- Email: [email protected]
Chemical composition of major VOC emission sources in the Seoul atmosphere
Chemosphere 55 (2004) 585–594 www.elsevier.com/locate/chemosphere
Chemical composition of major VOC emission sources in the Seoul atmosphere Kwangsam Na
a,* ,
Yong Pyo Kim b, Il Moon a, Kil-Choo Moon
c
a
c
Department of Chemical Engineering, Yonsei University, Seodaemun-Ku, Shinchon-Dong 134, Seoul 120-749, South Korea b Department of Environmental Science and Engineering, Ewha Womans University, Seoul 120-750, South Korea Global Environmental Research Center, Korea Institute of Science and Technology, 39-1 Seongbuk-Ku Hawolgok-Dong, Seoul 136-791 South Korea Received 5 February 2003; received in revised form 12 December 2003; accepted 6 January 2004
Abstract This paper describes a chemical analysis of volatile organic compounds (VOCs) for five emission sources in Seoul. The source categories included motor vehicle exhaust, gasoline evaporation, paint solvents, natural gas and liquefied petroleum gas (LPG). These sources were selected because they have been known to emit significant quantities of VOCs in the Seoul area (more than 5% of the total emission inventory). Chemical compositions of the five emission sources are presented for a group of 45 C2 –C9 VOCs. Motor vehicle exhaust profiles were developed by conducting an urban tunnel study. These emissions profiles were distinguished from the other emission profiles by a high weight percentage of butanes over seasons and propane in the wintertime. It was found that this is due to the wide use of butane-fueled vehicles. To obtain gasoline vapor profiles, gasoline samples from five major brands for each season were selected. The brands were blended on the basis of the marketshare of these brands in Seoul area. Raoult’s law was used to calculate gasoline evaporative compositions based on the liquid gasoline compositions. The measured and estimated gasoline vapor compositions were found to be in good agreement. Vehicle and gasoline evaporation profiles were made over seasons because of the seasonal change in their compositions. Paint solvent emissions profiles were produced based on a product-use survey and sales figures. These profiles are a composite of four major oil-based paints and thinning solvent. The source profile of natural gas was made on a methane-free basis. It was found that Ethane and propane were the most abundant compounds accounting for 95% of the natural gas composition. LPG was largely composed of propane and ethane and the remaining components were minor contributors. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Volatile organic compounds; Motor vehicle exhaust; Gasoline evaporation; Paint solvents; Natural gas; Liquefied petroleum gas
1. Introduction
*
Corresponding author. Present address: Center for Environmental Research and Technology, Bourns College of Engineering, University of California, Riverside, CA 92521, USA. Tel.: +1-909-781-5724; fax: +1-909-781-5790. E-mail address: [email protected] (K. Na).
In order to apply receptor models for the determination of the sources of the ambient air pollutant, the chemical composition of the pollutants at the point of emission from various emission sources has to be known. The composition pattern of species emitted from a source category is called source profile or can be usually expressed as the weight fraction of each compound relative
0045-6535/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2004.01.010
586
K. Na et al. / Chemosphere 55 (2004) 585–594
to the total mass of the compounds in the source emission. Since the variations of the strength of the sources along with meteorological conditions strongly influence diurnal, seasonal, and annual variation of the VOC (volatile organic compounds) concentrations, source profiles of VOCs are useful in understanding atmospheric features of VOCs and characterizing VOCs in the atmosphere. Additionally, accurately speciated source profiles that reliably reflect source composition of a given area are crucial to the success of the developed control strategies. VOC source profiles have been developed for several urban areas of the United State, Canada, Japan, and Australia (Nelson et al., 1983; Wadden et al., 1986; Conner et al., 1995; Mclaren et al., 1996). According to a calculated emission inventory of Seoul in 1997, vehicle emissions and evaporative emissions by use of solvents are two dominant sources of VOCs: Vehicle emissions were calculated to contribute 34.5%, gasoline evaporation 6.0%, paint solvents 46.2%, printing solvents 3.3%, dry cleaning 2.1%, and asphalt paving 3.3% (MOE, 2000). This emissions inventory suggests that controlling vehicle and solvent emissions is the most effective way to reduce VOC emissions and, thereby, the ambient level of VOCs. The selection of source types for receptor modeling is based on the emission inventory estimated in 1997 by the Ministry of Environment and examination of ambient VOC concentrations measured in the present study. The source categories include vehicle emissions, gasoline vapor, paint solvents, natural gas and liquefied petroleum gas (LPG). According to the reported emission inventory, the first three selected sources represent the major contributors (more than 5% of the total emissions) to the VOCs emitted to the environment. It has been thought that emissions from natural gas and LPG are minor contributors to ambient VOC concentrations. However, the ambient concentrations of ethane and propane closely related to natural gas and LPG, respectively, have been reported to be very high (the 5 most abundant compounds) in Seoul (Na and Kim, 2001). This suggests that natural gas and LPG usage are important contributors to ambient VOC concentrations. Thus, these two sources are considered in this study though the two sources were not included in the emission inventory made by the Ministry of Environment. Other than these sources, emissions from printing solvents, dry cleaning and asphalt paving account for less than 5% of the total VOC emissions. In this study, source profiles for 5 sources of VOCs were developed and evaluated. In addition, from a comparison of other studies, characteristics of source profiles for a group of 45 compounds in Seoul are discussed. The source profiles are presented. These compounds were selected for a variety of reasons including
usefulness in previous receptor modeling applications, ease of measurement in the ambient environment, compound toxicity and reactivity.
2. Source characterization of major emission sources The canisters were analyzed by a GC/FID (STAR 3600CX, Varian, USA) and GC/MS (3400CX GC & Saturn 2000 MS, Varian, USA) at Korea Institute of Science and Technology. A GC/FID was used to quantify C2 –C3 VOCs. The separation was achieved by a capillary column (60 m long, 0.32 mm I.D., 3.0 lm film thickness RTX-1 column). A GC/MS was used to identify C2 –C9 VOCs and quantify C4 –C9 VOCs. These VOCs were separated in a 60m long, 0.32 mm I.D., 1.0 lm film thickness DB-1 fused silica column (J&W Scientific, USA) coated with a polydimethylsiloxane. Precision, as determined from five replicate analyses of the standards and samples, is within ±15% for the compounds at the concentrations above 5 ppbC. The lower quantifiable limits were between 0.1 and 0.5 ppbC depending on component for the 200 ml of sample concentrated. A full description of the analytical methodology is given in an earlier paper (Na and Kim, 2001). 2.1. Motor vehicle exhaust There are two widely used methods to determine the vehicular emission profiles: dynamometer test on an individual vehicle and measurements in a tunnel. In the former approach, operating conditions and fuel composition can be controlled. However, it is disadvantageous in terms of cost and time and does not represent a composite of the large number and different types of onroad vehicles. The latter method, the one that we chose in this study, has been widely used to determine the VOC speciation of vehicle emissions in the past decade (Pierson et al., 1990; Haszpra and Szil agyi, 1994; Gertler et al., 1996; Mugica et al., 1998). The compositions of the VOC species in the tunnel air are believed to be representative of a large number of vehicles and fuel types used broadly in urban areas (Lonneman et al., 1986). Most tunnel studies till date have focused on highspeed driving conditions in highway tunnels (Gertler et al., 1996; Fraser et al., 1998; Touaty and Bonsang, 2000). This driving pattern may not be realistic enough to represent traffic-related pollution in urban areas because vehicle speeds differ during typical urban driving which includes the accelerating, cruising, and decelerating stages. In this study, to obtain actual vehicular emission profiles for VOC in Seoul, measurements were carried out from a tunnel under high and low speed
K. Na et al. / Chemosphere 55 (2004) 585–594
