Historical practices at firefighter-training areas involved repeated aqueous film-forming foams (AFFFs) applications, resulting in source zones characterized by high concentrations of perfluoroalkyl and polyfluoroalkyl substances (PFAS). Repeated applications of AFFF composed of 14 anionic and 23 zwitterionic perfluoroalkyl substances (PFAS) were conducted on a single one-dimensional saturated soil column to quantify PFAS retention. An electrofluorination-based (3M) Milspec AFFF, which was above the mixture's critical micelle concentration (CMC), was at application strength (3%, v/v). Retention and retardation of PFAS mass increased with each successive AFFF addition, although the PFAS concentration profiles for subsequent applications differed from the initial. Greater degree of mass retention and retardation correlated with longer PFAS carbon-fluorine chain length and charged-headgroup type and as a function of AFFF application number. Anionic PFAS were increasingly retained with each subsequent AFFF application, while zwitterionic PFAS exhibited an alternating pattern of sorption and desorption. Surfactant-surfactant adsorption and competition during repeat AFFF applications that are at concentrations above the CMC resulted in adsorbed PFAS from the first application, changing the nature of the soil surface with preferential sorption of anionic PFAS and release of zwitterionic PFAS due to competitive elution. Applying a polyparameter quantitative structure-property relationship developed to describe sorption of AFFF-derived PFAS to uncontaminated, saturated soil was attempted for our experimental conditions. The model had been derived for data where AFFF is below the apparent CMC and our experimental conditions that included the presence of mixed micelles (aggregates consisting of different kinds of surfactants that exhibit characteristics properties different from micelles composed of a single surfactant) resulted in overall PFAS mass retained by an average of 27.3% ± 2.7% (standard error) above the predicted values. The correlation was significantly improved by adding a "micelle parameter" to account for cases where the applied AFFF was above the apparent CMC. Our results highlight the importance of interactions between the AFFF components that can only be investigated by employing complex PFAS mixtures at concentrations present in actual AFFF at application strength, which are above their apparent CMC. In firefighter-training areas (AFFF source zones), competitive desorption of PFAS may result in downgradient PFAS retention when desorbed PFAS become resorbed to uncontaminated soil.
Keywords: 1D flow experiments; AFFF; PFAS; contaminant transport; surfactant−surfactant interactions.