Bioavailability of Polycyclic Aromatic Hydrocarbons in the Aquatic Environment.

Abstract

Polycyclic aromatic hydrocarbons (PAHs) are ubiquitous in the environment and have been shown to elicit toxicity in humans and other organisms. Therefore, it is important to monitor environmental concentrations of PAHs. Toxicologically, we are concerned not only with the total PAH concentration but, with that fraction available to partition into an organism (bioavailable fraction). This research fits within three areas concerning bioavailability of PAHs including; 1) development of methods to measure bioavailability in the field, 2) identification and characterization of mechanisms controlling bioavailability and, 3) development of models to predict bioavailability in the natural environment.

In the first phase of this research, the role of black carbon (BC) in the bioavailability of PAHs in soil and sediment was examined by measuring sorption in systems containing BC, natural organic matter (NOM), and microorganisms. A model was developed to predict the bioavailable fraction of PAHs and factors that may alter sorption in the natural environment from that predicted by laboratory models were examined. In the second phase of this research, a novel passive sampling device was developed to monitor truly dissolved PAH concentrations in water.

Sorption isotherms of pyrene-d10 were measured for diesel soot (DS), Suwannee River NOM, Leonardite humic acids (HA), DS previously exposed to NOM and HA, in binary systems containing both DS and NOM, and to DS exposed to lake water. When DS was previously exposed to NOM, competition for sorption sites was observed. However, when both pyrene-d10 and NOM were introduced to DS simultaneously, less competition occurred and sorption was predicted within 92% of observed values using additive sorption models (based on the unit-normalized Freundlich model and Polyani-Dubinin-Manes model). Weathering of DS significantly reduced adsorption capacity but many strong sorption sites still remained, possibly due to renewal of sorption sites by microorganisms. This research demonstrated that sorption in the natural environment may be altered from that predicted by laboratory models due to 1) competition of linear organic carbon for sorption sites on DS, and 2) the presence of microorganisms. This research has important implications for predicting bioavailability and ecotoxicological risk of organic contaminants in soils and sediments.

In the second phase of this research, I evaluated a novel passive sampling device, the polydimethyl(siloxane) (PDMS) integrative sampler, by 1) measuring the uptake and release kinetics of 48 PAHs from water into PDMS 2) comparing methods of loading performance reference compounds and 3) verifying the uptake kinetics by comparing PAH concentrations predicted by samplers to measured concentrations. Polycyclic aromatic hydrocarbons with a log K[subscript ow] 4.90 remained in the linear uptake phase for the duration of the exposure. Standard deployments of two weeks could be used for time-integrative monitoring of these compounds. Compounds with a log K[subscript ow] 4.38 reach equilibrium rapidly (T₅₀ 9 d) and the linear uptake model could not be used to predict analyte concentrations. Decreasing the surface area to volume ratio of the sampler would easily solve this issue. Surface area normalized sampling rates of PDMS samplers and semi-permeable membrane devices (SPMDs), the most commonly used passive sampling device, were similar, indicating that PDMS samplers are comparable to the SPMD. Concentrations of PAHs in the PDMS samplers predicted concentrations in water within a factor of two and on average within 30% of the actual concentration. Poly(dimethylsiloxane) samplers offer great potential for monitoring PAH exposure in the aquatic environment.