Concentration characteristics of atmospheric PCBs for urban and rural area, Korea

Science of the Total Environment 324 (2004) 261–270

Concentration characteristics of atmospheric PCBs for urban and rural area, Korea Hyun-Gu Yeoa,*, Minkyu Choib, Man-Young Chunc, Tae-Wook Kimc, Ki-Chul Chod, Young Sunwooa a

Department of Environmental Engineering, Konkuk University, 1 Hwayang-dong, Kwangjin-gu, Seoul 143-701, South Korea b Environment Management Division, National Fisheries Research and Development Institute, 408-1, Sirang-ri, Gijang-eup, Gijang-gun, Busan 619-902, South Korea c Department of Environmental Engineering, National Hankyong University, 67 Sukjong-dong, Ansung-city, Kyonggi-do 456-749, South Korea d Department of Environmental Science, Dongam Health College, 937 Jungia-dong, Changa-dong, Suwon 440-714, South Korea Received 28 June 2003; accepted 31 October 2003

Abstract Mean concentrations of total PCBs (gasqparticle) detected in urban and rural atmospheres were 130.41"62.57 pgym3 and 39.65"34.04 pgym3, respectively. The concentration distribution of PCB homologs in the urban and rural area decreased with increasing Cl substitutions and showed significant correlation coefficients (P)0.05) with the octanol-air partition coefficient (KOA ) and vapor pressure, respectively. The fractions (%) of total PCBs were 28% for tri-CBs, 25% for tetra-CBs and 24% for penta-CBs in urban air and 45% for tri-CBs, 24% for tetra-CBs, and 21% for penta-CBs in rural air. The sum of those homologs was 77% for urban and 90% for rural air. Therefore, these homologs were identified as the main components of PCB homologs compared to other homologs ()pentaCBs). The Clausius–Clapeyron (CC) plot was applied to atmospheric PCB data, relating PCB partial vapor pressure (logarithm P) to inverse absolute temperature (1yT). The slopes obtained from Clausius–Clapeyron plots were y 3888 (R 2s0.75, P-0.0001) for urban and y1902 (R 2s0.22, P-0.1) for rural air. The slope for urban air was approximately two times higher than that of rural air, possibly because the atmospheric concentration of lower molecular weight congeners in urban air may be predominantly influenced by local sources relative to rural air. 䊚 2003 Elsevier B.V. All rights reserved. Keywords: PCBs; Homologs; Temperature; KOA; Vapor pressure; Clausius–Clapeyron plot

1. Introduction A number of semivolatile organic compounds (SOCs) have been found to be widely distributed *Corresponding author. Tel.: q82-2-455-0129; fax: q82-2454-0428. E-mail address: [email protected] (H.-G. Yeo).

in the atmosphere because of their moderate vapor pressure, low solubility and low reactivity. Many of these compounds contain chlorine and they have the capability of bio-accumulating through the food chain. The presence of these chemicals in regions where they were not used is of concern, due to both ecological health reasons and because

0048-9697/04/$ - see front matter 䊚 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2003.10.031

262

H.-G. Yeo et al. / Science of the Total Environment 324 (2004) 261–270

of our inability to control the movement of these chemicals in the environment without having a better knowledge of the processes regarding their movement and fate (Hoff et al., 1998). The spatial distributions of polychlorinated biphenyls (PCBs) in the atmosphere have been studied in urban, rural and remote locations. The urban concentrations of PCBs in the atmosphere are generally higher than in rural and remote areas (Cotham and Bidleman, 1995). PCBs have a propensity to revolatilize from secondary source compartments such as soil, vegetation, water and atmospheric particles. Since the primary sources (i.e. manufacture and specific useyhanding of PCBs) have been largely reduced, outgassing from these compartments is believed to account for the contemporary atmospheric burden of PCBs. For these secondary sources, temperature is invoked as the primary controller for the condensationyvolatilization process, effectively accounting for the cycling of PCBs between air and surface. Evidence of seasonality in air concentration has been reported for many PCB congeners in year-round monitoring studies (Manchester-Neesvig and Andren, 1989; Hoff et al., 1992). Several studies have investigated this process utilizing the Clausius–Clayperon equation, whereby air concentrations expressed as partial pressure are plotted against inverse temperature (Hoff et al., 1998). Haugen et al. (1998) suggested that a steep slope of regression line between ln P (partial pressure) and (1yT) indicates that air concentration is controlled by re-evaporation from surface in the local surroundings of the sampling site, whereas a shallow slope indicates that advection of air is governing atmospheric concentration levels. By contrary, if the slope of regression line between ln P (partial pressure) and (1yT) is weak, the PCB concentration may be more affected by outside pollutant inflow rather than local sources (Hoff et al., 1998). A significant number of studies also exists for urban and industrialized areas (Franz et al., 1998; Hoff et al., 1998; Semb and Pacyna, 1998; Ashley and Baker, 1999; Bamford et al., 1999; Offenberg and Baker, 1999; Zhang et al., 1999). In addition to these studies, several important reviews have examined the spatial distribution of these chemicals.