driving conditions with both moving and stationary vehicles. In the present study, vehicle emission profiles were measured for each season because the composition of vehicle fuel and vehicle exhaust varies from season to season (Na et al., 2002a). The methods of obtaining vehicle emission source profiles have been described in detail elsewhere (Na et al., 2002a). Table 1 shows vehicle emission source profiles by species and chemical groups for spring, summer, and winter. Alkanes are the most abundant in vehicle emissions over all seasons, followed by aromatics and alkenes. The most pronounced differences among the seasons are the higher mass percentage for propane in winter and higher mass percentages of butanes and ethylene in summer. We analyzed the compositions of butane fuels for three seasons to investigate the reasons for the high percentage of propane in winter and high percentages of butanes in summer. As shown in Table 2, the compositions of butane fuel vary from season to season. In particular, the seasonal variation of propane composition is very high. It is because propane is added to the butane fuel in proportions ranging 5–30 wt.% during the cold months (from November to March). This is done in order to promote optimum vehicle start-up and driveability. As a result, butane fuel in winter has a higher volatility than in summer. Thus, the higher mass fraction for propane in winter and higher mass fractions of butanes in summer may be influenced by changes in composition of butane fuel. As of February 2001, the total number of vehicles in Seoul was over 2.3 million. Butane-fueled vehicles comprise approximately 10% (220 000) of the vehicle population in Seoul. It has been known that butanepowered vehicles are a major source of butane in the Seoul atmosphere (Na et al., 2002b). A similar study conducted at the Cassiar tunnel in Canada showed high propane mass fractions. This was explained by the presence of propane-powered vehicles in Canada (Rogak et al., 1998). This suggests that the pattern of fuel usage is an important factor affecting the VOC composition in vehicle emissions. The prominent features of vehicle emissions in Seoul are higher mass percentages of n=i-butanes and propane, and lower mass percentages of n=i-pentanes compared to cities in other parts of the world. 2.2. Gasoline evaporation Gasoline is a complex mixture of C4 –C13 hydrocarbons and oxygen-containing compounds such as methyl tertiary-butyl ether (C5 H12 O, MTBE). The most important factor affecting the degree of evaporative emissions is vapor pressure. The south Korean government has recently mandated seasonal vapor pressure limits for gasoline to reduce VOC emissions by gasoline
587
evaporation. Gasoline can enter the atmosphere by complete evaporation (as in spillage or vehicle ‘‘hotsoak’’ emission) in which case the composition will be that of vapor-phase gasoline, or by partial evaporation (as in storage tank evaporation or vehicle diurnal evaporation) in which case the composition will be that of the vapor in equilibrium with the gasoline at the relevant temperature. Evaporative emissions are divided into five types: (1) Diurnal evaporation, which occurs due to ambient temperature changes over a typical 24-h period when the vehicle is at rest (the portion of emissions at rest driven by the impact of temperature on the vapor pressure above the fuel). (2) Hot soak, which is driven by residual engine heat once a warmed-up vehicle is parked and the engine is shut off. (3) Running loss, which occurs when the vehicle is being driven. (4) Resting loss (the constant at rest evaporative emissions). (5) Refueling loss (displaced vapors and drippage resulting from refueling) (Sawyer et al., 2000). In this paper, emission profiles from gasoline evaporation is measured and estimated. To obtain the gasoline evaporative profiles, a 250 ml flask containing gasoline sample immersed in constant temperature bath maintained at 0 °C for the winter data. This temperature was selected to reflect the feature of the gasoline evaporation during the wintertime. The liquid gasoline samples were made of gasoline of five major brands for each season blended on the basis of the market share in Seoul. The analysis on the three gasoline mixtures for each season (winter, spring and summer) was carried out by Korea Petroleum Quality Inspection Institute. The gasoline profiles can be used to represent the emissions of unburned gasoline such as running losses from saturated fuel-injection system. To obtain a pure gasoline vapor composition in the flask, air in the headspace of a flask was completely replaced with pure nitrogen (99.999%). After a 20-min equilibrium period, 500-ll of the headspace air was taken from the flask and injected into evacuated (<0.1 Torr) 6 l SUMMA-polished canisters while filling the canister with pure nitrogen (99.999%) to 15 psig. All tubes connected with the canister were sufficiently purged with the pure nitrogen before the injection of the headspace sample. Gasoline vapor composition can be measured as described above. It can also be calculated from the liquid gasoline composition and vapor pressure data. The measured compositions of the liquid gasoline mixtures are presented in Table 2, and were used to calculate gasoline vapor compositions by means of Raoult’s Law. The compositions thus calculated accurately reflect the equilibrium vapor composition by measurements of the vapor over individual gasoline samples. Raoult’s law states that the mole fraction of component i (xi ) in an
588
K. Na et al. / Chemosphere 55 (2004) 585–594
Table 1 Chemical compositions of vehicle emission sources (wt.%) This study Spring
Summer
Winter
Cassiar (Canada)
Ethane Propane n-Butane i-Butane n-Pentane i-Pentane 2-Methylpentane 3-Methylpentane 2,2-Dimethylbutane 2,3-Dimethylbutane n-Hexane 2-Methylhexane 3-Methylhexane 2,3-Dimethylpentane 2,4-Dimethylpentane n-Heptane 2,2,4-Trimethylpentane 2,3,4-Trimethylpentane 2-Methylheptane 3-Methylheptane Octane Nonane Ethylene Propylene 1-Butene t-2-Butene c-2-Butene 1-Pentene Isoprene t-2-Pentene c-2-Pentene 2-Methyl-2-butene Acetylene Cyclopentane Methylcyclopentane Cyclohexane Methylcyclohexane Benzene Toluene Ethylbenzene m- + p-Xylene o-Xylene Styrene 1,2,4-Trimethylbenzene 1,3,5-Trimethylbenzene
1.3 ± 0.3 2.0 ± 0.4 12.4 ± 2.4 6.0 ± 1.2 2.9 ± 0.4 3.5 ± 0.6 2.8 ± 0.5 2.0 ± 0.4 0.4 ± 0.2 2.3 ± 0.5 2.1 ± 0.4 2.1 ± 0.3 0.7 ± 0.2 1.7 ± 0.5 0.4 ± 0.2 1.3 ± 0.5 0.5 ± 0.2 0.4 ± 0.2 0.5 ± 0.3 0.6 ± 0.2 0.4 ± 0.2 0.5 ± 0.1 7.6 ± 1.1 3.3 ± 0.5 3.2 ± 0.4 1.8 ± 0.4 0.9 ± 0.3 0.5 ± 0.2 0.3 ± 0.1 0.6 ± 0.2 0.9 ± 0.2 1.0 ± 0.2 2.7 ± 0.4 0.4 ± 0.3 1.2 ± 0.3 0.3 ± 0.1 0.5 ± 0.2 3.3 ± 0.7 8.4 ± 1.1 1.3 ± 0.5 5.6 ± 1.1 1.1 ± 0.3 0.3 ± 0.1 5.6 ± 1.0 1.1 ± 0.4
1.8 ± 0.4 3.5 ± 0.6 16.8 ± 2.7 8.6 ± 1.4 2.8 ± 0.4 3.1 ± 0.5 1.8 ± 0.3 1.3 ± 0.5 0.2 ± 0.1 2.8 ± 0.6 2.2 ± 0.6 0.9 ± 0.1 0.5 ± 0.2 1.3 ± 0.3 0.2 ± 0.1 0.6 ± 0.3 0.1 ± 0.1 0.1 ± 0.1 0.3 ± 0.1 0.4 ± 0.2 0.2 ± 0.1 0.2 ± 0.1 11.1 ± 2.8 5.2 ± 0.8 1.4 ± 0.3 1.2 ± 0.3 0.5 ± 0.2 0.3 ± 0.1 0.4 ± 0.2 0.5 ± 0.2 0.5 ± 0.3 0.6 ± 0.2 3.6 ± 0.4 0.1 ± 0.1 1.4 ± 0.6 0.2 ± 0.1 0.3 ± 0.1 2.2 ± 0.4 10.8 ± 1.8 1.4 ± 0.5 3.4 ± 1.0 1.7 ± 0.4 0.2 ± 0.1 3.1 ± 0.8 0.8 ± 0.5
3.4 ± 0.8 6.6 ± 1.9 12.7 ± 2.0 7.2 ± 1.5 2.8 ± 0.3 3.1 ± 0.5 2.6 ± 0.2 0.6 ± 0.3 0.0 ± 0.1 1.8 ± 0.3 1.7 ± 0.3 0.6 ± 0.2 0.1 ± 0.1 1.2 ± 0.4 0.2 ± 0.1 1.1 ± 0.4 0.2 ± 0.1 0.1 ± 0.1 0.3 ± 0.1 0.4 ± 0.2 0.4 ± 0.1 0.3 ± 0.1 9.4 ± 2.5 4.2 ± 0.5 2.8 ± 0.5 0.7 ± 0.2 0.5 ± 0.2 0.2 ± 0.1 0.1 ± 0.1 0.5 ± 0.2 0.2 ± 0.1 0.7 ± 0.2 4.1 ± 0.5 0.3 ± 0.1 1.2 ± 0.4 0.5 ± 0.2 0.4 ± 0.1 3.4 ± 0.7 9.5 ± 1.9 1.1 ± 0.4 4.5 ± 0.8 1.6 ± 0.5 0.0 ± 0.1 5.3 ± 1.2 1.1 ± 0.5
2.2 7.5 4.4 0.1 2.9 7.0 2.6 1.7 0.3 0.8 1.5 0.8 0.9 1.9 1.1 0.6 2.9 0.9 0.3 0.4 0.3 0.2 9.4 5.1 2.7 0.4 0.3 0.3 0.1 0.5 0.3 0.8 4.0 0.4 1.0 0.2 0.3 4.5 8.2 1.4 4.5 1.7 0.3 1.6 0.5
Alkanes Alkenes Alkynes Naphthenes Aromatics
46.5 ± 5.7 20.3 ± 3.7 2.6 ± 0.5 2.4 ± 0.4 28.2 ± 4.9
49.2 ± 6.7 21.6 ± 3.4 3.6 ± 0.4 2.0 ± 0.3 23.6 ± 3.8
47.5 ± 5.1 19.4 ± 2.8 4.1 ± 0.5 2.4 ± 0.3 26.6 ± 3.9
Tuscarora (US)
Mexican (Mexico)
2.7 0.5 2.4 0.7 3.9 11.2 3.9 2.3 1.1 1.1 1.8 1.4 1.2
0.6 2.0 2.4 0.9 4.5 5.7 2.7 1.7 0.3
0.7 3.0 1.0
0.3 0.1 11.2 4.5
3.0 1.5
7.1 11.0 2.2 8.2 3.1 4.1 1.3
3.0 1.0 1.1 0.4 0.3 1.3 1.3 0.6 0.6 0.6 0.9 0.8 4.1 1.7 0.6 0.3 0.3 0.5 0.3 0.2 0.5 0.3 7.2 0.2 0.9 0.3 0.5 2.6 5.4 1.4 5.0 1.8 0.3 1.7 1.0
Cassiar: Rogak et al. (1998); Tuscarora: Gertler et al. (1996); Mexican: Mugica et al. (1998). Note: Blank values means not reported or not analyzed; Result of Mexican tunnel study is expressed by ppbC%.
ideal solution is equal to the ratio of the partial pressure of i above the solution (Pipart ) to the vapor pressure of
pure component i (Pipure ), provided vapor behaves as an ideal gas (Smith and Van Ness, 1987):
Table 2 Seasonal compositions of butane fuels being used in Seoul (wt.%) Compounds
Spring
Summer
Winter
Ethane Propane n-Butane i-Butane n-Pentane i-Pentane Ethylene Propylene Acetylene
0.32 ± 0.07 4.37 ± 0.21 69.10 ± 9.46 26.02 ± 3.14 0.02 ± 0.01 0.12 ± 0.04 0.03 ± 0.01 0.02 ± 0.01 0.00 ± 0.01
0.04 ± 0.01 1.58 ± 0.12 68.33 ± 7.14 29.90 ± 1.87 0.04 ± 0.02 0.10 ± 0.02 0.01 ± 0.01 0.00 ± 0.01 0.00 ± 0.01
1.38 ± 0.21 35.01 ± 4.09 52.35 ± 6.18 10.42 ± 0.81 0.16 ± 0.03 0.48 ± 0.06 0.11 ± 0.03 0.08 ± 0.02 0.01 ± 0.01
Not listed compounds indicate that their concentrations are below the detection limit.
ð1Þ
If the mole fraction and the vapor pressure of each component i in gasoline are known, the above equation can be solved for the partial pressure of each component i above the gasoline. The partial pressures are proportional to the composition of the vapor above the gasoline. The vapor pressures of pure components were calculated for a specific temperature by using the Antoine equation, which relates vapor pressure (P , in Torr) and temperature (T , in K) as follows: B T þC
ð2Þ
where, A, B, and C are constants characteristic of each species (Reid et al., 1998). Gasoline evaporative compositions calculated from the Raoult’s law at 0 °C are compared with the measured compositions in Fig. 1 for 45 species. The measured vapor compositions are obtained from analysis of the compositions of headspace vapor of the flask filled with the gasoline mixtures. The profile for the gasoline headspace vapor reflects evaporative emissions due to refueling, diurnal emissions, and running losses. Profile abundance has been normalized for the species partial pressures calculated using Raoult’s law as described above. Agreement between calculated and measured headspace composition is very good (R2 ¼ 0:95). Relationship between the calculated and the measured results is as follows: Calculation ¼ 1:04 measurement 0:08
ð3Þ
The difference between the calculated and measured values for each species, as well as the difference between the calculated and regression-derived values, is less than
589
30 8
25
6 4
20
2 0
15
0
2
4
6
8
10
5
0
Pipart ¼ xi Pipure
log P ¼ A
Calculated headspace composition (wt%)
K. Na et al. / Chemosphere 55 (2004) 585–594
0
5
10
15
20
25
30
Measured headspace composition (wt%) Fig. 1. Gasoline headspace vapor composition calculated from Raoult’s law for 0 °C versus measured 0 °C headspace composition.
