Fs for Various Atmospheric

Environmental and Occupational Health, Medical College,. National Cheng Kung ... sources for different atmospheric environments in a highly industrial...
0 downloads 0 Views 206KB Size
Download PDF
Environ. Sci. Technol. 2004, 38, 4937-4944

Source Identification of PCDD/Fs for Various Atmospheric Environments in a Highly Industrialized City WEI-SHAN LEE,† GUO-PING CHANG-CHIEN,† LIN-CHI WANG,† WEN-JHY LEE,‡ P E R N G - J Y T S A I , * ,§ K U E N - Y U H W U , | A N D CHIEH LIN⊥ Department of Chemical Engineering, Cheng Shiu University, 840 Chengching Road, Kaohsiung 833, Taiwan, Department of Environmental Engineering, National Cheng Kung University, 1 University Road, Tainan 70101, Taiwan, Department of Environmental and Occupational Health, Medical College, National Cheng Kung University, 138 Sheng-Li Road, Tainan 70428, Taiwan, Division of Environmental Health and Occupational Medicine, National Health Research Institutes, 100 Shih-Chuan 1st Road, Kaohsiung 807, Taiwan, and Department of Environmental Engineering and Science, National Pingtung University of Science and Technology, Nei Pu 91207, Ping Tung, Taiwan

This study set out to identify possible PCDD/F emission sources for different atmospheric environments in a highly industrialized city located in southern Taiwan. We collected stack flue gas samples from five main stationary emission sources of the municipal solid waste incinerators (MSWIs), medical waste incinerators (MWIs), electric arc furnaces (EAFs), secondary aluminum smelters (ALSs), and sinter plants to assess the characteristics of their PCDD/F emissions. For mobile sources, congener profiles reported in U.S. EPA’s database for unleaded gas-fueled vehicles (UGFV) and diesel-fueled vehicles (DFV) were directly adopted owing to lack of local data. The congener profiles of the 2,3,7,8-substituted PCDD/Fs were selected as the signatures of these PCDD/F emission sources. We conducted PCDD/F samplings on atmospheric environments of four categories, including background, residential area, traffic area, and industrial area. Through PCA and cluster analyses, we found that traffic areas were most influenced by PCDD/F emissions from UGFV and DFV, while those of industrial areas were mainly influenced by metallurgical facilities and MWIs. The above results were further examined by using the methodology of the indicatory PCDD/Fs. We confirmed that traffic areas were contributed by traffic sources, but industrial areas were simply affected by metallurgical facilities rather than MWIs. In conclusion, besides the use of PCA and cluster analyses, the methodology of the indicatory PCDD/Fs should be conducted for further validation in order to prevent misjudgment. * Corresponding author phone: +886-6-2088390; fax: +886-62088391; e-mail: [email protected]. † Cheng Shiu University. ‡ Department of Environmental Engineering, National Cheng Kung University. § Medical College, National Cheng Kung University. | National Health Research Institutes. ⊥ National Pingtung University of Science and Technology. 10.1021/es0499795 CCC: $27.50 Published on Web 09/01/2004

