In-Vehicle Exposure to Aldehydes While Commuting on Real Commuter Routes in a Korean Urban Area

Environmental Research Section A 88, 44}51 (2002) doi:10.1006/enrs.2001.4313, available online at http://www.idealibrary.com on

In-Vehicle Exposure to Aldehydes While Commuting on Real Commuter Routes in a Korean Urban Area Wan-Kuen Jo1 and Jin-Woo Lee Department of Environmental Engineering, Kyungpook National University, Taegu 702-701, Korea Received October 18, 2001

and tert-butyl methyl ether (MTBE), to gasoline, the aldehyde vehicle emissions become even higher (Dumdei and O’Brien, 1984; Williams et al., 1990; Gorse et al., 1992; Hoekman, 1992; Gabele and Knapp, 1993; Stump et al., 1994). Consequently, the aldehyde vehicle emissions increase the aldehyde levels in roadway air, which then penetrate into vehicles, thereby elevating the in-vehicle aldehyde levels. According to the Korean Petroleum Corp., the gasoline manufactured by all 7ve Korean petroleum companies contains 6.0 to 8.0% MTBE by volume. MTBE has been added to gasoline in Korea for almost a decade to enhance the octane ratings. Accordingly, Korean vehicle occupants can be expected to experience an elevated health risk from apparent invehicle exposure to toxic aldehydes. With regard to the toxicity of the two most abundant aldehydes (formaldehyde and acetaldehyde) in vehicle exhaust emissions (Gorse et al., 1992), the U.S. Environmental Protection Agency (EPA) has classi7ed formaldehyde as group B1, probable human carcinogen (U.S. EPA, 1987a, 1991) and acetaldehyde as group B2, probable human carcinogen (U.S. EPA, 1987b). Moreover, the International Agency for Research on Cancer (IARC, 1987) concurs that formaldehyde is most likely carcinogenic to humans. However, only limited information is currently available in the scienti7c literature on in-vehicle exposure to aldehydes (Rodes et al., 1998), although there have been several reports on exposure to aldehydes in various microenvironments, such as residential indoor and outdoor air and workplaces (Grosjean, 1991; Grosjean et al., 1993, 1996; Zhang et al., 1994; Reiss et al., 1995; Jurvelin et al., 2001). Accordingly, the current study was undertaken to evaluate in-vehicle exposure to two toxic aldehydes (formaldehyde and acetaldehyde) with a focus on three factors (transportation mode, passenger car type, and commuting season) that could in8uence the amount of exposure to the two aldehydes.

This study evaluated in-vehicle exposure to formaldehyde and acetaldehyde on actual commuting routes, while focusing on three factors (transportation mode, passenger car type, and commuting season). A total of 40 passenger car commuters and 20 public bus commuters were recruited. The same commuters participated in both the summer and the winter studies. The transportation mode and passenger car type were found to have little effect on the in-vehicle aldehyde levels. Conversely, the commuting season did inBuence the in-vehicle aldehyde levels. Meanwhile, the mean formaldehyde-to-acetaldehyde concentration ratios were similar in both the passenger cars and the public buses, plus there were signiAcant correlations ( P:0.0001) between formaldehyde and acetaldehyde concentrations for both the passenger cars and the public buses. This study also conArmed that, under Korean commuting conditions, vehicle interiors are an important microenvironment for exposure to formaldehyde and acetaldehyde. The mean in-car concentrations were 20.0 and 8.9 ppb for formaldehyde and acetaldehyde, respectively. Similarly, the mean in-bus concentrations were 21.2 and 9.1 ppb for formaldehyde and acetaldehyde, respectively. Furthermore, the in-vehicle formaldehyde levels were higher than those of a previous California study.  2002 Elsevier Science

Key Words: aldehydes; in-vehicle exposure; passenger car type; season; transportation mode.

INTRODUCTION

Motor vehicle exhaust emissions are a major source of aldehydes in urban areas (Gorse et al., 1992; Leonard, 1992; Lorang, 1992), and with the addition of oxygenates, such as methanol, ethanol, 1

To whom correspondence should be addressed. Fax:#82-53950-6579. E-mail: [email protected]. 44 0013-9351/02 $35.00  2002 Elsevier Science All rights reserved.

