Subscriber access provided by UNIV OF MISSOURI COLUMBIA

Policy Analysis
Environmental Implications on the Oxygenation of Gasoline
with Ethanol in the Metropolitan Area of Mexico City
I. Schifter, M. Vera, L. Daz, E. Guzmn, F. Ramos, and E. Lpez-Salinas
Environ. Sci. Technol., 2001, 35 (10), 1893-1901• DOI: 10.1021/es001177w • Publication Date (Web): 10 April 2001
Downloaded from http://pubs.acs.org on March 14, 2009

More About This Article

Additional resources and features associated with this article are available within the HTML version:

• Supporting Information
• Access to high resolution figures
• Links to articles and content related to this article
• Copyright permission to reproduce figures and/or text from this article

Environmental Science & Technology is published by the American Chemical Society.

1155 Sixteenth Street N.W., Washington, DC 20036
 Policy Analysis
Environmental Implications on the
Oxygenation of Gasoline with
Ethanol in the Metropolitan Area of
Mexico City
I . S C H I F T E R , * M . V E R A , L . D IÄ A Z ,
E . G U Z M AÄ N , F . R A M O S , A N D
E. LO Ä PEZ-SALINAS
Subdirección de Proceso y Medio Ambiente,
Instituto Mexicano del Petróleo,
Eje Central L. Cárdenas 152, México, D.F., CP 07730
Motor vehicle emission tests were performed on 12 in-
use light duty vehicles, made up of the most representative
emission control technologies in Mexico City: no catalyst,
oxidative catalyst, and three way catalyst. Exhaust
regulated (CO, NOx, and hydrocarbons) and toxic (benzene,
formaldehyde, acetaldehyde, and 1,3-butadiene) emissions
were evaluated for MTBE (5 vol %)- and ethanol (3, 6,
and 10 vol %)-gasoline blends. The most significant overall
emissions variations derived from the use of 6 vol %
ethanol (relative to a 5% MTBE base gasoline) were 16%
decrease in CO, 28% reduction in formaldehyde, and
80% increase in acetaldehyde emissions. A 26% reduction
in CO emissions from the oldest fleet (<MY 1991, without
catalytic converter), which represents about 44% of the in-
use light duty vehicles in Mexico city, can be attained
when using 6 vol % ethanol-gasoline, without significant
variation in hydrocarbons and NOx emissions, when compared
with a 5% vol MTBE-gasoline. On the basis of the
emissions results, an estimation of the change in the
motor vehicle emissions of the metropolitan area of Mexico
city was calculated for the year 2010 if ethanol were to
be used instead of MTBE, and the outcome was a considerable
decrease in all regulated and toxic emissions, despite
the growing motor vehicle population.
Introduction
The use of oxygenated gasoline is mandatory in the met-
ropolitan area of Mexico City (MAMC) as a control strategy
to reduce ambient CO concentrations and improve urban
air quality. Methyl tertiary-butyl ether (MTBE) has been the
most important oxygenated compound blended in the two
types of gasoline sold in the country (Regular and Premium),
both provided by the state owned oil company. The impact
of substitution of MTBE by ethanol has not been evaluated
yet in Mexico, and is not likely to be identical to others
countries. That is especially true if one takes into account
the altitude, age distribution of the vehicular fleet and the
prevailing motor and emissions control technologies. The
MAMC is a closed valley largely forced to contain the
atmospheric contaminants generated by economic activities.
This region is especially sensitive to atmospheric contami-
* To whom correspondence should be addressed. Phone: +52 (5)
5333-8507. Fax: +52 (5) 5333-8484. E-mail: ishifter@imp.mx.
10.1021/es001177w CCC: $20.00  2001 American Chemical Society
Published on Web 04/10/2001
nation owing to its topographic, climatic, and urban condi-
tions. Being one of the most highly populated zones, one-
fifth of the Mexican population lives in an area of approxi-
mately 2300 km2. The University of Denver monitored the
on-road emissions using remote sensing techniques in the
Mexico City in 1991 and returned in the fall of 1994 and
revisited the same sites surveyed before (1, 2). They found
a large drop (around 50%) that occurred in CO and
hydrocarbons emissions. However, insufficient motor vehicle
information was available for the 1994 data set to fully explain
the large decrease. It should be noted however, that those
results account for emissions for only a single mode of vehicle
operation and therefore do not cover the full range of modes
included in the Federal Test Procedure (FTP).
In this paper, we present a comparative assessment of
the environmental impact in the MAMC of four different
fuel formulations and three different fuel-injection/catalyst-
control vehicle technologies to provide a quantitative basis
for future policy actions in controlling and reducing envi-
ronmental pollutant emissions from motor vehicles in the
MAMC. The context of this problem is the nature of the
pollution problem in the MAMC which is also one of the
largest urban centers in the world. In the MAMC ozone levels,
the most critical pollutant, surpass the Pollutant Standard
Index of the United States most of the year. Oxygenated (5%
vol MTBE) gasoline was introduced in the MAMC in 1989,
and the use of oxygenated gasoline became mandatory for
the MAMC, Guadalajara, and Monterrey, the largest urban
areas in Mexico. In 1997 fuel reformulation, which included
changing composition and physical properties, was imple-
mented and MTBE concentration was set to a maximal value
of 7.5 vol % along with restrictions to the aromatics, olefins,
and benzene content.
Using a representative fleet of the MAMC, exhaust
regulated emissions (CO, NOx, and hydrocarbons) were
evaluated for MTBE- and ethanol-gasoline blends. Also, in
this work we determined the emissions of toxic air pollutants
which at present time are unregulated but pose risk for human
health. The ozone forming potential values of the hydro-
carbons from the exhaust composition were determined and
compared.
Background
Ethanol is commonly used as an oxygenate for blending in
Federal reformulated gasoline in the U.S. (3). Reuter et al.
reported emissions from 20 1989 vehicles according to the
Federal Test Procedure (4). They found a reduction in CO,
total hydrocarbons (HC), and benzene emissions with all
fuel oxygenates, including MTBE, compared to the nonoxy-
genated fuels. The NOx emissions increased for all of the
oxygenates, but the result was statistically significant for the
ethanol fuels. Acetaldehyde emissions increased for ethanol-
gasoline blends. The fleet average CO benefit for the 7 test
oxygenate fuels was -3.8%/wt % O2.
Knepper et al. (5) worked with a group of seven MY 1986-
1987 vehicles selected for their high emission characteristics.
The oxyfuel effects were tested with a 3.65 wt % oxygen
ethanol fuel and 2.76 and 2.54 wt % oxygen MTBE fuels. The
emission levels varied greatly due to unstable air/fuel ratio.
The CO benefits are larger than observed in normally
operating vehicles, averaging about -9.7%/wt % oxygen. The
HC emissions averaging about -9.4%/wt % oxygen and NOx
about +9%. Tests conducted by the EPA (6, 7) on normal and
high emitting vehicles employed 20 fuels in 27 MY 1986-91
VOL. 35, NO. 10, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1893

