Per- and polyfluoroalkyl substances (PFAS) are anthropogenic surfactants that have recently been identified as persistent organic pollutants. These so called “Forever Chemicals” have been detected in drinking waters, ground waters, soils, and consumer and industrial products globally; with environmental impacts stretching into the artic, far from known PFAS sources. The increase in awareness regarding PFAS distribution in the environment has generated interest into how PFAS interact with humans, what PFAS specific properties may be involved, and what additional environmental compartments may they be found in.
In Chapter 2 we discuss the use of molecular dynamics (MD) modeling to screen for protein – PFAS binding affinity to inform experimental measurements of binding affinity via equilibrium dialysis (Eq D). The equilibrium dissociation constants (KD) of six perfluoroalkyl carboxylates (PFCAs) and three perfluoroalkyl sulfonates (PFSAs) to liver and intestinal fatty acid binding proteins (L- and I-FABPs) and peroxisome proliferator activated nuclear receptors (PPAR-α, - δ and - γ) were determined via liquid chromatography mass spectrometry. The MD models were found to predict relative and not absolute binding for all protein – PFAS combinations. This research was the first to identify sub micromolar binding between short chain PFAS (6 or less carbons) and PPAR-α and δ, which may have implications for the assumed safety of shorter chain PFAS due to rapid clearance. Chain length dependent binding was observed for L-FABP but not observed for PPAR proteins which means that for these proteins binding affinity cannot be inferred by PFAS chain length. Additionally, a comparison was made between KDs derived from EqD and other in-vitro approaches, using these experimental results and results from literature. It was discovered that KDs derived from EqD were lower (i.e. higher binding affinity) than other in-vitro approaches which has implications for comparisons between methodologies and raises an important question regarding which KDs should be considered most relevant in-vivo.
Research discussed in Chapter 3 surrounds the development of an extraction and analytical method to quantify PFAS in environmental non-aqueous phase liquids (NAPL). As mentioned above, PFAS are used in industrial products and one common group of industrial products that have been identified as the root cause for environmental PFAS contamination at U.S. military sites are aqueous film forming foams (AFFF). AFFF are complex mixtures known to contain high concentrations of many surfactants including PFAS. At U.S. military sites it is also common to encounter NAPL in the subsurface. Co-disposal of PFAS (AFFF) with NAPL has happened historically through intentional use (e.g. firefighting) or unintentionally at waste sites. In order to quantify PFAS within NAPL, a liquid-liquid extraction method was developed that could successfully extract anionic, cationic, and zwitterionic PFAS. This research discovered the presence of PFAS in recovered NAPL at microgram per liter concentrations at non-source zone sites. Concentrations of PFAS in NAPL are likely much higher at source zone sites and could have implications for NAPL remediation.
Chapter 4 discusses the partitioning and interfacial adsorption of PFAS into NAPL at environmentally relevant concentrations (i.e. nano – microgram per liter). Given the discovery of low microgram per liter concentrations of PFAS in recovered NAPL discussed in Chapter 3, it is relevant to investigate what partitioning and sorption processes are occurring at these concentrations. Current research in this area has focused on the NAPL – water interface and has done so at high concentrations, milli – gram per liter. Here we performed batch equilibrium experiments at low concentrations (2,000 – 100,000 ng/L) between jet fuel A (NAPL) and synthetic freshwater. Single point partition coefficients (Kn) were calculated for PFAS of carbon chain length 8-14 across the concentration range. Values for Kn decreased with increasing PFAS concentration indicating non-ideal partitioning, which become more evident with increasing chain length. Partitioning into jet fuel A was not observed for PFAS below eight carbons. Interfacial sorption (Knw) was estimated by mass difference and found to be orders of magnitude higher than previously reported literature values.