Removing per- and polyfluoroalkyl substances from groundwaters using activated carbon and ion exchange resin packed columns

Abstract

Human exposure to per- and polyfluoroalkyl substances (PFAS) in drinking water is of growing concern as a result of increasing reports of occurrence and potential regulation. Adsorption of PFAS by granular activated carbon (GAC) or ion exchange (IX) resin is a suitable treatment technique. However, few studies compare PFAS removal in continuous flow GAC or IX adsorption systems using real drinking water sources. In this study, rapid small-scale column tests (RSSCTs) were used to investigate the effects of PFAS type and chain length on adsorption by GAC and IX resin for six groundwaters used as drinking water supplies. Seven PFAS substances with chain lengths of C4–C9 were detected in the groundwaters with the sum of their concentrations (ΣPFAS) ranging from 156 to 7,044 ng/L. Adsorption capacities (qΣPFAS) were calculated to compare the removal capacity among different sorbents and qΣPFAS values ranged from 10.3 to 228 ng PFAS/mg sorbent after about 100,000 bed volumes treated. Coal-based GACs had higher adsorption capacity compared with coconutshell-based GAC, which was likely due to higher mesopore and macropore volumes. IX resins performed better than GAC in removing PFAS, but they were not effective in treating short-chain perfluorocarboxylic acids (PFCAs). Perfluorosulfonic acids (PFSAs) broke through later than PFCAs with the same chain length. Within PFSA or PFCA classes, shorter-chain PFAS species always broke through before longer-chain PFAS. Statistical analysis demonstrated that PFAS with higher hydrophobicity and molecular weight are more amenable to GAC adsorption. Empirical models were developed and predicted PFAS breakthrough. By quantifying PFAS selectivity and removal efficiency, this work provides benchmark data for commercially available treatment technologies and guidance toward specific PFAS classes for which new treatment technologies may be most beneficial.