Environ. Sci. Technol. 1999, 33, 3091-3099
Characterization of Polycyclic Aromatic Hydrocarbons in Motor Vehicle Fuels and Exhaust Emissions LINSEY C. MARR, THOMAS W. KIRCHSTETTER, AND ROBERT A. HARLEY* Department of Civil and Environmental Engineering, University of California, Berkeley, California 94720-1710 ANTONIO H. MIGUEL CE-CERT 022, University of California, Riverside, California 92521-0434 SUSANNE V. HERING Aerosol Dynamics Inc., 2329 Fourth Street, Berkeley, California 94710 S. KATHARINE HAMMOND School of Public Health, University of California, Berkeley, California 94720
Motor vehicles are a significant source of polycyclic aromatic hydrocarbon (PAH) emissions. Improved understanding of the relationship between fuel composition and PAH emissions is needed to determine whether fuel reformulation is a viable approach for reducing PAH emissions. PAH concentrations were quantified in gasoline and diesel fuel samples collected in summer 1997 in northern California. Naphthalene was the predominant PAH in both fuels, with concentrations of up to 2600 mg L-1 in gasoline and 1600 mg L-1 in diesel fuel. Particle-phase PAH size distributions and exhaust emission factors were measured in two bores of a roadway tunnel. Emission factors were determined separately for light-duty vehicles and for heavyduty diesel trucks, based on measurements of PAHs, CO, and CO2. Particle-phase emission factors, expressed per unit mass of fuel burned, ranged up to 21 µg kg-1 for benzo[ghi]perylene for light-duty vehicles and up to ∼1000 µg kg-1 for pyrene for heavy-duty diesel vehicles. Light-duty vehicles were found to be a significant source of heavier (four- and five-ring) PAHs, whereas heavy-duty diesel engines were the dominant source of three-ring PAHs, such as fluoranthene and pyrene. While no correlation between heavy-duty diesel truck PAH emission factors and PAH concentrations in diesel fuel was found, light-duty vehicle PAH emission factors were found to be correlated with PAH concentrations in gasoline, suggesting that gasoline reformulation may be effective in reducing PAH emissions from motor vehicles.
Introduction Polycyclic aromatic hydrocarbons (PAHs) constitute a class of organic compounds that include potent mutagenic and * Corresponding author phone: (510)643-9168; fax: (510)642-7483; e-mail:
[email protected]. 10.1021/es981227l CCC: $18.00 Published on Web 08/13/1999
1999 American Chemical Society
carcinogenic compounds (1) and are often associated with combustion processes. Motor vehicles are a significant source of PAHs in the atmosphere, accounting for about one-third of total PAH emissions in the United States (2), and particlephase PAHs measured in urban air and roadway tunnels have been found in the respirable size range (3-8). PAHs are found in both the gas and particle phases, depending on the vapor pressure of the PAH. Smaller PAHs are found predominantly in the gas phase, while PAHs with four or more rings are found mainly in the particle phase (9). There are a number of sources of PAHs in motor vehicle exhaust, including unburned fuel, lubricating oil, and pyrosynthesis from lower molecular weight aromatics. PAHs in unburned diesel fuel have been shown to be the primary contributor to particle-phase PAHs in the exhaust of a directinjection diesel engine (10). In radiotracer experiments, benzo[a]pyrene in the fuel was found to be the major source of benzo[a]pyrene in the exhaust of a diesel engine, whereas lubricating oil and pyrosynthesis combined were found to contribute no more than 20% to benzo[a]pyrene in exhaust emissions (11). Similarly, unburned fuel was the predominant source of particle-phase exhaust PAHs for two direct-injection diesel engines (12). This investigation also concluded that the pyrosynthesis of two- and three-ring PAHs may contribute to exhaust emissions of five-ring and larger PAHs and that lubricating oil affects the behavior of PAHs in the combustion process. Naphthalene surviving combustion has been identified as the source of 24% of naphthalene in diesel exhaust emissions, and the conversion of fuel 2-methylnaphthalene