IN-VEHICLE EXPOSURE OF REAL COMMUTERS TO ALDEHYDES

Furthermore, this study was conducted using actual commuting routes, not hypothetical routes as used in many previous in-vehicle exposure studies of volatile organic compounds (VOCs). The transportation mode included public buses and passenger cars, which are the two most popular forms of transport in the study area, as in many other developing countries. With regard to the passenger car type, the cars were classi7ed based on their size, model/year, and mileage. The seasonal factor was examined based on the temperature differences between winter and summer. The study area, Taegu, is the third largest city in Korea with a population of 2.49 million and population density of 2812/km2. Traf7c movement in the city is typically slow during the morning and evening rush hour periods. According to the statistical yearbook for Taegu published in July 2000, the numbers of vehicles registered in this area were as follows: 450,975 passenger cars, 18,731 taxicabs, 61,677 buses (3024 public buses), 136,344 trucks, 121,722 motorcycles, and 895 special cars. Most of these passenger cars and motorcycles consume gasoline, while the buses and taxis mainly use diesel fuel and liquid petroleum gas, respectively.

EXPERIMENTAL METHODS

Survey Protocol Two experimental designs were developed to measure the in-vehicle concentrations of carbon monoxide (CO), PM10 , and VOCs, including aldehydes, while commuting on real commuter routes. Both designs were used for the two seasonal temperature extremes: summer, June to August of 2000, and winter, November and December of 2000 and January of 2001. This paper focuses on the two most abundant aldehydes (formaldehyde and acetaldehyde) in vehicle exhaust emissions. The 7rst experiment was designed to measure the aldehyde concentrations inside the passenger cars, whereas the second one measured the aldehyde concentrations inside the public buses. A total of 40 actual passenger car commuters volunteered to take part in the 7rst experiment. The same commuters participated in both the summer and the winter studies. The commuters were classi7ed into two equal groups: drivers of large- and small-sized cars. The car size was determined based on the engine size: large was more than a 1800-cc engine and small was less than a 1500-cc engine. Plus, each group included 10 drivers of new cars (1998 or later models with less than 20,000 miles)

45

and 10 drivers of older cars (models between 1994 and 1997 with more than 40,000 miles). The information related to the cars is shown in Table 1. All the passenger cars were equipped with electronic fuel-injected engines and catalytic converters, used unleaded gasoline fuel, and were apparently functioning normally during the entire experimental period. Each car was sampled twice (morning and evening) in a single season. Another criterion for the commuter selection was the duration of the commute, between 30 and 60 min for a one-way commute, which was determined in the selection interviews for participants. The drivers were all undergraduate or graduate students or faculty members of Kyungpook National University (KNU) in Taegu. The samples were collected by trained technicians sitting in the front passenger seats. The commute routes were real, not hypothetical. The windows and vents of the passenger cars were kept closed for most of the sampling time, with the temperature level and blower speed set to the personal comfort level of the occupants. During most of the summer sampling hours, the drivers used air conditioning systems. To prevent any interference due to tobacco smoke, the participants of this experiment were asked not to smoke during sampling. Twenty public bus commuters to Kyungpook National University volunteered to take part in the second experiment. As with the car commuters, the same bus commuters participated in both the summer and the winter studies. All the public buses surveyed in this study were 40-seater diesels. As with the selection of car commuters, the public bus commuters were only selected if their traveling time for a one-way commute was between 30 and 60 min. This consistency assists in the systematic comparison of public and private transport modes for invehicle exposure to the target compounds. The in-bus samples were collected halfway down the length of the bus. The bus commuters were trained so that they could collect their own in-bus samples. A trained technician also accompanied the bus commuters to provide help. The same ventilation conditions as for the passenger cars were observed with the public buses for most of the sampling time, except that the bus doors were opened when passengers got on and/or off. Smoking is not allowed in public buses in Korea. The in-vehicle air samples were collected from the breathing zone of the passengers during rush hour on standard workdays (Monday through Friday). An aldehyde sample was collected during both a morning (7:00}9:00) and an evening (5:30}7:30) commute.