25% for this sample. This suggests that if the gasoline vapor composition can be reliably constructed entirely from a measured gasoline composition and the Raoult’s law calculations, the need for doing separate chemical analyses of the gasoline vapor could be eliminated. In theory, vapor composition could be calculated at any ambient temperature, within the vapor pressure calculation restrictions. This would represent considerable savings in samples required and in analyses performed. The calculated compositions of the gasoline vapor are listed in Table 3, together with the analyzed gasoline compositions and the results from other foreign studies. The temperatures for the spring and summer were 11 and 24 °C, respectively. These temperature values are the 30-year average temperatures of March–May and June–August for spring and summer, respectively. Calculated profiles that reflect the composition of the gasoline vapor above the liquid gasoline are quite different from the gasoline profiles. Alkanes and alkenes contents are higher in headspace vapors than those in gasoline, especially i-pentane, n-butane, n-pentane, i-butane, trans- and cis-2-butenes, 2-methyl-2-butene, and transand cis-2-pentenes. Aromatic content is much lower in the gasoline vapor profiles. The abundance of butanes is the lowest in summer, possibly due to changes in vehicle technology and efforts to reduce fuel Reid vapor pressure (RVP) that have occurred in recent years. At present, RVP is largely controlled by the contents of butanes in gasoline. The partial pressures of the compounds decrease as molecular weights increase beyond the C6 compounds (higher
590
Table 3 Chemical compositions of liquid gasolines and gasoline vapor emission sources (wt.%) This study
Australia
Gasoline
Canada
Japan
US
0.0 1.8 19.1 15.2 13.1 35.8 6.3 3.1
0.14 0.97 21.80 5.13 7.40 27.90 3.53 1.93 0.68 1.49 1.20 0.46 0.44 0.52 0.21 0.06 0.01 0.03 0.01 0.01 0.13 0.98 1.54 1.38 1.19 0.07 2.29 1.25 2.88 0.01 1.49 0.81 0.12 0.12 0.86 1.26 0.11 0.32 0.12
Gasoline vapor
Winter
Spring
Summer
Winter
Spring
Summer
Ethane Propane n-Butane i-Butane n-Pentane i-Pentane 2-Methylpentane 3-Methylpentane 2,2-Dimethylbutane 2,3-Dimethylbutane n-Hexane 2-Methylhexane 3-Methylhexane 2,4-Dmpentane n-Heptane 2-Methylheptane 3-Methylheptane Octane Nonane Ethylene Propylene 1-Butene t-2-Butene c-2-Butene 1-Pentene Isoprene t-2-Pentene c-2-Pentene 2-Methyl-2-butene Acetylene Cyclopentane Methylcyclopentane Cyclohexane Methylcyclohexane Benzene Toluene Ethylbenzene m-/p-Xylene o-Xylene
0.00 ± 0.01 0.08 ± 0.01 2.71 ± 0.31 1.32 ± 0.11 5.65 ± 0.44 11.73 ± 1.31 6.68 ± 0.70 4.45 ± 0.55 0.80 ± 0.06 1.38 ± 0.09 4.50 ± 0.38 4.96 ± 0.37 3.97 ± 0.44 0.85 ± 0.07 2.81 ± 0.11 1.07 ± 0.09 1.28 ± 0.09 1.05 ± 0.08 0.27 ± 0.04 0.00 ± 0.00 0.00 ± 0.01 0.26 ± 0.03 0.96 ± 0.08 0.83 ± 0.08 0.73 ± 0.09 0.00 ± 0.00 1.95 ± 0.21 1.04 ± 0.22 2.81 ± 0.30 0.00 ± 0.00 0.74 ± 0.11 2.72 ± 0.28 0.48 ± 0.06 1.13 ± 0.17 2.92 ± 0.34 12.86 ± 2.09 1.53 ± 0.11 5.20 ± 0.47 2.20 ± 0.31
0.00 ± 0.00 0.06 ± 0.01 3.14 ± 0.28 1.07 ± 0.20 4.75 ± 0.33 10.64 ± 0.98 6.04 ± 0.55 4.16 ± 0.37 0.78 ± 0.11 1.26 ± 0.10 4.08 ± 0.38 5.67 ± 0.56 4.69 ± 0.37 0.86 ± 0.01 3.51 ± 0.28 1.23 ± 0.11 1.46 ± 0.12 1.32 ± 0.09 0.34 ± 0.05 0.00 ± 0.00 0.00 ± 0.00 0.15 ± 0.02 0.53 ± 0.04 0.44 ± 0.04 0.63 ± 0.03 0.00 ± 0.00 1.65 ± 0.11 0.90 ± 0.12 2.60 ± 0.17 0.00 ± 0.00 0.37 ± 0.04 2.13 ± 0.27 0.26 ± 0.04 1.05 ± 0.08 1.99 ± 0.23 11.40 ± 1.73 1.54 ± 0.11 5.68 ± 0.64 2.68 ± 0.31
0.00 ± 0.00 0.04 ± 0.01 2.20 ± 0.24 0.72 ± 0.11 7.76 ± 0.68 12.39 ± 0.11 6.27 ± 0.78 4.36 ± 0.53 0.78 ± 0.08 1.29 ± 0.31 4.32 ± 0.22 5.37 ± 0.44 4.42 ± 0.39 0.84 ± 0.15 3.10 ± 0.27 1.15 ± 0.10 1.39 ± 0.24 1.25 ± 0.22 0.32 ± 0.05 0.00 ± 0.00 0.00 ± 0.00 0.13 ± 0.02 0.79 ± 0.11 0.35 ± 0.06 0.64 ± 0.05 0.00 ± 0.00 1.67 ± 0.20 0.90 ± 0.16 2.62 ± 0.25 0.00 ± 0.00 0.41 ± 0.13 2.13 ± 0.26 0.28 ± 0.30 1.03 ± 0.11 1.67 ± 0.21 10.18 ± 1.47 1.30 ± 0.11 4.50 ± 0.54 2.17 ± 0.36
0.00 ± 0.02 2.33 ± 0.30 17.34 ± 0.22 13.03 ± 0.15 8.21 ± 0.96 24.38 ± 4.41 3.48 ± 0.37 2.05 ± 0.22 0.69 ± 0.18 0.82 ± 0.17 1.56 ± 0.22 0.65 ± 0.11 0.48 ± 0.10 0.18 ± 0.04 0.23 ± 0.05 0.03 ± 0.01 0.04 ± 0.01 0.02 ± 0.01 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.01 2.05 ± 0.33 5.80 ± 0.41 4.49 ± 0.33 1.37 ± 0.11 0.00 ± 0.00 2.76 ± 0.17 1.44 ± 0.08 3.61 ± 0.44 0.00 ± 0.00 0.62 ± 0.09 0.85 ± 0.09 0.10 ± 0.02 0.10 ± 0.02 0.57 ± 0.17 0.62 ± 0.08 0.02 ± 0.01 0.06 ± 0.01 0.02 ± 0.01
0.00 ± 0.01 1.53 ± 0.21 20.70 ± 3.32 10.32 ± 1.14 8.27 ± 0.67 25.32 ± 0.24 4.20 ± 0.43 2.58 ± 0.31 0.84 ± 0.11 0.98 ± 0.15 1.99 ± 0.20 1.17 ± 0.18 0.89 ± 0.10 0.27 ± 0.04 0.49 ± 0.06 0.07 ± 0.02 0.08 ± 0.02 0.05 ± 0.02 0.00 ± 0.01 0.00 ± 0.01 0.00 ± 0.01 1.19 ± 0.22 3.06 ± 0.47 2.78 ± 0.24 1.39 ± 0.17 0.00 ± 0.00 2.83 ± 0.18 1.51 ± 0.17 4.08 ± 0.41 0.00 ± 0.00 0.39 ± 0.04 0.94 ± 0.20 0.08 ± 0.02 0.15 ± 0.03 0.59 ± 0.67 0.96 ± 0.14 0.04 ± 0.01 0.13 ± 0.04 0.05 ± 0.01
0.00 ± 0.01 1.40 ± 0.17 19.88 ± 4.01 9.73 ± 1.14 8.42 ± 1.47 25.32 ± 6.89 4.45 ± 1.01 2.74 ± 0.15 0.87 ± 0.17 1.02 ± 0.11 2.15 ± 0.36 1.31 ± 0.22 1.00 ± 0.09 0.27 ± 0.05 0.46 ± 0.07 0.08 ± 0.02 0.09 ± 0.02 0.06 ± 0.02 0.00 ± 0.01 0.00 ± 0.01 0.00 ± 0.01 0.92 ± 0.07 4.12 ± 0.53 2.99 ± 0.30 1.32 ± 0.19 0.00 ± 0.00 2.73 ± 0.34 1.46 ± 0.23 3.94 ± 0.41 0.00 ± 0.00 0.42 ± 0.05 0.95 ± 0.20 0.09 ± 0.02 0.15 ± 0.03 0.51 ± 0.06 0.93 ± 0.12 0.04 ± 0.01 0.12 ± 0.23 0.05 ± 0.01
0.0 1.5 18.7 11.1 10.7 25.4 3.5 2.2 0.6 1.1 1.9 0.7 0.5 0.2 0.3
0.0 0.0 0.0 0.0 1.6 3.7 2.9 0.7 1.5 0.9 2.6 0.0 0.6 0.9 0.3 0.2 0.9 1.0 0.1 0.2 0.1
19.92 5.68 12.30 25.65 3.83 2.20 0.33 1.22 2.27
3.2
0.47 0.21 0.05 0.05 0.03 0.00
0.72
0.13 0.93 1.27 0.07 0.27 0.08
0.9 1.0 0.1 0.3 0.1
K. Na et al. / Chemosphere 55 (2004) 585–594
Compounds
Australia: Nelson et al. (1983), Canada: Mclaren et al. (1996), Japan: Wadden et al. (1986), US: Conner et al. (1995). Blank values: not reported.
80.18 ± 11.37 16.47 ± 1.81 0.00 ± 0.00 1.61 ± 0.29 1.73 ± 0.23 79.76 ± 10.642 16.84 ± 21.75 0.00 ± 0.00 1.56 ± 0.18 1.83 ± 0.27 55.57 ± 6.23 8.57 ± 9.11 0.00 ± 0.00 5.07 ± 0.77 30.79 ± 4.45 Total Total Total Total Total
alkanes alkenes alkynes napthenes aromatics
55.06 ± 7.892 6.90 ± 0.82 0.00 ± 0.00 3.81 ± 0.45 34.23 ± 5.50
57.95 ± 7.019 7.11 ± 0.67 0.00 ± 0.00 3.86 ± 0.51 31.09 ± 4.67
75.50 ± 9.28 21.52 ± 3.41 0.00 ± 0.00 1.67 ± 0.21 1.31 ± 0.24
0.00 ± 0.00 0.06 ± 0.02 0.02 ± 0.01 0.00 ± 0.00 4.75 ± 0.64 1.33 ± 0.13 Styrene 1,2,4-Trimethylbenzene 1,3,5-Trimethylbenzene
0.00 ± 0.00 8.68 ± 0.77 2.25 ± 0.33
0.00 ± 0.00 8.83 ± 0.97 2.44 ± 0.32
0.00 ± 0.00 0.01 ± 0.01 0.00 ± 0.01
0.00 ± 0.00 0.05 ± 0.01 0.01 ± 0.01
0.0 0.0 0.0
0.01 0.11 0.04
K. Na et al. / Chemosphere 55 (2004) 585–594
591
boiling point). Thus, components beyond C6 constitute lesser fractions of the gasoline vapor compared with the gasoline, especially aromatics. Alkanes are the most abundant in the gasoline vapor profiles (77.4% on average), followed by alkenes (19.1%), and aromatics (1.7%). For the purpose of chemical mass balance modeling, Nelson et al. (1983) and Conner et al. (1995) calculated gasoline vapor compositions from their average gasoline compositions by means of Raoult’s law. They applied the calculated gasoline profiles to the apportionment of VOC to their sources. The calculated profiles would not fully represent the real world of the gasoline evaporative emissions. However, the calculated profiles have advantages of the savings of time and cost. Gasoline vapor composition in this study is generally in agreement with other studies, especially for aromatic hydrocarbons. Our result shows very good correlation (R2 ¼ 0:99) with studies conducted in Australia and US (R2 ¼ 0:94). It suggests that there is no big problem in mutual application of these profiles. Canada and US have lower i-butane compositions compared to the other countries. Japan shows somewhat different pattern in the compositions of C4 –C6 alkanes. 2.3. Paint solvents Solvents are used mainly in the process of painting, printing, dry cleaning and vapor degreasing. Among them, solvent usage by paint accounts for about 86% of the total solvent usage in Korea. To obtain the compositions of paint solvents, we used the results of paint compositions measured by National Institute of Environmental Research, Korea (NIER) (Kim, 1998). Since it is difficult to analyze the compositions of all kinds of paints, only four major paints were analyzed: urethane, varnish, archryl and thinner. In this study, the composite was calculated by applying sales figures of each paint solvent to the compositions of each paint solvent. The compositions are presented in Table 4 along with the results from other studies. In the present study, aromatics account for about 95% of paint solvents and alkanes account for the rest. Toluene (63%) is the most abundant component in paint solvents, followed by m-/p-xylene (19%) and oxylene (8%). Contribution from benzene is about 1%. This composition shows different pattern as compared to the results of Japan and US. However, all the profiles are largely composed of hydrocarbons greater than C5 . 2.4. Natural gas Natural gas is mainly used for heating and cooking in Korea. Emissions of natural gas into atmosphere are
592
K. Na et al. / Chemosphere 55 (2004) 585–594
largely derived from leakage and consumption related to combustion. The degree of leakage is dependent on the pressure of the system pipeline and number of leaks. Its emission can be controlled by pipe improvement. For most of foreign urban areas, leakage is known to be the most important factor in natural gas emissions (Derwent et al., 1996). However, in Korea, it was identified that consumption related to combustion is more important than natural gas leakage in the emission (Na and Kim, 2001). It was reported that the natural gas consumption in winter is about five times higher than that in summer. In this study, a profile of natural gas was generated from results of measurements of natural gas being used in Seoul. To obtain the profile, a 20 ml tygon tube was fully cleaned with purified nitrogen gas (purity: 99.9999%) for 10 min at a flow rate of 1 l min1 . Then, the tube was purged with natural gas for 5 min at a flow rate of 1 l min1 and both sides of the tube were tightly closed with a clip. Finally, a 1 ll sample was taken with a syringe and injected into evacuated (104 Torr) 6 l SUMMA-polished canisters with filling the canisters with the purified nitrogen gas to 15 psig. From the experiment, we obtained five duplicate samples. The natural gas source profile is presented in Table 5 along with the results from other studies. Here, the composition of natural gas in this study was calculated on a methane-free basis. Natural gas used in Korea is mainly made up of methane (89.1% by mass) and ethane (7.8% by mass). In this study, natural gas is largely composed of C2 –C5 saturated hydrocarbons. Ethane and propane are the major compounds and these account for 71% and 22% by mass, respectively. The natural gas emission profile obtained in this study is not significantly different from the other results. 2.5. Liquefied petroleum gas (LPG) LPG is used for cooking and heating in residential and commercial areas in Korea. LPG is known to be Table 4 Chemical compositions of paint solvents (wt.%) Compounds
This study
Japan
n-Butane i-Pentane n-Hexane n-Heptane Benzene Toluene Ethylbenzene m-/p-Xylene o-Xylene Styrene
0.12 ± 0.02 0.79 ± 0.14 0.26 ± 0.04 0.98 ± 0.15 0.96 ± 0.14 62.74 ± 9.41 6.20 ± 0.94 19.09 ± 2.86 7.86 ± 1.18 0.99 ± 0.15
0.0 0.0 0.0 0.0 0.0 25.7 32.5 30.3 11.5
Table 5 Chemical compositions of natural gas (wt.%; methane-free base) Compounds
This study
Egypt
Germany
US
Ethane Propane n-Butane i-Butane n-Pentane i-Pentane
70.88 ± 12.23 22.29 ± 4.78 2.92 ± 0.37 2.17 ± 0.34 0.77 ± 0.21 0.97 ± 0.28
61.98 26.15 3.29 3.35 1.26 1.93
64.00 24.20 5.00 3.40 0.90 1.00
69.40 21.30 3.10 2.10 0.70 0.70
Egypt: Doskey et al. (1999); Germany: Thijsse et al. (1999); US: Fujita et al. (1995).
