Zwitterionic, cationic, and anionic perfluoroalkyl and polyfluoroalkyl substances integrated into total oxidizable precursor assay of contaminated groundwater

Highlights

• PFAS were analyzed in contaminated groundwater samples (Σ₄₁PFAS up to 5 mg/L).

• AFFF-derived zwitterionic and cationic precursors were amenable to TOP conversion.

• The 6:2 FTAB was converted to C3-C5 PFCAs as major TOP products (minor: C6-C7).

• The expanded list of PFAA-precursors allowed to better explain ∆PFCA postoxidation.

• FHxSA, 6:2 FTAB and 6:2 FtS were major precursors at an active fire-training area.

Abstract

The total oxidizable precursor (TOP) assay can be useful for integrating precursors to perfluoroalkyl acids (pre-PFAAs) into the assessment of sites contaminated by per- and polyfluoroalkyl substances (PFAS). Current research gaps include risks of instrumental matrix effects due to the complexity of post-oxidation extracts, potential reproducibility issues during TOP itself, and limited information for zwitterionic and cationic pre-PFAAs. We first investigated a suitable method for the analysis of groundwater samples, using liquid chromatography high-resolution mass spectrometry (UHPLC-HRMS). Initial sample pre-treatment through filtration could affect the dissolved PFAS concentrations and was therefore avoided. Amending the postoxidation samples with methanol allowed for improved precision and low instrumental matrix effects. We also documented the oxidation yields of 23 anionic, neutral, zwitterionic, and cationic precursor compounds of PFAAs. These precursor compounds were amenable to TOP conversion. The total oxidative yield of 6:2 fluorotelomer sulfonamidoalkyl betaine (6:2 FTAB), for instance, was 80 mol%, with C₃-C₅ PFCAs as major oxidation products (minor: C₆-C₇ PFCAs). The method was applied to determine a wide range of PFAS (n = 41) without oxidation as well as ΔPFCA via persulfate oxidation in AFFF-impacted groundwater samples from fire-equipment testing sites in Ontario and Newfoundland, Canada. Summed PFAS concentrations as high as 5 mg L⁻¹ were reported before oxidation, and post-oxidation increases of PFCAs up to + 2300% were observed. A significant contribution of increases in individual PFCAs was attributed to precursors such as 6:2 FTAB, fluorotelomer sulfonates (6:2 FtS, 8:2 FtS), perfluorooctane sulfonamidoalkyl amine (PFOSAm), and perfluorohexane sulfonamide (FHxSA) at the active firefighting training site.

Introduction

The recent years have seen considerable improvements in the qualitative and quantitative analysis of per- and polyfluoroalkyl substances (PFAS) [1], [2], [3], environmental pollutants that can present bioaccumulative, persistent, and toxic properties. Monitoring activities have essentially targeted the perfluoroalkyl acids (PFAAs), including perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) that are on track to be phased out globally. Aqueous film-forming foams (AFFFs), dispersed in fire-equipment testing activities and fire emergency response at airports, military bases, and certain industrial facilities, represent a major source of PFAS contamination [4], [5], [6], [7], [8], [9], [10]. Managing the environmental and health risks posed not only by PFAAs, but also by co-occurring zwitterionic, cationic, neutral, and anionic PFAS that may be precursors to PFAAs (pre-PFAAs), has been articulated as a major challenge to address in the years to come [11], [12].

The pre-PFAAs reported in AFFF formulations and/or at impacted sites include well-studied compounds (e.g., 6:2 fluorotelomer sulfonate) and infrequently monitored PFAS including amines, betaines, amine oxides, and quaternary ammonium compounds [1], [2], [3], [6], [13], [14]. This wide structural diversity of precursors contributes to a great deal of unknown PFAS ‘dark matter’ escaping our grasp [12]. There exist many uncertainties regarding the analytical assessment of the novel PFAS, first, due to the paucity of available analytical standards. In addition, methods developed for historically-monitored PFAS may not always be transferable to the newly identified ones for various reasons [15], [16]. These rationales include, notably: i) low and/or variable recoveries according to matrix complexity [14], [16], [17], ii) increased risks of instrumental matrix effects due to the lack of suitable isotope-labelled internal standards [16], and iii) failure to implement quality control assays and the consequences thereof—risks of inconsistent accuracy performance.