TABLE 1. Technology of the MAMC Fleet and Vehicles
Employed in This Study
% of
LDGV vehicles odometer
MY fleet technology employeda (km)
<1990 44 carburator 1989 123 154
electronic ignition 1982 146 277
no catalytic converter 1984 74 108
1987 208 120
1990 183 869
1991-1992 16.4 carburator 1991 76 097
electronic ignition 1991 113 109
oxidative catalyst 1992 117 932
no oxygen sensor 1992 147 320
1994-1998 38.6 electronic ignition 1994 31 653
close-loop control 1998 24 954
fuel injection 1998 48 995
TWC
a One vehicle of the indicated MY.
normal emitters and 12 MY 1986-89 high emitters. Both
showed similar CO and HC emission benefits, estimated to
be in the range of 4 to 5% reduction in CO/wt % oxygen. The
oxyfuel effects on NOx emissions are small in all cases. A
large effect of S content on CO emissions was noted by
Mayotte et al. (6) who found a 13.8% reduction in the
combined normal and high emitting fleet, when S was
decreased from 324 to 112 ppm. Benson et al. (8) reported
13% CO, 16% HC, and 9% NO reductions from a low emitting
fleet, when the fuel sulfur content was decreased from 466
to 49 ppmw.
Hood and Farina (9) examined the effects of oxygenate
and Reid vapor pressure (RVP) on emissions from an 11
vehicle fleet MY 1981-1989 at several ambient temperatures.
The fuel oxygenates included 10% ethanol, 15% MTBE, and
17.1% ETBE. The fleet average CO emission levels were about
5, 7, and 15 g/mi at 35-80 °F, respectively. The HC emissions
were reduced by about 3.5%/wt % oxygen except for the
high oxygen fuel at 35 °F. The NO emissions were generally
higher. No distinct emission benefit of one oxygenate over
the others was identified. Most dynamometer studies report
the effects of oxyfuels on pollutant emissions other than CO.
An important effect of all oxygenate fuels is achieved through
dilution. Adding 8-15 vol % oxygenate to a base gasoline
lowers the benzene concentration proportionately. Since
some emissions are unburned fuel components, the dilution
effect directly lowers their levels.
In this work, emission tests were performed on 12 in-use
vehicles, which were obtained from owners in the MAMC
after being selected randomly using a Motor Vehicle owner-
ship database. Our facilities are audited by the Instituto
Nacional de Ecologı́a, an environmental federal institution.
Table 1 shows some characteristics of the MAMC fleet as
well as the technological families employed in the tests.
Considering their emission control technologies and for
simplicity, these fleets will be referred to as no catalyst [model-
year (MY) before 1990], oxidative catalyst (MY 1991-1992),
and three-way catalyst (MY 1993-1998). These vehicles
represent the emission control technology found in a large
segment of the present MAMC vehicle fleet.
Emissions from all vehicles were considered as “normal
emitters” using the criteria that their exhaust emissions are
below twice the certification standards. Four gasoline samples
were obtained and tested by blending 3, 6, and 10 vol % of
ethanol and 5 vol % of MTBE to the same hydrocarbon base
stock (Table 2). The MTBE blend is fully complying with the
requirements for MAMC’s gasoline regulation (10) indicated
also in Table 2. The key differences between the fuels are the
RVP and oxygen content. The MAMC gasoline regulations
1894 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 10, 2001
limit the RVP to 7.8 psi in order to control evaporative
emissions of hydrocarbons. At present, the oxygen cap limit
of 2.7 wt % in the MAMC regulations will restrict the amount
of ethanol blended into gasoline to about 7.7 vol %. The total
consumption of gasoline in the country for the year 1999
was 511 millions of barrels/day (90% Regular) of which 23%
was sold in the MAMC.
Experimental Methods
Each vehicle was submitted to a general inspection and only