 2004 American Chemical Society

Introduction After polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) were discovered in the flue gases and fly ash of municipal solid waste incinerators (MSWIs) in 1977 (1), PCDD/F emissions from various sources have become a serious issue in many countries, because of their toxicological effects and associated adverse health implications. PCDD/Fs released to the atmosphere are mainly from anthropogenic activities, including waste incineration, power/energy generation, other high-temperature sources, metallurgical processes, and chemical-industrial sources. In the Lombardy Region (a highly industrialized area located in the north of Italy which has about 9 million inhabitants), the main PCDD/F sources are the MSWIs (32% of total value), followed by the electric arc furnaces (EAFs) (26%). Other significant PCDD/F sources are diesel combustion in vehicles (10%) and secondary aluminum smelters (secondary ALSs, 9%) (2). Various research conducted for establishing PCDD/F inventory (3-6) also reveals that MSWIs, medical waste incinerators (MWIs), metal metallurgy (ferrous and nonferrous), and vehicles are the dominant PCDD/F emission sources in other countries. In principle, emission inventory provides useful information for assessing the amount of PCDD/Fs released to the atmosphere. But it should be noted that the impacts of various PCDD/F emission sources on a given atmospheric environment are strongly affected by their involved transportation and dilution processes. To further evaluate the influence of various PCDD/F emission sources directly posed on the given environment, it raises a question regarding how to identify emission sources that truly have main contributions to the given environment. The whole study was conducted in Kaohsiung city, a highly industrialized city located in southern Taiwan. The city has 1.5 million inhabitants, 0.4 million cars, and 1 million motorcycles. The main PCDD/F stationary emission sources of the city include MSWIs, MWIs, sinter plants, EAFs, secondary ALSs, and others. In this study, the characteristics of PCDD/F emission from various stationary emission sources were determined by directly collecting samples from their stack flue gases. To characterize PCDD/F emissions from mobile sources, data reported in U.S. EPA’s database for unleaded gas-fueled vehicles (UGFV) and diesel-fueled vehicles (DFV) (5) were chosen owing to lack of local data. The characteristics of PCDD/F concentrations in various atmospheric environments of the studied city were investigated by directly conducting PCDD/F samplings in the field. In this study, only the contents of 2,3,7,8-congeners were analyzed because of their toxicities and the compliance to law regulation in all countries. Finally, the significance of various emission sources to PCDD/F concentration of any given atmospheric environment was first assessed through the use of PCA and cluster analysis (7-10) and further validated by using the indicatory technique (11).

Experimental Section PCDD/F Sampling. For stationary PCDD/F emission sources, stack flue gas samples were collected from five main emission sources, including MSWIs (m = 13, n = 130), MWIs (m = 5, n = 15), EAFs (m = 6, n = 18), secondary ALSs (m = 13, n = 39), and sinter plants (m = 4, n = 63) according to U.S. EPA modified Method 23. Here, “m” denotes the number of the studied facilities, while “n” denotes the number of collected samples. The sampling train adopted in this study is comparable with that specified by U.S. EPA Modified Method 5. Prior to sampling, XAD-2 resin was spiked with PCDD/F surrogate standards prelabeled with isotopes. VOL. 38, NO. 19, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4937

FIGURE 1. Congener profiles of seventeen 2,3,7,8-PCDD/Fs of several PCDD/F emission sources. In this study, four categories of atmospheric environments were investigated, including background, residential area (R1-R2), traffic area (T1-T6), and industrial area (I1-I7). The background sampling site, the Kengting National Park, is situated at the southern end of Taiwan. Because this sampling site is far away from all possible pollution sources (50 km for stationary sources and 10 km for mobile sources) hence its PCDD/F concentration could be regarded as the 4938

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 19, 2004

background level. All categories were sampled once per month for four continuous months. Each ambient air sample was collected using a PS-1 sampler (Graseby Andersen, GA) according to the revised EPA Reference Method T09A. The sampling flow rate was specified at ∼0.225 m3 min-1. Each sample was collected continuously on three consecutive days (sampling volume = ∼972 m3). The PS-1 sampler was equipped with a quartz-fiber filter for sampling particle-phase

TABLE 1. Mean PCDD/F Concentrations Found in Ambient Air of the Background, Residential Area, Traffic Area, and Industrial Area background (m ) 1; n ) 4)

residential area (m ) 2; n ) 8)

traffic area (m ) 6; n ) 24)

industrial area (m ) 7; n ) 28)

PCDD/Fs

mean, pg Nm-3

RSD, %

mean, pg Nm-3

RSD, %

mean, pg Nm-3

RSD, %

mean, pg Nm-3

RSD, %

PCDDs PCDFs PCDFs/PCDDs ratio total PCDD/Fs total I-TEQ (pg I-TEQ Nm-3)