A possible solution to avoid detailed analysis of the many pre-PFAAs, all the while gaining a better understanding of the extent of PFAS contamination, resides in the implementation of a total oxidizable precursor (TOP) assay [18], [19]. The TOP assay is based on persulfate chemistry, an efficient use of the latter requiring initial activation to achieve fast kinetics—sulfate radical generated by thermolysis of persulfate is quickly converted to hydroxyl radical, which oxidizes the pre-PFAAs. The TOP assay works under alkaline conditions (pH > 12) in an aqueous medium. Under such conditions and when an elevated temperature (85 °C) is achieved, PFCAs and PFSAs typically remain intact, whereas the pre-PFAAs are converted into more easily measurable PFCAs after a sufficient reaction time [18]. By analyzing the concentration of PFCAs before and after oxidation, it is therefore possible to estimate the total concentration of pre-PFAAs in environmental samples [20], [21]. The TOP assay was recently applied to diverse environmental and consumer product samples [20], [21], [22], [23], [24], [25], [26], [27], [28], [29].

Analyses via the TOP assay comprise many subtleties and are not without pitfalls. Ye et al. [21] signaled different PFAS blank backgrounds between the two types of samples (i.e., samples submitted to TOP assay Vs. those analyzed without oxidation), but there exist other issues. For instance, instrumental matrix effects may be enhanced due to the high salt content of TOP samples. Therefore, post-oxidation samples may require either solid-phase extraction [21] or appropriate dilution prior to LC-MS analysis. Matrix complexity could also preclude a complete conversion at the earlier oxidation reaction stage, due to the presence of organic molecules competing for the oxidant [30]. To date, few of the newly-identified cationic, zwitterionic, and anionic precursors have been assessed for their oxidative yields to PFCAs, in part, due to the lack of available standards. A recent study by Xiao et al. [31] showed that two betaine and ammonium derivatives of perfluorooctane sulfonamide, PFOSB and PFOSAmS, could generate PFOS and PFOA upon water disinfection with ozone and chlorine. Investigating the TOP products of PFOSB, PFOSAmS, 6:2 FTAB and a few other model zwitterionic and cationic PFAS will be useful to confirm whether these compounds can be integrated into TOP assay analyses. The conclusions drawn can probably be extended to compounds with related structures. Since previous studies relating the initial precursors to the increase in PFCAs after TOP (ΔPFCAs) suggested an incomplete mass balance [20], [23], [24], [28], expanding the list of monitored pre-PFAAs could also help in substance prioritization for future monitoring surveys.

Here, we set out to address knowledge gaps relating to the selection of suitable pretreatment conditions, TOP applicability to a wider range of pre-PFAAs, and resilience to matrix effects. For this purpose, a robust procedure was validated for the analysis of a wide range of PFAS (n = 41) in AFFF-impacted groundwater. The TOP workflow used a small aliquot of the sample (1.2 mL or lower) and therefore a reduced consumption of oxidation reagents. The oxidized samples were aliquoted and subjected to ultra-high-performance liquid chromatography high-resolution mass spectrometry (UHPLC-HRMS). The method was validated for method blanks, linearity performance, instrumental detection limits and method reporting limits, instrumental matrix effects, and accuracy/precision profiles. We also examined the possibility of matrix effects during the oxidation step by comparing conversion yields in ultrapure water and those in real groundwater. For the first time, zwitterionic and cationic PFAS including betaines, amines, amine oxides, and quaternary ammonium compounds were confirmed to be amenable to TOP conversion. The method was demonstrated through the analysis of PFAS pre- and post- TOP assay in groundwater samples from fire-equipment testing sites in Canada. The TOP validation procedure could be applied to other matrix types in future efforts toward standardizing the oxidative assay.