Comparison of polymeric samplers for accurately assessing PCBs in pore waters.

Assessing the hazard posed by sediments contaminated with hydrophobic organic compounds is difficult, because measuring the freely dissolved porewater concentrations of such low-solubility chemicals can be challenging, and estimating their sediment-water partition coefficients remains quite uncertain. We suggest that more accurate site assessments can be achieved by employing sampling devices in which polymers, with known polymer-water partition coefficients, are used to absorb the contaminants from the sediment. To demonstrate the current accuracy and limitations of this approach, we compared use of three polymers, polydimethylsiloxane, polyoxymethylene, and polyethylene, exposed to a single sediment in two modes, one in which they were exhaustively mixed (tumbled) with the sediment and the other in which they were simply inserted into a static bed (passive). Comparing porewater concentrations of specific polychlorinated biphenyl (PCB) congeners with results obtained using air bridges, we found the results for tumbled polymers agreed within 20%, and the passive sampling agreed within a factor of 2. In contrast, porewater estimates based on sediment concentrations normalized to f(OC)K(OC), the weight fraction of organic carbon times the organic-carbon normalized partition coefficient, averaged a factor of 7 too high. We also found good correlations of each polymer's uptake of the PCBs with bioaccumulation by the polychaete, Neanthes arenaceodentata. Future improvements of the passive sampling mode will require devices that equilibrate faster and/or have some means such as performance reference compounds to estimate mass transfer limitations for individual deployments.