Table 6 Chemical compositions of liquefied petroleum gas (wt.%) Compounds
This study
Egypt
Mexico
US
Ethane Propane n-Butane i-Butane Propylene n-Pentane i-Pentane
2.28 ± 0.36 96.90 ± 14.21 0.22 ± 0.11 0.59 ± 0.15 0.01 ± 0.01 0.00 ± 0.01 0.00 ± 0.02
0.06 4.67 69.59 20.07
0.98 67.66 16.30 12.07
0.21 2.38
0.02 0.14
4.10 90.40 0.00 0.20 5.10 0.00 0.00
Blank values: not reported. Egypt: Doskey et al. (1999); Mexico: Vega et al. (2000); US: Fujita et al. (1995).
the major source of propane in the Seoul atmosphere (Na and Kim, 2001). The experimental method used to obtain the LPG emission source profile was similar to that used for determination of the natural gas profile. Table 6 lists the LPG profile measured in this study along with a profile developed in the US. LPG is largely composed of the light alkanes (C2 –C4 ). The LPG profiles for this study and US are similar with propane dominating the emission except that US LPG contains higher mass percent of propylene compared to that of this study. Percentages of butanes in LPG from Egypt and Mexico are higher than those of this study and US.
US 0.00 0.00 0.00 0.27 78.34 1.36 8.08 8.65
Blank values: not analyzed or not reported. Japan: Wadden et al. (1986); US: Scheff et al. (1989).
3. Summary Chemical compositions of five major VOC emission sources were determined. Vehicle exhaust profiles were developed by conducting an urban tunnel study. These emissions profiles were distinguished from the other emission profiles by a high weight percentage of butanes over seasons and propane in the wintertime. This was because of the addition of propane to butane-fueled vehicles in wintertime. The gasoline evaporation profiles were prepared by blending the gasoline samples of five
K. Na et al. / Chemosphere 55 (2004) 585–594
major brands for each season on the basis of the marketshare in Seoul area. We calculated gasoline evaporative compositions using Raoult’s law from the liquid gasoline compositions. Alkanes were the most abundant in the gasoline vapor profiles and the contribution of aromatics was minor. In gasoline vapor, i-pentane was the most abundant, followed by n-butane, n-pentane, ibutane, trans- and cis-2-butenes. Aromatic content was much lower in the headspace profiles. From the comparison between experimental and calculated compositions, we identified the fact that if the gasoline vapor composition can be reliably constructed entirely from the measured gasoline composition and Raoult’s law calculations, the need for doing separate chemical analyses of the gasoline vapor could be eliminated. For the compositions of paint solvents, aromatics accounts for about 95% of paint solvents and the remainder were alkanes. Toluene was the most abundant compound in paint solvent compositions, followed by m-/p-xylene and o-xylene. Benzene and styrene contributions were less than 1%. The source profile of natural gas is calculated on a methane-free basis. Ethane and propane were the major compounds in natural gas and these accounted for 95% of natural gas composition. LPG was largely composed of the light alkanes. Propane was the most abundant compound in LPG, followed by ethane.
Acknowledgements This work was supported in part by the National Research Laboratory Program of Korean Ministry of Science and Technology (2000-N-NL-01-C-184). The authors wish to thank to KH Jung in Korea Petroleum Quality Inspection Institute for the gasoline and diesel analysis.
References Conner, T.L., Lonneman, W.A., Seila, R.L., 1995. Transportation-related volatile hydrocarbon source profiles measured in Atlanta. Journal of the Air and Waste Management Association 45, 383–394. Derwent, R.G., Davies, T.J., Delaney, M., Dollard, G.J., Field, R.A., Dumitrean, P., Nason, P.D., Jones, B.M.R., Pepler, S.A., 1996. Analysis and interpretation of the continuous hourly monitoring data for 26 C2 –C8 hydrocarbons at 12 United Kingdom sites during 1996. Atmospheric Environment 34, 297–312. Doskey, P.V., Fukui, Y., Sultan, M., Maghraby, A.A., Taher, A., 1999. Source profiles for nonmethane organic compounds in the atmosphere of Cairo, Egypt. Journal of the Air and Waste Management Association 49, 814–822. Fraser, M.P., Cass, G.R., Simoneit, B.R.T., 1998. Gas-phase and particle-phase organic compounds emitted from motor vehicle traffic in a Los Angels roadway tunnel.
593
Environmental Science and Technology 32, 2051– 2060. Fujita, E.M., Watson, J.G., Chow, J.C., Magliano, K.L., 1995. Receptor model and emissions inventory source apportionments of nonmethane organic gases in California’s San Joaquin Valley and San Francisco Bay area. Atmospheric Environment 29, 3019–3035. Gertler, A.W., Fujita, E.M., Pierson, W.R., Wittorff, D.N., 1996. Apportionment of NMHC tailpipe vs nontailpipe emissions in the Fort McHenry and Tuscarora Mountain tunnels. Atmospheric Environment 30, 2297– 2305. Haszpra, L., Szilagyi, I., 1994. Non-methane hydrocarbon composition of car exhaust in Hungary. Atmospheric Environment 28, 2609–2614. Kim, S.Y., 1998. A study on the source characterization of VOC from major emission sources. Master Thesis, KonKuk University, Seoul, South Korea. Lonneman, W.A., Sella, R.L., Meeks, S.A., 1986. Non-methane organic composition in the Lincoln tunnel. Environmental Science and Technology 20, 790–796. Mclaren, R.M., Gertler, A.W., Wittorff, D.N., Belzer, W., Dann, T., Singleton, D.L., 1996. Real-world measurements of exhaust and evaporative emissions in the Cassiar tunnel predicted by chemical mass balance modeling. Environmental Science and Technology 30, 3001–3009. Ministry of Environment (MOE), 2000. Environment White Paper, Seoul, South Korea (in Korean). Mugica, V., Vega, E., Arriaga, J.L., Ruiz, M.E., 1998. Determination of motor vehicle profiles for non-methane organic compounds in the Mexico City metropolitan area. Journal of the Air and Waste Management Association 48, 1060–1068. Na, K., Kim, Y.P., 2001. Seasonal characteristics of ambient volatile organic compounds in Seoul. Atmospheric Environment 35, 2603–2614. Na, K., Kim, Y.P., Moon, K.C., 2002a. Seasonal variation of the C2 –C9 hydrocarbons concentrations and compositions emitted from motor vehicles in a Seoul tunnel. Atmospheric Environment 36, 1969–1978. Na, K., Kim, Y.P., Moon, K.C., 2002b. Diurnal characteristics of volatile organic compounds in the Seoul atmosphere. Atmospheric Environment 37, 733–742. Nelson, P.F., Quigley, S.M., Smith, M.Y., 1983. Sources of atmospheric hydrocarbons in Sydney; a quantitative determination using a source reconcilation technique. Atmospheric Environment 17, 439–449. Pierson, W.R., Gertler, A.W., Bradow, R.L., 1990. Comparison of the SCAQS tunnel study with other on-road vehicle emission data. Journal of the Air and Waste Management Association 40, 1495–1504. Reid, R.C., Prausnitz, J.M., Poling, B.E., 1998. The properties of Gases and Liquids, fourth ed. McGraw-Hill, New York. Rogak, S.N., Pott, U., Dann, T., Wang, D., 1998. Gaseous emissions from vehicles in a traffic tunnel in Vancouver, British Columbia. Journal of the Air and Waste Management Association 48, 604–615. Sawyer, R.F., Harley, S.H., Cadle, S.H., Norbeck, J.M., Slott, R., Bravo, H.A., 2000. Mobile sources critical review: 1998 NARSTO assessment. Atmospheric Environment 34, 2161– 2181.
594
K. Na et al. / Chemosphere 55 (2004) 585–594
Scheff, P.A., Wadden, R.A., Bates, B.A., 1989. Aronian PF. Source fingerprints for receptor modeling of volatile organics. Journal of Air Pollution Control Association 39, 469– 478. Smith, J.M., Van Ness, H.C., 1987. Introduction to chemical engineering thermodynamics, fourth ed. McGraw-Hill, Singapore. Thijsse, T.R., van Oss, R.F., Lenschow, P., 1999. Determination of source contributions to ambient volatile organic compound concentrations in Berlin. Journal of the Air and Waste Management Association 49, 1394–1404.
Touaty, M., Bonsang, B., 2000. Hydrocarbon emissions in a highway tunnel in the Paris area. Atmospheric Environment 34, 985–996. Vega, E., Mugica, V., Carmona, R., Valencia, E., 2000. Hydrocarbon source apportionment in Mexico city using the chemical mass balance receptor model. Atmospheric Environment 34, 4121–4129. Wadden, R.A., Uno, I., Wakamatsu, S., 1986. Source discrimination of short-term hydrocarbon samples measured aloft. Environmental Science and Technology 20, 473– 483.