those that needed minor repair were accepted. Once the
vehicle was approved for testing, an extensive procedure
was performed in order to minimize the fuel changeover
effects (11) and later the vehicle was tested following the
FTP-75 of the U.S. certification test procedure for light-duty
gasoline vehicles (12). Vehicles were evaluated on a Clayton
model ECE-50-250 chassis dynamometer with a direct-drive
variable-inertia flywheel system. Three samples of dilute
exhaust gas from the constant volume-sampling system were
collected during the FTP corresponding to the cold transient
phase (bag 1), the hot stabilized phase (bag 2), and the hot
transient phase (bag 3). The exhaust emissions samples were
analyzed for THC, NOx, and CO in our laboratory, as
prescribed by standard FTP certification. “Composite” emis-
sions were averaged over the entire test cycle. Tests were
performed by duplicate but a third test was conducted on
each vehicle if the difference between emission levels in back-
to-back tests exceeded 8%. Quantitative hydrocarbon analysis
was performed according to the procedures published
elsewhere (13). Fuel economy values were calculated from
carbon mass-balance measurements obtained on the vehicles
during chassis dynamometer testing (as opposed to being
in-use values). Calculations for fuel economy followed the
procedure published elsewhere (14).
Two types of evaporative losses were evaluated in our
facilities: Diurnal and "hot-soak” evaporative emissions.
Diurnal emissions occur while the vehicle is not operating
and result from the daily variation in ambient temperature.
As the ambient temperature increases throughout the day,
some gasoline in the fuel tank vaporizes, and some of the
vapors escape to the atmosphere through leaks in the fuel
system and emissions control system. Hot soak emissions
have historically originated primarily from the vehicle’s
carburetor; they occur immediately after the vehicle engine
is turned off when gasoline in the carburetor bowl vaporizes
due to the temperature increase of the carburetor. To perform
both evaluations, the protocol followed the procedure
published in ref 12.
The comparative ozone forming potential for the fuels
was determined using the Carter maximum incremental
reactivity (MIR) scale. The scale applies an individual
reactivity value to each nonmethane organic compound
measured in the exhaust or evaporative emissions from the
motor vehicles (15).
Results
Regulated Exhaust Emissions. The CO emissions variations
as a function of ethanol concentration, which were grouped
depending on vehicle’s emission control technology, are
shown in Figure 1. In the fleet without catalytic converter,
the addition of ethanol brought about considerable CO
emissions decrease in comparison with the MTBE base-
gasoline. For instance, CO emissions dropped about 27%
more with 6% ethanol than with the base-gasoline. In the
fleet with oxidative catalyst, a similar CO emissions decreasing
tendency was observed upon increasing ethanol concentra-
tion. Here, a 39% decrease in CO emissions (in comparison
with base-gasoline) was achieved with 10% ethanol. However,
in the fleet equipped with three-way catalytic converter (TWC)
TABLE 2. Relevant Properties of the Gasolines
MTBE ethanol (vol %) federal regulated
property (vol %) 5 3 6 10 propertiesa
gravity, 20/4 °C 0.7444 0.7436 0.7475 0.7478
Reid vapor pressure, (lb/in.2) 7.27 7.58 7.69 8.03 7.8 max
distillation ASTM, D-86 (°C)
10% evaporated 64.8 62.7 58.1 57.1
50% evaporated 108.5 110.0 108.7 105.3
90% evaporated 172.0 171.1 168.5 169.3
end point 212.2 213.3 210 209.7
sulfur, ppm 690 640 640 580
aromatics, (vol %) 25.0 25.0 24.8 27.1 25 max
olefins, (vol %) 10.2 11.8 10.9 11.7 10 max
benzene, (vol %) 1.1 1.2 1.1 1.1 1 max
(RON+MON/2) 87.5 87.8 87.2 88.3
oxygen, (vol %) 1.0 1.0 2.0 3.7 2-2.7
a MAMC,Guadalajara,Monterrey.