46

JO AND LEE

TABLE 1 Information on Passenger Cars Surveyed in Current Study

Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Manufacturing company

Engine size (cc)

Daewoo

2200 2000 2000 1800 1800 1800 1800 1500 1500 1300 1300 1300 800 800 800 2500 2500 2500 2000 1800 1800 1800 1800 1800 1500 1500 1500 1500 1500 1300 1300 1300 800 2000 2000 1800 1300 1300 1300 2200

Hyundai

Kia

Samsung

Model/year Brougham, 1996 Brougham, 1995 Brougham, 1994 Leganza, 1998 Prince, 1996 Prince, 1995 Prince, 1994 Nubira II, 2000 Nubira, 1998 Lanos, 1998 Lanos, 1998 Lanos, 1998 Matiz, 1998 Matiz, 1998 Tico, 1994 Grandeur, 2000 Grandeur, 1999 Grandeur, 1996 EF Sonata, 1999 Marcia, 1996 Sonata III, 1998 Sonata III, 1998 Sonata III, 1998 Sonata III, 1998 Avante, 1995 Avante, 1995 Excel, 1995 Verna, 2000 Verna, 1999 Accent, 1996 Accent, 1995 Excel, 1994 Atoz, 1998 Potentia, 1996 Potentia, 1995 Credos, 1998 Avella, 1995 Pride, 1994 Pride b, 1995 SM520V, 1999

Mileage

Classi7cation

51,875 75,200 81,500 19,883 47,510 50,110 68,720 3,120 19,300 18,530 19,300 9,370 19,980 6,250 77,500 4.060 7,860 25,625 12,500 68,750 17,950 19,300 18,769 18,350 71,800 68,734 75,600 1,250 6,870 70,000 75,150 65,600 12,500 35,625 67,890 19,900 41,800 85,700 68,700 4,000

Large-Old Large-Old Large-Old Large-New Large-Old Large-Old Large-Old Small-New Small-New Small-New Small-New Small-New Small-New Small-New Small-Old Large-New Large-New Large-Old Large-New Large-Old Large-New Large-New Large-New Large-New Small-Old Small-Old Small-Old Small-New Small-New Small-Old Small-Old Small-Old Small-New Large-Old Large-Old Large-New Small-Old Small-Old Small-Old Large-New

Note. Car classi7cation is de7ned in text.

Vehicle speed was calculated by dividing the travel distance by the travel time. For each transportation mode, median speed was calculated using the speed distribution of all commutes. The summer car speed per commute ranged from 13 to 51 km/h, with a median of 29 km/h, whereas the bus speed ranged from 20 to 45 km/h, with a median of 30 km/h. For the winter study, the car speed per commute ranged from 18 to 46 km/h, with a median of 30 km/h, whereas the bus speed ranged from 21 to 47 km/h, with a median of 33 km/h.

Sampling and Analytical Methods The sampling and analytical procedure for formaldehyde and acetaldehyde was a modi7cation of the U.S. EPA Method TO-11A (U.S. EPA, 1999). The formaldehyde and acetaldehyde were collected on Sep-Pak DNPH-silica cartridges (Waters, USA) using Handy Samplers (KIMOTO HS-7, Japan) at a nominal 8ow rate of 1.4 L/min. The 8ow rates were measured using a digital 8ow meter (Field-Cal 650; Humonics Inc., USA) prior to and following the

47

IN-VEHICLE EXPOSURE OF REAL COMMUTERS TO ALDEHYDES

collection of each sample. No samples departed by more than 10% from the initial 8ow rate during this study. The average of these two rates was then used as the sample 8ow rate in all volume calculations. The volume collected ranged from 43 to 89 L. The air volumes were large enough for the sensitivity of the analytical system and small enough to remain below the breakthrough volumes of the target aldehydes. To avoid any negative interference from O3 , a PTFE tube (1 cm inside diameter and 10 cm length) with KI coating was used as an O3 scrubber for all samples (U.S. EPA, 1984; Tejada, 1986; Arnts and Tejada, 1989). The analysis was usually done within 3 days of collection to minimize the potential artifact reaction of the sample. After sampling, the cartridges were capped and stored in a refrigerator prior to extraction. The aldehydes were extracted with 4 mL of acetonitrile (ACN) (HPLC grade) and then analyzed by a high-performance liquid chromatograph with UV detection (Shimadzu LC-10A, Japan) (Risner, 1995). The chromatic separations were carried out using a solvent isocratic elution at a 8ow rate of 1.5 mL/min in an analytical column (0.46;250 cm, Shimadzu CLC-ODS (M)), preceded by a C18 insert guard column. An isocratic run was used for over 25 min with 65% water, 30% ACN, and 5% tetrahydrofuran. The quality assurance program included 7eld blank cartridges, spiked samples and duplicate measurements, and interlaboratory comparisons. To check the quantitative response, known standards of formaldehyde and acetaldehyde prepared from purchased solutions of aldehyde DNPH derivatives (Supelco Inc., USA) were directly injected into a cartridge to transfer the target compounds to the HPLC. When the quantitative response differed by more than $10% from that predicted by the speci7ed calibration equation, a new calibration equation was determined. The mean concentrations measured in the 7eld blanks were 0.29 ppb with a range

