PFAS Data Guidelines and Implications to Fate through Thermal Processes

Introduction Per- and polyfluoroalkyl substances (PFAS) have dominated public attention recently given their environmental contamination. PFAS resist environmental degradation and can accumulate in the urban water cycle, including wastewater systems (Winchell et al., 2021). Furthermore, various PFAS will migrate to wastewater solids (termed biosolids when stabilized) where conventional treatment technologies have little to no effect on the PFAS content of the solids (Loos et al., 2018, Gallen et al., 2018, Kim Lazcano et al., 2019). PFAS in biosolids have the potential to derail beneficial reuse via land application due to the potential impacts to groundwater, and transfer into agricultural crops (Navarro et al., 2017, Ghisi et al., 2019). Thermal treatment is the only process currently identified to reliably destroy PFAS (USEPA, 2020), but destruction efficiency must be confirmed to avoid worsening atmospheric pollution and deposition (Shafer et al., 2020). Several research initiatives have commenced to document the fate of PFAS through thermal processes, including the Water Research Foundation (WRF) Project 5111 focused on sewage sludge incinerators (SSIs). The treatment of the data when reviewing these fate studies has profound impacts on the defensible conclusions offered by researchers. This paper will present the procedures implemented for the quantifiable data from the WRF project soon to be published (Winchell et al., in review). Data Usage Guidelines Data collected as part of any research project must undergo a strict quality assurance and control program to support defensible conclusions. Data from the WRF study was subjected to a four-step assessment process (Table 1). Data were verified, validated, and qualified prior to the final assessment on usability. The research team then reviewed the quality control information and deemed reported values usable or not. This final step is critical in advancing scientific knowledge on sound information. Molar Basis Moles matter. Too often researchers sum multiple PFAS compounds on a mass basis. While convenient this can lead inappropriate assessments of process performance. For example, consider a scenario where a mixture of PFBA (MW = 214.04 g/mol) and PFHxDA (MW = 814.13 g/mol), two analytes found in the WRF study, are reported. Let's assume 100 grams of each compound entered a process, while 1 and 90 grams exit, irrespective of compound, the summed mass removal would be 55%. On a molar basis, if PFBA were the 1 gram emission the summed removal increases to 80%. Or, if PFBA were the 90 gram emission the summed removal molar removal decreases to 29%. Each researcher must consider whether the arbitrary combined mass of PFAS is critical compared to the number of molecules present. Destruction and Removal Efficiency (DRE) Researchers and regulators often characterize process pollution control in terms of DRE. Summed DRE characterizations suffer from several complications in addition to the molar discussion previously. First, current analytical capabilities can only quantify roughly 0.62% of suspected PFAS compounds (Eurofins, 2022; USEPA, 2021). Second, confident DRE assessments require reportable values of the targeted pollutant in each input and output stream. Samples absent reportable values, possible values ranging from zero to the limit of quantification, can significantly impact the significant figures in the DRE calculation. A difference between 99% and 99.9% DRE can determine regulatory compliance. Last, some input or output streams can misleadingly impact a DRE assessment as discussed in the following section. Given these complications the WRF study did not attempt to report a DRE. Instead, the average molar load of targeted PFAS were compared in the dewatered sewage sludge and stack (air emissions). From the multiple hearth furnace (MHF), only 12 µmols escape the stack while none left fluidized bed furnace (FBF), resulting in over 95% removal of PFAS, see Figure 1. Consequential Impacts of Inconsequential Streams Some input and output streams in an SSI can have impractical impacts on DRE assessments if not evaluated with scrutiny. In the WRF study, the reportable levels of PFAS in the wet scrubber supply water and drain influence DRE estimates and bias the environmental relevance of the SSI performance. Perhaps the best illustration of this paradox comes from the observation of PFHxS in the wet scrubber supply and drain, but in no other samples. The calculated DREs for each run range from 4—25%. Because the values in the wet scrubber supply slightly exceed that measured in the drain, though not statistically significant (2-tailed, α = 0.05, paired), one might attribute the entire SSI with a low DRE for this compound if applying a narrow interpretation. However, given a wet scrubber is not intended to treat PFAS contained in the supply water, the resulting DREs for PFHxS mischaracterize the performance of the thermal process. Conclusions PFAS and the data collected thereon is complicated and misuse can lead to misleading interpretations. Establishing the data base upon which confident conclusions can be supported requires application of the discussed usage guidelines. Even with a defensible dataset researchers must be wary of misleading interpretations. If properly applied, the approach discussed herein will lead to sound scientific advancement to support industry initiatives.