FIGURE 1. Effect of ethanol concentration and vehicles’ emission control technology on CO emissions.

there was not any significant variation in CO emissions in
all the ethanol concentration range (3-10%) tested.
Fuel vapor pressure and gasoline sulfur content affect
the CO and HC emission levels of vehicles. Mayotte et al. (6)
reported 11.3% reduction in CO emissions in the combined
normal and high emitting fleet, when the fuel RVP was
reduced from 8.3 to 7.6.psi. A large effect of sulfur content
on CO emissions was noted by Mayotte et al. (7) who found
a 13.8% reduction in CO emissions in the combined normal
and high emitting fleet, when fuel sulfur was decreased from
324 to 112 ppm by weight. The Auto/Oil AQIRP study reported
13% CO, 16% HC, and 9% NO reductions from a 10 vehicle,
MY 1989, low emitting fleet, when the fuel sulfur content
was decreased from 466 ppm by weight to 49 ppm.
The effect of ethanol concentration on THC emissions is
shown in Figure 2. In the vehicle fleet without catalytic device,
there is a significant increase in THC emissions when using
3% ethanol-containing gasoline, and higher ethanol con-
centrations resulted in a marginal decrease, about 7% (for
10% ethanol) in comparison with the base-gasoline. The fleet
with oxidative catalyst showed a gradual THC emissions
decrease with increasing ethanol concentration. However
and similar to the CO emissions’ case, the THC emissions
did not decrease significantly in the TWC-equipped fleet,
within the ethanol concentration range tested. Separately,
the Office of Science and Technology Policy completed the
most comprehensive study regarding oxygenated fuels, in
1997 (16). This report concluded that the percentage reduc-
tions in CO and THC emissions from the use of fuel
oxygenates were found to be smaller in the newer technology
vehicles compared to older technology and higher emitting
vehicles. High-emitting vehicles (malfunctioning or operating
in a fuel-rich mode) generally also have larger CO reduction
benefits from oxyfuels.
The behavior of NOx emissions upon different ethanol
concentrations is shown in Figure 3. For both the oxidative
catalyst and TWC vehicles there is a tendency to generate
more NOx as the concentration of ethanol increases. For 3
or 6% ethanol, the TWC-equipped fleet showed almost no
variation in NOx emissions, in comparison with the base
gasoline.
Fuel Economy, Diurnal, and Hot-Soak Evaporative
Emissions. The effect of ethanol concentration on the
evaporative emissions (as g of THC/test) and fuel economy
for the three fleet-types is shown in Table 3. A comparison
of the hot-soak evaporative emissions from the three fleets

VOL. 35, NO. 10, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1895
FIGURE 2. Effect of ethanol concentration and vehicles’ emission control technology on THC emissions.

FIGURE 3. Effect of ethanol concentration and vehicles’ emission control technology on NOx emissions.

clearly shows a considerable decrease related to the ad-
vancement of the vehicles technology. For instance, the TWC
fleet shows an average of 5.5 and 4.0 times less hot-soak
emissions than the no-catalyst and oxidative catalyst fleets,
respectively. Due to the incorporation of fuel injection devices
in this fleet, less open parts are exposed to the atmosphere,
as opposed to the carburetor-equipped fleets. On the other
hand, the technological content of the fleets affects to a lesser
extent the diurnal evaporative emissions.
In the three fleets, increasing ethanol concentrations
brought about higher hot-soak emissions. Notwithstanding,
the variation of hot-soak emissions in the TWC fleet when
using 3 and 6% ethanol was negligible in comparison with
that of the base-gasoline. The effect of ethanol concentration
on the diurnal evaporative emissions was not clear.
When using 6% ethanol, the fuel economy values were 7,
2.5, and 2% higher than the base-gasoline, for no-catalyst,
oxidative catalyst, and TWC fleets, respectively.
The environmental standard for total evaporative emis-
sions in new vehicles sold in Mexico is 2 g/test for light-duty
vehicles. Running losses, not measured in this work, are
probably modest at low RVP and medium temperatures, but
increase substantially at high RVP and high temperature.
Toxic Emissions. Table 4 shows the variation of toxic
emissions as a function of ethanol concentration and
vehicular emissions control technology. Clearly, the four toxic
compounds: benzene 1,3-butadiene, formaldehyde, and
acetaldehyde have been greatly diminished with the ad-
vancement of vehicles technology. In fact, the substitution
of oxidative catalyst by TWC brought about 4 and 6 times