However, research of SOCs in different environmental matrices is still at an early stage in Korea. A few studies have been conducted for PCDDyFs (Park and Kim, 2002), PCBs (Yeo et al., 2003a,b) in the atmosphere and municipal and industrial waste incinerators (Chang et al., 1999; Ikonomou et al., 2002) of Korea. Breivik et al. (2002) investigated the estimated cumulative global consumption pattern for total PCBs, and includes estimates for 114 individual counties. It is estimated that USA has been responsible for as much as approximately 46% of the total historical global PCB consumption. Other major consuming countries are Russia (7.9%), Germany (7.1%), Japan (4.1%), France (4.1%), Canada (3.0%), Ukraine (2.4%), Italy (2.1%), UK (2.0%) and China (0.6%). Unfortunately, we had only initially imported in Japan from 1954 and no reports of production and usage amount in Korea Our objective is to investigate the concentration distribution and to use the Clausius–Clayperon process to understand the effects of the air temperature on atmospheric PCBs in the urban and rural area, which have the different source characteristics of PCBs. 2. Experiment 2.1. Sampling programs Atmospheric samples were taken from July, 1999 to January, 2000. The sampling program was conducted in the urban area of Seoul and rural area of Kyonggi-do. A low-volume polyurethane form (PUF) and glass fiber filter (GFFs) sampler was used to collect atmospheric PCB samples. The urban site is located on the roof of the college of engineering building of Konkuk University (15 m height above ground) and the rural site is located within a small shelter (0.5 m height above ground). The rural site is located at the National Hankyong University campus, which is approximately 1 km northeast of downtown Ansung, and 10 km west of Kyongbu-express way. Urban site (Seoul) is the largest city of Korea, and rural site (Ansung) with agricultural and industrial region is located in 100 km on the south side of Seoul. The distance between two sites is approximately 150 km.

H.-G. Yeo et al. / Science of the Total Environment 324 (2004) 261–270

263

Fig. 1. Sampling sites for urban and rural atmosphere.

The population is approximately 9 853 972 and population density (populationykm2) is 16 342 at the urban site. The population is approximately 130 000 and population density (populationykm2) is 882 and agricultural fields make up 33% of the total area in the rural site. There is a complex mix between agricultural and industrial regions in the south side of Seoul (Fig. 1). Air samples were taken with a low-volume PUF sampler modified GPS-1 PUF sampler with dual sampling module which comes with two main parts, filter holder and glass cartridge, located on top of the sampler (General Metal Works Inc., Ohio). The overall average volume was approximately 600 m3 (average flow rate: 30 lymin) and each sample was a 2-week composite. Fourteen samples were taken during the sampling period. The air was drawn through a glass fiber filter (GFF, length 5 cm) to collect particles and then through a polyurethane form (PUF) plug (length 8.0 cm, diameter 7.5 cm) to collect compounds present in the gas phase. Meteorological data such as temperature, wind

speed, wind direction and relative humidity were obtained from Korea Meteorological Administration located in Seoul and a meteorological tower (AWS) located on top of a building located on the Hankyong University campus. 2.2. Chemical analysis The GFFs were precombusted at 450 8C for 24 h in loosely wrapped aluminum foil envelopes and then they were sealed and stored at 4 8C until sampling. The PUFs were precleaned by Soxhlet with hexane: dichloromethane (9:1 vyv) and then placed in a vacuum dry oven to dry and stored in sealed glassware at 4 8C until sampling. After sampling, the GFFs and PUFs were all wrapped in glassware and stored at y26 8C until analysis. PUFs were spiked with PCBs surrogate standards prior to sampling to determine analytical recovery. The surrogate standards (13C12) were made up of PCB 28, 52, 101, 138, 153, 180, 209. The 4,49dibromooctaflurobiphenyls (Supelco, USA) were