of 0.23 to 0.38 ppb and 0.38 ppb with a range of 0.29 to 0.47 ppb (using an assumed sampling volume of 55 L). The detection limits, de7ned as three times the standard deviation of the 7eld blanks, were 0.14 ppb for formaldehyde and 0.18 ppb for acetaldehyde. The laboratories of two institutes, KNU and Kyungpook Province Health Institute, compared their analytical techniques by analyzing the same 4-mL ACN extracts of 7ve samples. The analytical results were within $10% for all 7ve samples. The recovery ef7ciency was calculated to be 94% for formaldehyde and 95% for acetaldehyde. Twenty duplicate samples were collected for each sampling period to test the precision of the sampling and analytical techniques. The mean relative standard deviations of the duplicate measurements were less than 10% for the two aldehydes. Statistical Analyses The statistical analyses were performed using the SAS program (Version 6.1) on a personal computer. The paired sample means were analyzed using a nonparametric test (Wilcoxon test), as a statistical test of normality (Shapiro}Wilk statistics) indicated that the data sets were nonnormally distributed. Spearman correlation coef7cients were calculated to examine the relationship between the formaldehyde and the acetaldehyde concentrations. The criterion for signi7cance in the procedures was P:0.05. RESULTS AND DISCUSSION

Concentration Levels According to Transportation Mode The aldehyde concentrations measured in the passenger cars and public buses during actual commuting are summarized in Table 2. Unlike in-vehicle aromatic VOC levels, previously reported on by Jo and Park (1999), the in-vehicle aldehyde levels were not found to be dependent on the transportation

TABLE 2 Summary of Aldehyde Concentrations (ppb) inside Passenger Cars and Public Buses Passenger car

Public bus

Compound

Min

Median

Max

Mean

S.D.

F/Aa

Min

Median

Max

Mean

S.D.

F/Aa

Formaldehyde Acetaldehyde

8.8 1.4

19.1 7.0

39.8 29.1

20.0 8.9

7.3 5.7

2.2

8.7 2.5

20.1 7.5

28.8 15.5

21.2 9.1

5.3 3.2

2.3

Note. Number of samples: N"80 for passenger car; N"40 for public bus. a Mean formaldehyde-to-acetaldehyde ratios; signi7cant correlation between formaldehyde and acetaldehyde concentrations with P:0.0001 for both passenger cars (Spearman correlation coef7cient"0.61) and public buses (Spearman correlation coef7cient"0.58).