1896 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 10, 2001
TABLE 3. Fuel Economy and Evaporative Emissions (Diurnal and hot-soak) as a Function of Vehicles’ Technology and Ethanol
Concentration
no catalyst fleet oxidative catalyst fleet TWC fleet
evaporative emissions evaporative emissions evaporative emissions
FEa (g/test) FEa (g/test) FEa (g/test)
gasoline (km/l) diurnal hot-soak (km/l) diurnal hot-soak (km/l) diurnal hot-soak
5% vol MTBE 10.73 6.17 2.70 10.99 4.00 1.56 11.47 5.22 0.46
3% vol ethanol 10.88 6.34 2.51 10.65 4.37 1.94 11.56 4.14 0.46
6% vol ethanol 11.48 6.17 2.92 11.27 5.10 2.30 11.74 3.77 0.44
10% vol ethanol 11.43 5.88 3.15 11.23 5.00 2.43 11.50 4.60 0.68
a FE: fuel economy.

TABLE 4. Variation in Toxic Emissions as a Function of Vehicles’ Technology and Ethanol Concentrationa
no catalyst fleet (mg/km) oxidative catalyst fleet (mg/km) TWC fleet (mg/km)
gasoline Bz 1,3C42- FrAld AcAld Bz 1,3C42- FrAld AcAld Bz 1,3C42- FrAld AcAld
5% vol MTBE 51.67 20.81 23.76 7.96 40.35 7.08 1.43 3.45 6.54 0.89 1.45 0.67
3% vol ethanol 46.88 16.91 24.47 9.99 35.47 5.67 1.97 5.30 7.22 0.83 1.32 1.12
6% vol ethanol 49.51 21.36 26.71 16.26 32.05 4.73 0.71 5.10 7.38 0.77 0.78 1.25
10% vol ethanol 47.54 18.35 30.29 15.3 28.70 4.50 2.58 7.00 8.11 0.83 1.01 1.62
a Bz, benzene; 1,3C42-, 1,3-butadiene; FrAld, formaldehyde; AcAld, acetaldehyde.

FIGURE 4. Specific reactivity toward O3 formation as a function of ethanol concentration and vehicles’ emission control technology.
average lower benzene and 1,3-butadiene emissions, re- different emission control technology in the three fleets.
spectively, for the ethanol-containing gasolines. Evidently Vehicle exhaust emissions data have shown that acet-
from Table 4, the use of different types of oxygenates and aldehyde emissions can increase by as much as 100% with
oxygen levels in advanced emission control technology fleet the use of 2.0 wt % ethanol oxygenated gasoline, part of which
has small mass effects on aldehydes emissions. The most undergoes photochemical reactions in the atmosphere to
detrimental effect of ethanol was the considerable increase make peroxyacetyl nitrate (PAN) (17). Reuter et al. (4) reported
of acetaldehyde emissions (80-104%), regardless of the that benzene emissions are reduced by fuel oxygenate.

VOL. 35, NO. 10, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1897
FIGURE 5. Comparison of CO emissions variations for the years 1999 and 2010 in the MAMC.

FIGURE 6. Comparison of THC, NOx, and evaporative emissions variations for the years 1999 and 2010 in the MAMC.
However, in this work, benzene seems to increase among Butadiene emissions decreased more in the fleet with
the ethanol fuelled vehicles equipped with TWC. Benzene oxidative catalyst than in the TWC-equipped fleet. The reason
exhaust emissions are dependent on fuel and catalyst type of this might be the greater oxidizing efficiency of the oxidative
used in the vehicle, for that reason individual variations are catalyst, than that of the TWC. Most of the other emissions
present due to car to car technology and also fuel formulation remained practically without variation. The 1,3-butadiene
make comparison difficult. On the other hand, ethanol emissions are reduced but not significantly except for the
decreased formaldehyde emissions (about 46-50%); par- ethanol fuel with the highest oxygenate content. The
ticularly in the fleets with oxidative catalyst or TWC. 1,3- formaldehyde emissions has been reported not change

