Filters and treatment

Which filters actually remove PFAS

Most filters sold in American supermarkets do nothing whatsoever about PFAS. The ones that work say so in a specific, checkable way.

Why most filters do nothing

The filter in a typical supermarket pitcher is designed to improve taste. It removes chlorine, which is the main thing people notice, and it does that well and cheaply. PFAS molecules are a different problem: they are present at parts per trillion, they are chemically inert, and removing them requires either a large mass of carbon with a long contact time or a membrane fine enough to physically exclude them.

A pitcher filter has neither. Water passes through a small cartridge in seconds. That is adequate for chlorine and inadequate for PFOA, and a manufacturer that does not claim PFAS reduction is not being deceptive — people simply assume the claim extends further than it does.

The three technologies that work

Granular activated carbon

Carbon processed to have an enormous internal surface area, which PFAS molecules adsorb onto. Effective when there is enough carbon and enough contact time, which in practice means an under-sink or whole-house unit rather than a pitcher. Performance declines as the carbon loads up, so the replacement schedule is not optional — an exhausted cartridge can release what it captured.

Ion exchange resin

A resin that swaps PFAS ions for harmless ones. Generally more effective than carbon on short-chain PFAS compounds, which are the harder ones to capture, and often combined with carbon in a single treatment train.

Reverse osmosis

Water is forced through a semi-permeable membrane that physically excludes molecules above a certain size. The most thorough of the three, and the most demanding: it is slower, it wastes water to drain, it needs adequate pressure, and it strips beneficial minerals along with everything else. Usually installed under a sink for drinking and cooking water only.

How to verify a claim

Certification is specific to the contaminant. This is the single most useful thing to understand before buying anything.

A product can be certified to Standard 53 for lead and say nothing about PFAS. Certification is granted claim by claim, so the question is never "is it NSF certified" but "what is it certified to reduce".

The older protocol NSF P473, written specifically for PFOA and PFOS, has been retired and folded into Standards 53 and 58. If you see P473 referenced on packaging, the product is using outdated language. The current total PFAS reduction claim covers PFOA, PFOS, PFHpA, PFHxS, PFNA and PFBS to a combined 20 parts per trillion.

Choosing by what you actually have

Filtration is worth matching to a measured problem rather than a general worry. Start by checking what federal testing found in your system, then read your annual water quality report for the contaminants that testing did not cover.

• Nothing detected, and you dislike the taste. A Standard 42 pitcher or faucet filter is the right and cheapest answer.
• PFAS detected below the limit. A Standard 53 under-sink carbon unit with a named PFAS claim, on the drinking tap.
• PFAS above a limit, or lead in older plumbing. Reverse osmosis under the sink, or a certified carbon unit sized for the household. If lead is also a concern, confirm the certification names both.
• A private well. Test first. Filtration chosen without a test result is guesswork, and wells can carry arsenic, nitrate or bacteria that need entirely different treatment.

What filtration cannot fix

A filter is a point solution at the end of a long chain. It does not change what is in the source water, and it does not help anyone in your neighborhood who has not bought one. Where a system is above a federal limit, the durable remedy is treatment at the plant, which is what the compliance deadline exists to compel.