Polyethylene Passive Samplers for Measuring Hydrophobic Organic Chemical Concentrations in Sediment Porewaters

In order to determine the hazards posed by hydrophobic organic compounds (HOCs) in sediment beds, the following areas of research were explored: (1) the use of polyethylene (PE) sheets as passive sampling devices in sediments, using performance reference compounds (PRCs) in order to reduce deployment times to the order of days, (2) the use of limited PRC data to calibrate mass-transfer models for the exchange of a suite of compounds between polymer strips and sedimentary porous media, and (3) the use of in-situ passive sampling methods to deduce chemical activities of HOCs in sediments and the tissues of soft shelled clams (Mya arenaria) , in order to measure bioaccumulation potential. First, the use of PE passive samplers, in-situ, to measure freely dissolved HOC concentrations, is demonstrated. PRCs, impregnated into the PE before use, allow porewater concentrations to be deduced after exposure times much shorter than would be required for sampler equilibration (days instead of months). Next, the method is expanded for measuring suites of compounds of the same class. A one-dimensional diffusion model of chemical exchange between a polymer sheet of finite thickness and an unmixed sediment bed is employed. Porewater concentrations for seventeen polycyclic aromatic hydrocarbons (PAHs) are measured using samplers deployed for 3 to 10 days in homogenized sediment from a coal-tar contaminated site. Finally, the samplers are used to determine the potential for HOCs to bioaccumulate in Mya arenaria in sediments from six sites near Boston, MA. PE-deduced porewater PAH concentrations are compared to lipid-normalized tissue PAH concentrations in samples taken from twelve stations distributed throughout the sites. Additionally, tissue concentrations are compared to bulk sediment concentrations and porewater concentrations deduced from equilibrium partitioning models that include sorption to both organic carbon and black carbon fractions. Results show correlations only between PE-deduced porewater concentrations and tissue concentrations, illustrating the usefulness of PE passive samplers for gauging risk to benthic organisms associated with HOC contaminated sediments. Also, porewater concentrations estimated using equilibrium partitioning methods overestimated porewater concentrations by one to three orders of magnitude at all but one site. Thesis Supervisor: Philip M. Gschwend Title: Professor of Civil and Environmental Engineering