1898 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 10, 2001
FIGURE 7. Comparison of toxic emissions variations for the years 1999 and 2010 in the MAMC.
significantly, although most studies find MTBE increases
formaldehyde emissions (9). Baez et. al reported average air
concentrations of formaldehyde and acetaldehyde for Mexico
City of 14.71 and 0.11 ppbV, respectively (18).
Specific Reactivity. The MIR represents the predicted
impact of the respective constituent on urban atmospheric
ozone formation, expressed as grams of ozone per gram of
the constituent. Ozone forming potential for a specific fuel
was computed by incorporating the MIR values for all
constituents measured in the exhaust from a particular fuel.
Specific reactivity (SR) is a measure of how much ozone could
be produced by a unit mass of emissions and is typically
expressed as gram of O3 per gram of nonmethane organic
gases (NMOG). The potential of emitted compounds to form
O3 from ethanol-containing gasoline is shown in Figure 4.
For the fleet with oxidative catalyst, there is only a marginal
decrease in the ozone formation with 6 and 10% ethanol. In
the case of the TWC-equipped fleet, however, the effect of
adding 3-10% vol ethanol to the gasoline in the ozone
formation was negligible.
The National Research Council (NRC) found that the
ozone formation potential (as measured with the California
MIR scale) of exhaust emissions from motor vehicles operat-
ing on ethanol-blended gasoline were lowersbut not sig-
nificantly lowersthan from motor vehicles using MTBE-
blended gasoline. The NRC also concluded that the mass
and ozone formation potential (per mile) of evaporative
emissions from ethanol-blended gasoline were significantly
higher, and that this increase would be detrimental to air
quality in terms of ozone (19). This latter finding was
attributed to an RVP increase of 1 psi from the ethanol-
containing gasoline (19). The available data on exhaust and
evaporative emissions from older vehicles show that the small
benefit in exhaust emissions due to the higher oxygen con-
tent is not sufficient to overcome the large increase in
evaporative emissions (19). The increase in evaporative mass
emissions from a 1 psi increase in RVP is so substantial that
even considering the lower reactivity of evaporative emis-
sions, the resultant ozone forming potential significantly
increases when exhaust and evaporative emissions are
combined (20).
Discussion
Turnover rate of the Mexican vehicle fleet is considerably
longer than in the U.S. (15 and 4 years, respectively).
Additionally, the incorporation of emission controls in new
vehicles was delayed in Mexico compared to those in the
U.S.. Thus, older, less fuel-efficient and more polluting
vehicles still comprise a large percentage of the current
vehicular fleet in the MAMC. Approximately, 44% of the
vehicular fleet is not equipped with a catalytic converter since
the introduction of these emission control systems did not
occur until the MY 1991 for new vehicles.
The contribution by each vehicle type for the year 1999
and 2010 was calculated. In the case of the extrapolation for
2010, the estimation of the total vehicle population for a
given calendar year was used. To determine annual sales we
consulted the historical data of the Mexican Manufacturers
Association of Vehicles and projected a vehicle sales annual
growth rate of 4% for 1999-2010. Scrappage is defined as the
number of vehicles scrapped or otherwise removed from
circulation in any given year, this value is reported as
percentage of the number of vehicles in use. To determine
the effects of annual vehicle scrappage, we consulted rates
reported recently by the Inspection/Maintenance Program,
mandatory twice per year in the MAMC, of vehicles that failed
to obtain the permission to circulate. Higher rates were apply
to old vehicles but we estimated an average scrappage rate
of 2.7% annually. The general vehicle count formula is
VCx ) (VCx - 1 + Sales x) (1 - SRx) (1)
where VC is the vehicle count, SR is the percent of in-use
fleet that is scrapped, and x is the current calendar year.
Figure 5 shows an estimation of the total emissions in ton
per year for CO for the year 1999 and extrapolation to the
VOL. 35, NO. 10, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1899
TABLE 5. Estimated Regulated Emissions Variationa
CO THC NOx
thousand (t/yr)
3% EtOH -26 20.6 7.7
6% EtOH -60 14.6 9.7
10% EtOH -76 14.6 13.7
a Difference in estimated total fleet emissions for 2010 fleet (with
