Environ. Sci. Technol. 2005, 39, 7380-7388
Congener-Specific Survey for Polychlorinated Biphenlys in Sediments of Industrialized Bays in Korea: Regional Characteristics and Pollution Sources SANG HEE HONG,* UN HYUK YIM, WON JOON SHIM, AND JAE RYOUNG OH South Sea Institute,Korea Ocean Research & Development Institute, 391 Jangmok-ri, Jangmok-myon, Geoje-shi, Gyungsangnamdo, 656-834, Republic of Korea
Areas of the Korean coastline with heavy industry and major harbors were investigated for polychlorinated biphenyl (PCB) pollution. This investigation paid attention to variations in the PCB congener patterns for a possible source of contamination. Surface sediments from 49 sites were sampled. Although the occurrence of PCBs in coastal marine environment correlates well with shipping and industrial activities, the contribution from shipping activities is considerable because of its enormous economical importance in Korea. The highest concentrations were found in harbors with heavy ship traffic and ship construction. Principal component analysis (PCA) of congener-specific composition of PCBs revealed distinct regional patterns, especially in a harbor and steel manufacturing area. PCB signatures with enhanced higher chlorinated congeners were typical for harbors with shipping activities and correlated well with commercial formulations that were formerly used in ship painting. Lower chlorinated congeners with up to five chlorines were significantly abundant in steel works zones which differed from harbor zones. This distinction was consistent with the congener patterns in the ambient air and the effluent of the steel works as well as in the nearby surface sediments. This study identified steel manufacturing as a recent and ongoing emission source of PCBs in Korea’s coastal zone.
Introduction Polychlorinated biphenyls, produced commercially since 1929, are now ubiquitous pollutants in the environment. By virtue of their chemical and physical stability, they were used in a wide variety of applications (e.g., dielectric fluids in capacitors and transformers, print inks, paints, and pesticides), resulting in global environmental contamination. An estimated 1.3 million tons were produced globally from 1930 to 1993 (1). Although the production of PCBs has been regulated and banned worldwide since the 1970s, they are still in use in closed systems; hence, ongoing loss of PCBs to the environment is a major international issue. It is therefore essential to evaluate the regional status of PCB contamination. This survey has been a part of constructing a nationwide database for persistent organic pollution in the Republic of Korea. * Corresponding author phone: +82-55-639-8674; fax: +82-55639-8689; e-mail:
[email protected]. 7380
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 19, 2005
In the process of Korea’s rapid industrialization and urbanization since the 1970s, a large amount of PCB has been used. Although an inventory of PCB usage is unfortunately not available, an estimated 560 tons of PCBs were used in Korea from 1975 to 1984 (2). The coastal region is important for shipping and other sectors of industrial growth in Korea. Consequently, developmental pressure on the coastline is always increasing. A nationwide monitoring study of organochlorine compounds using marine bivalves showed elevated levels of PCBs at stations in industrialized regions (3). Other regional studies also found PCBs at industrialized and urbanized coastal zones (4-7). However, the earlier studies were limited in extent, with only a few sampling stations in selected areas. For example, Busan Bay, Gwangyang Bay, and Youngil Bay, which are highly industrialized, were not included in earlier studies. In this investigation, we have undertaken a congener-specific approach to PCB contamination in five major bays, relating contamination levels and their characteristic regional signatures to respective industrial and harbor activities. Point source identification of ubiquitous pollutants such as PCBs is complex. However, a thorough statistical analysis of the data matrix, including principal component analysis, reveals a hidden relationship between regional PCB signatures and possible point sources. Such an approach will be very helpful in regulating PCBs in Korea.
Materials and Methods Study Areas and Sampling Strategy. Surface sediments were collected from five major industrialized bays (Kyeonggi Bay, Gwangyang Bay, Busan Bay, Youngil Bay, and Ulsan Bay) from February to March 2000 (Figure 1). We used a Van Veen grab deployed from a research vessel to take 49 surface sediment samples. Approximately the top 2 cm of sediments was taken and placed in a precleaned amber glass jar using a stainless steel spoon. The collected samples were immediately frozen and stored at -20 °C until analysis. Kyeonggi Bay (Figure 1a) is located on the western coast of the Korean peninsula. Industrial complexes such as Namdong, Shihwa, and Banweol are situated along the coast. The second largest harbor in Korea is located in the northern part of the bay. Contaminated waters are periodically discharged during low tide in this bay from a wastewater storage reservoir (K3) and artificial Lake Shihwa (K13), both of which receive a heavy pollution load from lakeside industrial complexes. We collected 13 sediment samples in the Incheon Harbor area, near industrial complexes, and offshore. Gwangyang Bay, a semi-enclosed bay on the mid-southern coast, was rapidly industrialized during the past three decades. A huge steel industry occupies a point of land projecting into the middle of the bay and several industrial complexes are located on the north and south shores of the bay (Figure 1b). To extend industrial facilities and port facilities, land is being reclaimed on the western and central parts of the bay. Sediment samples were taken from six sites that were away from heavy dredging. Busan Bay is a representative shipping area. Busan Harbor is located inside the bay, which is the foremost port in Korea as well as the world’s third largest container port (Figure 1c). Since 1876, Busan Harbor has been the marine gateway to Korea. It processes 40% of Korea’s marine export cargos and 81% of the country’s container cargos as well as 42% of the domestically produced marine products. Most of the facilities are inside a breakwater that transects the bay. Midscale shipyards are located on the west shore of the harbor. 10.1021/es050397c CCC: $30.25