264

H.-G. Yeo et al. / Science of the Total Environment 324 (2004) 261–270

used as the internal standard. The GFF and PUF samples were extracted and analyzed together in order to study the concentration distribution of total PCBs (particleqgas phase). The samples were extracted in Soxhlet apparatus with hexane: dichloromethane (9:1 vyv) for 24 h. Extracts were concentrated to approximately 2 ml in a rotary evaporator (BUCHI, R-124). The first and second clean-up of PCBs were conducted with a silica column (activated silica 3 g at 350 8C overnight, sodium sulfate at 450 8C overnight) and GPC (gel permeation chromatography, BUCHI, B-688) column to remove any polar organics that might interfere with the analysis (Yeo et al., 2003a,b). The samples were then concentrated under a gentle stream of purified nitrogen concentrator to 50 ml by using dodecane (SIGMA, D-4259) which was added to the extracts to prevent losses during the volume reduction to 50 ml using N2 concentration and analyzed on a Hewlett Packard 6890 equipped with MS detector (HP 5973). PCBs were analyzed by GCyMS with HP-5MS (5% diphenyl and 95% dimethypolysiloxane) column w30 m (length), 0.25 mm (i.d.), 0.25 mm (film thickness)x. The temperature program was as follows: 150 8C for 2 min, 30 8Cymin to 170 8C, 4 8Cymin to 200 8C for 13 min, 4 8Cymin to 268 8C and 70 8Cymin to 320 8C for 4.1 min. 2.3. Quality controlyquality assurance The analytical quality of the data was determined using LOD (limit of detection), recovery, reproducibility, linearity and by checking sampling artifacts. LOD was calculated with spiking a blank with the calibration standard at a concentration at 1–5 times the expected LOD. Analyze it seven times. The LOD are then calculated as follows: LODsmean"3=S.D. These LODs used in air samples with gas and particle, respectively. The detection limit of the PCB standards ranged from 0.005 (PCB 18) to 0.04 pgym3 (PCB 52) with normalizing average sampling volume (600 m3). The materials used in the present study were assessed for possible analytical contaminants. The purpose of the blank test is to check whether and how much contamination of PCBs can be found in experiment materials. Due to the many experi-

mental materials used in this study, it was necessary to check the background concentration level of PCBs in a considerable number of individual matrices. The solvent-cleaned and pre-combusted sampling materials, glass fiber filters, filters for PUF samplers and PUF plug, were Soxhlet extracted with hexane:dichloromethane (9:1). The results of the blank tests did not show any interference peaks that would inhibit quantifying the PCB congeners in PUFs and GFFs. The recoveries (ns 5) of PCBs surrogate standards were 77.2"5.5% for PCB 28, 62.9"7.3% for PCB 52, 78.0"9.2% for PCB 101, 100.8"9.7% for PCB 153, 106.1"8.2% for PCB 138, 116.6"10.2% and 116.0"13.4% for PCB 209. The linearities of the calibration standard (Ultra Scientific Inc.) were calculated by regression analysis with values ranging from 0.99 to 1.00 (R 2). Sampling artifacts associated with the GFF and PUF can affect the apparent gas-particle distributions of PCBs. The GFF may exhibit two such artifacts with counteracting effects on the distribution. First, gas phase PCBs may adsorb to the filter surface and particles collected on the filter (McDow and Huntzicker, 1990; Hart et al., 1992; Hart and Pankow, 1994). Second, the more volatile compounds may be stripped from the filter by continuing gas flow if the gas phase concentration decreases, the temperature increases during the sampling periods or due to gas phase reactions on the filter (Katz and Chan, 1980; Zhang and McMurry, 1991). The extent of gas adsorption is often estimated using a second filter. A backup filter was used on five samples collected in Ansung city. The percent mass on the second filter for individual PCBs was either below 5% or not detected at all. Therefore, the mass from the backup filter was neither subtracted from the particle phase concentration nor added to the gas phase concentration. Also, volatilization from the filter has the opposite effect. This was not determined, but Zhang and McMurry (1991) have suggested that this impact is usually less than or equal to 10%. Spilt PUFs were collected to assess gas phase breakthrough. The bottom half of the spilt PUFs contained an average of 12% (ns3) of the total mass, indicating minimal breakthrough. Therefore,