0.033 0.042 1.29 0.076 0.0063

19.8 30.9 32.8 21.5 40.6

0.50 0.85 1.7 1.35 0.088

39.5 41.2 9.9 40.2 32.5

0.49 0.71 1.51 1.2 0.073

57.5 43.0 22.2 43.8 38.0

0.75 1.4 2.00 2.1 0.15

47.5 47.4 49.6 40.0 43.6

TABLE 2. Indicatory PCDD/Fs of Several PCDD/F Emission Sources PCDD/F emission sources MSWIs MWIs EAFs (carbon steel) EAFs (steel) secondary ALSs 1

FIGURE 2. Congener profiles of seventeen 2,3,7,8-PCDD/Fs in background atmospheric environment.

secondary ALSs 2 secondary ALSs 3

PCDD/Fs and followed by a glass cartridge for sampling gasphase PCDD/Fs, respectively. A known amount of surrogate standard was spiked to the glass cartridge in the laboratory prior to the field sampling was conducted. Details are similar to that given in our previous work (12). In this study, we directly adopt data reported in U.S. EPA’s database to characterize PCDD/F emissions from mobile sources for both UGFV and DFV mainly owing to a lack of local data. But according to recent research conducted on characterizing the emission of PCDD/Fs from diesel engines by using the U.S. heavy-duty federal test procedure (transient cycle test), our preliminary results (unpublished) suggest that the most abundant congeners contained in the exhaust of diesel engines were similar to that presented in U.S. EPA’s database. Although no local data are available for UGFV, we find that the PCDD/F congener profile presented in U.S. EPA’s database for UGFV is consistent with that reported by Hagenmaier et al. (13). Based on these, we assume that the congener profiles of UGFV and DFV presented in U.S. EPA’s database would be representative to local emissions. Analyses of PCDD/Fs. Analyses of stack flue gas and ambient air samples followed the U.S. EPA modified method 23 and EPA Reference Method T09A, respectively. All chemical analyses were carried out by the Super Micro Mass Research and Technology Center in Cheng Shiu University;the accredited laboratory in Taiwan for PCDD/F analyses. The sample analyses were performed according to the standard procedures. Two high-resolution gas chomatographs/high-resolution mass spectrometers (HRGC/HRMS) were used for PCDD/Fs analyses (one for analyzing stack flue gas samples and the other for ambient air samples). The HRGC (Hewlett-Packard 6970 Series gas, CA) was equipped with a DB-5 MS fused

secondary ALSs 4 sinter plants 1 sinter plants 2 UGFV DFV

indicatory PCDD/Fs (the highest three ratio values) OCDD (12.6) 1,2,3,4,6,7,8-HpCDD (8.1) OCDF (6.4) 1,2,3,7,8,9-HxCDF (11.4) 1,2,3,4,7,8,9-HpCDF (7.8) 1,2,3,7,8-PeCDF (7.2) 1,2,3,7,8-PeCDF (29.3) 1,2,3,7,8,9-HxCDF (21.7) 2,3,7,8-TeCDF (19.6) 1,2,3,7,8,9-HxCDF (28.0) 1,2,3,7,8-PeCDF (20.8) 2,3,4,7,8-PeCDF (14.2) 2,3,7,8-TeCDF (18.4) 1,2,3,7,8-PeCDF (13.5) 2,3,4,7,8-PeCDF (8.9) 1,2,3,7,8-PeCDF (13.1) 2,3,4,7,8-PeCDF (11.7) OCDF (9.8) 1,2,3,7,8-PeCDF (16.6) 2,3,4,7,8-PeCDF (13.1) 2,3,7,8-TeCDF (9.3) 1,2,3,4,6,7,8-HpCDD (7.7) 1,2,3,7,8,9-HxCDD (6.5) 1,2,3,6,7,8-HxCDD (6.2) 1,2,3,7,8-PeCDF (22.9) 2,3,4,7,8-PeCDF (15.0) 1,2,3,7,8,9-HxCDF (12.6) 1,2,3,7,8,9-HxCDF (19.3) 1,2,3,7,8-PeCDF (15.2) 2,3,4,7,8-PeCDF (11.0) OCDD (18.6) OCDF (11.8) 2,3,7,8-TeCDD (5.7) OCDD (23.4) 1,2,3,4,6,7,8-HpCDD (5.9) 1,2,3,7,8,9-HxCDD (5.4)

