Search Results

Exhaust emissions were measured as a function of gasoline composition in two fleets of vehicles - 20 1989 vehicles and 14 1983-1985 vehicles. Eighteen different gasolines were tested which varied in aromatic, olefin, and MTBE content and in the 90 percent distillation temperature (T90). Subject to the cautions and qualifications described in the body of this paper, mass exhaust emissions in both fleets of vehicles were affected by changes in fuel composition. Responses to changes in MTBE and olefins were similar in both fleets: adding MTBE reduced emissions of HC and CO, and reducing olefins lowered emissions of NOx while raising emissions of HC. In the current fleet, reducing aromatics lowered HC and CO, while in the older fleet, reducing aromatics raised HC and lowered NOx. In the current fleet, lowering T90 reduced HC over 20%, while raising NOx slightly. In the older fleet, lowering T90 reduced HC by only 6%.

This paper presents results derived from Phase I of the Auto/Oil Air Quality Improvement Research Program. The Clean Air Act-defined mobile source toxic air pollutants benzene, 1,3-butadiene, formaldehyde and acetaldehyde have been measured in exhaust from twenty current model vehicles and fourteen older model vehicles during testing with 18 gasolines of varying composition. The gasoline fuel compositional variables evaluated included aromatic content, methyl tertiary-butyl ether (MTBE) content, olefin content, and the 90% distillation temperature (T90). The four fuel parameters were varied at target values of 45 and 20 vol % total aromatics, 0 and 15 vol % MTBE, 20 and 5 vol % total olefins and 360 and 280 °F 90% distillation temperature. An industry average fuel and an emissions certification test fuel were tested as reference fuels. In the current fleet, benzene levels were lowered when either fuel aromatics or T90 were reduced.

Prior studies of the effect of gasoline composition and physical properties on automotive exhaust and evaporative emissions have been reviewed. The prior work shows that the parameters selected for investigation in the Auto/Oil Air Quality Improvement Research Program (AQIRP) - gasoline aromatics content, addition of oxygenated compounds, olefins content, 90% distillation temperature, Reid vapor pressure, and sulfur content - can affect emissions. Effects have been observed on the mass of hydrocarbon, CO, and NOx emissions; on the reactivity of emissions toward ozone formation; and on the emissions of designated toxic air pollutants. The individual effects of some of the AQIRP parameters have been studied extensively in modern vehicles, but the most comprehensive studies of gasoline composition were conducted in early 1970 vehicles, and comparing the various studies shows that fuel effects can vary among vehicles with different control technology.

In this portion of the Auto/Oil Air Quality Improvement Research Program, ten 1989 model vehicles were tested using two fuels with different sulfur levels. These tests were run to determine instantaneous effects on exhaust emissions, not long-term durability effects. The high- and low-sulfur fuels contained 466 ppm and 49 ppm sulfur, respectively. Mass exhaust emissions of the fleet decreased as fuel sulfur level was reduced. Overall, HC, CO, and NOx were reduced by 16, 13, and 9 percent, respectively, when fuel sulfur level decreased. This effect appeared to be immediately reversible. Engine-out mass emissions were unaffected by changes in the fuel sulfur content, therefore, tailpipe emissions reductions were attributed to increased catalyst activity as the sulfur level was reduced.

An overview of Phase 1 of the Auto/Oil Air Quality Improvement Research Program is presented. Specific information is provided on each of the individual test fuel matrices that were conducted to investigate vehiclelfuel “system” effects on emissions. Procedures for sampling exhaust, evaporative, and running loss vehicle emissions are described, as well as techniques developed for speciation of individual hydrocarbons. Air quality models to project ozone reduction potential of reformulated gasolines and methanol, and economic studies to estimate the relative cost-effectiveness of the vehiclelfuel alternatives are also briefly explained.