H.-G. Yeo et al. / Science of the Total Environment 324 (2004) 261–270

265

Table 1 PCB congeners routinely quantified using GCyMSD (SIM) analysis PCB congeners PCB 18, 27, 28, 33, 38 PCB 44, 47, 49, 52, 60, 61q74, 77, 79 PCB 87, 101, 105, 110, 118, 114, 123, 126 PCB 138, 141, 149, 150, 151, 153, 156, 157, 167, 169 PCB 170, 180q193, 183, 187, 189 PCB 194q205 PCB 209

Monitoring ions (myz) 256a, 258, 186 292a, 290, 220 326a, 324, 256 360a, 362, 290 394a, 396, 324 430a, 428, 360 498a, 496, 428

PCB surrogate standards PCB 28 PCB 52 PCB 101 PCB 138, 153 PCB 180 PCB 209

268a, 304a, 338a, 372a, 406a, 510a,

270, 302, 340, 374, 408, 512,

198 232 266 302 336 440

a

Target ion.

the gas phase concentrations in this study were adjusted accordingly. Simcik et al. (1998) examined gas phase breakthrough in the same manner during their study in Chicago and over Lake Michigan in July when the ambient temperature was highest. Individual PCBs on the bottom half of the spilt PUFs were all less than 15% of their total PUF masses. We concluded that the GFFand PUF-associated concentrations of PCBs are a good estimate of the true gas and particle concentrations in this study. PCBs were identified by the mass to charge (my z) ratios from a full scan, then compared to the myz of standards or matched to the computer database (NIST library). Once the compounds were identified, selected ion monitoring (SIM) was used for routine analysis, because: (1) SIM is more sensitive than full scan for ions; and (2) full scan data always occupy large amounts of disk space. Table 1 gives the monitored myz ratios and the retention time of the compounds. 3. Results and discussion 3.1. Concentration distribution of total PCBs (gasqparticle) and homologs During the sampling period, the mean temperature showed in the urban (13.6 8C) and rural area (13.9 8C). Mean wind speeds indicated 1.4 mys

for the urban area and 0.5 mys for the rural area. However, total precipitation was 675 mm in urban and 1737 mm in rural area. Fig. 2 shows the distribution of total PCBs concentration and homologs, which was detected in the urban and rural air during the sampling period. The calibration standards of 38 individual PCB congeners frequently detected in the atmosphere of Korea were investigated during this study. Twenty-four PCBs were detected in the rural atmosphere while thirtytwo PCBs were detected at the urban air. Twenty PCB congeners were detected in both the rural and urban area. The mean of total PCBs detected in the urban and rural areas were 130.41"62.57 pgym3 and 39.65"34.04 pgym3, respectively. The mean concentration in the urban area is 3.3 times higher than that of the rural area. As I mentioned previously, the total precipitation amounts in the rural area were approximately three times higher than those of the urban site. However, there were no significant correlation coefficients between total PCBs concentration and precipitation amount in urban and rural area, respectively. The highest concentration was 289.80 pgym3 for urban and 143.76 pgym3 for rural during the summer season. Also, PCB homologs concentrations were distributed in the order of tri-CBs)tetra-CBs)pentaCBs)hexa-CBs)hepta-CBs)octa-CBs)decaCBs in the two areas. That is, the atmospheric concentration of PCBs decreased with increasing

266

H.-G. Yeo et al. / Science of the Total Environment 324 (2004) 261–270

Cl substitutions. As shown in these data, with increasing number of Cl ()penta-CBs), the log KOA also increased (Table 2). The PCB homologs ()penta-CBs) easily deposit on plants, soil and layers of water and evaporate less into the atmosphere due to lower vapor pressure compared to tri- and tetra-CBs. Because PCB homologs with high atmospheric concentrations have higher vapor pressure, most of the PCBs in the atmosphere were tri- and tetra-CBs, which mostly exists as gases in the atmosphere. As we can find out from the above result, the distribution of PCB homologs in the atmosphere can be explained by log KOA (rsy 0.85, P-0.01 for urban and rsy0.73, P-0.05 for rural) and vapor pressure (rs0.63, P-0.05 for urban and rs0.91, P-0.01 for rural area) which are physicochemical characteristics of PCBs. 3.2. Fraction (%) of individual PCB congeners in total PCBs Fig. 3 shows the fractions (%) of individual PCB congeners out of total PCBs, which were detected during the sampling period. PCB 28 had the highest value of 12.3% in the urban air, but, in the rural air, PCB 38 showed the highest value of 23.5%. The fraction of tri-CBs, tetra-CBs and

