Improved methods are needed for sampling and analysis of per- and polyfluoroalkyl substances (PFAS) at Department of Defense (DoD) sites to facilitate more accurate and precise assessments of the extent of PFAS contamination. While grab-sampling is an option for characterizing contaminated sites, this process is labor-intensive, cost-prohibitive, and difficult to manage effectively. Passive sampling methods have the potential to provide time-weighted average (TWA) concentrations of contaminants in aquatic systems. One such method, based on the diffusive gradients in thin-films (DGT) technique, functions independently of hydrodynamic conditions at a site, as analyte mass transport into the DGT passive sampling device (PSD) is controlled solely by molecular diffusion. The objective of this research project is to develop and validate a DGT-PSD capable of accurately quantifying (± 20%) TWA concentrations of 24 target PFAS in water.
The project consists of three hypothesis-driven and laboratory-based research tasks, crafted to develop and validate a DGT-PSD for the twenty-four target PFAS in waters of varying temperature, pH, conductivity, and dissolved organic carbon. Semi-quantitative PFAS analyses will be completed at the Arkansas Statewide Mass Spectrometry facility and aim to ensure experiments are successfully executed prior to definitive quantitative PFAS analyses at a DoD-accredited laboratory.
In Task 1, PFAS diffusion coefficients are measured as a function of temperature and water chemistry using a two-compartment diffusion cell.
In Task 2, DGT components are identified, including the housing, diffusive gel type, and membrane type, each of which must neither be a source or sink of the target PFAS. Next, the dual binding layer is developed, consisting of an outer layer for the more hydrophobic long-chain PFAS and an inner layer for lesser hydrophobic short-chain PFAS. If necessary, binding layers can be amended with additional sorbents to minimize remaining competitive sorption effects. The diffusive gel layer thickness is then determined to produce linear PFAS mass accumulation profiles over a period of several days, resulting in a prototype DGT-PSD. PFAS diffusion coefficients are then re-measured, this time in the prototype and compared to their values determined in Task 1 to validate its design, and, if necessary, guide further improvements.
In Task 3, PFAS method detection levels and concentration ranges are determined for the DGT-PSD. Competitive sorption effects will be assessed using PFAS mixtures at widely varying concentrations, from low ng×L-1 through µg×L-1 levels. The DGT binding layers will be modified as needed to minimize any remaining competitive sorption effects among the target PFAS.
PFAS diffusion coefficients will be measured as a function of water temperature and solution chemistry. Appropriate DGT components will be identified, including the DGT housing, diffusive gel layer type, and membrane filter type. Dual binding layers will be configured to minimize competitive sorption effects in the uptake of long- and short-chain PFAS and must have abundant PFAS sorption capacity and rapid uptake kinetics. The diffusive gel layer must be of appropriate thickness to facilitate linear increases in PFAS uptake over several days that can be used to calculate aqueous phase TWA-PFAS concentrations.
The final deliverable of this research project is a validated DGT-PSD that can be deployed by end-users at DoD sites to accurately quantify TWA-concentrations of short- and long-chain PFAS. Expected benefits to the DoD include reduced labor associated with site profiling, reduced analytical lab costs, and increased precision and accuracy of the extent of PFAS contamination. Expected benefits to the scientific community include an increased understanding of PFAS diffusion in aqueous systems and competitive sorption effects among long- and short-chain PFAS on engineered materials. Such fundamental knowledge could be applied broadly to advance PFAS treatment protocols at DoD sites and in drinking water filtration applications.