Chemical composition of major VOC emission sources in the Seoul atmosphere Kwangsam Na
a,* ,
Yong Pyo Kim b, Il Moon a, Kil-Choo Moon
c
a
c
Department of Chemical Engineering, Yonsei University, Seodaemun-Ku, Shinchon-Dong 134, Seoul 120-749, South Korea b Department of Environmental Science and Engineering, Ewha Womans University, Seoul 120-750, South Korea Global Environmental Research Center, Korea Institute of Science and Technology, 39-1 Seongbuk-Ku Hawolgok-Dong, Seoul 136-791 South Korea Received 5 February 2003; received in revised form 12 December 2003; accepted 6 January 2004
Abstract This paper describes a chemical analysis of volatile organic compounds (VOCs) for five emission sources in Seoul. The source categories included motor vehicle exhaust, gasoline evaporation, paint solvents, natural gas and liquefied petroleum gas (LPG). These sources were selected because they have been known to emit significant quantities of VOCs in the Seoul area (more than 5% of the total emission inventory). Chemical compositions of the five emission sources are presented for a group of 45 C2 –C9 VOCs. Motor vehicle exhaust profiles were developed by conducting an urban tunnel study. These emissions profiles were distinguished from the other emission profiles by a high weight percentage of butanes over seasons and propane in the wintertime. It was found that this is due to the wide use of butane-fueled vehicles. To obtain gasoline vapor profiles, gasoline samples from five major brands for each season were selected. The brands were blended on the basis of the marketshare of these brands in Seoul area. Raoult’s law was used to calculate gasoline evaporative compositions based on the liquid gasoline compositions. The measured and estimated gasoline vapor compositions were found to be in good agreement. Vehicle and gasoline evaporation profiles were made over seasons because of the seasonal change in their compositions. Paint solvent emissions profiles were produced based on a product-use survey and sales figures. These profiles are a composite of four major oil-based paints and thinning solvent. The source profile of natural gas was made on a methane-free basis. It was found that Ethane and propane were the most abundant compounds accounting for 95% of the natural gas composition. LPG was largely composed of propane and ethane and the remaining components were minor contributors. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Volatile organic compounds; Motor vehicle exhaust; Gasoline evaporation; Paint solvents; Natural gas; Liquefied petroleum gas
1. Introduction
*
Corresponding author. Present address: Center for Environmental Research and Technology, Bourns College of Engineering, University of California, Riverside, CA 92521, USA. Tel.: +1-909-781-5724; fax: +1-909-781-5790. E-mail address: [email protected] (K. Na).
In order to apply receptor models for the determination of the sources of the ambient air pollutant, the chemical composition of the pollutants at the point of emission from various emission sources has to be known. The composition pattern of species emitted from a source category is called source profile or can be usually expressed as the weight fraction of each compound relative
0045-6535/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2004.01.010
586
K. Na et al. / Chemosphere 55 (2004) 585–594
to the total mass of the compounds in the source emission. Since the variations of the strength of the sources along with meteorological conditions strongly influence diurnal, seasonal, and annual variation of the VOC (volatile organic compounds) concentrations, source profiles of VOCs are useful in understanding atmospheric features of VOCs and characterizing VOCs in the atmosphere. Additionally, accurately speciated source profiles that reliably reflect source composition of a given area are crucial to the success of the developed control strategies. VOC source profiles have been developed for several urban areas of the United State, Canada, Japan, and Australia (Nelson et al., 1983; Wadden et al., 1986; Conner et al., 1995; Mclaren et al., 1996). According to a calculated emission inventory of Seoul in 1997, vehicle emissions and evaporative emissions by use of solvents are two dominant sources of VOCs: Vehicle emissions were calculated to contribute 34.5%, gasoline evaporation 6.0%, paint solvents 46.2%, printing solvents 3.3%, dry cleaning 2.1%, and asphalt paving 3.3% (MOE, 2000). This emissions inventory suggests that controlling vehicle and solvent emissions is the most effective way to reduce VOC emissions and, thereby, the ambient level of VOCs. The selection of source types for receptor modeling is based on the emission inventory estimated in 1997 by the Ministry of Environment and examination of ambient VOC concentrations measured in the present study. The source categories include vehicle emissions, gasoline vapor, paint solvents, natural gas and liquefied petroleum gas (LPG). According to the reported emission inventory, the first three selected sources represent the major contributors (more than 5% of the total emissions) to the VOCs emitted to the environment. It has been thought that emissions from natural gas and LPG are minor contributors to ambient VOC concentrations. However, the ambient concentrations of ethane and propane closely related to natural gas and LPG, respectively, have been reported to be very high (the 5 most abundant compounds) in Seoul (Na and Kim, 2001). This suggests that natural gas and LPG usage are important contributors to ambient VOC concentrations. Thus, these two sources are considered in this study though the two sources were not included in the emission inventory made by the Ministry of Environment. Other than these sources, emissions from printing solvents, dry cleaning and asphalt paving account for less than 5% of the total VOC emissions. In this study, source profiles for 5 sources of VOCs were developed and evaluated. In addition, from a comparison of other studies, characteristics of source profiles for a group of 45 compounds in Seoul are discussed. The source profiles are presented. These compounds were selected for a variety of reasons including
usefulness in previous receptor modeling applications, ease of measurement in the ambient environment, compound toxicity and reactivity.
2. Source characterization of major emission sources The canisters were analyzed by a GC/FID (STAR 3600CX, Varian, USA) and GC/MS (3400CX GC & Saturn 2000 MS, Varian, USA) at Korea Institute of Science and Technology. A GC/FID was used to quantify C2 –C3 VOCs. The separation was achieved by a capillary column (60 m long, 0.32 mm I.D., 3.0 lm film thickness RTX-1 column). A GC/MS was used to identify C2 –C9 VOCs and quantify C4 –C9 VOCs. These VOCs were separated in a 60m long, 0.32 mm I.D., 1.0 lm film thickness DB-1 fused silica column (J&W Scientific, USA) coated with a polydimethylsiloxane. Precision, as determined from five replicate analyses of the standards and samples, is within ±15% for the compounds at the concentrations above 5 ppbC. The lower quantifiable limits were between 0.1 and 0.5 ppbC depending on component for the 200 ml of sample concentrated. A full description of the analytical methodology is given in an earlier paper (Na and Kim, 2001). 2.1. Motor vehicle exhaust There are two widely used methods to determine the vehicular emission profiles: dynamometer test on an individual vehicle and measurements in a tunnel. In the former approach, operating conditions and fuel composition can be controlled. However, it is disadvantageous in terms of cost and time and does not represent a composite of the large number and different types of onroad vehicles. The latter method, the one that we chose in this study, has been widely used to determine the VOC speciation of vehicle emissions in the past decade (Pierson et al., 1990; Haszpra and Szil agyi, 1994; Gertler et al., 1996; Mugica et al., 1998). The compositions of the VOC species in the tunnel air are believed to be representative of a large number of vehicles and fuel types used broadly in urban areas (Lonneman et al., 1986). Most tunnel studies till date have focused on highspeed driving conditions in highway tunnels (Gertler et al., 1996; Fraser et al., 1998; Touaty and Bonsang, 2000). This driving pattern may not be realistic enough to represent traffic-related pollution in urban areas because vehicle speeds differ during typical urban driving which includes the accelerating, cruising, and decelerating stages. In this study, to obtain actual vehicular emission profiles for VOC in Seoul, measurements were carried out from a tunnel under high and low speed
K. Na et al. / Chemosphere 55 (2004) 585–594