Table 2 The physical–chemical properties of PCB homologs (Mackay et al., 1992) Homologs

log KOA

Vapor pressure (Pa at 25 8C)

Tri-CBs Tetra-CBs Penta-CBs Hexa-CBs Hepta-CBs Octa-CBs Deca-CBs

8.19 8.68 9.50 9.94 10.3 11.2 14.0

3.8=10y2 6.4=10y3 1.1=10y3 1.8=10-4 – 2.8=10y5 –

penta-CBs in the urban area were 28%, 25% and 24%, respectively. The sum of those homologs is 77% of total PCBs. In the rural area, the fraction of tri-CBs, tetra-CBs and penta-CBs were 45%, 24% and 21%, respectively. Therefore, these homologs were identified as the main components of PCB homologs at these two sites compared to other homologs ()penta-CBs). The concentration fraction of low molecular PCB congeners (e.g. triCBs) showed higher contributions in rural rather than urban area. That is, low molecular PCB congeners that are mostly influenced by long-range transport (Oehme et al., 1996) had higher concentrations in rural area. In this study, also concentrations of PCBs in rural area may be more affected by long-range transport than in urban area.

Fig. 2. Concentration distribution of PCB homologs and total PCBs for urban (left) and rural (right) atmosphere.

H.-G. Yeo et al. / Science of the Total Environment 324 (2004) 261–270

267

Fig. 3. Mean fraction (%) of individual PCB congeners as percentage of the total PCBs in urban and rural atmosphere.

Twenty PCB congeners, which were detected simultaneously in rural and urban area were used for the regression analysis. The regression line indicated a statistically significant correlation (slope: 0.97, P-0.001) and may be written as follows: wRuralxs0.97 wUrbanx. The slope of the regression line is almost 1 and the fraction of PCB congeners was showed very similar between the rural and urban area. It is expected that these results were influenced by physicochemical properties of PCB congeners rather than differences of local sources in rural and urban area. The concentrations of ICES (International Council for the Explanation of the Seas) congeners, which are frequently detected worldwide,

were compared to data from other countries (Table 3). The urban concentrations of ICES congeners researched in this study showed slightly higher concentrations compared to other rural areas of Greece and the UK. However, ICES concentrations in this study showed lower levels (ranging from 4 to 10 times) compared to urban areas in foreign countries (USA, Greece). Also, in the case of rural area, the concentration of ICES congeners indicated lower values than that of rural and urban areas in foreign countries, and similar to levels in polar area. The SICES congeners in this study showed 42.17 pgym3 for urban which was ranged from 1.6 to 3.3 times lower than of those measured in Greece and the USA. That is, in general, the

Table 3 Summary of PCBs concentration (ICES congeners) in this study and other researchers (pgym3) Congener

28 52 101 118 138 153 180 S ICES

This study

Rural area

Urban area

Urban

Rural

Lake Districta UK

Coastal Woodlanda UK

Coastal areab Greece

Athensb Greece

Baltimorec USA

14.34 7.90 10.74 2.82 2.11 3.26 1.00 42.17

8.39 2.00 4.37 0.80 NAe 0.55 0.19 16.3

5.79 2.26 1.10 0.66 0.60 0.77 0.40 11.58

22.4 4.18 3.59 1.09 1.76 2.70 0.51 36.23

14.33 7.00 16.27i 9.03 3.01j 6.39 NA 41.7

30.46 25.13 21.04i 9.94 3.36j 7.04 NA 66.51

55.5 55.9f 32.1 NA 18.4g 31.5h 4.6 142.5

Background Arcticd Norway 4.30 2.47 1.28 0.53 0.54 0.61 0.16 9.89

a Halsall et al., (1999); bMandalakis et al., (2002); c Brunciak et al., (2001); d Oehme et al., (1996); e NA: not analyzed; f52q43; 138q163; h153q132; i101q90; j138q163q164.

