Polychlorinated Biphenyl Congeners in River Sediments: Distribution and Source Identification Using Multivariate Factor Analysis

Abstract

To assess the level of polychlorinated biphenyl (PCB) contamination and identify their sources, surface sediments were collected from selected locations along Nakdong River, Korea, and analyzed for 209 PCB congeners using high-resolution gas chromatography/high-resolution mass spectroscopy. PCB levels ranged from 0.124 to 79.2ng/g dry weight (coplanar PCBs 0.295 to 5720pg/g dry), which were similar to those of three other major rivers (Han, Geum, and Youngsan rivers) in Korea but slightly lower than those in neighboring countries. Regarding homologue composition, tetra-CBs were most abundant in most samples, but some samples with much higher PCBs concentrations had relatively lower proportions of tetra-CBs and higher proportions of penta- to hepta-CBs. To identify the sources of PCBs in sediment samples, principal component analysis/absolute principal component scores (PCA/APCS), positive matrix factorization (PMF), and multiple linear regression (MLR) were used with the congener composition of aroclors (1242, 1248, 1254, and 1260) and the flue gas of waste incinerators (data obtained from a previous article) as source profiles. Results showed that the three models showed similar source apportionments. Most sediment samples with lower PCB concentrations had higher proportions of incineration-derived materials, and some sediment samples with much higher PCB concentrations had higher proportions of aroclor 1260. This occurred because many industrial facilities, such as landfill leachate-treatment facilities, were gathered around sampling points with high PCB concentrations, and high-chlorinated PCBs are more stable in the elution process of landfill leachate than the incineration process. PCB concentrations estimated by APCS, PMF, and MLR were similar to the measured values with coefficients of determination ranging from 0.77 to 0.99.