to naphthalene in the combustion chamber has also been confirmed by radiotracer experiments (13). Less is known about the origin of PAHs in gasoline engine exhaust, despite the fact that gasoline engines are more important than diesels as a source of some PAHs (9, 14-16). A study of four different fuels with one fuel-injected, gasolinepowered engine, using the U. S. Federal Test Procedure, found that a linear relationship existed between the concentrations of individual PAHs in the different fuels and their concentrations in the exhaust emissions (17). Numerous studies have quantified PAHs in motor vehicle exhaust (8, 9, 15-20), but substantially fewer have examined PAH concentrations in fuels used by these vehicles. Naphthalene, fluorene, and phenanthrene were found to be the predominant PAHs in French diesel fuel (21). Analysis of eight diesel fuels in Sweden found phenanthrene, methylphenanthrene, and 2-methylanthracene to be the most abundant PAHs in the fuels (18). Another study involving two diesel fuels in England found that total PAH content of the fuels ranged from 1.7 to 4.5% and that one fuel had at least 20 times the benzo[a]pyrene content of the other (10). Fuel made by a Swedish refinery to correspond to U. S. unleaded regular gasoline in 1988 contained 27 mg L-1 of phenanthrene, 8.9 mg L-1 of pyrene, and lesser amounts of other PAHs (17). Mi et al. (20) found that PAH concentrations ranged up to 11.5 mg L-1 in a sample of Taiwanese unleaded gasoline. In general, less is known about PAH concentrations in gasoline than in diesel fuel. In the past, control strategies for motor vehicles emphasized improvements in engine and emission control technologies. More recently, fuels have been reformulated to reduce vehicle emissions. For example, oxygenated compounds, such as methyl tert-butyl ether (MTBE), have been added to gasoline (22), and the sulfur and aromatic contents of diesel fuel have been reduced (23). It is possible that some types of fuel reformulation might help to reduce PAH emissions from motor vehicles (18). Improved understanding VOL. 33, NO. 18, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3091
TABLE 1. Selected Ion Monitoring Program for PAHs Measured in Gasoline and Diesel Fuel species name
abbreviation
molecular weight
number of rings
time after injection (min)
ions monitored
naphthalene acenaphthylene acenaphthene fluorene anthracene phenanthrene fluoranthene pyrene benz[a]anthracene chrysene benzo[b]fluoranthene benzo[k]fluoranthene benzo[a]pyrene benzo[ghi]perylene indeno[1,2,3-cd]pyrene dibenz[a,h]anthracene
NAP ACY ACE FLU ANT PHE FLT PYR BAA CRY BBF BKF BAP BGP IND DBA
128 152 154 166 178 178 202 202 228 228 252 252 252 276 276 278
2 3 3 3 3 3 4 4 4 4 5 5 5 6 6 5
10.0-16.0 16.0-18.4 18.4-19.5 19.5-23.0 23.0-27.0 23.0-27.0 27.0-32.0 27.0-32.0 32.0-37.0 32.0-37.0 37.0-47.0 37.0-47.0 37.0-47.0 51.0-64.0 47.0-50.0 50.0-51.0
128, 102 152, 126 153, 126 166, 83 178, 152 178, 152 202, 101 202, 101 228, 113 228, 113 252, 126 252, 126 252, 126 276, 138 276, 138 278, 139
TABLE 2. Sampling Conditions at Tunnel, Summer 1997
a
date
bore sampleda
traffic volume (veh h-1)
% HD diesel
tunnel
July 21 July 22 July 23 July 24 July 31 August 1 August 4 August 5
1 1 1 1 2 2 2 2
2191 2333 2299 2517 3897 4139 4191 4216
4.8 3.6 4.6 3.9 0.33 0.29 0.35 0.35
720 735 747 779 1008 946 1053 1090
CO2 (ppm) ambient 369 370 366 384 365 369 387 384
tunnel 19.0 16.9 19.4 19.4 27.5 26.1 27.5 27.6
CO (ppm) ambient 1.2 0.9 1.1 1.3 0.8 0.9 1.3 0.8
Bore 2 was reserved for use by light-duty vehicles.
is, therefore, needed of the relationship between fuel composition and particle-phase PAH emissions for both gasoline and diesel engines. The objectives of this study were (1) to quantify PAH concentrations in commercial gasoline and diesel fuel samples, (2) to measure on-road particle-phase PAH emission factors for light-duty (mainly gasoline) and heavy-duty diesel vehicles, and (3) to determine the size distribution of PAH emissions in the particle phase. These measurements were made to gain insight into the origin of PAHs in motor vehicle exhaust.
