Activated carbon filters: what they remove and when they stop

An activated carbon filter removes certain gases and odors from air by holding their molecules on the surface of the carbon. It is good at some chemicals, poor at others, and it stops working once its surface is full, usually with no sign you can see. EPA’s Guide to Air Cleaners in the Home says these filters “can be effective, provided that there is a large amount of material used in the filter.” That proviso decides most of what a carbon filter in a home purifier can do, because many hold very little.

How activated carbon works

EPA calls activated carbon the most common adsorbent used in HVAC systems and portable air cleaners to remove gaseous contaminants. As air passes through, gas molecules touch the carbon’s surface and cling to it. The process is adsorption, with a d: molecules stick to a surface and are not soaked up into the material as a sponge takes up water.

Two consequences follow, and both are in EPA’s Technical Summary:

• The surface can fill. Adsorbent filters “have a finite capacity for adsorption and therefore must contain sufficient sorbent media for the application and must be replaced regularly.”
• The hold is not permanent. EPA states that “physical adsorption is a reversible process, meaning pollutants may not be permanently captured.”

Carbon also has a limit in the other direction: it is not a particle filter. A purifier meant to deal with both needs a separate particle stage such as a HEPA filter. EPA says gas-phase filters containing sorbents “should generally be located downstream of particle air filters”.

Packaging often says “charcoal” and “activated carbon” as though the words were interchangeable. For a buyer the name matters less than the amount.

What carbon captures well and what it misses

EPA’s Technical Summary says activated carbon “has the potential to remove most hydrocarbons, many aldehydes, organic acids through adsorption, and ozone through chemisorption.” The next sentence matters as much:

“However, activated carbon is not especially effective against oxides of sulfur, hydrogen sulfide, low molecular weight aldehydes (e.g., formaldehyde), ammonia, and nitrogen oxide.”

Formaldehyde is the hard case on this list. Laboratory work at Lawrence Berkeley National Laboratory on activated carbon fiber illustrates the gap. The researchers tested a small specimen against a seven-chemical mixture in an HVAC-style setup, and after heated regeneration they reported “consistent removal efficiencies of 70 - 80% for most VOCs”, but for formaldehyde “a maximum removal of 25-30%” with heated regeneration. That test used carbon fiber in a lab, at 20 to 30 ppb upstream, and not a consumer air purifier, so it shows the pattern and does not predict what any home unit will do.

EPA adds a broader warning: sorbent filters may remove specific pollutants, but “none is expected to remove adequately all of the gaseous pollutants in a typical home.” For how these limits play out against particular smells and chemicals, see the guide to odors and VOCs.

EPA’s position on radon is also direct: it “does not recommend air cleaning to reduce the health risks associated with radon and the decay products of radon gas.”

What real-world measurements show

Measurements of gas removal in homes are scarce. EPA’s Technical Summary says “only a few studies have investigated the effects of gas-phase air cleaners in homes.”

One test came from NIST researchers, published in 2008. They ran a sorption-based in-duct unit and a sorption-based portable unit against one chemical, decane, in a single-zone test house. Across 24 experiments, directly measured single-pass removal efficiencies ranged “from 24% to 56%”, and removal based on a whole-building mass balance ranged “from 30% to 44%.”

Those numbers are removal measurements for one chemical in one test house. They do not tell you what happens with a different gas, and they say nothing about health.

How much carbon is enough

The documents cited here give no answer in pounds. EPA says a carbon filter works when there is “a large amount of material used in the filter (the thicker the better)”, but the EPA documents cited here give no minimum weight or depth for a home unit. Treat a specific weight threshold seen elsewhere as a rule of thumb unless it names the test behind it.

EPA does explain why thin is a problem. Of consumer-grade systems it says: “Effectiveness of many consumer-grade systems with small amounts of activated carbon is unknown.” Thin layers “can become quickly saturated, and the filter can become a source of previously adsorbed pollutants.”

Filters impregnated with sorbent are described in the Technical Summary as generally ranging “from 1/8 inch to 2 inches thick.” EPA’s assessment of them is that breakthrough “can take place very quickly in the thin layer impregnated with sorbents, resulting in a short service life for the filter, which must be replaced frequently. Thus, these devices usually have limited effectiveness in removing odors.”

Thickness alone does not rescue the design either. EPA describes tests of three VOCs at 100 to 200 ppb on three carbon samples, assuming 150 ppb in the air, a 50 ppb exit concentration and 100 cfm across a 2-foot by 2-foot filter. The conclusion, citing Ramanathan and colleagues in 1988, was that “breakthrough of these chemicals would occur quickly in 6-inch deep carbon filters used for odor control.” From that it concluded that “the thicker the media, the more efficient the filter will be for longer periods of time.” EPA goes further about whole systems: in its 2018 summary, a properly designed and built gas-phase system “is too big for a typical residential HVAC system or portable air cleaner.”

Why there is no CADR for carbon

CADR, the familiar number on air purifier boxes, covers particles only. EPA says so outright: “there are no widely used performance rating systems for portable air cleaners or filters designed to remove gases. The CADR rating system is for particles only.” AHAM, whose test method produces CADR numbers, says “there is no universal test method yet accepted for measuring performance of air cleaners to clean gases or volatile organic compounds.” The NIST paper noted the same absence of standard methods in 2008, though EPA’s 2018 summary does name two ASHRAE test standards, 145.1 and 145.2, for gas-phase media.

The practical result is that a buyer cannot compare two carbon filters the way two particle CADRs can be compared. Words like “odor eliminator” on the box have no agreed test behind them, and what is left to compare is limited: how much carbon, in what form, and how often the maker says to replace it.

Saturation, breakthrough and replacement

Saturation means the carbon’s surface is full. Breakthrough means pollutant is passing through the filter and back into the room. Neither is visible, and a carbon filter that has stopped working looks exactly like one that has not. EPA says all adsorbents “have limited capacities and thus require frequent maintenance.” That frequency depends on what the filter is exposed to, and how hard it works, so a filter in a kitchen is likely to be used up sooner than the same filter in a spare room.

A few signals are useful:

1. Follow the maker’s replacement interval as a ceiling, not a target.
2. Replace sooner if smells return with the source unchanged and the unit running.
3. Remember that a smell-free room does not prove the filter is still working, because plenty of gases have no odor.
4. Do not count on an indicator light to know about the carbon. Check the manual for what it measures.

The steps for changing it, and for handling a used filter, are in the guide to replacing air purifier filters.

When a carbon filter is the wrong tool

EPA’s advice on gases starts with the source. Its Technical Summary says reducing or eliminating avoidable sources, then exhausting unavoidable pollutants outdoors, is “usually” the most economical and effective way to address indoor air pollution. For a chemical smell from a new product, ventilation comes first and a carbon stage after it. EPA also says research has not yet shown that plasma, photocatalytic oxidation or ultraviolet light can remove gases effectively in portable residential air cleaners, so those are not a substitute.

No air purifier cures or prevents any disease, and an activated carbon stage does not make a room safe from a chemical. Anyone with symptoms they connect to indoor air should see a clinician.