silica capillary column (L = 60 m, ID = 0.25 mm, film thickness = 0.25 µm) (J&W Scientific, CA) and with a splitless injection. Helium was used as the carrier gas. The HRMS (Micromass Autospec Ultima, Manchester, UK) was equipped with a positive electron impact (EI+) source. The analyzer mode of the selected ion monitoring (SIM) was used with resolving power at 10 000. The electron energy and source temperature were specified at 35 eV and 250 °C, respectively. Details of analytical procedures and QA/QC results are given in our previous work (12).

Results and Discussion PCDD/F Congener Profiles of Emission Sources and Atmospheric Environments. The congener profiles of the 2,3,7,8-substituted PCDD/Fs were selected as the signatures of these PCDD/F emission sources. Each selected congener VOL. 38, NO. 19, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4939

FIGURE 3. Congener profiles of seventeen 2,3,7,8-PCDD/Fs in ambient air of residential area.

FIGURE 4. Congener profiles of seventeen 2,3,7,8-PCDD/Fs in ambient air of traffic area. was normalized by reference to the total weight of all 2,3,7,8congeners. Figure 1 shows the congener profiles of the seventeen 2,3,7,8-chlorinated substituted PCDD/Fs (mean ( SD) detected from the stack flue gases for the investigated stationary emission sources and UGFV and DFV as reported in U.S. EPA’s database for mobile emission sources (5). For the investigated stationary emissions sources, the congener profiles of EAFs were further classified into two types based on their final products (i.e., carbon steel for EAFs 4940

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 19, 2004

1 and stainless steel for EAFs 2). We found that EAFs 1 were dominated by 2,3,7,8-TeCDF, 1,2,3,7,8-PeCDF, and 2,3,4,7,8PeCDF and this is consistent with the results obtained from ref 14, while EAFs 2 were dominated by 2,3,7,8-TeCDF, 1,2,3,7,8-PeCDF, 2,3,4,7,8-PeCDF, and 1,2,3,4,6,7,8-HpCDF. Sinter plants were classified into two types based on their air pollution control devices (i.e., with selective catalytic reduction (SCR) for sinter plants 1 and without SCR for sinter plants 2). We found that sinter plants 1 were dominated by 2,3,7,8TeCDF, 1,2,3,7,8-PeCDF, 2,3,4,7,8-PeCDF, and 1,2,3,4,6,7,8-

FIGURE 5. Congener profiles of seventeen 2,3,7,8-PCDD/Fs in ambient air of industrial area. HpCDF, and this is consistent with the results obtained from ref 14. Results for sinter plants 2 were consistent with that obtained from ref 15 with the most dominant congeners being 2,3,4,7,8-PeCDF and 1,2,3,4,6,7,8-HpCDF. As for secondary ALSs, four different congener profiles are obtained from 13 secondary ALSs. We found that secondary ALSs 1 were dominated by 2,3,7,8-TeCDF and 1,2,3,4,6,7,8-HpCDF, and this is consistent with the results obtained from ref 16. Results for secondary ALSs 4 were consistent with that obtained from ref 17 with the most dominant congeners being 1,2,3,4,6,7,8HpCDD, OCDD, and 1,2,3,4,6,7,8-HpCDF. Table 1 shows the mean PCDD/F concentrations of the background, residential area, traffic area, and industrial area were 0.076, 1.4, 1.2, and 2.1 pg Nm-3, respectively; and the corresponding I-TEQ concentrations were 0.0063, 0.088, 0.073, and 0.15 pg I-TEQ Nm-3, respectively. The low

