	3000 4000	5000

Figure 2
M = NflPS-Z
N . A.
2. OL

N = 8

N . A.
5: ICR
2 . OL

N = 8

S = SOF IM T/C 01
8 1 0481
2. 5L

N = 7

V = VW
I ~ I
1 . 5 L

N = 9

— — -
500	1 500 2500	3500	4500
1 000	2000 3000	4000	5000

Figure 3

Figure 4


	* *	X

¦OX' —	&	
28 -|

Figure 5
-• 1.5L HRCC engine—Gasoline
-A 1-6L Standard engine—Gasoline
-X 1-5L HRCC Engine—Methanol
(Gasoline Energy-Equivalent results)
1.5-2.5 L Diesel
Fuel consumption (kg/h)