driving conditions with both moving and stationary vehicles. In the present study, vehicle emission profiles were measured for each season because the composition of vehicle fuel and vehicle exhaust varies from season to season (Na et al., 2002a). The methods of obtaining vehicle emission source profiles have been described in detail elsewhere (Na et al., 2002a). Table 1 shows vehicle emission source profiles by species and chemical groups for spring, summer, and winter. Alkanes are the most abundant in vehicle emissions over all seasons, followed by aromatics and alkenes. The most pronounced differences among the seasons are the higher mass percentage for propane in winter and higher mass percentages of butanes and ethylene in summer. We analyzed the compositions of butane fuels for three seasons to investigate the reasons for the high percentage of propane in winter and high percentages of butanes in summer. As shown in Table 2, the compositions of butane fuel vary from season to season. In particular, the seasonal variation of propane composition is very high. It is because propane is added to the butane fuel in proportions ranging 5–30 wt.% during the cold months (from November to March). This is done in order to promote optimum vehicle start-up and driveability. As a result, butane fuel in winter has a higher volatility than in summer. Thus, the higher mass fraction for propane in winter and higher mass fractions of butanes in summer may be influenced by changes in composition of butane fuel. As of February 2001, the total number of vehicles in Seoul was over 2.3 million. Butane-fueled vehicles comprise approximately 10% (220 000) of the vehicle population in Seoul. It has been known that butanepowered vehicles are a major source of butane in the Seoul atmosphere (Na et al., 2002b). A similar study conducted at the Cassiar tunnel in Canada showed high propane mass fractions. This was explained by the presence of propane-powered vehicles in Canada (Rogak et al., 1998). This suggests that the pattern of fuel usage is an important factor affecting the VOC composition in vehicle emissions. The prominent features of vehicle emissions in Seoul are higher mass percentages of n=i-butanes and propane, and lower mass percentages of n=i-pentanes compared to cities in other parts of the world. 2.2. Gasoline evaporation Gasoline is a complex mixture of C4 –C13 hydrocarbons and oxygen-containing compounds such as methyl tertiary-butyl ether (C5 H12 O, MTBE). The most important factor affecting the degree of evaporative emissions is vapor pressure. The south Korean government has recently mandated seasonal vapor pressure limits for gasoline to reduce VOC emissions by gasoline
587
evaporation. Gasoline can enter the atmosphere by complete evaporation (as in spillage or vehicle ‘‘hotsoak’’ emission) in which case the composition will be that of vapor-phase gasoline, or by partial evaporation (as in storage tank evaporation or vehicle diurnal evaporation) in which case the composition will be that of the vapor in equilibrium with the gasoline at the relevant temperature. Evaporative emissions are divided into five types: (1) Diurnal evaporation, which occurs due to ambient temperature changes over a typical 24-h period when the vehicle is at rest (the portion of emissions at rest driven by the impact of temperature on the vapor pressure above the fuel). (2) Hot soak, which is driven by residual engine heat once a warmed-up vehicle is parked and the engine is shut off. (3) Running loss, which occurs when the vehicle is being driven. (4) Resting loss (the constant at rest evaporative emissions). (5) Refueling loss (displaced vapors and drippage resulting from refueling) (Sawyer et al., 2000). In this paper, emission profiles from gasoline evaporation is measured and estimated. To obtain the gasoline evaporative profiles, a 250 ml flask containing gasoline sample immersed in constant temperature bath maintained at 0 °C for the winter data. This temperature was selected to reflect the feature of the gasoline evaporation during the wintertime. The liquid gasoline samples were made of gasoline of five major brands for each season blended on the basis of the market share in Seoul. The analysis on the three gasoline mixtures for each season (winter, spring and summer) was carried out by Korea Petroleum Quality Inspection Institute. The gasoline profiles can be used to represent the emissions of unburned gasoline such as running losses from saturated fuel-injection system. To obtain a pure gasoline vapor composition in the flask, air in the headspace of a flask was completely replaced with pure nitrogen (99.999%). After a 20-min equilibrium period, 500-ll of the headspace air was taken from the flask and injected into evacuated (<0.1 Torr) 6 l SUMMA-polished canisters while filling the canister with pure nitrogen (99.999%) to 15 psig. All tubes connected with the canister were sufficiently purged with the pure nitrogen before the injection of the headspace sample. Gasoline vapor composition can be measured as described above. It can also be calculated from the liquid gasoline composition and vapor pressure data. The measured compositions of the liquid gasoline mixtures are presented in Table 2, and were used to calculate gasoline vapor compositions by means of Raoult’s Law. The compositions thus calculated accurately reflect the equilibrium vapor composition by measurements of the vapor over individual gasoline samples. Raoult’s law states that the mole fraction of component i (xi ) in an
588
K. Na et al. / Chemosphere 55 (2004) 585–594
Table 1 Chemical compositions of vehicle emission sources (wt.%) This study Spring
Summer
Winter
Cassiar (Canada)
Ethane Propane n-Butane i-Butane n-Pentane i-Pentane 2-Methylpentane 3-Methylpentane 2,2-Dimethylbutane 2,3-Dimethylbutane n-Hexane 2-Methylhexane 3-Methylhexane 2,3-Dimethylpentane 2,4-Dimethylpentane n-Heptane 2,2,4-Trimethylpentane 2,3,4-Trimethylpentane 2-Methylheptane 3-Methylheptane Octane Nonane Ethylene Propylene 1-Butene t-2-Butene c-2-Butene 1-Pentene Isoprene t-2-Pentene c-2-Pentene 2-Methyl-2-butene Acetylene Cyclopentane Methylcyclopentane Cyclohexane Methylcyclohexane Benzene Toluene Ethylbenzene m- + p-Xylene o-Xylene Styrene 1,2,4-Trimethylbenzene 1,3,5-Trimethylbenzene
1.3 ± 0.3 2.0 ± 0.4 12.4 ± 2.4 6.0 ± 1.2 2.9 ± 0.4 3.5 ± 0.6 2.8 ± 0.5 2.0 ± 0.4 0.4 ± 0.2 2.3 ± 0.5 2.1 ± 0.4 2.1 ± 0.3 0.7 ± 0.2 1.7 ± 0.5 0.4 ± 0.2 1.3 ± 0.5 0.5 ± 0.2 0.4 ± 0.2 0.5 ± 0.3 0.6 ± 0.2 0.4 ± 0.2 0.5 ± 0.1 7.6 ± 1.1 3.3 ± 0.5 3.2 ± 0.4 1.8 ± 0.4 0.9 ± 0.3 0.5 ± 0.2 0.3 ± 0.1 0.6 ± 0.2 0.9 ± 0.2 1.0 ± 0.2 2.7 ± 0.4 0.4 ± 0.3 1.2 ± 0.3 0.3 ± 0.1 0.5 ± 0.2 3.3 ± 0.7 8.4 ± 1.1 1.3 ± 0.5 5.6 ± 1.1 1.1 ± 0.3 0.3 ± 0.1 5.6 ± 1.0 1.1 ± 0.4
1.8 ± 0.4 3.5 ± 0.6 16.8 ± 2.7 8.6 ± 1.4 2.8 ± 0.4 3.1 ± 0.5 1.8 ± 0.3 1.3 ± 0.5 0.2 ± 0.1 2.8 ± 0.6 2.2 ± 0.6 0.9 ± 0.1 0.5 ± 0.2 1.3 ± 0.3 0.2 ± 0.1 0.6 ± 0.3 0.1 ± 0.1 0.1 ± 0.1 0.3 ± 0.1 0.4 ± 0.2 0.2 ± 0.1 0.2 ± 0.1 11.1 ± 2.8 5.2 ± 0.8 1.4 ± 0.3 1.2 ± 0.3 0.5 ± 0.2 0.3 ± 0.1 0.4 ± 0.2 0.5 ± 0.2 0.5 ± 0.3 0.6 ± 0.2 3.6 ± 0.4 0.1 ± 0.1 1.4 ± 0.6 0.2 ± 0.1 0.3 ± 0.1 2.2 ± 0.4 10.8 ± 1.8 1.4 ± 0.5 3.4 ± 1.0 1.7 ± 0.4 0.2 ± 0.1 3.1 ± 0.8 0.8 ± 0.5
3.4 ± 0.8 6.6 ± 1.9 12.7 ± 2.0 7.2 ± 1.5 2.8 ± 0.3 3.1 ± 0.5 2.6 ± 0.2 0.6 ± 0.3 0.0 ± 0.1 1.8 ± 0.3 1.7 ± 0.3 0.6 ± 0.2 0.1 ± 0.1 1.2 ± 0.4 0.2 ± 0.1 1.1 ± 0.4 0.2 ± 0.1 0.1 ± 0.1 0.3 ± 0.1 0.4 ± 0.2 0.4 ± 0.1 0.3 ± 0.1 9.4 ± 2.5 4.2 ± 0.5 2.8 ± 0.5 0.7 ± 0.2 0.5 ± 0.2 0.2 ± 0.1 0.1 ± 0.1 0.5 ± 0.2 0.2 ± 0.1 0.7 ± 0.2 4.1 ± 0.5 0.3 ± 0.1 1.2 ± 0.4 0.5 ± 0.2 0.4 ± 0.1 3.4 ± 0.7 9.5 ± 1.9 1.1 ± 0.4 4.5 ± 0.8 1.6 ± 0.5 0.0 ± 0.1 5.3 ± 1.2 1.1 ± 0.5
2.2 7.5 4.4 0.1 2.9 7.0 2.6 1.7 0.3 0.8 1.5 0.8 0.9 1.9 1.1 0.6 2.9 0.9 0.3 0.4 0.3 0.2 9.4 5.1 2.7 0.4 0.3 0.3 0.1 0.5 0.3 0.8 4.0 0.4 1.0 0.2 0.3 4.5 8.2 1.4 4.5 1.7 0.3 1.6 0.5
Alkanes Alkenes Alkynes Naphthenes Aromatics
46.5 ± 5.7 20.3 ± 3.7 2.6 ± 0.5 2.4 ± 0.4 28.2 ± 4.9
49.2 ± 6.7 21.6 ± 3.4 3.6 ± 0.4 2.0 ± 0.3 23.6 ± 3.8
47.5 ± 5.1 19.4 ± 2.8 4.1 ± 0.5 2.4 ± 0.3 26.6 ± 3.9
Tuscarora (US)
Mexican (Mexico)
2.7 0.5 2.4 0.7 3.9 11.2 3.9 2.3 1.1 1.1 1.8 1.4 1.2
0.6 2.0 2.4 0.9 4.5 5.7 2.7 1.7 0.3
0.7 3.0 1.0
0.3 0.1 11.2 4.5
3.0 1.5
7.1 11.0 2.2 8.2 3.1 4.1 1.3
3.0 1.0 1.1 0.4 0.3 1.3 1.3 0.6 0.6 0.6 0.9 0.8 4.1 1.7 0.6 0.3 0.3 0.5 0.3 0.2 0.5 0.3 7.2 0.2 0.9 0.3 0.5 2.6 5.4 1.4 5.0 1.8 0.3 1.7 1.0
Cassiar: Rogak et al. (1998); Tuscarora: Gertler et al. (1996); Mexican: Mugica et al. (1998). Note: Blank values means not reported or not analyzed; Result of Mexican tunnel study is expressed by ppbC%.
ideal solution is equal to the ratio of the partial pressure of i above the solution (Pipart ) to the vapor pressure of
pure component i (Pipure ), provided vapor behaves as an ideal gas (Smith and Van Ness, 1987):
Table 2 Seasonal compositions of butane fuels being used in Seoul (wt.%) Compounds
Spring
Summer
Winter
Ethane Propane n-Butane i-Butane n-Pentane i-Pentane Ethylene Propylene Acetylene
0.32 ± 0.07 4.37 ± 0.21 69.10 ± 9.46 26.02 ± 3.14 0.02 ± 0.01 0.12 ± 0.04 0.03 ± 0.01 0.02 ± 0.01 0.00 ± 0.01
0.04 ± 0.01 1.58 ± 0.12 68.33 ± 7.14 29.90 ± 1.87 0.04 ± 0.02 0.10 ± 0.02 0.01 ± 0.01 0.00 ± 0.01 0.00 ± 0.01
1.38 ± 0.21 35.01 ± 4.09 52.35 ± 6.18 10.42 ± 0.81 0.16 ± 0.03 0.48 ± 0.06 0.11 ± 0.03 0.08 ± 0.02 0.01 ± 0.01
Not listed compounds indicate that their concentrations are below the detection limit.
ð1Þ
If the mole fraction and the vapor pressure of each component i in gasoline are known, the above equation can be solved for the partial pressure of each component i above the gasoline. The partial pressures are proportional to the composition of the vapor above the gasoline. The vapor pressures of pure components were calculated for a specific temperature by using the Antoine equation, which relates vapor pressure (P , in Torr) and temperature (T , in K) as follows: B T þC
ð2Þ
where, A, B, and C are constants characteristic of each species (Reid et al., 1998). Gasoline evaporative compositions calculated from the Raoult’s law at 0 °C are compared with the measured compositions in Fig. 1 for 45 species. The measured vapor compositions are obtained from analysis of the compositions of headspace vapor of the flask filled with the gasoline mixtures. The profile for the gasoline headspace vapor reflects evaporative emissions due to refueling, diurnal emissions, and running losses. Profile abundance has been normalized for the species partial pressures calculated using Raoult’s law as described above. Agreement between calculated and measured headspace composition is very good (R2 ¼ 0:95). Relationship between the calculated and the measured results is as follows: Calculation ¼ 1:04 measurement 0:08
ð3Þ
The difference between the calculated and measured values for each species, as well as the difference between the calculated and regression-derived values, is less than
589
30 8
25
6 4
20
2 0
15
0
2
4
6
8
10
5
0
Pipart ¼ xi Pipure
log P ¼ A
Calculated headspace composition (wt%)
K. Na et al. / Chemosphere 55 (2004) 585–594
0
5
10
15
20
25
30
Measured headspace composition (wt%) Fig. 1. Gasoline headspace vapor composition calculated from Raoult’s law for 0 °C versus measured 0 °C headspace composition.
