Water Research Journal: Occurrence of PFAS Substances in U.S. WRRFs

Reposted with permission
Emerging Issues

View CWEA’s PFAS topics page >

Research by Charles E. Schaefer, Jennifer L. Hooper, Laurel E. Strom, Ibrahim Abusallout, Eric R.V. Dickenson, Kyle A. Thompson, Gayathri Ram Mohan, Dina Drennan, Ke Wu, and Jennifer L. Guelfo (please see article for affiliations)


Both quantifiable and semi-quantifiable poly- and perfluoroalkyl substances (PFAS) were evaluated in the influent, effluent, and biosolids of 38 wastewater treatment plants. PFAS were detected in all streams at all facilities. For the means of the sums of detected, quantifiable PFAS concentrations were 98 ± 28 ng/L, 80 ± 24 ng/L, and 160,000 ± 46,000 ng/kg (dry weight basis) in the influent, effluent, and biosolids (respectively). In the aqueous influent and effluent streams this quantifiable PFAS mass was typically associated with perfluoroalkyl acids (PFAAs). In contrast, quantifiable PFAS in the biosolids were primarily polyfluoroalkyl substances that potentially serve as precursors to the more recalcitrant PFAAs. Results of the total oxidizable precursor (TOP) assay on select influent and effluent samples showed that semi-quantified (or, unidentified) precursors accounted for a substantial portion (21 to 88%) of the fluorine mass compared to that associated with quantified PFAS, and that this fluorine precursor mass was not appreciably transformed to perfluoroalkyl acids within the WWTPs, as influent and effluent precursor concentrations via the TOP assay were statistically identical. Evaluation of semi-quantified PFAS, consistent with results of the TOP assay, showed the presence of several classes of precursors in the influent, effluent, and biosolids; perfluorophosphonic acids (PFPAs) and fluorotelomer phosphate diesters (di-PAPs) occurred in 100 and 92% of biosolid samples, respectively. Analysis of mass flows showed that, for both quantified (on a fluorine mass basis) and semi-quantified PFAS, the majority of PFAS exited WWTPs through the aqueous effluent compared to the biosolids stream. Overall, these results highlight the importance of semi-quantified PFAS precursors in WWTPs, and the need to further understand the impacts of their ultimate fate in the environment.


Poly- and perfluoroalkyl substances (PFAS) are known to occur in groundwater, surface water, wastewater, soil, and air (Gawor et al., 2014; Munoz et al., 2017; Roth et al., 2020; Sepulvado et al., 2011; Wei et al., 2018; Xiao et al., 2012). PFAS occur widely in the environment due to their use in numerous commercial and industrial applications, including fire-fighting foams, metal finishing, textiles, food packaging materials, and consumer products (Buck et al., 2011; Kempisty et al., 2019). The widespread use of these products has resulted in their occurrence in wastewater treatment plants (WWTPs) from commercial and industrial discharges, and potentially from domestic use of consumer products (Kurwadkar et al., 2022; Lenka et al., 2021; Letcher et al., 2020). Although transformation from parent PFAS to intermediate and terminal PFAS products can occur, conventional wastewater treatment generally does not remove or destroy PFAS, resulting in their release through the effluent or biosolids (Houtz et al., 2018, 2016). For example, a meta-analysis concluded perfluorooctane sulfonic acid (PFOS) did not significantly change between influent and effluent, while perfluorooctanoic acid (PFOA) tended to be 6 ng/L higher in effluent than influent (Thompson et al., 2022).

Several perfluoroalkyl acid (PFAA) precursors, including perfluoroalkane sulfonamides (FASAs), fluorotelomer alcohols (FTOHs), and fluorotelomer sulfonates (FTSs), have been reported to transform to PFAAs in wastewater (Benskin et al., 2012; Wang et al., 2011; Zhang et al., 2013). The enhanced biological process (e.g., aeration, activated sludge process) that occur in WWTPs are likely to facilitate and enhance such reactions. The recurring detection of PFAA precursors in wastewater might explain the observed increase of PFAAs in WWTP effluents relative to influents. For example, Xiao et al. (2012) observed increases in PFHxA in effluents relative to influents in 60% of Minnesota’s 37 WWTPs, while the concentration of quantifiable PFAA precursors decreased from the influent to the effluent. PFAA precursor transformation in WWTPs is dependent on treatment processes used and process temperature. Higher PFAA formation rates have been observed in activated sludge systems operating at longer retention times and higher temperatures (Guerra et al., 2014). Furthermore, Houtz et al. (2018) observed new classes of potential PFAA precursors, including fluorotelomer thioether surfactants (FTSHCs) and fluorotelomer thioether alkylamido hydroxyl carboxylates (FTSAHCs), in WWTP influents but not in the effluent. This suggests transformation during treatment to PFAAs or strong sorption to sludge by precursors. These results indicate that WWTP processes may cause an increase in the loading of PFAAs into the environment. Future monitoring of precursors will likely become increasingly important since the annual production of precursors has surpassed the production of other PFAS (Buck et al., 2011).