ethanol gasoline) compared to that of 1999 fleet (with MTBE base
gasoline).
TABLE 6. Estimated Toxic Emissions Variationa
benzene 1,3-butadiene FrAld
(t/yr)
3% EtOH -343 -219 -178
6% EtOH -356 -200 -199
10% EtOH -358 -228 -137
a Difference in estimated total fleet emissions for 2010 fleet (with
ethanol gasoline) compared to that of 1999 fleet (with MTBE base
gasoline).
year 2010 for the four fuels tested. Figure 6 shows similar
results for THC, NOx, and evaporative emissions. These
calculations were carried out taking into account the average
daily traveled distance and the experimental emission factors
calculated based on this study (see Supporting Information).
The “evaporative” emissions, as the pondered sum of diurnal
and hot-soak emissions divided by the daily distance, are
shown in Figure 6. For MY 2001 and newer fleets, TIER-I
emission factors were used in the calculations based on
128 000 km certified emissions.
From Figures 5-7 and in spite of the motor vehicle
population growth, it is clear that all emissions (regulated
and toxic) will be lower in the year 2010 than in the year
1999, due to lower emissions factors (of newer MY vehicles)
and disappearance of old highly polluting vehicles. There is
a considerable reduction trend in CO emissions as the ethanol
concentration rise. THC emissions decreased above 6 vol %
ethanol. Nevertheless, NOx and evaporative hydrocarbon
emissions increased with ethanol concentration. A com-
promise between reduced CO emissions and moderate NOx
emissions could be attained when using 6% ethanol gasoline.
Clearly, ethanol incorporation in current gasolines should
be accompanied by a more efficient catalytic converter,
particularly in its performance on NOx reduction.
Figure 7 shows the total variations in toxic emissions for
the year 1999 and 2010. As discussed in previous sections,
one of the clearest detrimental effects of replacing MTBE for
ethanol is the considerable increase in acetaldehyde emis-
sions. Total formaldehyde emissions would also increase for
the year 2010 in the case of the 10% ethanol gasoline. Last,
Tables 5 and 6 show the estimated regulated and toxic
emissions variation, respectively, for the year 2010 obtained
with the ethanol addition with respect to those of the 1999
MTBE base gasoline.
It is clear that the mere substitution of MTBE for ethanol
in Mexican gasolines will not upgrade MAMC current air
quality if it is not accompanied by (1) acceleration of old
vehicles scrappage rates (particularly those without a catalytic
device), (2) improved catalytic devices (to cope with toxic
compounds), and (3) imposition of new regulations on toxic
emissions.
The oxygenate problem has a different scale in Mexico in
comparison with that in the U.S., as it is shown in Table 7.
The key differences in environmental factors are that per
capita consumption of regular and oxygenated gasoline is
much lower in Mexico. Also, there are a limited number of
1900 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 10, 2001
TABLE 7. Comparison of the Different Scales on the
Oxygenated Gasolines Issue in Mexico and the U.S.
Evap
highlights Mexico U.S.
-4.8 area of the country (km2/103) 1959 9364
-4.5 population (millions) 94 268
-4 1998 gasoline consumption 512 8184
(barrels/dy103)
1998 oxygenated gasoline, 192 3260
consumption (barrels/dy103)
1998 per capita oxygenated gasoline 2 12
consumption (barrels/dy103)
significant natural water reservoirs limited high
around metropolitan areas
AcAld other consumption (Jet-skis,etc) limited moderate
amount of vehicles (/106) 13.5 129
vehicle average age (yr) 15 4
7 program to detect leaks centralized disperse
104
133
water bodies, around the major metropolitan areas in Mexico
in comparison with those in the U.S. The nonautomotive
sources of contaminants are much lower in Mexico. But the
most remarkable difference from Table 7 is the vehicle average
age: 15 yr in Mexico versus 4 yr in the U.S. Before any actions
to remove MTBE from reformulated gasoline sold in Mexico’s
main metropolitan areas were to be taken, studies to assess
MTBE content into groundwater are urgently needed.
Groundwater contamination has been the key environmental
problem associated with MTBE-containing gasoline in the
U.S., and therefore, the potential for groundwater contami-
nation in the MAMC should be considered a relevant factor.
The official decision to ban MTBE, not later than
December 31, 2002, from the gasoline sold in the state of
