Method Selection for PFAS-removal : An Investigation of Pilot Studies at Water Treatment Plants

New threshold values for per- and polyfluorinated alkyl substances (PFAS) pose challenges for Swedish water treatment plants. To select a PFAS removal method, pilot tests are often conducted, but these require time and resources. In this study, results from pilot trials at six water treatment plants in Sweden have been analyzed and compared. The aim was to investigate the suitability of different methods, the impact of water chemistry on PFAS removal, and to what extent future pilot studies are needed. The methods evaluated included filtration using granular activated carbon (GAC), ion exchange resin (IEX), and nanofiltration (NF). To identify the most suitable technique, material costs were compared across the different plants and treatment methods. The least squares method was used to fit a tanh function to the PFAS removal curves for GAC and IEX, which facilitated analysis of how water quality affects PFAS removal. The impact of water quality was also investigated using a physical model. The results indicate that NF is the most effective method for PFAS removal. However, management of the produced reject water remains a practical challenge and was not investigated in this study. GAC and IEX also performed well and were found to be comparable with each other in terms of effectiveness. The need for future pilot studies was discussed. The curve fittings could not be used to extrapolate results from incomplete pilot studies, but they were used to qualitatively examine the effects of water chemistry on PFAS-removal. A clear relationship was identified between higher levels of dissolved organic carbon (DOC) and reduced PFAS removal with earlier breakthrough when using GAC. No corresponding relationship could be established for ion exchange. For NF, the results generally aligned well with previous studies where such data were available, suggesting that future pilot tests for NF might be replaced by existing data. In the absence of prior studies, a modeling tool was used to estimate removal efficiency, but this proved to be less reliable. There is a need for increased coordination between water utilities to make future pilot studies with GAC and ion exchange more comparable. This study contributes to a broader understanding, but additional analysis of more consistent and comparable pilot tests could yield deeper insights and reduce the need for new studies in the future.