Abstract
The adsorption process, known for its cost-effectiveness and high efficiency, has been extensively investigated at the laboratory scale for removing per- and polyfluoroalkyl substances (PFAS) from non-conventional irrigation water. However, a syringe filtration step is commonly used when quantifying PFAS removal during this adsorption process, potentially leading to PFAS retention onto the filters and an overestimate of adsorption removal efficiency. Here, we assessed the retention of three prevalent PFAS (i.e., perfluorooctanoic acid [PFOA], perfluorooctane sulfonic acid [PFOS], and perfluorobutanoic acid [PFBA]) on six syringe filters. When filtering distilled deionized water spiked with 1 µg/L and 100 µg/L of each PFAS, we observed the highest and lowest PFAS recovery percentages by mixed cellulose ester (MCE) (0.20 µm, 25 mm; 97 ± 11%, 101 ± 4.8%) and polytetrafluoroethylene (0.45 µm, 13 mm; 61 ± 37%, 80 ± 28%), respectively. Under the initial concentration of 1 µg/L and 100 µg/L, PFOS had recovery percentages of 55 ± 25% and 68 ± 24%, significantly lower than 96 ± 12% and 99 ± 5% for PFOA and 95 ± 8% and 97 ± 4% for PFBA, highlighting the importance of PFAS functional groups. PFAS recovery percentage increased with filtration volume in the order of 80 ± 28% (1 mL) < 85 ± 21% (5 mL) < 90 ± 18% (10 mL). Using MCE to filter treated municipal wastewater spiked with 1 µg/L and 100 µg/L of each PFAS, we found recovery percentages >90% for all three PFAS. Our study underscores the significance of syringe filter selection and potential overestimate of PFAS removal efficacy by the lab-scale adsorption processes.