California (21), and the recent announcement from U.S. EPA
administrator to “begin regulatory action aimed at eliminat-
ing MTBE from gasoline” and boost ethanol use (22), serve
as a serious warning that soon MTBE will be banned
throughout the U.S., and will probably infiltrate into Mexico
too. Current ethanol production in Mexico derives mainly
from sugar cane crops. If MTBE were to be banned in Mexico,
and ethanol were to be the substituting oxygenate, this could
mean a great opportunity to rescue the sugar cane industry
and a variety of renewable feedstocks.
Supporting Information Available
A more detailed description of the regulated and toxic
emission calculations for the years 1999 and 2010 is available
free of charge via the Internet at http://pubs.acs.org.
Literature Cited
(1) Beaton, S. P.; Bishop, G. A.; Stedman, D. H. J. Air Waste Manage.
Assoc. 1992, 42, 1424.
(2) Bishop, G. A.; Stedman, D. H.; De la Garza Castro, J.; Davalos,
F. J. Enivron. Sci. Technol. 1991, 31, 3505.
(3) Howard, C.; Russell, A.; Atkinson, R.; Calvert, J. Air Quality Effects
of the Winter Oxyfuel Program; White House Office of Science
and Technology Policy, Washington, DC, July 2, 1997.
(4) Reuter, R. M.; Gorse, R. A., Jr.; Painter, L. J.; Benson, J. D.;
Hochhauser, A. M.; Rippon, B. H.; Burns, V. R.; Koehl, W. J.;
Rutherford, J. A. Auto/Oil Air Quality Improvement Program;
SAE Technical Paper Series No. 920326, Society of Automotive
Engineers, Warrendale, PA, 1992.
(5) Knepper, J. C.; Koehl, W. J.; Benson J. D.; Burns, V. R.; Gorse,
R. A., Jr.; Hochhauser, A. M.; Leppard, W. R.; Rapp, L. A.; Reuter,
R. M. SAE Technical Paper Series No. 930137; Society of
Automotive Engineers, Warrendale, PA, 1993.
(6) Mayotte, S. C.; Lindhjem, C. E.; Rao, V.; Sklar, M. S. SAE Technical
Paper Series No. 941973; Society of Automotive Engineers,
Warrendale, PA, 1994.
(7) Mayotte, S. C.; Lindhjem, C. E.; Rao, V.; Sklar, M. S. SAE Technical
Paper Series No. 941974; Society of Automotive Engineers,
Warrendale, PA, 1994.
 (8) Benson, J. D.; Burns, V.; Gorse, R. A.; Hochhauser, A. M.; Koehl,
W. J.; Painter, L. J.; Reuter, R. M. Auto/OilAir Quality Improve-
ment Research Program; SAE Technical Paper Series No. 912323,
Society of Automotive Engineers, Warrendale, PA, 1991.
(9) Hood, J.; Farina, R. SAE Draft for Fuels and Lubricants Meeting
in Toronto; SAE Paper No. 952403, Society of Automotive
Engineers, Warrendale, PA, 1995.
(10) Normatividad Ambiental y Emisiones Vehiculares en Mexico
(Environmental Standards and Vehicular Emissions in Mexico);
Report from the CESPEDES (Centro de Estudios del Sector
Privado para el Desarrollo Sustentable), Chapter II, México, D.
F., September 1998.
(11) Burns, V.; Benson, J.; Hoehkuser, A.; Koehl, J.; Dreucher, M.;
Reuter, R. SAE Technical Paper: 912320, Society of Automotive
Engineers, Warrendale, PA, 1992.
(12) Office of the Federal Register, Code of Federal Regulations, Title
40, Part 86, Washington, DC, U.S. Government Printing Office,
1993.
(13) Newkirk, M.; Smith, L.; Payne, M. E.; Segal, J. S. SAE Technical
Paper No. 96199; Society of Automotive Engineers, Warrendale,
PA., 1996.
(14) Kellner, T. A.; Neusen, K.; Bresenham, D.; Pike, M.; Rose, D. SAE
Technical Paper No. 960855; Society of Automotive Engineers,
Warrendale, PA, 1996.
(15) Carter, W. P. J. Air Waste Manage. Assoc. 1994, 44, 881.
(16) Interagency Assessment of Oxygenate Fuels; Office of Science
and Technology Policy, National Science and Technology
Council, Washington, DC, 1997.
(17) Kirchner, F.; Thuneer, L. P.; Barnes, I.; Becker, K. H.; Donner,
B.; Zabel, F. Environ. Sci. Technol. 1997, 31, 1801.
(18) Baez, A. P.; Belmont, R.; Padilla, H. Environ. Poll. 1995, 89, 163.
(19) Ozone-Forming Potential of Reformulated Gasoline; National
Research Council, National Academy Press: Washington, DC,
1999.
(20) Proposed Determination Pursuant to Health and Safety Code
Section 43830(g) of the Ozone Forming Potential of Elevated
RVP Gasoline Containing 10 Percent Ethanol; California Envi-
ronmental Protection Agency, November 1998.
(21) Executive Order D-5-99, Gray Davis, Governor of the State of
California, March 25, 1999.
(22) Press Conference of EPA Adminstrator Carol M. Browner
Remarks, Available from USEPA, Washington, DC, Office of
Communications, Education and Public Affairs, March 20, 2000.
Received for review April 12, 2000. Revised manuscript re-
ceived January 19, 2001. Accepted April 12, 2000.
ES001177W
VOL. 35, NO. 10, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1901