Quantifying the adsorption of PFAS and hydrocarbon surfactants at the air–water interface: A systematic review and meta-analysis of surface-science measurements, molecular-modeling simulations, and environmental-application results

The adsorption of PFAS at the air–water interface is a topic of current interest that affects several aspects of PFAS transport and remediation. Numerous studies over the past several years have investigated the air–water interfacial adsorption (AWIA) of PFAS, with a subset focusing on quantifying the magnitudes of adsorption and assessing the linearity of adsorption at low concentrations. Different methods have been used in these studies to measure AWIA, and there has been to date limited evaluation of the consistency of results across the methods. Furthermore, differing outcomes have been reported regarding AWIA linearity. To address these knowledge gaps, measured AWIA data sets were compiled from the literature for both PFAS and hydrocarbon surfactants, and subjected to detailed comparative analyses. The data sets represent all three primary approaches for quantifying AWIA, specifically advanced surface-science methods, molecular-modeling simulations, and various physical-based methods, and comprise more than 70 studies in total. With the exception of four studies, measured adsorbed concentrations and K_aw values were consistent across the different methods for a given surfactant, indicating good correspondence among the methods. The aggregated data from these studies clearly demonstrate that K_aw attains a constant, maximum value at concentrations below the critical reference concentration for both PFAS and hydrocarbon surfactants. This behavior is consistent with linear AWIA at low concentrations. In addition, the results demonstrate that the Langmuir isotherm provides an excellent description of AWIA and that the Freundlich isotherm does not. The observation of such consistency across all of the different methods, which vary greatly in approach, is a compelling outcome. Factors that may affect the robustness of measurements and cause potential deviations are discussed. It is anticipated that these results will be useful for planning laboratory and field studies, conducting risk assessments, and implementing modeling-based investigations.