48

JO AND LEE

mode. A nonparametric test showed that the invehicle aldehyde concentrations were not signi7cantly different between the passenger cars and the public buses. Conversely, Jo and Park (1999) reported that the in-vehicle levels of six aromatic VOCs were signi7cantly higher (P:0.05) in passenger cars than in public buses. The main reason for the difference between the results from the present study and those from Jo and Park’s (1999) study would appear to be the sources of the compounds that penetrated into the vehicles. With regard to the aromatic VOCs, the in-car levels re8ect the combined effect of the car engine running loss emissions (including possible evaporation losses from tank and engine) and roadway air levels, whereas the in-bus levels only re8ect the roadway air levels as, since buses are diesel-fueled, the bus engine running loss emissions are not a source (Jo and Park, 1999). As such, higher in-vehicle aromatic VOC levels in passenger cars can be expected. However, in the current study, the in-vehicle aldehyde levels in both the passenger cars and the public buses re8ected the same major source, i.e., the roadway air aldehyde levels, which depend on in situ photochemical processes and exhaust emissions, rather than engine running loss emissions (Gorse et al., 1992; Leonard, 1992; Lorang, 1992). This was also supported by the similar mean formaldehyde-to-acetaldehyde concentration ratios in the two transportation modes (Table 2). The ratios were 2.2 and 2.3 for the passenger cars and public buses, respectively. Furthermore, there were signi7cant correlations between the formaldehyde and the acetaldehyde concentrations with P:0.0001 for both the passenger cars (Spearman correlation coef7cient"0.61) and the public buses (Spearman correlation coef7cient"0.58). The mean formaldehyde-to-acetaldehyde concentration ratios measured in-vehicle were very close to those of ambient air (2.3), as reported by Hwang et al. (1996) who measured ambient air concentrations at 36 public facility sites located in the same area as the present study. As with in-vehicle levels, since public facilities are typically located next to a road, the ambient air levels are primarily in8uenced by the roadway aldehyde levels. In contrast, the mean formaldehyde-to-acetaldehyde concentration ratios measured in-vehicle were different from those of other microenvironments, as reported by Jurvelin et al. (2001). In their study, the mean formaldehyde-to-acetaldehyde concentration ratios measured in Helsinki were 1.69 in outdoor residential areas, 3.46 in indoor residential areas, and 5.93 in workplaces.

The in-vehicle levels measured in the present study were much higher than the ambient air levels measured in other studies. The mean in-car concentrations were 20.0 and 8.9 ppb for formaldehyde and acetaldehyde, respectively. Similarly, the mean inbus concentrations were 21.2 and 9.1 ppb for formaldehyde and acetaldehyde, respectively. Meanwhile, Hwang et al. (1996) reported on a mean range of formaldehyde and acetaldehyde concentrations from 3.5 to 10.3 ppb and from 2.3 to 4.7 ppb, respectively. The most probable cause for the concentration difference between the Hwang et al. (1996) study and the present study would appear to be the dilution of the target compounds relative to the distance from the emission source, i.e., roadway vehicles. In the previous study carried out in Helsinki ( Jurvelin et al., 2001), the mean ambient air concentrations measured in residential areas were 2.6 and 1.5 ppb for formaldehyde and acetaldehyde, respectively. Rodes et al. (1998) reported that the mean range of ambient air formaldehyde concentrations was 1.6 to 3.2 ppb in Sacramento and (5.7 to 15.4 ppb in Los Angeles. The in-vehicle formaldehyde levels in the current study were still higher than those reported by Rodes et al. (1998), where the mean range of in-vehicle formaldehyde concentrations was 4.1 to 11.4 ppb in Sacramento and less than the quanti7cation limit to 17.9 ppb in Los Angeles. The lg/m3 unit employed in this study was converted to ‘‘ppb’’ for comparison. The concentration difference between the Rodes et al. (1998) study and the present study would be a re8ection of the combined effects of driving parameters such as vehicle type, driving route, driving period, vehicle ventilation, driving speed, and fuel type and regional environmental conditions such as temperature and dispersion or turbulence variability. Concentration Levels According to Passenger Car Type Table 3 shows the median in-vehicle aldehyde concentrations measured inside the passenger cars based on their size, model/year, and mileage. Information on the passenger cars and car types is presented in Table 1. There are various possible sources of formaldehyde in the interior of a vehicle, including insulating materials and fabrics (Humfrey et al., 1996), plus a higher formaldehyde emission would be expected from the interior materials in a new car compared to an older model. In addition, the in-vehicle ventilation was relatively low even during the summer experimental period, as indicated under Experimental Methods. However,

