Atmospheric monitoring of persistent organic pollutants using passive air samplers: Source-receptor relationship and long-range transport

Abstract

Polychlorinated biphenyls (PCBs), polychlorinated naphthalenes (PCNs), polybrominated diphenyl ethers (PBDEs), and organochlorine pesticides (OCPs) have all been used for commercial purposes. Because of their physicochemical properties, they are persistent in the environment and undergo long-range atmospheric transport (LRAT) via repeated evaporation and deposition. They can thus be found distant from their sources, and therefore it is difficult to identify the locations and contributions of specific sources of these compounds. In order to investigate local sources, fate, and movement of these contaminants, this study monitored these compounds in various environments using passive air samplers. Firstly, coplanar polychlorinated biphenyls (PCBs), polychlorinated naphthalenes (PCNs), and polybrominated diphenyl ethers (PBDEs) were sampled using polyurethane foam (PUF) disk passive air samplers (PAS) at 19 sites in a heavily industrialized region of South Korea for six months (January 2006 - July 2006). The levels and spatial distribution of these three chemical groups were investigated to identify potential sources and transport in the study area, which can be divided into five regions: a steel manufacturing complex, a residential area near the steel complex, a rural area, a semi-industrial area, and a petrochemical manufacturing complex. Air concentrations (pg&#8226

m-3) were estimated using an average sampling rate of 3.0 m3&#8226

day-1 and ranged as follows: coplanar PCBs (0.8 - 16 pg&#8226

m-3), PCNs (1.7 - 35 pg&#8226

m-3), and PBDEs (3.8 - 24 pg&#8226

m-3). The levels of coplanar PCBs and PBDEs were found to be the highest in the steel complex, followed by the petrochemical complex and the semi-industrial area. In addition, a high level of PCNs was measured near a petrochemical processing plant. However, the residential area near the steel complex and the rural area showed relatively low concentrations of these chemicals, suggesting that the steel and petrochemical industries are probably important sources in the study area, but these potential sources do not strongly influence the surrounding areas.Secondly, four consecutive passive air samplings (September 2006 - July 2007) were conducted at 15 sites around an iron and steel making plant in Pohang, Korea to investigate the spatial and seasonal variations of polychlorinated biphenyls (PCBs) and ultimately the source-receptor relationships. Annual mean values of Σ8PCBs (IUPAC number 8, 28, 52, 101, 118, 138, 153, 180) were in the range 15.1 - 166 pg&#8226

m-3 with an average of 53.0 pg&#8226

m-3. The spatial distribution of PCBs for each sampling period clearly suggests that the steel complex is a major source of PCBs in this area, and the prevailing winds facilitated the atmospheric transport and dispersion of PCBs from the steel complex to the surrounding areas. Seasonal patterns of PCBs were observed clearly, which were influenced by meteorological conditions

the highest levels of PCBs were observed with the highest average air temperature, and the influence of rainfall (i.e. wet scavenging) was also observed. In addition, PCB 11, a non-Aroclor congener, was detected in high concentrations at all sites, implying that the sources of PCB 11 are both unique and ubiquitous. This study confirms that passive air sampling is a useful tool to obtain seasonal and spatial distributions of time-averaged POPs data at a local scale. Thirdly, PAS was deployed for one year at 8 cities in Mongolia, China, and South Korea to investigate country-specific sources and long-rage transport of polychlorinated biphenyls (PCBs), polychlorinated naphthalenes (PCNs) and organochlorine pesticides (OCPs). There was strong correlation between PCB level and population of each city (r2=0.859, p<0.05). Especially in the case of South Korea, PCB levels have highly correlated with populations (r2=0.986, p<0.05) and population density (r2=0.984, p<0.05) of each city, respectively. Unexpectedly, Mongolian cities had significantly high concentrations of PCBs considering population. Distribution of PCNs also showed strong relationship with population of each city (r2=0.773, p<0.05). This means human activities are responsible for contamination of PCNs and PCBs. Even though almost OCPs for agriculture had been banned in the world, they were still detected at high concentrations and the region- and country-specific source of each OCP was observed. The levels of OCPs were dependent on the past use of those chemicals. Many OCPs, HCHs, DDTs, and mirex, were detected at the highest level at sites of China. Interestingly, significantly high endosulfan was measured in all sites of South Korea. Mirex was detected at all sites, which was not registered in South Korea and Mongolia. This suggests the influence by LRAT.Finally, XAD-2 resin based passive air samplers (PAS) were deployed for three one-year periods at the Korean polar and South Pacific research stations at Ny-&Aring

lesund (2005-2009), King George Island (2005-2007), and Chuuk (2006-2009) to investigate long-range transport, local sources, and temporal trends of organochlorine pesticides (OCPs) and polychlorinated biphenyls (PCBs). The highest hexachlocyclohexane (HCH) concentration (35.2 - 78.9 pg&#8226

m-3) over the entire sampling period was detected at Ny-&Aring

lesund, in the Arctic. α-HCH was the dominant OCP (31.1 - 67.1 pg&#8226

m-3), contributing about 50% of the total OCP load. Additionally, a high and consistent α/γ-HCH ratio was observed at Ny-&Aring

lesund. HCHs might reach Arctic sites more easily than other OCPs from surrounding countries through long-range atmospheric transport (LRAT). Interestingly, high levels of the current-use OCP endosulfan - particularly endosulfan-I -&#8211

were detected at almost all sampling sites, including in Antarctica, ranging 12.2 - 88.5, 17.7 - 130, and ND - 59.7 pg&#8226

m-3 at King George Island, Ny-&Aring

lesund, and Chuuk, respectively. Specific OCP and PCB patterns, such as low trans/cis-chlordane ratios and a prevalence of lighter PCB congeners, were observed in all three regions (excepting one site at Ny-&Aring

lesund and one site in the South Pacific affected by local sources) during all sampling periods. This indicates that these Polar and remote South Pacific sites are mainly influenced by LRAT. Over the entire sampling period, a decreasing trend of HCHs (α- and γ-HCH) and an increasing trend of endosulfan-I were observed at the Ny-&Aring

lesund sites.