Do Water Filters Remove Microplastics?

Written and reviewed by the BWF Team, a certified water specialist, a former municipal plant operator, a licensed plumber and a former system dealer.

Yes, and it is a size problem rather than a chemistry one. A plastic fragment is suspended in the water rather than dissolved in it, so it is removed by being too big to get through the pore. That makes this an unusually honest category: the number that decides the outcome is a number manufacturers publish.

There is also a certification for it, which almost nobody mentions. Microplastics Reduction is one of 16 separate claims under NSF/ANSI 401 and 1,099 models hold it. Because those claims are granted one test at a time, a filter "certified to NSF/ANSI 401" may or may not have been tested for microplastics, and the claim list underneath the badge is what tells you.

Last updated

It is a size problem

Microplastics do not react, dissolve or break down on any useful timescale. They are small pieces of solid plastic drifting in the water, and a filter deals with them the way a sieve deals with gravel. That is the entire mechanism, and it means one specification decides the answer.

What Size What it means here
A human hair 70 microns For scale. Everything here handles this.
Microplastics, as drinking water studies count them Down to about 1 micron Below roughly a micron the standard counting methods stop being reliable, so this is the floor of what gets measured rather than the floor of what exists.
A supermarket jug cartridge Usually unstated Granular carbon built for chlorine and taste. If the box gives no micron figure, that is the answer to your question.
A good carbon block 0.5 microns Half the size of the smallest fragment routinely counted.
Hollow fibre, as used in camping filters 0.1 microns absolute Fine enough to strain out bacteria, which are larger than the pore.
A reverse osmosis membrane About 0.0001 microns Rejects dissolved salt ions. A plastic fragment is thousands of times larger than the gap.

The practical version: if a filter states a pore size below one micron, microplastics are within what it can physically catch. If it states no pore size at all, that silence is the most informative thing on the box. A jug cartridge built for chlorine and taste is not a fine strainer and is not sold as one.

There is a certification, and it is granted claim by claim

Nearly every filter now says something about microplastics, and it is fair to assume most of those sentences rest on nothing in particular. A good number of them can rest on something. Microplastics reduction is a certifiable claim: it sits under NSF/ANSI 401, the emerging contaminants standard, and 1,099 certified models held it when we checked, across 683 distinct product lines. This is an established credential, not a novelty.

The structure underneath is the part worth carrying into a shop. NSF/ANSI 401 contains 16 separate claims: pharmaceuticals such as ibuprofen and naproxen, hormones such as estrone, industrial chemicals such as bisphenol A, several pesticides, and microplastics. Each one is granted on its own test. A manufacturer submits a product for the claims they want and is listed for those.

So the standard number on a box is the wrong thing to compare, and it is usually the only thing printed. Two filters can both be certified to NSF/ANSI 401 and have been measured for entirely different contaminants. The claim list is published for every certified product, in the certifier's own database, searchable by model number.

Certified products exist in every format a home would use. The listing carries sections for pour-through jugs, refrigerator filters, faucet mounts and plumbed-in systems, alongside commercial equipment. Whatever kind of filter you are looking at, there is a certified example of it to compare against.

How to check any filter in two minutes

This works for any filter from any brand, including ones we have never written about, and it is the single most useful habit in this category.

  1. Find the model code, not the product name. Certifier databases index by the string the manufacturer registered, and it is rarely what is on the front of the box. Brondell's reverse osmosis system is sold as the Circle and listed as RC100. iSpring's RCC7AK is listed as RCC7AK-BN. Search the marketing name and you will find nothing and conclude the wrong thing.
  2. Search all four certifiers, not just the famous one. A product missing from NSF is very often certified by WQA or IAPMO instead, and their databases do not carry each other's listings.
  3. Read the claim list under standard 401. Not the standard number, the list. You are looking for the word microplastics in it.
CertifierWhere to search
NSF International info.nsf.org/Certified/DWTU
WQA find.wqa.org/find-products
IAPMO R&T pld.iapmo.org
CSA Group CSA product listings

If a filter turns out not to hold the claim, that is not evidence of anything much on its own. Certification costs money and is applied for claim by claim, and plenty of capable filters were simply never submitted for this one. What the check buys you is the difference between a marketing sentence and a measured result, which is worth two minutes.

