HEPA vs Hydroxyl vs Activated Carbon vs Ionizer vs Ozone Generator: How They Differ

What they do

  • HEPA filtration — catches particles

    • A fan pulls room air through a dense mat of fibers. Particles stick to the fibers; the air passes through.
    • A filter that meets the U.S. Department of Energy standard removes 99.97% of particles at 0.3 micrometers — the hardest size to catch, which is why it is the size the standard uses. Larger and smaller particles are captured more easily.
    • What it leaves: gases. A fiber filter has nothing to grip a molecule with, so VOCs, odors and off-gassing pass straight through unless the unit also carries activated carbon. It also only ever treats the air that reaches the intake — nothing on your surfaces, nothing in the next room.

  • Ionizers and electrostatic cleaners — charge and settle

    • These apply an electrical charge to particles in the air. Charged particles attract each other and clump, grow heavier, and fall out of the air — onto a collector plate in the unit, or onto your floors and furniture.
    • They are usually silent and need no filter. The trade-off is the by-product: generating that charge can also generate ozone. The standard that governs them is UL 867, Electrostatic Air Cleaners, whose test caps a household unit at 0.05 ppm of ozone.
    • What it leaves: the particles, physically — they are relocated, not removed. Gases are largely untouched. And the ozone figure varies enormously by device, which is why it is the number to ask for.

  • Ozone generators — oxidize, at a cost

    • These intentionally produce ozone, sometimes marketed as “activated oxygen.” Ozone is a strong oxidant and does react with odors and contaminants.
    • The California Air Resources Board’s position is blunt: ozone is only partially effective at cleaning indoor air at concentrations high enough to be a serious health risk, and CARB recommends these devices not be used in occupied spaces at all. It publishes a list of hazardous ozone generators sold as air purifiers.
    • What it leaves: ozone, which is itself a respiratory irritant. This is the category WellisAir is most often confused with, but our innovation Ozone Safe status separates us from this group.
    Read CARB’s guidance 

  • Hydroxyl radicals — break molecules apart

    WellisAir releases hydroxyl radicals (OH) into the room. A hydroxyl takes a hydrogen atom away from a pollutant molecule. The hydroxyl becomes water; the pollutant, now missing a hydrogen, is unstable and breaks down further.

    Because the radicals travel with the air, they reach the whole room and the surfaces in it, not only what passes an intake. No filter, no fan drawing air in — a cartridge of hydrogen-peroxide solution, replaced every 90 days.

    What it leaves: particles. Dust, pollen and dander are physical objects, and breaking molecular bonds does not remove them. For particles you want a filter — see below.

    How hydroxyls work 

The measurements behind our column

  • Ozone: 0.002–0.006 ppm, measured by Intertek

    Intertek measured the unit’s ozone output across its operating range. The result sits between 0.002 and 0.006 ppm against a 0.05 ppm regulatory limit.

    Intertek test report — the ozone page sets out what was measured and how.

    The ozone test, in full 

  • VOCs: Up to 99.5% reduction in two hours

    Korea Conformity Laboratories exposed the unit to single compounds and measured what was left:

    Toluene — 99.5% in 2 hours.
    Ammonia — 80.0% in 4 hours.
    Formaldehyde — 65.0% in 4 hours.

    Korea Conformity Laboratories, test report 026961 — test summary (PDF).

  • Mold and mildew: a peer-reviewed clinic result

    In a dental clinic over a working day, airborne molds and yeasts fell 66.7% in the treated rooms compared with the controls (p=0.03) — a published, peer-reviewed result rather than a manufacturer’s test.

    Mold and mildew, in depth 

  • Why half a second is long enough

    A single hydroxyl radical survives about half a second, which is the most common objection to the whole approach. The answer is that the pool is continuously regenerated — roughly every 28 seconds — and persists for minutes, with each radical recycled about four times before the chain ends.

    Monks, P. S. (2005). Gas-phase radical chemistry in the troposphere. Chemical Society Reviews, 34(5), 376–395.
Every source we cite

Comparison questions, answered

Is WellisAir an ionizer?

Not in the sense the word usually carries. WellisAir is sold and tested as a hydroxyl generator, and it does not work the way an electrostatic cleaner does — it is not charging particles so they clump and fall. What does the cleaning is chemical: hydroxyl radicals breaking pollutant molecules apart.

It is fair to say the categories overlap at the hardware level. The unit’s own Intertek test documentation describes an ionizing stage in the emitter. We have asked the manufacturer for a fuller engineering description and will add it here when we have it.

Which is why the useful question is not which category the device belongs in. It is what comes out of it, and that is measurable: 0.002 to 0.006 ppm of ozone, measured by Intertek, against the 0.05 ppm ceiling California sets for an air cleaner used in an occupied space. Ask any device you are comparing for the same number.

Does WellisAir produce ozone?

A very small, measured amount. Intertek put it between 0.002 and 0.006 ppm. The regulatory ceiling for an air cleaner in an occupied space is 0.05 ppm, tested to the UL 867 method, so the top of our range is about one eighth of that limit and the bottom is close to the detection floor of the test. The unit is CARB certified.

A number can be verified; “ozone free” cannot.

Is a HEPA filter better than hydroxyl technology?

For particles, a HEPA filter removes 99.97% of particles at 0.3 micrometers.

For gases and VOCs, no. A fiber filter has nothing to hold a molecule with, which is why gas removal in a filter unit depends on activated carbon rather than on the HEPA element. The two technologies address different problems.

Do air purifiers remove VOCs?

Most don’t, and the EPA is direct about why: “most filters are designed to filter either particles or gases.” A HEPA filter is a particle filter, and its CADR rating is, in the EPA’s words, “for particles only.” For gases the EPA points you to activated carbon or another gas-specific approach.

Hydroxyls work on the gas side — they break the molecule apart rather than trapping it. Korea Conformity Laboratories measured 99.5% of toluene gone in two hours on the WellisAir unit.

Why doesn’t WellisAir publish a CADR rating?

Because CADR measures one thing: how fast a unit clears particles from a chamber. It is the right yardstick for a filter and a meaningless one here — a hydroxyl radical does not capture particles, so a CADR number for this unit would be close to nothing and would tell you nothing useful.

Asking for a CADR rating is the right instinct applied to the wrong category. The equivalent question for this technology is which compounds were reduced, by how much, over what time, and by whom — which is what our test reports state.

Are ozone generators dangerous?

The California Air Resources Board recommends they not be used in occupied spaces, and publishes a list of hazardous ozone generators sold as air purifiers. Its assessment is that ozone only partially cleans indoor air at concentrations high enough to pose a serious health risk, and that inhaling ozone is itself a risk to people and pets.

Watch for “activated oxygen” in marketing — CARB specifically flags it as a way of making ozone sound healthy.

Can I run WellisAir and a HEPA purifier together?

Yes, and if you have both a particle problem and an odor or off-gassing problem, that is the sensible setup. They do not interfere with each other: one is pulling particles onto a filter, the other is releasing molecules that break down gases. Nothing about running a fan-and-filter unit consumes or blocks hydroxyl radicals.

How does a hydroxyl radical clean a whole room if it only lasts half a second?

Because the pool is replenished rather than spent. When a hydroxyl takes a hydrogen from a pollutant it becomes water, and the onward chemistry of the broken pollutant forms a new hydroxyl — so the population persists even though no individual radical does. Published measurements put the regeneration at roughly every 28 seconds, with the pool persisting for minutes and each radical recycled about four times before the chain terminates.

Monks, P. S. (2005). Gas-phase radical chemistry in the troposphere. Chemical Society Reviews, 34(5), 376–395.