concentration of the background suggests the feasibility of using it for characterizing unaffected ambient air. The mean I-TEQ concentration of the industrial area was 23.5-, 1.7-, and 2.0-fold higher than that of the background, residential area, and traffic area, respectively. It reveals that PCDD/F emission from various sources did play different roles in atmospheric concentrations at different sites in Kaohsiung city. Figures 2-5 show the congener profiles of PCDD/Fs of the background, residential area, traffic area, and industrial area, respectively. All four categories show that the most abundant congeners in common were 1,2,3,4,6,7,8-HpCDD, OCDD, 1,2,3,4,6,7,8-HpCDF, and OCDF, which were consistent with those found in other studies (18-21). Principal Component Analysis (PCA) and Cluster Analysis. To identify the possible pollution sources for all selected ambient environments, their PCDD/F congener profiles VOL. 38, NO. 19, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4941

FIGURE 6. Score plot from PCA. together with that obtained from each facility of the five investigated stationary emission sources and two mobile emission sources were analyzed by PCA by using the mass fractions of 2,3,7,8-congeners (congener profile) as the variables. Figure 6 presents the score plot from PCA, which factor 1 explains 43.7% of the total variance, while factor 2 explains 19.3% of the total variance and both account for 63.0% of the total variance. The score plot reveals that the data points were clustered into five groups, that is, Group 1 (MSWIs, UGFV, DFV), Group 2 (background ambient air, MWIs and two facilities of the secondary ALSs 4), Group 3 (metallurgical facilities, including sinter plants, EAFs and secondary ALSs), Group 4 (ambient air of traffic area, except for T1 and T2), and Group 5 (ambient air of industrial area, except for I2 and I3). The data points with similar congener profiles were closely located, while those which had divergent patterns were located further apart, and the data points were classified according to the position of their corresponding coordinates with respect to the factor axis (i.e., the congener profiles of MSWIs, UGFV, and DFV were similar to each other and were different from those of 4942

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 19, 2004

metallurgical facilities). The data points of the selected ambient environments are classified in the same way. The data points of the traffic area and industrial area, which originally clustered to those of background, moved toward the data points of Group 1 (MSWIs, UGFV, DFV), Group 2 (MWIs), and Group 3 (metallurgical facilities), respectively (i.e., the score plot reveals that the traffic area was influenced by MSWIs, UGFV, and DFV, while those of the industrial area were influenced by metallurgical facilities and MWIs). Because PCDD/F emissions from MSWIs had been proven no influence on the atmosphere according to the relationship found between PCDD/F concentration isopleths and wind rose (22). For practical consideration, we ignored the influence of MSWIs and proposed the influence on traffic area might mainly result from UGFV and DFV. The result of cluster analysis (Figure 7) was similar to that of PCA. The data points of ambient air, even for that of background, clustered first with those of MWIs and then with other emission sources. But it seems to be impossible for background ambient air to have any relation with MWIs. Therefore, we extrapolated that after undergoing several

FIGURE 7. Dendrogram from cluster analysis.

TABLE 3. Indicatory PCDD/Fs of Ambient Air Sampling Sites and Its Suspected PCDD/F Emission Sources sampling sites

indicatory PCDD/Fs (the highest three ratio values)