A rigidly controlled vehicle test program was conducted to assess the impact of MMT fuel additive on the operation of Low Emission Vehicles (LEVs). Two pairs of each of five vehicle models were tested over extended mileage (75,000 to 100,000 miles). Vehicles were driven on a test track using a customer-type driving cycle and emission tested at regular intervals throughout the program. One vehicle of each pair used a Clear base fuel and the other used the same base fuel with the addition of MMT at a concentration of 8.3 mg Mn/L (0.031 or 1/32 g Mn/US gal). For the four light-duty vehicle models, seven of the eight MMT-fueled vehicles exceeded the NMOG emission certification standards; one Clear-fueled vehicle of one model exceeded the standards, but all other Clear-fueled vehicles met the standards. All four vehicles of the one medium-duty vehicle model met the certification standards, which are higher than those for light-duty vehicles.

Speciated exhaust emission data from Phase I of the Auto/Oil Air Quality Improvement Research Program are presented and analyzed. Eighteen fuels were tested which varied in four fuel parameters: aromatics, MTBE content, olefins, and T90. These fuels were tested in two fleets of vehicles. One consisted of twenty 1989 vehicles and the other consisted of fourteen 1983-1985 vehicles. The 1990 version of Carter reactivity factors were used to calculate reactivities for each of these tests. Two types of reactivities were calculated. The first was Specific Reactivity and has units of grams ozone per gram NMOG (non-methane organic gas). The second was Ozone Forming Potential and has units of grams ozone per mile. Both types of reactivities were calculated using Carter's MIR (Maximum Incremental Reactivity) as well as MOR (Maximum Ozone Reactivity) factors.

This report presents the Auto/Oil AQIRP results of a methanol fueled vehicle emission study. Nineteen early prototype flexible/variable fueled vehicles (FFV/VFV) were emission tested with industry average gasoline (M0), an 85% methanol-gasoline blend (M85), and a splash-blend of M85 with M0 (gasoline) giving 10% methanol (M10). Vehicle emissions were analyzed for the FTP exhaust emissions, SHED diurnal and hot soak evaporative emissions, and running loss evaporative emissions. Measurements were made for HC, CO and NOx emissions and up to 151 organic emission species, including air toxic components. M0 and M10 emissions were very similar except for elevated M10 evaporative emissions resulting from the high M10 fuel vapor pressure. M85 showed lower exhaust emissions than M0 for NMHC (non-methane hydrocarbon), OMHCE (organic material hydrocarbon equivalent), CO and most species. M85 had higher exhaust emissions for NMOG (non-methane organic gases), NOx, methanol and formaldehyde.

Exhaust and evaporative emissions were measured as a function of gasoline composition and fuel vapor pressure in a fleet of 20 1989 vehicles. Eleven fuels were evaluated; four hydrocarbon only, four splash blended ethanol fuels (10 vol %), two methyl tertiary-butyl ether (MTBE) blends (15 vol %) and one ethyl tertiary-butyl ether (ETBE) blend (17 vol %). Reid vapor pressures were between 7.8 and 9.6 psi. Exhaust emission results indicated that a reduction in fuel Reid vapor pressure of one psi reduced exhaust HC and CO. Adding oxygenates reduced exhaust HC and CO but increased NOx. Results of evaporative emissions tests on nineteen vehicles indicated a reduction in diurnal emissions with reduced Reid vapor pressure in the non-oxygenated and ethanol blended fuels. However, no reduction in diurnal emissions with the MTBE fuel due to Reid vapor pressure reduction was observed. Reducing Reid vapor pressure had no statistically significant effect on hot soak emissions.

Fuel economy measurements from portions of Phase I of the Auto/Oil Air Quality Improvement Research Program were analyzed. The following fuel variables were examined: aromatics, olefins, T90, RVP, and various oxygenates (MTBE, ETBE and ethanol). Two vehicle fleets were tested: twenty 1989 vehicles and fourteen 1983-1985 vehicles. Three measures of fuel economy were analyzed. EPA Fuel Economy used the calculation defined in the Federal Register and is an attempt to correct for changes in fuel properties. Volumetric Fuel Economy is based on a carbon balance calculation and is a measure of the actual volume of gasoline burned. Energy Specific Fuel Economy is a measure of fuel economy based on energy content. The following fuel changes resulted in reductions of Volumetric Fuel Economy in both fleets: reduced aromatics, reduced olefins, reduced T90, and addition of oxygenates. Changes in RVP did not have a significant effect on fuel economy.