g

268

H.-G. Yeo et al. / Science of the Total Environment 324 (2004) 261–270

concentration levels researched in this study showed slightly lower values than those of foreign countries. Also, the concentrations of ICES congeners measured in other European and US urban areas where higher than those determined in the present study. Namely, for congeners 28, 52, 118 and 153 the corresponding concentrations in London (Halsall et al., 1995) were 557 pgym3, 531 pgym3, 53 pgym3 and 26 pgym3, respectively, while in Chicago (Simcik et al., 1998) for congener 28, 52, 101, 118, 138, 153 and 180 showed 158 pgym3, 123 pgym3, 100 pgym3, 46 pgym3, 115 pgym3, 96 pgym3 and 78 pgym3, respectively. The higher PCBs atmospheric concentrations of the USA and the UK are probably due to the fact that much larger amounts of PCBs commercial mixtures were produced and consumed in these countries than in Korea (Breivik et al., 2002). 3.3. Temperature dependence of ICES congeners The relationship between PCBs concentration (ln P wpartial pressurex) and temperature (1yT) in the atmosphere has been researched in urban, rural and background areas by many researchers (Hoff et al., 1992; Halsall et al., 1995; Hornbuckle and Eisenreich, 1996; Panshin and Hites, 1994; Kaupp et al., 1996; Berg et al., 1996). Also, atmospheric PCB concentrations as a function of temperature have been investigated in urban locations, rural sites in cold temperature regions, as well as very remote sites. Because the air–surface exchange is significantly affected by physicochemical properties, it is more appropriate to analyze the temperature dependence for individual congeners rather than the sum of PCBs (Wania et al., 1998). In this study, we examined the temperature dependence in urban and rural area using lower molecular weight PCB congeners, which dominantly exist as gas-phase PCBs in the atmosphere. The slope of the regression line between ln P wpartial pressurex and 1yT was a very important indicator for understanding PCB sources (Wania et al., 1998). The slope and regression coefficient (R 2 ) is a good indicator for evaluating whether PCBs are mostly affected by long-range transport or volatilization from local sources (e.g. soil, water system and vegetation). If the amount of PCBs present in the

atmosphere is partly controlled by temperaturedependent re-volatilization from soils, vegetation and water systems, the measured concentrations should be related to ambient temperature (Hoff et al., 1998). Haugen et al. (1998) suggested that a steep regression line between ln P and 1yT in the atmosphere indicates that air concentration is mostly controlled by re-evaporation from the surface in the local surroundings of the sampling site, whereas a shallow slope indicates that advection of air is governing atmospheric concentration levels. Namely, if the slope of the regression line between ln P and 1yT is shallow, the PCBs concentration may be affected by outside pollutant inflow rather than local sources (Hoff et al., 1998). In this study, the slope of regression line for total PCBs (tri-q tetra-CBs) between ln P and temperature (1yT) was y3888 (R 2s0.75, P-0.0001) in the urban area and y1902 (R 2s0.22, P-0.1) in the rural area (Table 4). It is rather lower than that in foreign countries’ data (Wania et al., 1998). For instance, the slope in urban Bloomington is 6000– 7000 and exceeds the slope observed in the Great Lakes (4000–5000), which further exceeds the slope observed in Arctic Canada (Wania et al., 1998). The rural sites (Lista, Gardsjon and Rovik) appear to have slightly steeper slopes than the urban locations (Augsburg, Manchester, Wania et al., 1998). The Arctic site (Ny Alesund) has a very low slope. Consequently, the slope of the urban area, which was researched in the study showed a rather shallower slope lower than those of foreign urban areas. It may be presented that our urban area is affected mostly by local sources rather than by transport. However, the slope for the rural area was very similar to that of the North Pole, which is affected more by pollutant inflow (i.e. long range transport) than by local sources. 4. Conclusion Total PCBs concentrations at the urban site were three times higher than those of the rural site and thus the urban area is more polluted than the rural area with respect to PCBs. The summation fraction of tri-CBs, tetra-CBs and penta-CBs out of total PCBs was 77% for urban and 90% for rural. In this study, the concentration of PCBs in the rural

H.-G. Yeo et al. / Science of the Total Environment 324 (2004) 261–270

269

Table 4 Slope, regression coefficient (R 2 ), significant levels (P) with 95% confidence limits using Clausius–Clapeyron equation for low molecular PCB congeners (tri-CBs and tetra-CBs) Congener