suspected PCDD/F emission sources

R1 R2 T1 T2 T3 T4 T5 T6 I1 I2 I3 I4 I5 I6 I7

2,3,7,8-TeCDF (1.5), 1,2,3,4,7,8,9-HpCDF (1.4), 1,2,3,4,6,7,8-HpCDD (1.4) 2,3,7,8-TeCDF (1.6), 1,2,3,4,7,8,9-HpCDF (1.4), 1,2,3,7,8-PeCDD (1.4) 2,3,7,8-TeCDF (1.7), 1,2,3,4,6,7,8-HpCDD (1.4), OCDD (1.4) 2,3,7,8-TeCDF (1.7), OCDD (1.4), 1,2,3,4,6,7,8-HpCDD (1.4) OCDD (1.7), 1,2,3,4,6,7,8-HpCDD (1.4), 1,2,3,7,8,9-HxCDF (1.2) OCDD (1.7), 2,3,7,8-TeCDF (1.5), OCDF (1.3) 2,3,7,8-TeCDF (1.7), OCDD (1.5), OCDF (1.3) OCDD (1.9), 1,2,3,4,7,8-HxCDD (1.3), 1,2,3,4,6,7,8-HpCDD (1.3) 2,3,7,8-TeCDF (1.6), 1,2,3,6,7,8-HxCDD (1.5), 1,2,3,7,8-PeCDD (1.5) OCDD (1.7), 1,2,3,4,6,7,8-HpCDD (1.5), 2,3,7,8-TeCDF (1.4) 2,3,7,8-TeCDF (1.7), OCDD (1.6), 1,2,3,4,6,7,8-HpCDD (1.6) 2,3,7,8-TeCDF (2.2), 1,2,3,4,6,7,8-HpCDD (1.5), 2,3,4,7,8-PeCDF (1.4) 2,3,7,8-TeCDF (2.3), 1,2,3,4,7,8,9-HpCDF (1.6), 2,3,4,7,8-PeCDF (1.5) 2,3,7,8-TeCDF (1.5), 1,2,3,4,7,8-HxCDD (1.5), 1,2,3,4,6,7,8-HpCDD (1.4) 2,3,7,8-TeCDF (2.3), 1,2,3,4,7,8,9-HpCDF (1.7), 2,3,4,7,8-PeCDF (1.6)

EAFs, ALSs, MWIs EAFs, ALSs, MWIs EAFs, ALSs, (DFV or MSWIs) EAFs, ALSs, (DFV or MSWIs) (DFV or MSWIs) UGFV, EAFs, ALSs EAFs, ALSs, UGFV (DFV or MSWIs) EAFs, ALSs (DFV or MSWIs), EAFs, ALSs EAFs, ALSs, (DFV or MSWIs) EAFs, ALSs, sinter plants EAFs, ALSs, MWIs, sinter plants EAFs, ALSs EAFs, ALSs, MWIs, sinter plants

degradation/loss processes, such as reactions with hydroxyl radicals, photolysis, and wet/dry deposition. We found that PCDD/Fs in the atmosphere own their “inherent” characteristics with the most abundant congeners 1,2,3,4,6,7,8HpCDD, OCDD, 1,2,3,4,6,7,8-HpCDF, and OCDF. Surprisingly, the above “inherent” characteristics were quite similar to the congener profile of MWIs. In principle, the use of PCA and cluster analysis for identification of PCDD/F sources is simply based on the similarity in the congener profile of emission sources and receptors. But it should be used with caution to avoid misjudgment because of the inherent similarities in characteristics of emission sources and receptors (for example, the congener profile of MWIs and background atmosphere in this study). Because of this, the method of the indicatory PCDD/Fs analysis was adopted in this study for further validation purposes according to the relative contributions of various emission sources to different receptors. Indicatory PCDD/Fs. For indicatory PCDD/Fs analysis, the following formula was used to define the indicatory PCDD/Fs (11):

ratioji )

∑X) (X /∑X) (Xi/ i

j

min

Here, the numerator represents the mass fraction of the ith congener of the emission source j, and the denominator represents the minimum value of the mass fraction of the ith congener among all emission sources. A higher value of ratioji means that the ith congener of emission source j is the one with more contributions than that of other emission sources. The PCDD/F emission sources in Figure 1 were selected to determine indicatory PCDD/Fs of PCDD/F emission sources. The highest three ratio values of PCDD/Fs for each source were recognized as the indicatory PCDD/Fs and were listed in Table 2. Among them, we found that the indicatory PCDD/Fs of sinter plants, EAFs and secondary ALSs were quite similar. Their indicatory PCDD/Fs include 2,3,7,8TeCDF, 1,2,3,7,8-PeCDF, 2,3,4,7,8-PeCDF, and 1,2,3,7,8,9HxCDF. On the same basis, we combined the atmosphere samples to determine indicatory PCDD/Fs for each selected ambient environment. The above results were compared with indicatory PCDD/Fs of emission sources (Table 2), and we were able to determine the suspected pollution sources for all investigated ambient environments as listed in Table 3. In principle, the above results are very similar to that obtained from PCA with the exception of the influence of MWIs which become not so obvious. It shows that the most suspected PCDD/F emission sources for the ambient atmosphere were metallurgical facilities, including EAFs, secondary ALSs, and sinter plants. In some dense traffic areas, we found that their VOL. 38, NO. 19, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4943

PCDD/Fs were also contributed by traffic sources. The above results were theoretically plausible because engine emissions were very close to the investigated ambient environments.