2005 American Chemical Society Published on Web 09/01/2005
FIGURE 1. Sampling locations of surface sediments from (a) Kyeonggi Bay, (b) Gwangyang Bay, (c) Busan Bay, (d) Ulsan Bay, and (e) Youngil Bay. Sediment samples were taken from seven stations inside the harbor zone (B1, B2, B3, B4, B5, B6, and B9) and three outside with increasing distance from the harbor. Ulsan Bay is a mixed zone with both industrial and shipping activities. Sediment samples were taken from inner to outer part of the bay (Figure 1d). An industrial port, Ulsan Harbor, and a small fishing harbor are located in front of stations U2 and U4, respectively. A shipyard and shiprepairing docks are at the east side of the bay. Most importantly, numerous industrial facilities (e.g., Youseung Chemical, Petrochemical, and Onsan Industrial Complexes) are sited all around the bay. Station U8 is on the lower part of a stream crossing an industrial zone. The inner part of Youngil Bay has a steel manufacturing facility, the largest in Korea, that was founded in 1968 and produces about 28 million tons of steel products per year. Five sampling stations (Y1-Y5) were selected inside this steel works region (Figure 1e). Ambient air and effluent samples were collected for pollution source identification. We used a high-volume air sampler equipped with glass-fiber filter (GF/F) to collect the particulate compounds and polyurethane foam (PUF) to collect the gas-phase compounds. The samplers were operated for 12 h and the total volume sampled was 450 m3. We also collected 20 L effluent samples directly from the outlets. Chemical Analysis. The sediment samples were prepared for PCB analysis in accordance with U.S. Ocean and Atmospheric Administration (NOAA) standard methods (8) but with some minor modifications (7). A 20-g sediment sample was Soxhlet extracted for 16 h with 200 mL of dichloromethane. Subsamples (∼2 g) were taken to measure water content. The extracts were concentrated to a volume of 2-3 mL using a water bath. Activated copper granules were used to remove elemental sulfur, which interferes with PCB determination. Sulfur-removed extracts were cleaned up by 20 g of 5% deactivated silica gel and 10 g of 1% deactivated alumina in a multilayer column. Further cleanup for concentrated sediment extracts was carried out using
high-performance liquid chromatography (HPLC; 250 × 22.5 mm i.d. size-exclusion column packing with Phenogel 100 Å, Phenomenex Co.). The HPLC system consisted of a solvent delivery system (M930, Young-Lin Instrument Co. Ltd.), a UV detector (DFW-20, D-Start Instrument), a fraction collector (Model 201, Gilson), a column heater (CH460, Eppendorf), and an injector (Model 7725i, Rheodyne) fitted with a 2-mL sample loop. Dichloromethane was used as the mobile phase with a flow rate of 7 mL/min. A mixture of dibromooctafluorobiphenyl and perlyene was used as a calibration standard for PCB elution. When the retention time of the test mixture was stabilized, the real samples were loaded into the HPLC. The final concentrated extracts in hexane were analyzed using a Hewlett-Packard 6890 GC with µ-ECD. A fused silica capillary column (DB-5, J & W, 30 m × 0.25 mm i.d. × 0.25 µm thickness) was used. Helium and argon: methane (95:5) were used as a carrier and a makeup gas, respectively. The temperature program was 100 °C for 1 min, 100 °C to 140 °C at 5 °C/min, 140 °C for 1 min, 140 °C to 250 °C at 1.5 °C/min, 250 °C for 1 min, 250 °C to 300 °C at 10 °C/min, and 300 °C for 5 min. The injector and detector temperatures were 275 °C and 300 °C, respectively. Twentytwo individual PCB congeners (IUPAC numbers 8, 18, 28, 29, 44, 52, 66, 87, 101, 105, 110, 118, 128, 138, 153, 170, 180, 187, 195, 200, 206, and 209) purchased from Ultra Scientific Co. were used as quantitative mixture. In this study, the sum of 22 PCB congeners is presented as total PCBs. All concentration data are based on dry weight. To observe in detail the different congener patterns of PCBs found in samples from Busan Bay and Youngil Bay, along with the 22 congeners noted above, 89 additional PCB peaks were quantified using Aroclor mixtures (1016:1242:1254:1260 ) 1:1:1:1). A surrogate mixture containing dibromooctafluorobiphenyl (DBOFB), PCB103, and PCB169 was spiked to the sample prior to extraction to measure recovery. As a GCinternal standard, tetrachloro-m-xylene (TCMX) was spiked to each sample prior to GC analysis. Air samples were treated in the same way as sediment samples except that PUF sample VOL. 39, NO. 19, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
7381
FIGURE 2. Concentrations of total PCBs (ng/g dry wt) in surface sediments collected from five bays: (a) Kyeonggi Bay, (b) Gwangyang Bay, (c) Busan Bay, (d) Ulsan Bay, and (e) Youngil Bay. was extracted using high-volume Soxhlet apparatus. Effluent samples were liquid-liquid extracted using dichloromethane, and the rest of the steps are similar to sediment samples. The mean recoveries (n ) 49) of the added surrogates were 71 ( 12%, 84 ( 9%, and 96 ( 21% for DBOFB, PCB103, and PCB169, respectively. A series of laboratory blanks containing 30 g of anhydrous sodium sulfate was prepared similar to the sediment samples to assess background levels of PCBs associated with the chemical analysis. Duplicate samples exhibited
