The agency must treat PFAS as a class of chemicals regulated in the same manner as the European Union and other countries.
To great applause, the U.S. Environmental Protection Agency last month proposed “maximum contamination levels” in drinking water for six per- and polyfluoroalkyl substances (PFAS) chemicals. This was the first such action that EPA had taken in more than 30 years on any drinking contaminant.
Unfortunately, despite the magnitude of this action for this handful of PFAS, there are at least 12,000 PFAS variations.
Before the applause for EPA could die down, this month, a study found 26 types of PFAS in drinking water samples from 16 states. Notably, EPA has no pending proposed standard for 20 of these PFAS. More disturbing, 12 of these PFAS are not included in EPA’s current monitoring—in other words, EPA does not even currently test for them.
Furthermore, three of these PFAS fall outside the “working definition” for PFAS that EPA adopted without any outside review in 2021. This means that EPA is not considering regulating them.
In fact, EPA’s working definition is far narrower than those adopted by other entities, such as the intergovernmental Organisation for Economic Co-operation and Development (OECD) and members of the European Union. Nor is it as broad as the definition adopted by numerous states that have started regulating PFAS on their own, frustrated by the slow pace set by EPA. It is also substantially narrower than the definition EPA itself uses for research purposes and non-regulatory estimates.
Why would EPA summarily adopt a definition that leaves out thousands of PFAS?
Public Employees for Environmental Responsibility (PEER), the organization I work for, asked the agency this very question. Just four months after EPA published this working definition in 2021, we submitted a Freedom of Information Act (FOIA) request for documents that would provide a scientific explanation or justification for this new definition.
Source: https://www.commondreams.org/opinion/epa-pfas-forever-chemicals-failure