Activated carbon removes the long-chain PFAS like PFOS and PFOA well but is weak on short-chain types. Reverse osmosis (RO) beats it at home. PFAS is now found in water almost everywhere, including rain on the Tibetan Plateau.
PFAS in water worldwide
[Ackerman Grunfeld et al., Nature Geoscience (8 Apr 2024)](https://www.nature.com/articles/s41561-024-01402-8 (opens in a new tab)), a UNSW-led study, pooled more than 45,000 samples. Even groundwater with no known contamination source often went over the proposed US limits: 50% for PFOS and 40% for PFOA. Near firefighting-foam sites it was 72% and 63%.
[Cousins et al. 2022](https://pubs.acs.org/doi/10.1021/acs.est.2c02765 (opens in a new tab)) found that PFOA in rainwater, even on the Tibetan Plateau, is about 14 times the EPA's 2022 health advisory.
[USGS (Jul 2023)](https://www.usgs.gov/news/national-news-release/tap-water-study-detects-pfas-forever-chemicals-across-us (opens in a new tab)) estimated that at least 45% of US tap water contains at least one PFAS.
In Australia, contamination is mostly from firefighting foam, at places like Williamtown NSW, Oakey and Katherine.
Drinking-water limits
[Australia's NHMRC guidelines (Jun 2025)](https://www.nhmrc.gov.au/about-us/news-centre/updated-australian-drinking-water-guidelines (opens in a new tab)) set PFOS at 8, PFOA at 200, PFHxS at 30 and PFBS at 1000 ng/L. These are guideline values, not enforceable limits.
The US limit is 4 ng/L each for PFOA and PFOS. The EPA has proposed dropping the limits for other PFAS and delaying compliance to 2031, but neither is final.
The EU limits apply from 12 Jan 2026.
The WHO has no final guideline value yet.
How well activated carbon works
[Herkert et al. 2020](https://pubs.acs.org/doi/10.1021/acs.estlett.0c00004 (opens in a new tab)) tested home filters for PFOA:
Pitcher filters removed about 67% and fridge filters about 71%.
Under-sink RO removed over 92%, and two-stage filters over 99%.
Whole-house carbon systems removed only 19%. In half of them, more short-chain PFAS came out than went in, most likely because the old media was saturated.
Short-chain PFAS like PFBA get through carbon quickly. RO handles them.
[A lab study of reed-straw biochar](https://par.nsf.gov/servlets/purl/10292330 (opens in a new tab)) found it removed 80% of PFBA versus 13% for commercial carbon. I found no study showing homemade or wood charcoal works on PFAS, so treat that as unproven.
Used filter media still holds the PFAS. The [EPA's disposal guidance](https://www.epa.gov/system/files/documents/2024-04/2024-interim-guidance-on-pfas-destruction-and-disposal.pdf (opens in a new tab)) says destroying it may need temperatures above 1,100°C, so burning old filters at home doesn't reliably destroy it.
To check a filter, look for NSF/ANSI 53 (carbon) or 58 (RO) certification on [the EPA's certified-filter list](https://www.epa.gov/water-research/identifying-drinking-water-filters-certified-reduce-pfas (opens in a new tab)). Older certifications only tested down to 70 ng/L, which is above Australia's PFOS guideline of 8.
Seen & read (0) — sign in to acknowledge
Sign in to acknowledge. This is voluntary.