chevron-down Created with Sketch Beta.
June 21, 2023

Regulating PFAS at the Edge of Detection

Linda Cook and Kirk O’Reilly

On March 29, 2023, the U.S. Environmental Protection Agency (EPA) published proposed regulations  (88 FR 18638) for per- and polyfluoroalkyl substances (PFAS) drinking water standards. While stakeholders requested an extension, the comment period ended on May 30, 2023. If implemented, the regulations would establish limits for perfluorooctane-sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA), along with four additional PFAS. 

EPA proposes to set the maximum contaminant level goal (MCLG) for PFOA and PFOS at zero. A MCLG is a risk-based value defined as the maximum level of a contaminant in drinking water at which no known or anticipated adverse effect on the health of persons would occur, allowing an adequate margin of safety. An MCLG is used in the process of establishing MCL, which is a regulatory target that is as close to the MCLG as possible but incorporates considerations of technical feasibility and cost. Considering feasibility, including available analytical methods and treatment technologies, EPA proposes an MCL of 4 parts per trillion (ppt) or nanograms per liter (ng/L) for both PFOA and PFOS. These MCLs would set enforceable limits on the concentrations of these chemicals in public drinking water systems. The only chemical with a lower MCL is 2,3,7,8-TCDD, which has an MCL of 0.03 ppt.

Additionally, EPA proposes non-enforceable MCLGs for four additional PFAS hexafluoropropylene oxide-dimer acid (GenX), perfluoro-1-butanesulfonic acid (PFBS), perfluorononanoic acid (PFNA), and Perfluoro-1-hexanesulfonic acid (PFHxS) as a mixture using a Hazard Index (HI) approach, which considers the individual and additive toxicity of co-occurring chemicals. While exact concentrations thresholds are not provided for the four additional PFAS, the health-based values used to calculate the HIs are 9 ppt for PFHxS, 10 ppt for GenX and PFNA, and 2000 ppt for PFBS.

Under the proposed rule, water systems have three years to demonstrate to their primacy agency that each regulated PFAS is not present in the drinking water supply or, if present, it does not exceed its proposed MCLs. Initial compliance will be determined based on annual concentrations calculated from four quarterly monitoring samples. Once compliance is obtained, the agency may reduce sampling frequency.

Setting MCLs at a reported detection limit creates technical and regulatory challenges. This is especially true for PFAS, as analytical methods are still being refined, and the number of certified laboratories is limited. EPA notes that PFOS and PFOA “can be reliably measured” at the 4 ppt concentration level. We would argue that this assumption is only true when all procedures, from sample collection to sample analysis, are executed flawlessly.

Sample collection is the foundational step in obtaining accurate data. PFAS are ubiquitous in our daily lives, making it difficult to acquire water samples free of potential interfering contamination. PFAS are present in certain water-resistant clothing, fabric softeners, cosmetics, hand cream, sunscreen, notepads, markers, Teflon, food packaging, and more. Field samplers need to be aware of all possible PFAS sources and ensure that none are present during sample collection. Combined with what the field samplers could potentially introduce, the environment in which the sample is collected must also be PFAS-free. The potential for introducing contamination into a sample at the time of collection is a constant challenge considering the trace levels of these proposed MCLs.

The next challenge in achieving accurate and precise PFOS and PFOA measurements at limits near 4 ppt is finding a reliable and competent laboratory to analyze the water samples. EPA published the first PFAS method, Method 537, in 2009 to support the UCMR3 (Third Unregulated Contaminant Monitoring Rule) program with reporting limits in the 10 to 90 ppt range. Method 537.1 Revision 2 (2020) lists the minimum reporting limits (MRLs) for PFOS and PFOA as 2.7 and 0.82 ppt, respectively. EPA Method 533 was published in 2019 and lists the MRLs for PFOS and PFOA as 4.4 and 3.4 ppt, respectively. According to text of Methods 533 and 537, the MRLs in both cases are from a single laboratory and do not represent what other laboratories can meet. While the UCMR5 Laboratory Approval Program found that most participating laboratories reported MRLs less than 4 ppt, not all laboratories can meet these limits. Similar to field sampling concerns, many items containing PFAS exist in the laboratory environment. Laboratories must set up dedicated sample preparation areas, laboratories, equipment, and instruments and ensure that all are free of PFAS-containing materials.

With the surge of PFAS testing required these days, environmental testing laboratories are rushing into this market to support the need and bolster business growth. Unfortunately, not all these laboratories have the experience and expertise to accurately and precisely analyze these samples.

Assessing compliance with a theoretical detection limit-based MCL raises the possibility of false positives and negatives. False positives may occur if PFAS is introduced during sample collection, while false negatives may occur if available laboratories cannot meet expected detection limits. Under the proposed rule, detections below a laboratory’s practical quantitation limit will be treated as non-detects when calculating annual average concentrations. The proposed regulation does not describe how regulators and water suppliers should respond to these analytical concerns.

Given the challenges of collecting and analyzing drinking water samples for PFOA, PFOS, and other PFAS, care and oversight are required to obtain high-quality and accurate data generation. Where possible, staff experienced in PFAS sample collection should be available to oversee sample collection efforts. Additionally, laboratories should be audited prior to the initiation of projects, and the laboratory-generated data should be inspected and validated by trained staff to assess the competency of the laboratory and the accuracy and precision of the results.

The material in all ABA publications is copyrighted and may be reprinted by permission only. Request reprint permission here.

Linda Cook and Kirk O’Reilly


Linda Cook is a managing scientist with Exponent’s Environmental and Earth Sciences practice.

Kirk O’Reilly JD PhD, a senior managing scientist with Exponent’s Environmental Sciences practice, has 34 years of experience assisting clients manage liabilities associated with contaminated waters, soils, sediments. Dr. O’Reilly participates in projects that involve the collaborative efforts of industry, regulatory agencies, and academic institutions. A member of the Washington State Bar, he has served various leadership roles on ABA’s Superfund and NRD Litigation Committee.