49

IN-VEHICLE EXPOSURE OF REAL COMMUTERS TO ALDEHYDES

although higher in-vehicle formaldehyde concentrations were anticipated in the new passenger cars compared to the older ones, neither the car size nor the model year and mileage created any signi7cant difference in the in-vehicle concentrations of formaldehyde. These results imply that the penetration of roadway formaldehyde into the vehicles exceeded any potential formaldehyde emission from the interior materials. Plus, the vehicle ventilation, which varied slightly between the car types, also did not appear to be a signi7cant factor for the in-vehicle aldehyde levels. The current results are consistent with those of Rodes et al. (1998), who also found no signi7cant effect related to vehicle type on in-vehicle levels of several pollutants, including formaldehyde. As with the in-vehicle formaldehyde levels, the vehicle type was also shown to have a minimal effect on the in-vehicle acetaldehyde levels in the current study. Concentration Levels According to Season The seasonal differences in the aldehyde levels inside the passenger cars and public buses were examined. Table 4 shows the median in-vehicle aldehyde concentrations measured during the two extreme temperature seasons, winter and summer. The in-vehicle aldehyde levels in both the passenger cars and the public buses were found to depend on the commute season. A nonparametric test showed that the in-vehicle concentrations were signi7cantly different between winter and summer commutes for each travel mode. The winter to summer ratios of the median formaldehyde concentrations were 1.4 and 1.5 for the passenger cars and public buses, respectively, and for acetaldehyde they were 2.2 and 1.4 for the passenger cars and public buses, respectively. This result is not consistent with that of Weisel et al. (1992), which measured in-vehicle VOC concentrations while idling in winter and summer. They

TABLE 4 Median Concentrations (ppb) inside Passenger Cars and Public Buses Based on Season Passenger cars

Compound

Formaldehyde Acetaldehyde

Winter Summer

22.6 13.0

15.9 5.9

Win/ Sum

Public buses

Winter Summer

1.4a 2.2a

22.9 7.8

15.3 5.6

Win/ Sum

1.5b 1.4b

Note. Number of samples: N"40 for passenger car-Winter; N"40 for passenger car-Summer; N"20 for public bus-Winter; N"20 for public bus-Summer. a A signi7cant difference between two seasons at P:0.0001. b A signi7cant difference between two seasons at P:0.0004.

reported that in-vehicle concentrations measured while idling were greater in summer than in winter due to a higher evaporation level of the measured components from the fuel tank and engine in the summer. As previously indicated for VOCs, in-vehicle levels re8ect the combined effect of engine running loss emissions and roadway air levels, whereas in-vehicle aldehyde levels only re8ect roadway air levels. Accordingly, the explanation used by Weisel et al. (1992) is inapplicable to the results of this study. A possible cause for the seasonal difference found in the present study may have been the lower combustion ef7ciency of gasoline in winter, thereby causing higher aldehyde concentrations in the roadway air. This is supported by Bruetsch’s (1981) study, which reported that high tailpipe emissions from motor vehicles are associated with a cold ambient temperature. Other possible parameters for the seasonal differences include wind speed, inversion, mixing height, and driving speed. On the other hand, according to the Korean Petroleum Association, the gasoline constituents were not signi7cantly different in the summer and winter seasons, indicating that the fuel composition was not an important parameter for the seasonal difference.

TABLE 3 Median Concentrations (ppb) inside Passenger Cars Based on Vehicle Type Small

Large

Compound

New

Old

New

Old

Formaldehyde Acetaldehyde

19.9 7.1

17.7 6.9

22.0 7.2

19.3 6.9

Note. Vehicle type is de7ned under Experimental Methods; number of samples: N"20 for Small-New; N"20 for Small-Old; N"20 for Large-New; N"20 for Large-Old.

CONCLUSIONS

The lack of available information on in-vehicle exposure to two toxic aldehydes, formaldehyde and acetaldehyde, motivated the present evaluation of in-vehicle exposure. The current study was conducted using actual commuting routes and focused on assessing three factors (transportation mode, passenger car type, and season) potentially in8uencing the level of exposure to the two aldehydes. The

50

JO AND LEE

transportation modes included public buses and passenger cars. The passenger car types re8ected the car size, model/year, and mileage. The last factor, season, involved the two temperature extremes, winter and summer. The transportation mode and passenger car type or age were found to have little effect on the in-vehicle aldehyde levels, whereas the season did in8uence the in-vehicle aldehyde levels. This seasonal difference was possibly due to the lower combustion ef7ciency of gasoline in winter, thereby causing higher aldehyde concentrations in the roadway air. Meanwhile, it was found that the in-vehicle aldehyde levels in both the passenger cars and the public buses re8ected the same major source, i.e., the roadway air aldehyde levels, which depend on in situ photochemical processes and exhaust emissions rather than engine running loss emissions. This was also supported by the similar mean formaldehyde-to-acetaldehyde concentration ratios in the two transportation modes. Plus, there were signi7cant correlations between the formaldehyde and the acetaldehyde concentrations for both the passenger cars and the public buses. Under Korean commuting conditions, vehicle interiors were con7rmed to be an important microenvironment for exposure to formaldehyde and acetaldehyde. This was supported by the 7nding that the in-vehicle air levels were much higher than ambient air levels reported by other studies. ACKNOWLEDGMENTS This study could not have been accomplished without the dedicated support of 60 (40 passenger car commuters and 20 public bus commuters) volunteers. We also thank 7ve graduate students (K. H. Park, J. W. Oh, G. W. Oh, J. H. Park, and G. K. Kim) from the Department of Environmental Engineering, Kyungpook National University, for their sample collecting and/or analyses. This work was supported by Korea Research Foundation Grant (KRF-2000-041-E00434).