PCB 18 PCB 28 PCB 33 PCB 47 PCB 49 PCB 52 PCB 60 PCB 61y74 PCB 66 S PCB

Urban

Rural 2

Slope

R

y5673 y3466 y2025 y1490 y3609 y3744 y4459 y4900 y5665 y3888

0.82 0.79 0.31 0.17 0.40 0.44 0.81 0.78 0.85 0.75

P

Slope

R2

P

P-0.0001 P-0.0001 P-0.05 Ps0.159 P-0.05 P-0.01 P-0.0001 P-0.0001 P-0.0001 P-0.0001

– y3413 y600 y5372a y7940 y1569 289 2737 y3051 y1902

– 0.55 0.10 0.41 0.62 0.17 0.01 0.31 0.39 0.22

– P-0.01 Ps0.460 Ps0.089 P-0.01 Ps0.144 Ps0.694 P-0.05 P-0.05 P-0.1

a

PCB 38.

area may be affected more by inflow from ‘outside’ pollutants compared to the urban area. In terms of relationship between total PCBs and temperature, correlation coefficients (r) in urban and rural area were 0.80 and 0.64, respectively. The slope of the regression lines for partial pressure (ln P) of lower molecular weight PCB congeners and temperature (1yT) using Clausius– Clayperon equation indicated that the slope of the CC plot in the urban site was approximately two times higher than the rural area. It means that the PCB concentration in the urban area was affected more by local sources than the rural area. References Ashley JTF, Baker JE. Hydrophobic organic contaminants in surficial sediments of Baltimore harbor: inventories and sources. Environ Toxicol Chem 1999;18:838 –849. Bamford HA, Offenberg JH, Larsen RK, Ko FC, Baker JE. Diffusive exchange of polycyclic aromatic hydrocarbons across the air–water interface of the Patapsco River, an urbanized subestuary of the Chesapeake Bay. Environ Sci Technol 1999;33:2138 –2144. Berg T, Hjellbrekke AG, Skjelmoen JE. Heavy metals and POPs within the ECE region. NILU EMEPyCCC Report 8y 96, Kjeller, Norway:Norwegian Institute for Air Research, 1996. Breivik K, Sweetman A, Pacyna JM, Jones KC. Toward a global historical emission inventory for selected PCB congeners—a mass balance approach. I. Global production and consumption. Sci Total Environ 2002;90:181 –198. Brunciak PA, Dachs J, Franz TP, Gigliotti CL, Nelson ED, Turpin BJ, Eisenreich SJ. Polychlorinated biphenyls and

particulate organicyelemental carbon in the atmosphere of Chesapeake Bay, USA. Atmos Environ 2001;35:5663 –5677. Cotham WE, Bidleman TF. Polychlorinated aromatic hydrocarbons and polychlorinated biphenyls in air at an urban and a rural site near Lake Michigan. Environ Sci Technol 1995;29:2782 –2789. Chang YS, Kong SB, Ikonomou MG. PCBs contributions to the total TEQ released from Korean municipal and industrial waste incinerator. Chemosphere 1999;39:2629 –2640. Franz TP, Eisenreich SJ, Holsen TM. Dry deposition of particulate polychlorinated biphenyls and polycyclic aromatic hydrocarbons to Lake Michigan. Environ Sci Technol 1998;32:3681 –3688. Halsall CJ, Lee RGM, Coleman PJ, Burnett V, Harding Jones P. PCBs in UK urban air. Environ Sci Technol 1995;29:2368 –2376. Halsall CJ, Gevao B, Howsam MR, Le GM, Ockenden WA, Jones KC. Temperature dependence of PCBs in the UK atmosphere. Atmos Environ 1999;33:541 –552. Hart KM, Isabelle LM, Pankow JF. High-volume air sampler for particle and gas sampling. 1. Design and gas sampling performance. Environ Sci Technol 1992;26:1048 –1052. Hart KM, Pankow JF. High-volume air sampler for particle and gas sampling. 2. Use of backup filter to correct for the adsorption of gas-phase polycyclic aromatic hydrocarbons to the front filter. Environ Sci Technol 1994;28:655 –661. Haugen JE, Wania F, Ritter N, Schlabach M. Hexachlorocyclohexanes in air in Southern Norway temporal variation, source allocation and temperature dependence. Environ Sci Technol 1998;32:217 –224. Hoff RM, Muir DCG, Grift NP. Annual cycle of polychlorinated biphenyls and Organohalogen pesticides in air in Southern Ontario: atmospheric transport and sources. Environ Sci Technol 1992;26:276 –283. Hoff RM, Brice KA, Halsall CJ. Non-linearity in the slope of Clausius–Clapeyronplots for SVOCs. Environ Sci Technol 1998;32:1793 –1798.

