Sources of polychlorinated biphenyl blank contamination and their impact on fingerprinting

The development of high-resolution mass spectrometry methods for the measurement of polychlorinated biphenyls (PCBs) in environmental samples has dramatically reduced detection limits, leading to problems obtaining clean blanks. When PCBs are detected in samples at concentrations similar to the blanks, blank contamination must be addressed before fingerprinting and source apportionment through positive matrix factorization (PMF) can be successful. We tested a variety of blank correction methods using data from a study of water column concentrations of PCBs in the Spokane River, where concentrations of Σ₂₀₉PCBs in whole water samples averaged 171 pg/L without blank correction and Σ₂₀₉PCBs in the blanks averaged 88 pg/L. The results suggest that subtracting blank masses from sample masses can lead to erroneous results. Instead, censoring at one times the batch-specific blank level is a better approach. The sources of PCBs in field and method blanks were investigated by examining their congener profiles via PMF. The results suggest that commercial PCB formulations (in the US, Aroclors) continue to be the primary source of PCBs in blanks forty years after PCB production and use were banned in the United States. Other sources of PCBs to blanks include PCB 11 from pigments; PCBs 44 + 47 + 65, 45 + 51, and 68 arising from polymers cured using bis(2,4-dichlorobenzoyl) peroxide; and PCBs 1, 2, 3, 4, 8, 15 and others arising from silicone products derived from phenylsiloxanes, such as silicone-based adhesives.