Processes Contributing to the Retention and Transport of Per and Polyfluoroalkyl Substances (PFAS) in Soils and Groundwater

Abstract

The widespread occurrence of per and polyfluoroalkyl substances (PFAS) in the environment and their potential to cause adverse human health impacts has resulted in the emergence of PFAS as major contaminants of concern and the focus of various research activities. High PFAS concentrations in soils and groundwater have been observed at many sites, particularly at fire-training sites that have used aqueous film-forming foams. Thus, understanding the conditions that impact PFAS retention and transport through subsurface environments is crucial for conducting accurate risk assessments and planning effective remedial actions. Most previous research has focused on solid-phase sorption as a retention mechanism. The majority of solid-phase sorption coefficients are quantified with batch experiments. However, few studies have used column experiments to quantify solid-phase sorption coefficients, and even fewer have compared the two methods. As sorption coefficients are often employed in model parameterization used in decision making, it is necessary to determine the representativeness of sorption coefficients determined from different experiment methods specifically for PFAS. In this study, sorption coefficients and isotherms obtained from batch and two column experiment approaches are compared. The results indicate that batch and column experiment determinations of sorption coefficients and isotherms give consistent results, despite rate-limited and nonlinear sorption influences of the PFAS studied. However, appropriate experimental and data analysis methods must be employed. For example, batch solids-solution ratios and mixing conditions and column data resolution and analysis methods may impact experimental results. Other processes may also impact PFAS transport when other fluid phases are present. Many sites have observed or suspected co-contamination of non-aqueous phase liquids (NAPLs). In this study the relative contributions of partitioning to the bulk NAPL phase and NAPL-water interfacial adsorption on PFAS retention are delineated through interfacial-tension measurements, batch partitioning experiments, and column experiments. The results indicate that NAPL-water interfacial adsorption significantly contributed to PFAS retention in column experiments (~70%), but that partitioning to NAPL had minimal contribution. Additionally, retention increased in the presence of greater residual NAPL saturations. In this study the first quantitative-structure/property-relationship (QSPR) focused on NAPL-water interfacial adsorption coefficients is provided, giving consistent results with previous studies despite different fluid-fluid systems and solution characteristics. This study also provides the first initial investigation to determine whether PFAS, surfactants, may impact NAPL distribution. The results indicate that high concentrations of PFAS may induce NAPL dissolution, but not mobilization under the conditions studied herein. However, for systems with PFAS mixtures and course porous media, mobilization may be more likely to occur. Further research is needed to determine to what extend PFAS may impact NAPL behavior. The presence of air may also impact PFAS retention, particularly through adsorption to the air-water interface. In this study, PFAS surface-tension data sets are provided for additional compounds and electrolyte solutions that have not been previously reported. The surface activity and interfacial adsorption coefficient (Ki) values for the PFAS measured follow trends observed in previous studies, such as increased carbon chain length corresponding to increased surface activity. Solution characteristics were also observed to impact surface activity, including synthetic groundwater (SGW) solutions corresponding to greater Ki values compared to NaCl solutions, due to the different electrolyte composition of SGW versus NaCl. The presence of background PFAS in mixture systems also increased surface activity compared to single-PFAS systems, with the most surface-active compound generally controlling the overall surface activity of the system. The magnitude of ionic strength for non-zero ionic strength systems was determined to have minimal impact on interfacial adsorption coefficients. However, interfacial adsorption coefficients exhibited concentration dependence. Additionally, QSPR analyses for PFAS air-water and combined air-water and NAPL-water interfacial adsorption coefficients are provided. The QSPR analyses provided consistent results with previous QSPR analyses, with molar volume providing a good descriptor of surface activity for log Ki values over a range of nearly four log units. Overall, the results reported here support previous findings that PFAS are highly surface active and are impacted by molecular and solution characteristics. The results of this study indicate the presence of solid, water, air, and NAPL phases can each significantly impact PFAS transport through subsurface environments, particularly through solid-phase sorption and air-water and NAPL-water interfacial adsorption. Thus, each phase should be carefully considered at sites with PFAS contamination. Further research is needed to determine how complex, heterogeneous, mixed-contaminant systems will impact PFAS transport.