REFERENCES Arnts, R. R., and Tejada, S. B. (1989). 2,4-Dinitrophenylhydrazine-coated silica gel cartridge method for determination of formaldehyde in air: Identi7cation of an ozone interference. Environ. Sci. Technol. 23, 1428}1430. Bruetsch, R. I. (1981). ‘‘Carbon Monoxide and Non-FTP Ambient Temperature.’’ Of7ce of Mobile Source Air Pollution Control, U. S. Environmental Protection Agency, Ann Arbor, MI. Dumdei, B. E., and O’Brien, R. J. (1984). Toluene degradation products in simulated atmospheric conditions. Nature 311, 248}250. Gabele, P. A., and Knapp, K. T. (1993). A characterization of emissions from an early model 8exible-fuel vehicles. J. Air Waste Manage. Assoc. 43, 851}858.

Gorse, R. A., Benson, J. D., Jr., Burns, V. R., Hochhauser, A. M., Koehl, W. J., Painter, L. J., Reuter, R. M., Rippon, B. H., and Rutherford, J. A. (1992). Toxic air pollutant vehicle exhaust emissions with reformulated gasolines. In ‘‘VIP 23 Toxic Air Pollutants from Mobile Sources: Emissions and Health Effects,’’ pp. 55}81. Proceedings of a U.S. EPA/A&WMA International Specialty Conference, Air & Waste Management Association, Pittsburgh, PA. Grosjean, D. (1991). Ambient levels of formaldehyde, acetaldehyde, and formic acid in Southern California: Results of a one-year base-line study. Environ. Sci. Technol. 25, 710}715. Grosjean, E., Grosjean, D., Fraser, M. P., and Cass, G. R. (1996). Air quality model evaluation data for organics. 2. C1}C14 carbonyls in Los Angeles air. Environ. Sci. Technol. 30, 2687}2703. Grosjean, E., Williams, E. L., and Grosjean, D. (1993). Ambient levels of formaldehyde and acetaldehyde in Atlanta, Georgia. J. Air Waste Manage. Assoc. 43, 469}474. Hoekman, S. K. (1992). Speciated measurements and calculated reactivities of vehicle exhaust emissions from conventional and reformulated gasoline. Environ. Sci. Technol. 26, 1206}1216. Humfrey, C., Shuker, L., and Harrison, P. (1996). ‘‘IEH Assessment on Indoor Air Quality in Home,’’ Assessment A2; Medical Research Council, Institute for Environment and Health, Leicester, UK. Hwang, Y. J., Park, S. K., and Baek, S. O. (1996). Measurement of carbonyl compounds in ambient air using a DNPH cartridge coupled with HPLC method. J. Korea Air Poll. Res. Assoc. 12, 199}209. IARC. (1987). ‘‘Monographs on the Evaluation of Carcinogenic Risk of Chemicals to Humans. Supplement 7. Overall Evaluations of Carcinogenicity: An Updating of IARC Monographs Volumes 1 to 42.’’ pp. 211}216. International Agency for Research on Cancer, World Health Organization, Lyon, France. Jo, W. K., and Park, K. H. (1999). Commuter exposure to volatile organic compounds under different driving conditions. Atmos. Environ. 33, 409}417. Jurvelin, J., Vartiainen, M., Jantunen, M., and Pasanen, P. (2001). Personal exposure levels and microenvironmental concentrations of formaldehyde and acetaldehyde in the Helsinki metropolitan area, Finland. J. Air Waste Manage. Assoc. 51, 17}24. Leonard, A. A. (1992). Motor vehicle toxics;The risk in perspective. In ‘‘VIP 23 Toxic Air Pollutants from Mobile Sources: Emissions and Health Effects,’’ pp. 17}27. Proceedings of a U.S. EPA/A&WMA International Specialty Conference, Air & Waste Management Association, Pittsburgh, PA. Lorang, P. (1992). Air toxics from mobile sources and the requirements of the 1990 Clean Air Act Amendments. In ‘‘VIP 23 Toxic Air Pollutants from Mobile Sources: Emissions and Health Effects,’’ pp. 5}14. Proceedings of a U.S. EPA/A&WMA International Specialty Conference, Air & Waste Management Association, Pittsburgh, PA. Reiss, R., Ryan, P. B., Tibbetts, S. J., and Koutrakis, P. (1995). Measurement of organic acids, aldehydes, and ketones in residential environments and their relation to ozone. J. Air Waste Manage. Assoc. 45, 811}822. Risner, C. H. (1995). High-performance liquid chromatographic determination of major carbonyl compounds from various sources in ambient air. J. Chrom. Sci. 33, 168}176.