25% for this sample. This suggests that if the gasoline vapor composition can be reliably constructed entirely from a measured gasoline composition and the Raoult’s law calculations, the need for doing separate chemical analyses of the gasoline vapor could be eliminated. In theory, vapor composition could be calculated at any ambient temperature, within the vapor pressure calculation restrictions. This would represent considerable savings in samples required and in analyses performed. The calculated compositions of the gasoline vapor are listed in Table 3, together with the analyzed gasoline compositions and the results from other foreign studies. The temperatures for the spring and summer were 11 and 24 °C, respectively. These temperature values are the 30-year average temperatures of March–May and June–August for spring and summer, respectively. Calculated profiles that reflect the composition of the gasoline vapor above the liquid gasoline are quite different from the gasoline profiles. Alkanes and alkenes contents are higher in headspace vapors than those in gasoline, especially i-pentane, n-butane, n-pentane, i-butane, trans- and cis-2-butenes, 2-methyl-2-butene, and transand cis-2-pentenes. Aromatic content is much lower in the gasoline vapor profiles. The abundance of butanes is the lowest in summer, possibly due to changes in vehicle technology and efforts to reduce fuel Reid vapor pressure (RVP) that have occurred in recent years. At present, RVP is largely controlled by the contents of butanes in gasoline. The partial pressures of the compounds decrease as molecular weights increase beyond the C6 compounds (higher
590
Table 3 Chemical compositions of liquid gasolines and gasoline vapor emission sources (wt.%) This study
Australia
Gasoline
Canada
Japan
US
0.0 1.8 19.1 15.2 13.1 35.8 6.3 3.1
0.14 0.97 21.80 5.13 7.40 27.90 3.53 1.93 0.68 1.49 1.20 0.46 0.44 0.52 0.21 0.06 0.01 0.03 0.01 0.01 0.13 0.98 1.54 1.38 1.19 0.07 2.29 1.25 2.88 0.01 1.49 0.81 0.12 0.12 0.86 1.26 0.11 0.32 0.12
Gasoline vapor
Winter
Spring
Summer
Winter
Spring
Summer
Ethane Propane n-Butane i-Butane n-Pentane i-Pentane 2-Methylpentane 3-Methylpentane 2,2-Dimethylbutane 2,3-Dimethylbutane n-Hexane 2-Methylhexane 3-Methylhexane 2,4-Dmpentane n-Heptane 2-Methylheptane 3-Methylheptane Octane Nonane Ethylene Propylene 1-Butene t-2-Butene c-2-Butene 1-Pentene Isoprene t-2-Pentene c-2-Pentene 2-Methyl-2-butene Acetylene Cyclopentane Methylcyclopentane Cyclohexane Methylcyclohexane Benzene Toluene Ethylbenzene m-/p-Xylene o-Xylene
0.00 ± 0.01 0.08 ± 0.01 2.71 ± 0.31 1.32 ± 0.11 5.65 ± 0.44 11.73 ± 1.31 6.68 ± 0.70 4.45 ± 0.55 0.80 ± 0.06 1.38 ± 0.09 4.50 ± 0.38 4.96 ± 0.37 3.97 ± 0.44 0.85 ± 0.07 2.81 ± 0.11 1.07 ± 0.09 1.28 ± 0.09 1.05 ± 0.08 0.27 ± 0.04 0.00 ± 0.00 0.00 ± 0.01 0.26 ± 0.03 0.96 ± 0.08 0.83 ± 0.08 0.73 ± 0.09 0.00 ± 0.00 1.95 ± 0.21 1.04 ± 0.22 2.81 ± 0.30 0.00 ± 0.00 0.74 ± 0.11 2.72 ± 0.28 0.48 ± 0.06 1.13 ± 0.17 2.92 ± 0.34 12.86 ± 2.09 1.53 ± 0.11 5.20 ± 0.47 2.20 ± 0.31
0.00 ± 0.00 0.06 ± 0.01 3.14 ± 0.28 1.07 ± 0.20 4.75 ± 0.33 10.64 ± 0.98 6.04 ± 0.55 4.16 ± 0.37 0.78 ± 0.11 1.26 ± 0.10 4.08 ± 0.38 5.67 ± 0.56 4.69 ± 0.37 0.86 ± 0.01 3.51 ± 0.28 1.23 ± 0.11 1.46 ± 0.12 1.32 ± 0.09 0.34 ± 0.05 0.00 ± 0.00 0.00 ± 0.00 0.15 ± 0.02 0.53 ± 0.04 0.44 ± 0.04 0.63 ± 0.03 0.00 ± 0.00 1.65 ± 0.11 0.90 ± 0.12 2.60 ± 0.17 0.00 ± 0.00 0.37 ± 0.04 2.13 ± 0.27 0.26 ± 0.04 1.05 ± 0.08 1.99 ± 0.23 11.40 ± 1.73 1.54 ± 0.11 5.68 ± 0.64 2.68 ± 0.31
0.00 ± 0.00 0.04 ± 0.01 2.20 ± 0.24 0.72 ± 0.11 7.76 ± 0.68 12.39 ± 0.11 6.27 ± 0.78 4.36 ± 0.53 0.78 ± 0.08 1.29 ± 0.31 4.32 ± 0.22 5.37 ± 0.44 4.42 ± 0.39 0.84 ± 0.15 3.10 ± 0.27 1.15 ± 0.10 1.39 ± 0.24 1.25 ± 0.22 0.32 ± 0.05 0.00 ± 0.00 0.00 ± 0.00 0.13 ± 0.02 0.79 ± 0.11 0.35 ± 0.06 0.64 ± 0.05 0.00 ± 0.00 1.67 ± 0.20 0.90 ± 0.16 2.62 ± 0.25 0.00 ± 0.00 0.41 ± 0.13 2.13 ± 0.26 0.28 ± 0.30 1.03 ± 0.11 1.67 ± 0.21 10.18 ± 1.47 1.30 ± 0.11 4.50 ± 0.54 2.17 ± 0.36
0.00 ± 0.02 2.33 ± 0.30 17.34 ± 0.22 13.03 ± 0.15 8.21 ± 0.96 24.38 ± 4.41 3.48 ± 0.37 2.05 ± 0.22 0.69 ± 0.18 0.82 ± 0.17 1.56 ± 0.22 0.65 ± 0.11 0.48 ± 0.10 0.18 ± 0.04 0.23 ± 0.05 0.03 ± 0.01 0.04 ± 0.01 0.02 ± 0.01 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.01 2.05 ± 0.33 5.80 ± 0.41 4.49 ± 0.33 1.37 ± 0.11 0.00 ± 0.00 2.76 ± 0.17 1.44 ± 0.08 3.61 ± 0.44 0.00 ± 0.00 0.62 ± 0.09 0.85 ± 0.09 0.10 ± 0.02 0.10 ± 0.02 0.57 ± 0.17 0.62 ± 0.08 0.02 ± 0.01 0.06 ± 0.01 0.02 ± 0.01
0.00 ± 0.01 1.53 ± 0.21 20.70 ± 3.32 10.32 ± 1.14 8.27 ± 0.67 25.32 ± 0.24 4.20 ± 0.43 2.58 ± 0.31 0.84 ± 0.11 0.98 ± 0.15 1.99 ± 0.20 1.17 ± 0.18 0.89 ± 0.10 0.27 ± 0.04 0.49 ± 0.06 0.07 ± 0.02 0.08 ± 0.02 0.05 ± 0.02 0.00 ± 0.01 0.00 ± 0.01 0.00 ± 0.01 1.19 ± 0.22 3.06 ± 0.47 2.78 ± 0.24 1.39 ± 0.17 0.00 ± 0.00 2.83 ± 0.18 1.51 ± 0.17 4.08 ± 0.41 0.00 ± 0.00 0.39 ± 0.04 0.94 ± 0.20 0.08 ± 0.02 0.15 ± 0.03 0.59 ± 0.67 0.96 ± 0.14 0.04 ± 0.01 0.13 ± 0.04 0.05 ± 0.01
0.00 ± 0.01 1.40 ± 0.17 19.88 ± 4.01 9.73 ± 1.14 8.42 ± 1.47 25.32 ± 6.89 4.45 ± 1.01 2.74 ± 0.15 0.87 ± 0.17 1.02 ± 0.11 2.15 ± 0.36 1.31 ± 0.22 1.00 ± 0.09 0.27 ± 0.05 0.46 ± 0.07 0.08 ± 0.02 0.09 ± 0.02 0.06 ± 0.02 0.00 ± 0.01 0.00 ± 0.01 0.00 ± 0.01 0.92 ± 0.07 4.12 ± 0.53 2.99 ± 0.30 1.32 ± 0.19 0.00 ± 0.00 2.73 ± 0.34 1.46 ± 0.23 3.94 ± 0.41 0.00 ± 0.00 0.42 ± 0.05 0.95 ± 0.20 0.09 ± 0.02 0.15 ± 0.03 0.51 ± 0.06 0.93 ± 0.12 0.04 ± 0.01 0.12 ± 0.23 0.05 ± 0.01
0.0 1.5 18.7 11.1 10.7 25.4 3.5 2.2 0.6 1.1 1.9 0.7 0.5 0.2 0.3
0.0 0.0 0.0 0.0 1.6 3.7 2.9 0.7 1.5 0.9 2.6 0.0 0.6 0.9 0.3 0.2 0.9 1.0 0.1 0.2 0.1
19.92 5.68 12.30 25.65 3.83 2.20 0.33 1.22 2.27
3.2
0.47 0.21 0.05 0.05 0.03 0.00
0.72
0.13 0.93 1.27 0.07 0.27 0.08
0.9 1.0 0.1 0.3 0.1
K. Na et al. / Chemosphere 55 (2004) 585–594
Compounds
Australia: Nelson et al. (1983), Canada: Mclaren et al. (1996), Japan: Wadden et al. (1986), US: Conner et al. (1995). Blank values: not reported.
80.18 ± 11.37 16.47 ± 1.81 0.00 ± 0.00 1.61 ± 0.29 1.73 ± 0.23 79.76 ± 10.642 16.84 ± 21.75 0.00 ± 0.00 1.56 ± 0.18 1.83 ± 0.27 55.57 ± 6.23 8.57 ± 9.11 0.00 ± 0.00 5.07 ± 0.77 30.79 ± 4.45 Total Total Total Total Total
alkanes alkenes alkynes napthenes aromatics
55.06 ± 7.892 6.90 ± 0.82 0.00 ± 0.00 3.81 ± 0.45 34.23 ± 5.50
57.95 ± 7.019 7.11 ± 0.67 0.00 ± 0.00 3.86 ± 0.51 31.09 ± 4.67
75.50 ± 9.28 21.52 ± 3.41 0.00 ± 0.00 1.67 ± 0.21 1.31 ± 0.24
0.00 ± 0.00 0.06 ± 0.02 0.02 ± 0.01 0.00 ± 0.00 4.75 ± 0.64 1.33 ± 0.13 Styrene 1,2,4-Trimethylbenzene 1,3,5-Trimethylbenzene
0.00 ± 0.00 8.68 ± 0.77 2.25 ± 0.33
0.00 ± 0.00 8.83 ± 0.97 2.44 ± 0.32
0.00 ± 0.00 0.01 ± 0.01 0.00 ± 0.01
0.00 ± 0.00 0.05 ± 0.01 0.01 ± 0.01
0.0 0.0 0.0
0.01 0.11 0.04
K. Na et al. / Chemosphere 55 (2004) 585–594
591
boiling point). Thus, components beyond C6 constitute lesser fractions of the gasoline vapor compared with the gasoline, especially aromatics. Alkanes are the most abundant in the gasoline vapor profiles (77.4% on average), followed by alkenes (19.1%), and aromatics (1.7%). For the purpose of chemical mass balance modeling, Nelson et al. (1983) and Conner et al. (1995) calculated gasoline vapor compositions from their average gasoline compositions by means of Raoult’s law. They applied the calculated gasoline profiles to the apportionment of VOC to their sources. The calculated profiles would not fully represent the real world of the gasoline evaporative emissions. However, the calculated profiles have advantages of the savings of time and cost. Gasoline vapor composition in this study is generally in agreement with other studies, especially for aromatic hydrocarbons. Our result shows very good correlation (R2 ¼ 0:99) with studies conducted in Australia and US (R2 ¼ 0:94). It suggests that there is no big problem in mutual application of these profiles. Canada and US have lower i-butane compositions compared to the other countries. Japan shows somewhat different pattern in the compositions of C4 –C6 alkanes. 2.3. Paint solvents Solvents are used mainly in the process of painting, printing, dry cleaning and vapor degreasing. Among them, solvent usage by paint accounts for about 86% of the total solvent usage in Korea. To obtain the compositions of paint solvents, we used the results of paint compositions measured by National Institute of Environmental Research, Korea (NIER) (Kim, 1998). Since it is difficult to analyze the compositions of all kinds of paints, only four major paints were analyzed: urethane, varnish, archryl and thinner. In this study, the composite was calculated by applying sales figures of each paint solvent to the compositions of each paint solvent. The compositions are presented in Table 4 along with the results from other studies. In the present study, aromatics account for about 95% of paint solvents and alkanes account for the rest. Toluene (63%) is the most abundant component in paint solvents, followed by m-/p-xylene (19%) and oxylene (8%). Contribution from benzene is about 1%. This composition shows different pattern as compared to the results of Japan and US. However, all the profiles are largely composed of hydrocarbons greater than C5 . 2.4. Natural gas Natural gas is mainly used for heating and cooking in Korea. Emissions of natural gas into atmosphere are
592
K. Na et al. / Chemosphere 55 (2004) 585–594
largely derived from leakage and consumption related to combustion. The degree of leakage is dependent on the pressure of the system pipeline and number of leaks. Its emission can be controlled by pipe improvement. For most of foreign urban areas, leakage is known to be the most important factor in natural gas emissions (Derwent et al., 1996). However, in Korea, it was identified that consumption related to combustion is more important than natural gas leakage in the emission (Na and Kim, 2001). It was reported that the natural gas consumption in winter is about five times higher than that in summer. In this study, a profile of natural gas was generated from results of measurements of natural gas being used in Seoul. To obtain the profile, a 20 ml tygon tube was fully cleaned with purified nitrogen gas (purity: 99.9999%) for 10 min at a flow rate of 1 l min1 . Then, the tube was purged with natural gas for 5 min at a flow rate of 1 l min1 and both sides of the tube were tightly closed with a clip. Finally, a 1 ll sample was taken with a syringe and injected into evacuated (104 Torr) 6 l SUMMA-polished canisters with filling the canisters with the purified nitrogen gas to 15 psig. From the experiment, we obtained five duplicate samples. The natural gas source profile is presented in Table 5 along with the results from other studies. Here, the composition of natural gas in this study was calculated on a methane-free basis. Natural gas used in Korea is mainly made up of methane (89.1% by mass) and ethane (7.8% by mass). In this study, natural gas is largely composed of C2 –C5 saturated hydrocarbons. Ethane and propane are the major compounds and these account for 71% and 22% by mass, respectively. The natural gas emission profile obtained in this study is not significantly different from the other results. 2.5. Liquefied petroleum gas (LPG) LPG is used for cooking and heating in residential and commercial areas in Korea. LPG is known to be Table 4 Chemical compositions of paint solvents (wt.%) Compounds
This study
Japan
n-Butane i-Pentane n-Hexane n-Heptane Benzene Toluene Ethylbenzene m-/p-Xylene o-Xylene Styrene
0.12 ± 0.02 0.79 ± 0.14 0.26 ± 0.04 0.98 ± 0.15 0.96 ± 0.14 62.74 ± 9.41 6.20 ± 0.94 19.09 ± 2.86 7.86 ± 1.18 0.99 ± 0.15
0.0 0.0 0.0 0.0 0.0 25.7 32.5 30.3 11.5
Table 5 Chemical compositions of natural gas (wt.%; methane-free base) Compounds
This study
Egypt
Germany
US
Ethane Propane n-Butane i-Butane n-Pentane i-Pentane
70.88 ± 12.23 22.29 ± 4.78 2.92 ± 0.37 2.17 ± 0.34 0.77 ± 0.21 0.97 ± 0.28
61.98 26.15 3.29 3.35 1.26 1.93
64.00 24.20 5.00 3.40 0.90 1.00
69.40 21.30 3.10 2.10 0.70 0.70
Egypt: Doskey et al. (1999); Germany: Thijsse et al. (1999); US: Fujita et al. (1995).