Fuel effects on exhaust emissions of a sample of seven high emitting vehicles were studied. The vehicles had various mechanical problems and all ran fuel rich. The degree of enrichment varied between tests, and strongly affected mass emissions. Variable enrichment can cause incorrect apparent fuel effects to be calculated if not accounted for in data analysis. After variable enrichment was compensated for, the percentage effects of fuel oxygen, RVP, and olefins were largely in agreement with prior findings for normally emitting vehicles. Reducing fuel sulfur and T90 may have less benefit on hydrocarbon emissions in these high emitters than in normal emitters, and reducing sulfur may have less benefit on CO emissions. Reducing aromatics may be somewhat more helpful in reducing hydrocarbon and CO emissions in the high emitters.

Effects of gasoline sulfur content on emissions were measured in a fleet of ten 1989 model year vehicles. Two ranges of sulfur content were examined. In a set of five fuels, reducing sulfur from 450 to 50 ppm, reduced fleet average tailpipe emissions of HC, NMHC and CO each by about 18%, and reduced NOx 8%. The largest effect on HC and CO emissions was observed in FTP Bag 2. This and the absence of any significant effect on engine emissions indicate that sulfur affected the performance of the catalytic converters. The response of HC and NMHC to fuel sulfur content was non-linear and increased as sulfur level was reduced. In the second set of three fuels, reducing sulfur from 50 to 10 ppm reduced HC and NMHC by 6% and CO by 10%, but had no significant effect on NOx. The effects on HC, NMHC and NOx were not significantly different from predictions based on the prior fuel set. The reduction in CO was larger than predicted.

The analysis of data from the Auto/Oil Air Quality Improvement Research Program (AQIRP) study of the effect of aromatics, MTBE, olefins, and T90 on mass exhaust emissions from current (1989) vehicles was extended to include individual vehicles during individual operating modes. The results of the modal data analysis agree with and complement results which have been reported previously by AQIRP. Beyond this, attention is focused on three fuel compositional changes where the effect on emissions shows a reversal in sign depending on the vehicle operating mode chosen.

Species analyses have been performed on engine-out and tailpipe hydrocarbon mass emissions to help understand why fuels with increasing amounts of heavy hydrocarbon constituents produce significantly higher tailpipe hydrocarbon emissions. Mass and speciated hydrocarbon emissions were acquired for a fleet of ten 1989 model year vehicles operating on twenty-six fuels of differing heavy hydrocarbon composition. These fuels formed two statistically designed matrices: one examining the effects of medium, heavy, and tail reformate and medium and heavy catalytically cracked components; and the other examining the effects of heavy paraffinic versus heavy aromatic components and the effects of the 50% distillation temperature. In this paper the fates of fuel species were traced across the engine and across the catalyst, and correlations were developed between engine-out and tailpipe hydrocarbon species emissions and fuel composition.

Modal analyses have been performed on engine-out and tailpipe hydrocarbon and carbon monoxide mass emissions to help understand why fuels with increasing amounts of heavy hydrocarbon constituents produce significantly higher tailpipe hydrocarbon emissions, yet do not produce significantly higher tailpipe carbon monoxide emissions. Mass emissions were acquired for a fleet of ten 1989 model year vehicles operating on twenty six fuels of differing heavy hydrocarbon composition. These fuels formed two statistically designed matrices: one examining the effects of medium, heavy, and tail reformate and medium and heavy catalytically cracked components; and the other examining the effects of heavy paraffinic versus heavy aromatic components and the effects of the 50% distillation temperature.

