No abstract

Carbonyls can be toxic and highly reactive in the atmosphere. To quantify trends in carbonyl emissions from light-duty (LD) vehicles, measurements were made in a San Francisco Bay area highwaytunnel bore containing essentially all LD vehicles during the summers of 1999, 2001, and 2006. The LD vehicle emission factor for formaldehyde, the most abundant carbonyl, did not change between 1999 and 2001, then decreased by 61 +/- 7% between 2001 and 2006. This reduction was due to fleet turnover and the removal of MTBE from gasoline. Acetaldehyde emissions decreased by 19 +/- 2% between 1999 and 2001 and by the same amount between 2001 and 2006. Absent the increased use of ethanol in gasoline after 2003, acetaldehyde emissions would have further decreased by 2006. Carbonyl emission factors for medium- (MD) and heavy-duty (HD) diesel trucks were measured in 2006 in a separate mixed-traffic bore of the tunnel. Emission factors for diesel trucks were higher than those for LD vehicles for all reported carbonyls. Diesel engine exhaust dominates over gasoline engines as a direct source of carbonyl emissions in California. Carbonyl concentrations were also measured in liquid-gasoline samples and were found to be low (< 20 ppm). The gasoline brands that contained ethanol showed higher concentrations of acetaldehyde in unburned fuel versus gasoline that was formulated without ethanol. Measurements of NO2 showed a yearly rate of decrease for LD vehicle emissions similar to that of total NOx in this study. The observed NO2/NOx ratio was 1.2 +/- 0.3% and 3.7 +/- 0.3% for LD vehicles and diesel trucks, respectively.

Ammonia is the primary alkaline gas in the atmosphere and contributes to fine particle mass, visibility problems, and dry and wet deposition. The objective of this research was to measure ammonia and other exhaust emissions from a large sample of on-road vehicles using California phase 2 reformulated gasoline with low sulfur content (∼10 ppm by weight). Vehicle emissions of ammonia, NO x , CO, and CO 2 were measured in the center bore of a San Francisco Bay area highway tunnel on eight 2-h afternoon sampling periods during summer 1999. Ammonia concentrations were divided by total carbon (mainly CO 2 ) concentrations to compute an emission factor of 475 ( 29 mg L -1 (95% C.I.). The molar ratio of nitrogen emitted in the tunnel in the form of ammonia to that emitted in the form of NO x was 0.27 ( 0.01. Emissions of NO x and CO have been measured at this tunnel sampling location since 1994. From 1994 to 1999, emissions decreased by 41 ( 4% for NO x and 54 ( 6% for CO. These reductions include the impacts of turnover in the vehicle fleet and the use of reformulated gasoline. Between 1997 and 1999, when fuel properties did not change significantly, emissions of NO x and CO decreased by 26 ( 2% and 31 ( 3%, respectively. While use of three-way catalytic converters has contributed to decreases in NO x and CO emissions, their use, in combination with fuel-rich engine operation, is the likely cause of the ammonia emissions from motor vehicles observed during this study.

[1] A chemical mass balance approach is used to determine the relative contributions of evaporative versus tailpipe sources to motor vehicle volatile organic compound (VOC) emissions. Contributions were determined by reconciling time-resolved ambient VOC concentrations measured downwind of Sacramento, California, in summer 2001 with source speciation profiles. A composite liquid fuel speciation profile was determined from gasoline samples collected at Sacramento area service stations. Vapor-liquid equilibrium relationships were used to determine the corresponding headspace vapor composition. VOC concentrations measured in a highway tunnel were used to define the composition of running vehicle emissions. The chemical mass balance analysis indicated that headspace vapor contributions ranged from 7 to 29% of total vehicle-related VOC depending on time of day and day of week, with a mean daytime contribution of 17.0 ± 0.9% (mean ± 95% CI). A positive association between the headspace vapor contribution and ambient air temperature was found for afternoon hours. We estimate a 6.5 ± 2.5% increase in vapor pressure-driven evaporative emissions and at least a 1.3 ± 0.4% increase in daily total (exhaust plus evaporative) VOC emissions from motor vehicles per degree Celsius increase in maximum temperature.

Laboratory studies have provided a foundation of knowledge regarding vehicle emissions, but questions remain regarding the relationship between on-road vehicle emissions and changes in vehicle speed and engine load that occur as driving conditions change. Light-duty vehicle emissions of CO, NO x , and NMHC were quantified as functions of vehicle speed and engine load in a California highway tunnel for downhill and uphill traffic on a ∼4% grade. Emissions were measured throughout the day; average speed decreased inside the tunnel as traffic volume increased. Emissions of CO were typically 16-34 g L -1 (i.e., grams of CO emitted per liter of gasoline consumed) during downhill driving and ranged from 27 to 75 g L -1 during uphill driving. Downhill driving and moderate-speed uphill driving resulted in similar CO emission factors. The factor of 2 increase in CO emissions observed during higherspeed uphill driving is likely evidence of enriched engine fuel/ air ratios; this was unexpected because uphill driving observed in this study occurred at moderate engine loads within the range experienced during the city driving cycle of the U.S. emissions certification test. Emissions of NO x (as NO 2 ) were typically 1.1-3.3 g L -1 for downhill driving and varied between 3.8 and 5.3 g L -1 for uphill driving. Unlike observations for CO, all uphill driving conditions resulted in higher NO x emission factors as compared to downhill driving. NO x emissions increased with vehicle speed for uphill driving but not as strongly as CO emissions. Emissions of CO and NO x are functions of both vehicle speed and specific power; neither parameter alone captures all the relevant effects on emissions. In contrast to results for CO and NO x reported here and results for NMHC reported previously by Pierson et al. (Atmos. Environ. 1996, 30, 2233-2256, emissions of NMHC per unit of fuel burned for downhill driving were over 3 times greater than NMHC emissions for uphill driving. Emission rates of CO and NO x varied more with driving conditions when expressed per unit distance traveled rather than per unit fuel burned while NMHC emission rates normalized to distance traveled were approximately constant for uphill versus downhill driving during peak traffic periods.