A growing number of studies have highlighted the important role of “unknown” PFAS which can only currently be identified using high resolution mass spectrometry (HRMS) coupled with semi-quantitative concentration estimates due to a lack of analytical standards for these compounds (Miaz et al., 2020; Nickerson et al., 2020; Wang et al., 2018). Semi-quantified PFAS are often polyfluorinated and PFAA precursors. In a recent occurrence survey in China, 30% of the total PFAS detected in WWTP influents and effluents were semi-quantified PFAS. These semi-quantified PFAS were primarily represented by hydro substituted perfluoro carboxylic acids (H-PFCAs), hydro substituted perfluoroalkyl ether sulfonates (H-PFESAs) and FASAs (Wang et al., 2020). Several studies also identified another class of semi-quantifiable PFAS, polyfluoroalkyl phosphoric acid diesters (di-PAPs), that were observed in wastewater sludge and biosolids (D’eon et al., 2009; Lee et al., 2014; Schaefer et al., 2022). Di-PAPs were also reported to transform into PFAAs, suggesting that di-PAPs may transform through WWTP processes and facilitate increased PFAA discharges (Lewis et al., 2016). A recent study focusing on the influent of 16 Belgian WWTPs identified 18 semi-quantified PFAS (Jeong et al., 2022).

In addition to HRMS-based techniques, another analytical tool to quantify the extent of “unknown” PFAA precursors is the total oxidizable precursor (TOP) assay. A recent study applying the TOP assay showed that unknown precursors accounted for 2- to 6-times the identified PFAS mass present following primary treatment in a survey of 6 WWTPs (Tavasoli et al., 2021). While these studies provide some critical initial insights into the role of semi-quantifiable PFAS, a considerable data gap still exists regarding the occurrence and fate of these semi-quantifiable compounds in WWTPs.

Gallen et al. (2018) conducted a PFAS mass balance at 14 WWTPs in Australia. Mass loading from the individual effluents reached up to 2.4 lbs per year each for PFOA and PFOS, and the annual load from biosolids were 0.81 and 1.4 lbs for PFOA and PFOS, respectively. In another PFAS mass balance study performed at a single WWTP by Schultz et al. (2006), 2.5 times more PFOS mass per day exited the plant via the finished effluent than in the anaerobic sludge. While these studies were useful in providing some initial insights into mass flows, only a limited number of PFAS and WWTPs were examined, leaving a substantial data gap with respect to the PFAS-related fluorine mass balance.

While the occurrence and persistence of PFAAs in WWTPs have been well demonstrated, there remains only limited information on the nature and fate of the wide range of PFAA precursors in WWTPs, particularly for semi-quantified PFAS. In addition, PFAS phase distribution, and the subsequent impacts on WWTP aqueous and solid mass flows, remains poorly understood. Typical PFAS have changed since the time of previous studies due to long-term trends and changes in industrial practice (Thompson et al., 2022). Regulatory actions relating to PFAS and WWTPs currently in consideration by the agencies such as the USEPA should thus consider up-to-date scientific data (US EPA, 2021). The purpose of this study was to examine the occurrence, along with the overall transformation and mass flows, in aqueous and. solid streams of PFAS (including quantified and non-quantified precursors), at 38 WWTPs across 23 states. While larger numbers of WWTPs have been sampled for PFAS within certain states (California State Water Resources Control Board, 2022; EGLE, 2020; “Public Hearing, 2021), those datasets may not be representative nationally. To our knowledge, this study is the largest nationwide WWTP PFAS sampling campaign to date. While this effort focuses on the mass flows, semi-quantitative PFAS/precursors, and fluorine balances, a more detailed statistical evaluation of this dataset relating PFAS occurrence to WWTP characteristics and processes will be the focus of a subsequent study.

Section snippets

Participating utilities

There were 38 participating facilities in this study located in 23 states within the US. The facilities represented a diverse range of sizes with average flows from 0.019 × 106 to 2.5 × 106 m3 d  1. Collection systems represented both small communities with primarily residential and light commercial sources to metropolitan areas with heavy industrial dischargers. Table S1a summarizes the flow and wastewater type for each facility. The primary potential sources of PFAS to these facilities (via

Occurrence of quantifiable pfas

Quantifiable PFAS were detected in the influent and effluent (Fig. 1), as well as the finished biosolids, at all 38 facilities. The average sum of quantifiable PFAS detected across all facilities was 98 ± 28 ng/L for the influent and 80 ± 24 ng/L for the effluent, indicating PFAS concentrations typically resided within a relatively narrow range for these facilities, despite differences in their geographic locations, sizes, dischargers, and treatment schemes. Total PFAS concentrations were


Results of this study, consistent with prior studies on the occurrence of PFAS in WWTPs, demonstrated the widespread presence of PFAS both entering and exiting WWTPs. Herein, it is shown that precursors accounted for 66% and 51% of the PFAS-based fluorine entering and leaving WWTPs in the aqueous flows (respectively, on average), and that the majority of precursors (based on a comparison of TOP results for the influent and effluent) were not transformed to PFAAs during the treatment processes.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.


CDM Smith gratefully acknowledges that The Water Research Foundation are funders of certain technical information upon which this manuscript is based under project 5031. CDM Smith thanks The Water Research Foundation for their financial, technical, and administrative assistance in funding the project through which this information was discovered. This material does not necessarily reflect the views and policies of the funders and any mention of trade names or commercial products does not…

View and download the complete research article on the science direct website.

Reprinted from Water Research Volume 233  April 15 2023, 119724, Occurrence of quantifiable and semi-quantifiable poly- and perfluoroalkyl substances in united states wastewater treatment plants, with permission from Elsevier.