Draft objective for per- and polyfluoroalkyl substances in Canadian drinking water: Analytical considerations
This proposed objective for PFAS in drinking water refers to 2 validated, standardized U.S. EPA methods that are available for the quantitation of a combined total of 29 compounds: EPA Methods 533 and 537.1 (respectively, U.S. EPA 2019, 2020). These analytical methods are specific, sensitive and practical for application in commercial laboratories. The MRLs established under the 5th cycle of the UCMR (UCMR 5) (U.S. EPA, 2021b) for the analysis of PFAS using these methods range from 2 to 20 ng/L and are listed in Appendix A. A jurisdiction could also validate and apply an alternate analytical method that quantifies a minimum of 18 PFAS.
EPA Method 533 is an isotope-dilution/anion exchange/solid-phase extraction (SPE) liquid chromatography/tandem mass spectrometry (LC-MS/MS) method for the determination of select PFAS in drinking water. It requires the use of MS/MS in multiple-reaction-monitoring (MRM) mode to enhance selectivity and specificity for compounds of interest. EPA Method 533 measures 25 specific PFAS (none greater than C12 chain length), including perfluorinated acids, sulfonates, fluorotelomer sulfonates and poly/perfluorinated ether carboxylic acids.
EPA Method 537.1 is an isotope-dilution/hydrophobic SPE LC-MS/MS method for the determination of select PFAS in drinking water. This method measures 18 specific PFAS, including perfluorocarboxylic acids up to 14 carbons in length. It differs from Method 533 in that the concentration technique used relies on hydrophobic interactions, and as such is not suitable for the more hydrophilic shorter carbon chain PFAS, such as PFBA and PFPeA.
EPA Method 533 complements EPA Method 537.1, with 14 PFAS in common and no inherent differences in MRLs. However, Method 533 can measure 11 additional PFAS compared with Method 537.1, including more hydrophilic chemicals such as PFBA. By combining both methods, a total of 29 unique PFAS can be effectively measured in drinking water. Many laboratories can reliably report at 2 ng/L for most PFAS and 5 ng/L for the rest (see Appendix A).
Because they rely on different sample concentration techniques, the 2 methods require separate sample preparation procedures and cannot be combined into a single analysis. In Canada, laboratories are generally accredited for EPA Method 537.1 (CALA, 2022). However, it is recognized that EPA Method 533 would provide better coverage of the PFAS observed in Canadian data noted in the exposure considerations section.
Given the overlap between the 2 EPA methods and to avoid duplication, the authorities responsible for drinking water may specify 1 method be used. Utilities should take into consideration the potential sources of PFAS and select a method that will provide analyses that include PFAS that may be present in the drinking water. Guidance on site characterization is available elsewhere (U.S. EPA, 2022a; ITRC, 2020, 2022a).
Total PFAS should include all PFAS listed in a method and detected in a sample. Any value above the MDL should be included in the summation for total PFAS, recognizing that a value of zero is assigned for any value below the detection limit.
Where possible, utilities should strive to analyze as many PFAS as possible to gain a better understanding of the PFAS present in the drinking water in order to inform the selection of treatment that will reduce exposure to the greatest extent possible.
Any detected PFAS, from all analyses undertaken, should be summed, and this sum should still not exceed the proposed objective value of 30 ng/L. When more than 1 method is used for the analysis, it is not necessary to do a duplicate analysis where there is overlap between methods. However, if duplicate analysis occurs, the highest of the duplicate individual PFAS results should be taken for the summation.
PFAS precursors can degrade to perfluoroalkyl acids (PFAAs) under the right environmental conditions. The Total Oxidizable Precursors (TOP) assay oxidizes PFAS precursors into their corresponding PFAAs, which can then be measured using the EPA methods or other methodologies. It is a useful screening tool that can help provide a better understanding of the amount of PFAS in a sample including unknown precursor species that might otherwise be missed (that is, total PFAS load). Because the TOP assay does not identify individual precursors, data typically are reported as the net change in PFAA concentrations before and after oxidation (Rodowa et al., 2020). The TOP assay may under-quantify short-chain PFAA precursors that are telomer-based (ITRC, 2022b).
Another common surrogate analysis for PFAS is the Total Organic Fluorine (TOF) analysis, which can be used for drinking water. While the TOF analysis can be useful, it is indiscriminate and may capture fluorine from non-PFAS compounds. To date, there has been no demonstrated method that avoids having to employ sample preparation steps and that loses a portion of the TOF. The U.S. EPA (2022b) has released a draft Adsorbable Organic Fluorine (AOF) method (EPA Draft Method 1621) for wastewaters that uses carbon adsorption to prepare the sample for the fluoride wash and the final combustion ion chromatography process. However, the application to drinking water is limited because the minimum detection limit is well above concentrations typically seen in drinking water sources.
The TOP assay and TOF analysis can more comprehensively assess the concentration of PFAS beyond the 29 listed in the above methods. However, they are qualitative techniques and not yet standardized, nor have they undergone multi-laboratory validation. Despite their limitations, these assays can provide a better understanding of which PFAS are or may be present in water and their impact on the treatment system's operations. These additional PFAS may break through more rapidly, necessitating more frequent media change-out or regeneration. Data from these assays could be used to augment the data from quantitative methods. (ITRC, 2022b).
Report a problem or mistake on this page
- Date modified: