Vol.:(0123456789) 1 3 Archives of Environmental Contamination and Toxicology https://doi.org/10.1007/s00244-019-00640-x Source Apportionment of Polychlorinated Biphenyls in Atmospheric Deposition in the Seattle, WA, USA Area Measured with Method 1668 Lisa A. Rodenburg 1  · Iris Winstanley 2  · Jennifer M. Wallin 2 Received: 24 October 2018 / Accepted: 21 May 2019 © Springer Science+Business Media, LLC, part of Springer Nature 2019 Abstract Atmospheric deposition can be an important pathway for the delivery of toxic polychlorinated biphenyls (PCBs) to ecosys- tems, especially in remote areas. Determining the sources of atmospheric PCBs can be difcult, because PCBs may travel long distances to reach the monitoring location, allowing for a variety of weathering processes that may alter PCB fnger- prints. Previous eforts to determine the sources of atmospheric PCBs have been hampered by the electron capture detection methods used to measure PCBs. In this work, EPA method 1668, which is capable of measuring all 209 congeners, was used to measure PCBs in bulk atmospheric deposition at seven locations in the Green-Duwamish River watershed in and near Seattle, WA. Analysis of this data set via Positive Matrix Factorization allowed the identifcation of six factors that represent PCB sources. Four factors, representing approximately 88% of all PCB mass, are strikingly similar to unweathered Aroclors, suggesting minimal weathering during transport and/or local PCB sources at some sites. A ffth factor contained virtually all of the PCB 11 mass and represents PCBs from pigments. It explained approximately 39% of the Toxic Equivalency Quotient in the atmospheric deposition samples. The remaining factor contained non-Aroclor PCBs and may be related to silicone. Atmospheric deposition can be an important pathway for the delivery of persistent organic pollutants (POPs), including polychlorinated biphenyls (PCBs), to ecosystems, especially in remote areas (Choi et al. 2008; Hung et al. 2005, 2010; Li et al. 2012; Simcik et al. 1997; Totten et al. 2004, 2006). Determining the sources of atmospheric PCBs can be dif- fcult, because PCBs may travel long distances to reach the monitoring location (Choi et al. 2008; Hung et al. 2005, 2010), allowing for a variety of weathering processes, such as volatilization, condensation, and reaction with hydroxyl radicals (Anderson and Hites 1996; Totten et al. 2001), which may alter PCB fngerprints during transport. Source apportionment of the PCB signal is only possi- ble when many PCB congeners are measured and is most successful when the methods to measure PCBs are the most sensitive and accurate available. The most sensitive method in the United States is the EPA method 1668 (U.S. EPA 1999; U.S. EPA 2010), which uses high-resolution mass spectrometry and is capable of measuring all 209 PCB con- geners and of resolving the dioxin-like congeners, but this method has not been widely used in atmospheric deposition studies. Two of the main atmospheric deposition monitoring networks in the United States—the Integrated Atmospheric Deposition Network (IADN) in the Great Lakes (Basu et al. 2009; Rodenburg and Meng 2013; Sun et al. 2007) and the Delaware Atmospheric Deposition Network (DADN) in New Jersey, Pennsylvania, and Delaware (Praipipat et al. 2017)— have primarily used electron capture detection (ECD) meth- ods to measured approximately 90 PCB congeners that are typically found in the Aroclor PCB formulations produced in the United States by the Monsanto Company. Several studies have attempted to apportion atmospheric PCB sources using data from these two networks (Du et al. 2009; Praipipat et al. 2017; Rodenburg and Meng 2013). These studies have generally focused on the gas phase and have observed PCB fngerprints that resemble low molecular weight PCB for- mulations, such as Aroclor 1242, as well as fngerprints that resemble weathered Aroclors. Notably, these source appor- tionment studies have suggested that PCB concentrations in Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00244-019-00640-x) contains supplementary material, which is available to authorized users. * Lisa A. Rodenburg rodenburg@envsci.rutgers.edu 1 Department of Environmental Sciences, Rutgers University, 14 College Farm Road, New Brunswick, NJ 08901, USA 2 Leidos, Environmental Planning & Restoration Portfolio, 18912 North Creek Parkway, Suite 101, Bothell, WA 98011, USA