Methods Fuel Sampling. Gasoline and diesel fuel samples were collected in August 1997 using standard sampling procedures developed by the California Air Resources Board. Gasoline samples were collected from service stations in Berkeley and Oakland, representing the five major brands of gasoline sold in the San Francisco Bay Area. The gasoline brands were designated with the letters A through E. Regular and premium grades were collected at four of the service stations, and midgrade and premium gasoline were sampled at one station. Approximately 750 mL of each gasoline sample was dispensed from service station pumps into 1 L steel canisters, and these samples were stored in a refrigerator at 4 °C. Average properties of gasoline sold in the San Francisco Bay Area during the summer of 1997 can be found in Kirchstetter et al. (24). The Compliance Division of the Air Resources Board collected diesel fuel samples from oil refineries in the San Francisco Bay Area. The diesel samples were designated with the numbers 1 through 5. These were also stored in 1 L steel canisters at 4 °C. The California Air Resources Board (25) measured sulfur, aromatic, and total PAH content of these samples, and these fuel properties are reported in Table 4. 3092
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 18, 1999
Fuel Analysis. Gasoline samples were diluted 10 to 2000 times in n-heptane and were analyzed on a Hewlett-Packard 5890 gas chromatograph connected to a Hewlett-Packard 5872 mass spectrometer (GC/MS). The autoinjector was programmed to make 1 µL splitless injections of all samples and standards onto a 30 m, 25 µm inner diameter, 86% dimethyl, 14% cyanopropyl column (J&W Scientific DB-1701, #05 600 367). The oven temperature was increased from 40 to 280 °C at a rate of 7 °C min-1, and then held at 280 °C for 30 min for a total run time of 64 min. The mass spectrometer was programmed in single ion mode, and Table 1 shows the program timing and ions monitored for each PAH. PAHs were identified by comparing retention times and ion ratios to those of a reference standard. A standard containing 16 priority PAHs (Radian ERS-010) shown in Table 1 was used to develop calibration curves. These PAHs incorporate a range of molecular weights of PAHs commonly found in the environment, some of which are both mutagenic and carcinogenic. The limit of detection, determined as the concentration of the standard at which the signal-to-noise ratio was at least five, for PAH analysis by GC/MS was ∼0.005 mg L-1, with slightly lower limits for the low molecular weight PAHs and slightly higher limits for the high molecular weight PAHs. The coefficient of variation for repeat injections of a standard was less than 5% for all sixteen PAHs, and for the samples, it was ∼10%, including the uncertainty in the manual integration of peaks. Blanks and standards were included for analysis at regular intervals during sample runs on the GC/MS. Because PAH peaks in diesel fuel samples were difficult to distinguish from other fuel components, a solid phase extraction technique was used to separate PAHs from other diesel fuel components. The method for PAH separation used in this study was modified slightly from a published method (26). PAHs in diesel fuel samples were separated by using
TABLE 3. PAH Concentrations in Ten Bay Area Gasoline Samples, Summer 1997 PAH (mg L-1)
rega
NAP ACY ACE FLU ANT PHE FLT PYR BAA CRY BBF BKF BAP BGP IND DBA
800 0.21 3.3 5.1 13 7.2 0.90 1.1 0.40 0.19 0.092 0.026 0.18 0.14 0.040 < lod
A
a Regular grade gasoline. mg L-1.
b
premb
midc
2100 0.29 6.3 7.8 20 14 2.6 3.1 1.4 0.82 0.28 0.16 0.50 0.55 0.074 < lod
1800 0.27 6.8 7.2 23 14 2.4 3.4 1.6 0.83 0.34 0.24 0.95 0.68 0.10 < lod
B
reg
2400 0.27 7.2 7.9 25 16 3.3 3.9 2.0 1.3 0.34 0.21 0.84 0.74 0.16 < lod
790 0.28 3.6 4.4 14 8.0 1.0 1.2 0.44 0.20 0.11 0.050 0.29 0.18 0.051 < lod
L-1)
NAP (mg ACY (mg L-1) ACE (mg L-1) FLU (mg L-1) PHE (mg L-1) FLT (mg L-1) PYR (mg L-1) CRY (mg L-1) BGP (mg L-1) sulfur (ppmw) aromatics (weight %) total PAH (weight %)
prem
reg
900 0.21 4.8 7.4 19 9.8 1.7 1.9 0.42 0.33 0.22 0.10 0.50 0.36 0.14 < lod
1900 0.27 7.4 7.8 10 14 2.6 4.8 1.1 0.91 0.29 0.20 0.47 0.76 0.096 < lod
D
prem
reg
2600 0.32 6.2 6.0 18 15 3.5 4.1 2.0 1.5 0.44 0.27 0.71 0.85 0.17 < lod
120 0.12 0.25 0.76 0.23 0.51 0.022 0.030 < lodd < lod < lod < lod < lod < lod < lod < lod
Premium grade gasoline. c Midgrade gasoline; regular grade was not sampled.
TABLE 4. Properties of Bay Area Diesel Fuel Samples, Summer 1997 fuel propertya
C
prem
1
2
3
4
5
1600 1.4 59 150 240 3.5 39