Effects of methyl tertiary-butyl ether (MTBE) and tertiary-amyl methyl ether (TAME) on emissions were compared in a fleet of ten 1989 model year vehicles. Test fuels containing 11.5 vol.% MTBE or 12.7 vol.% TAME were blended in a base fuel representing federal emission certification fuel. The oxygen content of both fuels was about 2.0 wt.%. No significant differences were found between the two fuels in exhaust mass HC, NMHC, CO, or NOx; in exhaust or evaporative toxic air pollutants, benzene, 1,3-butadiene, acetaldehyde, or total toxic emissions; or in evaporative hot soak emissions. The only differences found to be significant at the 95% level were in mass and estimated reactivity-weighted diurnal evaporative emissions, for both of which the TAME fuel was about 24% lower than the MTBE fuel; and in formaldehyde emissions, which were 28% higher with the TAME fuel.

Emission effects of gasoline hydrocarbon components distilling above 300°F were investigated to determine whether the effect of 90% distillation temperature (T90) found in an earlier Auto/Oil Program study is due to fuel distillation properties or to hydrocarbon composition, and also to determine whether the T90 effect is linear. Twenty-six fuels were tested in two sets. In Matrix A, the independent variables were catalytically cracked (FCC) and reformate stocks with nominal distillation ranges of 300 to 350, 350 to 400 and 400+°F. In Matrix B, the independent variables were a reformate stock (320 to 370°F), a heavy alkylate (330 to 475°F), and a light alkylate distilling below 300°F, which was used to vary fuel T50 at fixed levels of T90. Exhaust mass and speciation were measured using ten 1989 vehicles of the Auto/Oil Current Fleet. Tailpipe hydrocarbon emissions were found to increase nonlinearly with progressive addition of the heavier components.

Exhaust and hot soak evaporative emissions were measured in a fleet of 1993 production flexible/variable-fueled vehicles on methanol fuels blended with a reformulated gasoline. A fleet of 1993 California Tier 1 gasoline vehicles was also tested on the same reformulated gasoline blended to meet the specifications of California Phase 2 fuel. Ozone-forming reactivity, expressed as reactivity weighted emissions and specific reactivity, was calculated using 1991 SAPRC and 1994 CBM MIR and MOR factors. Within the FFV/VFV fleet, FTP exhaust and reactivity weighted emissions were significantly lower by 18 to 32% with Phase 2 gasoline relative to Industry Average gasoline. With the exception of greater NMOG emissions with the M85 blends, and lower OMHCE emissions with M85 blended with Industry Average gasoline, exhaust organic emissions, CO and NOx with the methanol fuels were not significantly different than their base gasolines.

Three-day diurnal SHED evaporative emissions were measured in a fleet of ten Auto/Oil current (1989) and 2 older (1984) vehicles using Auto/Oil Industry Average fuel. SHED temperature cycled each 24-hour period from 72 to 96 F (22.2 to 35.5C). Measurements included speciation of individual hydrocarbons in the SHED as well as total mass emissions at the end of each of the three 24-hour test periods. Previous evaporative emission studies provided evidence that permeation and/or fuel seepage could contribute significantly to the mass of diurnal and hot soak emissions. Data from this investigation were used to quantify the contribution of liquid fuel to total SHED emissions during diurnal testing. A calculation method, based on the concentration of 29 select hydrocarbons in the SHED, is presented to apportion SHED emissions between those associated with liquid fuel losses and those associated with fuel tank head space vapor losses.

Species analyses have been performed on engine-out and tailpipe hydrocarbon mass emissions to help understand why fuels with higher T50 and/or T90 distillation temperatures produce higher engine-out and tailpipe hydrocarbon emissions and why fuels with higher T90 distillation temperatures produce higher engine-out and tailpipe specific reactivities. Species analyses were also performed to examine the effects of fuel sulfur level on engine-out and tailpipe species and specific reactivities. These analyses were performed on three different test-vehicle fleets representing varying levels of emissions control technology and the effect of emissions control technology was examined. Individual hydrocarbon species concentrations in both the engine-out and tailpipe were found to correlate linearly with the concentrations of the same species in the fuel, implying that a small fraction of the fuel escapes the combustion process and conversion over the catalyst.