270

H.-G. Yeo et al. / Science of the Total Environment 324 (2004) 261–270

Hornbuckle KC, Eisenreich SJ. Dynamics of gaseous semivolatile organic compounds in terrestrial ecosystem effects of diurnal and seasonal climate variations. Atmos Environ 1996;30:3935 –3945. Ikonomou MG, Sather P, Oh JE, Choi WY, Chang YS. PCB levels and congener patterns from Korean municipal waste incinerator stack emissions. Chemosphere 2002;49:205 – 216. Katz M, Chan C. Comparative distribution of eight polycyclic aromatic hydrocarbons in airborne particulates collected by conventional high-volume sampling and by size fractionation. Environ Sci Technol 1980;14:838 –843. Kaupp H, Dorr G, Hippelein M, McLachlan MS, Hutzinger O. Baseline contamination assessment for a new resource recovery facility in Germany part IV: atmospheric concentrations of polychlorinated biphenyls and hexachlorobenzene. Chemosphere 1996;32:2029 –2042. Mackay D, Shiu WY, Ma KC. Illustrated handbook of physical–chemical properties and environmental fate for organic chemicals. Monoaromatic hydrocarbons, chlorobenezens and PCBs. CheiseayMI: Lewis Publishers, 1992. Manchester-Neesvig JB, Andren AW. Seasonal variation in the atmospheric concentration of polychlorinated biphenyls congeners. Environ Sci Technol 1989;23:1138 –1148. Mandalakis M, Tsapakis M, Tsoga A, Stephanou EG. Gasparticle concentrations and distribution of aliphatic hydrocarbons, PAHs, PCBs and PCDDyFs in the atmosphere of Athens (Greece). Atmos Environ 2002;36:4023 –4035. McDow SR, Huntzicker JJ. Effect of composition and state of organic compounds on polycyclic aromatic hydrocarbons decay in atmosphere aerosols. Atmos Environ 1990;24: 2563 –2572. Oehme M, Haugen JE, Schlabach M. Seasonal changes and relation between levels of organochlorines in a Arctic ambient air: first results of an all-year-round at Ny-lesund, Svalbard, Norway. Environ Sci Technol 1996;30:2294 – 2304.

Offenberg JH, Baker JE. Influence of Baltimore’s urban atmosphere on organic contaminants over the Northern Chesapeake Bay. J Air Waste Manage Assoc 1999;49:959 – 965. Panshin SY, Hites RA. Atmospheric concentrations of polychlorinated biphenyls at Bloomington, IN. Environ Sci Technol 1994;28:2008 –2013. Park JS, Kim JG. Regional measurements of PCDDyFs concentrations in Korea atmosphere and comparison with gasparticle partitioning models. Chemosphere 2002;49:755 – 764. Semb A, Pacyna JM. Toxic trace elements and chlorinated hydrocarbon, source, atmospheric transport and deposition: Report NORD. Copenhangen, Denmark: Nordic Council of Minssters, 1998. Simcik MF, Franz ThP, Zhang H, Eisenreich StJ. Gas-particle partitioning of PCBs and PAHs in the Chicago urban and adjacent coastal atmosphere: states of equilibrium. Environ Sci Technol 1998;32:251 –257. Wania F, Haugen JE, Lei YD, Mackay D. Temperature dependence of atmospheric concentration of semivolatile organic compounds. Environ Sci Technol 1998;32:1013 –1021. Yeo HG, Choi MK, Chun MY, Young S. Gasyparticle concentration and partitioning of PCBs in the atmosphere of Korea. Atmos Environ 2003a;37:3561 –3570. Yeo HG, Choi MK, Chun MY, Young S. Concentration distribution of polychlorinated biphenyls and organochlorine pesticides and their relationship with temperature in rural air. Atmos Environ 2003b;37:3831 –3839. Zhang X, McMurry PH. Theoretical analysis of evaporative losses of adsorbed of absorbed species during atmospheric aerosol sampling. Environ Sci Technol 1991;25:456 –459. Zhang H, Eisenreich SJ, Franz TR, Baker JE, Offenberg JH. Evidence for increased gaseous PCB fluxes to Lake Michigan from Chicago. Environ Sci Technol 1999;33:2129 – 2137.