IN-VEHICLE EXPOSURE OF REAL COMMUTERS TO ALDEHYDES Rodes, C., Sheldon, L., Whitaker, D., Clayton, A., Fitzgerald, K., Flanagan, J., DiGenova, F., Hering, S., and Frazier, C. (1998). ‘‘Measuring Concentrations of Selected Air Pollutants Inside California Vehicles.’’ Final report, ARB Contract No. 95}339, California Air Resources Board, CA. Stump, F. D., Knapp, K. T., Ray, W. D., Siudak, P. D., and Snow, R. F. (1994). In8uence of oxygenated fuels on the emissions from three pre- 1985 light-duty passenger vehicles. J. Air Waste Manage. Assoc. 44, 781}786. Tejada, S. B. (1986). Evaluation of silica gel cartridges coated in situ with acidi7ed 2,4-dinitrophenylhydrazine for sampling aldehydes and ketones in air. Int. J. Environ. Chem. 26, 167}185. Thomas, K. W., Pellizzari, E. D., Clayton, C. A., Whitaker, D. A., and Shores, R. C. (1993). Particle total exposure assessment methodology (PTEAM) 1990 study: Method performance and data quality for personal, indoor, and outdoor monitoring. J. Expos. Anal. Environ. Epidemiol. 3, 203}226. U.S. EPA (1984). ‘‘Compendium of Methods for Determination of Toxic Organic Compounds in Ambient Air.’’ Method 10-11, EPA-600/4-84-041, U.S. Environmental Protection Agency, Research Triangle Park, NC. U.S. EPA (1987a). ‘‘Assessment of Health Risks to Garment Workers and Certain Home Residents from Exposure to For

51

maldehyde.’’ Of7ce of Pesticides and Toxic Substances, U.S. Environmental Protection Agency, Washington, DC. U.S. EPA (1987b). ‘‘Health Assessment Document for Acetaldehyde.’’ Of7ce of Health and Environmental Assessment, U.S. Environmental Protection Agency, Washington, DC. U.S. EPA (1991). ‘‘Formaldehyde Risk Assessment Update: Final Draft.’’ Of7ce of Toxic Substances, U.S. Environmental Protection Agency, Washington, DC. U.S. EPA (1999). ‘‘Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air,’’ 2nd ed. Compendium Method TO-11A, EPA/625/R-96/010b, Center for Environmental Research Information, Of7ce of Research and Development, U.S. Environmental Protection Agency, Cincinnati, OH. Weisel, C. P., Nicholas, N. J., and Lioy, P. J. (1992). Exposure to emissions from gasoline within automobile cabins. J. Expos. Anal. Environ. Epidemiol. 2, 79}96. Williams, R. L., Lipari, F., and Potter, R. A. (1990). Formaldehyde, methanol and hydrocarbon emissions from methanolfueled cars. J. Air Waste Manage. Assoc. 40, 747}756. Zhang, J., He, Q., and Lioy, P. J. (1994). Characteristics of aldehydes: Concentrations, sources, and exposures for indoor and outdoor residential microenvironments. Environ. Sci. Technol. 28, 146}152.