Literature Cited (1) Olie, K.; Vermeulen, P. L.; Hutzinger, O. Chemosphere 1977, 6, 455. (2) Caserini, S.; Monguzzi, A. M. Chemosphere 2002, 48, 779. (3) Japan Ministry of the Environment. The Environmental Monitoring Report on the Persistent Organic Pollutants (POPs) in Japan. 2002, 40. (4) Quass, U.; Fermann, M. W.; Bro¨ker, G. Chemosphere 2000, 40, 1125. (5) U.S. EPA. Database of Sources of Environmental Releases of Dioxin like Compounds in the United States, EPA/600/C-01/ 012, 2001. (6) Alcock, R. E.; Gemmill, R.; Jones, K. C. Chemosphere 1999, 38, 759. (7) Masunaga, S.; Yao, Y.; Ogura, I.; Sakurai, T.; Nakanishi, J. Chemosphere 2003, 53, 315. (8) Go¨tz, R.; Steiner, B.; Friesel, P.; Roch, K.; Walkow, F.; Maass, V.; Reincke, H.; Stachel, B. Chemosphere 1998, 37, 1987. (9) Koch, M.; Knoth, W.; Rotard, W. Chemosphere 2001, 43, 737. (10) Helm, P.; Bidleman, T. Environ. Sci. Technol. 2003, 37, 1075. (11) Yang, H. H.; Lee, W. J.; Chen, S. J.; Lai, S. O. J. Hazard. Mater. 1998, 60, 159. (12) Wang, L. C.; Lee, W. J.; Lee, W. S.; Chang-Chien, G. P.; Tsai, P. J. Environ. Sci. Technol. 2003, 37, 62.

4944

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 19, 2004

(13) Hagenmaier, H.; Dawidowsky, V.; Weber, U. B.; Hutzinger, O.; Schwind, K. H.; Thoma, H.; Essers, U.; Buhler, B.; Greiner, R. Organohalogen Compd. 1990, 2, 329. (14) Hofstadler, K.; Friedacher, A.; Gebert, W.; Lanzerstorfer, C. Organohalogen Compd. 2000, 46, 66. (15) Anderson, D. R.; Fisher, R. Chemosphere 2002, 46, 371. (16) Buekens, A.;d0 Cornelis, E.; Huang, H.; Dewettinck, T. Chemosphere 2000, 40, 1021. (17) Cleverly, D.; Schaum, J.; Schweer, G.; Becker, J.; Winters, D. Organohalogen Compd. 1997, 32, 430. (18) Coleman, P. J.; Lee, R. G. M.; Alcock, R. E.; Jones, K. C. Environ. Sci. Technol. 1997, 31, 2120-2124. (19) Lee, R. G. M.; Green, N. J. L.; Lohmann, R.; Jones, K. C. Environ. Sci. Technol. 1999, 33, 2864. (20) Sin, D. W. M.; Choi, J. Y. Y.; Louie, P. K. K. Chemosphere 2002, 47, 647. (21) Abad, E.; Caixach, J.; Rivera, J. Chemosphere 1997, 35, 453. (22) Wang, L. C. Characteristics of Polychlorinated Dibenzo-p-dioxins and Dibenzofurans from Emission Sources and in the Atmosphere. Dissertation for Doctor of Philosophy, Department of Environmental Engineering, National Cheng Kung University, Taiwan. 2003, In Chinese.

Received for review January 1, 2004. Revised manuscript received April 14, 2004. Accepted April 22, 2004. ES0499795