Table 6 Chemical compositions of liquefied petroleum gas (wt.%) Compounds
This study
Egypt
Mexico
US
Ethane Propane n-Butane i-Butane Propylene n-Pentane i-Pentane
2.28 ± 0.36 96.90 ± 14.21 0.22 ± 0.11 0.59 ± 0.15 0.01 ± 0.01 0.00 ± 0.01 0.00 ± 0.02
0.06 4.67 69.59 20.07
0.98 67.66 16.30 12.07
0.21 2.38
0.02 0.14
4.10 90.40 0.00 0.20 5.10 0.00 0.00
Blank values: not reported. Egypt: Doskey et al. (1999); Mexico: Vega et al. (2000); US: Fujita et al. (1995).
the major source of propane in the Seoul atmosphere (Na and Kim, 2001). The experimental method used to obtain the LPG emission source profile was similar to that used for determination of the natural gas profile. Table 6 lists the LPG profile measured in this study along with a profile developed in the US. LPG is largely composed of the light alkanes (C2 –C4 ). The LPG profiles for this study and US are similar with propane dominating the emission except that US LPG contains higher mass percent of propylene compared to that of this study. Percentages of butanes in LPG from Egypt and Mexico are higher than those of this study and US.
US 0.00 0.00 0.00 0.27 78.34 1.36 8.08 8.65
Blank values: not analyzed or not reported. Japan: Wadden et al. (1986); US: Scheff et al. (1989).
3. Summary Chemical compositions of five major VOC emission sources were determined. Vehicle exhaust profiles were developed by conducting an urban tunnel study. These emissions profiles were distinguished from the other emission profiles by a high weight percentage of butanes over seasons and propane in the wintertime. This was because of the addition of propane to butane-fueled vehicles in wintertime. The gasoline evaporation profiles were prepared by blending the gasoline samples of five
K. Na et al. / Chemosphere 55 (2004) 585–594
major brands for each season on the basis of the marketshare in Seoul area. We calculated gasoline evaporative compositions using Raoult’s law from the liquid gasoline compositions. Alkanes were the most abundant in the gasoline vapor profiles and the contribution of aromatics was minor. In gasoline vapor, i-pentane was the most abundant, followed by n-butane, n-pentane, ibutane, trans- and cis-2-butenes. Aromatic content was much lower in the headspace profiles. From the comparison between experimental and calculated compositions, we identified the fact that if the gasoline vapor composition can be reliably constructed entirely from the measured gasoline composition and Raoult’s law calculations, the need for doing separate chemical analyses of the gasoline vapor could be eliminated. For the compositions of paint solvents, aromatics accounts for about 95% of paint solvents and the remainder were alkanes. Toluene was the most abundant compound in paint solvent compositions, followed by m-/p-xylene and o-xylene. Benzene and styrene contributions were less than 1%. The source profile of natural gas is calculated on a methane-free basis. Ethane and propane were the major compounds in natural gas and these accounted for 95% of natural gas composition. LPG was largely composed of the light alkanes. Propane was the most abundant compound in LPG, followed by ethane.
Acknowledgements This work was supported in part by the National Research Laboratory Program of Korean Ministry of Science and Technology (2000-N-NL-01-C-184). The authors wish to thank to KH Jung in Korea Petroleum Quality Inspection Institute for the gasoline and diesel analysis.
References Conner, T.L., Lonneman, W.A., Seila, R.L., 1995. Transportation-related volatile hydrocarbon source profiles measured in Atlanta. Journal of the Air and Waste Management Association 45, 383–394. Derwent, R.G., Davies, T.J., Delaney, M., Dollard, G.J., Field, R.A., Dumitrean, P., Nason, P.D., Jones, B.M.R., Pepler, S.A., 1996. Analysis and interpretation of the continuous hourly monitoring data for 26 C2 –C8 hydrocarbons at 12 United Kingdom sites during 1996. Atmospheric Environment 34, 297–312. Doskey, P.V., Fukui, Y., Sultan, M., Maghraby, A.A., Taher, A., 1999. Source profiles for nonmethane organic compounds in the atmosphere of Cairo, Egypt. Journal of the Air and Waste Management Association 49, 814–822. Fraser, M.P., Cass, G.R., Simoneit, B.R.T., 1998. Gas-phase and particle-phase organic compounds emitted from motor vehicle traffic in a Los Angels roadway tunnel.
593
Environmental Science and Technology 32, 2051– 2060. Fujita, E.M., Watson, J.G., Chow, J.C., Magliano, K.L., 1995. Receptor model and emissions inventory source apportionments of nonmethane organic gases in California’s San Joaquin Valley and San Francisco Bay area. Atmospheric Environment 29, 3019–3035. Gertler, A.W., Fujita, E.M., Pierson, W.R., Wittorff, D.N., 1996. Apportionment of NMHC tailpipe vs nontailpipe emissions in the Fort McHenry and Tuscarora Mountain tunnels. Atmospheric Environment 30, 2297– 2305. Haszpra, L., Szilagyi, I., 1994. Non-methane hydrocarbon composition of car exhaust in Hungary. Atmospheric Environment 28, 2609–2614. Kim, S.Y., 1998. A study on the source characterization of VOC from major emission sources. Master Thesis, KonKuk University, Seoul, South Korea. Lonneman, W.A., Sella, R.L., Meeks, S.A., 1986. Non-methane organic composition in the Lincoln tunnel. Environmental Science and Technology 20, 790–796. Mclaren, R.M., Gertler, A.W., Wittorff, D.N., Belzer, W., Dann, T., Singleton, D.L., 1996. Real-world measurements of exhaust and evaporative emissions in the Cassiar tunnel predicted by chemical mass balance modeling. Environmental Science and Technology 30, 3001–3009. Ministry of Environment (MOE), 2000. Environment White Paper, Seoul, South Korea (in Korean). Mugica, V., Vega, E., Arriaga, J.L., Ruiz, M.E., 1998. Determination of motor vehicle profiles for non-methane organic compounds in the Mexico City metropolitan area. Journal of the Air and Waste Management Association 48, 1060–1068. Na, K., Kim, Y.P., 2001. Seasonal characteristics of ambient volatile organic compounds in Seoul. Atmospheric Environment 35, 2603–2614. Na, K., Kim, Y.P., Moon, K.C., 2002a. Seasonal variation of the C2 –C9 hydrocarbons concentrations and compositions emitted from motor vehicles in a Seoul tunnel. Atmospheric Environment 36, 1969–1978. Na, K., Kim, Y.P., Moon, K.C., 2002b. Diurnal characteristics of volatile organic compounds in the Seoul atmosphere. Atmospheric Environment 37, 733–742. Nelson, P.F., Quigley, S.M., Smith, M.Y., 1983. Sources of atmospheric hydrocarbons in Sydney; a quantitative determination using a source reconcilation technique. Atmospheric Environment 17, 439–449. Pierson, W.R., Gertler, A.W., Bradow, R.L., 1990. Comparison of the SCAQS tunnel study with other on-road vehicle emission data. Journal of the Air and Waste Management Association 40, 1495–1504. Reid, R.C., Prausnitz, J.M., Poling, B.E., 1998. The properties of Gases and Liquids, fourth ed. McGraw-Hill, New York. Rogak, S.N., Pott, U., Dann, T., Wang, D., 1998. Gaseous emissions from vehicles in a traffic tunnel in Vancouver, British Columbia. Journal of the Air and Waste Management Association 48, 604–615. Sawyer, R.F., Harley, S.H., Cadle, S.H., Norbeck, J.M., Slott, R., Bravo, H.A., 2000. Mobile sources critical review: 1998 NARSTO assessment. Atmospheric Environment 34, 2161– 2181.
594
K. Na et al. / Chemosphere 55 (2004) 585–594
Scheff, P.A., Wadden, R.A., Bates, B.A., 1989. Aronian PF. Source fingerprints for receptor modeling of volatile organics. Journal of Air Pollution Control Association 39, 469– 478. Smith, J.M., Van Ness, H.C., 1987. Introduction to chemical engineering thermodynamics, fourth ed. McGraw-Hill, Singapore. Thijsse, T.R., van Oss, R.F., Lenschow, P., 1999. Determination of source contributions to ambient volatile organic compound concentrations in Berlin. Journal of the Air and Waste Management Association 49, 1394–1404.
Touaty, M., Bonsang, B., 2000. Hydrocarbon emissions in a highway tunnel in the Paris area. Atmospheric Environment 34, 985–996. Vega, E., Mugica, V., Carmona, R., Valencia, E., 2000. Hydrocarbon source apportionment in Mexico city using the chemical mass balance receptor model. Atmospheric Environment 34, 4121–4129. Wadden, R.A., Uno, I., Wakamatsu, S., 1986. Source discrimination of short-term hydrocarbon samples measured aloft. Environmental Science and Technology 20, 473– 483.