What does not work

Boiling. It kills organisms. A plastic fragment is not alive, does not evaporate, and is still there when the water cools. There is published research suggesting that boiling hard water traps some nanoplastics in the limescale that forms, which then has to be filtered out, but that is a laboratory result about hard water and not a method to rely on at home.

Letting it stand, or freezing it. Neither changes the size of a particle or the fact that it is suspended.

Switching to bottled water. Published studies have found microplastic particles in bottled water at rates that make it no better than filtered tap water on this measure, and it adds a plastic container the water has been sitting in for months. If microplastics are the reason for buying bottled water, a filter with a stated pore size does the same job at a fraction of the running cost.

Which kind of filter to get

Every option below is small enough. The choice is really about what else is in your water, because that is where they genuinely differ.

Your water supplier publishes an annual report telling you what is actually in the supply, and it is free. That report, not a product page, is what should decide which half of this list you are shopping in.

Common questions

Do water filters remove microplastics?
Yes, if the pore is smaller than the particle, and that is the whole mechanism. A plastic fragment is suspended rather than dissolved, so it is removed by being too big to pass rather than by any chemistry. Reverse osmosis membranes, carbon blocks rated at half a micron and hollow fibre filters at a tenth of a micron are all well below the size at which drinking water studies can reliably count a fragment. An ordinary jug cartridge built for chlorine and taste is a different thing and usually states no pore size at all.
Is there a certification for microplastics removal?
Yes. Microplastics Reduction is one of the sixteen claims under NSF/ANSI 401, the emerging contaminants standard, and 1,099 certified models held it when we checked in August 2026. The important part is that those sixteen claims are granted separately, one test at a time. So "certified to NSF/ANSI 401" tells you which rulebook was used and not what the filter was tested for, and two products can both hold 401 having been measured for entirely different things. The claim list is published for every certified product and it is the thing to read.
Does boiling water remove microplastics?
No. Boiling kills organisms, and a plastic fragment is not alive and does not evaporate. There is published research suggesting that boiling hard water traps some nanoplastics in the limescale that forms, which then has to be filtered out, but that is a laboratory result about hard water rather than a method to rely on at home. Removal is physical: something has to strain the particle out of the water.
Does a Brita remove microplastics?
It depends entirely which cartridge, and the way to settle it is to read the claim list rather than the box. Brita certifies through WQA and IAPMO rather than NSF, and their filters differ from each other: the Elite is certified under 42, 53 and 401 while the standard white cartridge is certified under 42 alone. Search the model code in the certifying body's own database and look at what is listed under 401. That is a two-minute job and it answers the question for any brand, not only this one.
Is bottled water free of microplastics?
It is not an escape from the question, and it has a problem tap water does not: the bottle. Published studies have found microplastic particles in bottled water at rates that make it no safer than filtered tap water on this measure, and a bottle is a plastic container the water sits in for months. If microplastics are your reason for buying bottled water, a filter with a stated pore size does the same job for a fraction of the running cost.
What size are microplastics in drinking water?
Studies count them down to roughly one micron, which is where the standard analytical methods stop being reliable. That figure is a limit of measurement rather than a limit of what exists, and smaller fragments are usually described as nanoplastics. For choosing a filter this matters less than it sounds: a half-micron carbon block, a tenth-micron hollow fibre and a reverse osmosis membrane are all below the measurable range already.

How we checked

The counts on this page were read from NSF's own database on 6 August 2026 and can be re-run. The claim vocabulary comes from NSF's search form, which is the authority on what can be certified at all and lists 151 claims across seven standards. Querying for microplastics reduction returned 1,099 certified models.

That query carried a control, because a count is only meaningful if the same method finds things elsewhere: chlorine reduction under standard 42, run the same way in the same session, returned 5,553 models. The method finds products where products exist, so the microplastics figure is a measurement rather than a search that failed.

Related