Clinically Proven

A qualitative test of how far the unit’s oxidizing activity actually travels. Potassium iodide test strips were placed at the device’s air outlet and 12 meters away, and read after 8 hours of continuous running.

Both strips oxidized. The control strip did not. A control strip left in the same aerated room with the devices switched off showed no reaction at all after 8 hours, which is what rules out the room itself as the cause. Both cartridge types — hydrogen peroxide and d-limonene — produced the same result.

By color-matching against peroxide solutions of known purity, the laboratory put the unit’s average oxidative capacity as equivalent to hydrogen peroxide at 0.5% purity.

This is a qualitative colorimetric test, not a quantitative measurement, and the report says so. What it demonstrates is reach: measurable oxidizing activity 12 meters from the unit, where a filter only ever treats the air drawn into it.

Monagas Asensio, P. & Falcó Grasa, M., approved by Josep Garcia Raurich. Laboratory CRESCA-UPC, Terrassa, Barcelona, 24/03/2020 — read the full report (PDF).

Half a stick of incense was burned in a 225 m³ chamber and the air measured at 30 minutes, 1 hour and 2 hours — against a control run with no device at all, so the figures are a comparison at the same moment rather than a before-and-after.

At 30 minutes, with the d-limonene cartridge:

Total VOCs — 139 µg/m³ against the control’s 2,432. A 94.3% difference.
Formaldehyde — 33 µg/m³ against the control’s 500. A 93.4% difference.

With the hydrogen-peroxide cartridge the figures were lower again (104 and 25 µg/m³), but the report is careful to say the gap between the two cartridge types is not statistically meaningful, and that running both devices at once was worse than running either alone. The control itself falls sharply after the first reading, because a room that size disperses smoke on its own — which is why the comparison is made at matched times.

A laboratory test on the unit, not a peer-reviewed study.

Salmerón Delgado, A. M. & Picón Torres, J., approved by Josep Garcia Raurich. CRESCA, Universitat Politècnica de Catalunya, report 20190930, Terrassa, Barcelona — read the full report (PDF).

This is background science, not a test of the WellisAir unit. It is a review article by atmospheric chemists at New Zealand’s National Institute of Water and Atmospheric Research, explaining how hydroxyl radicals clean the outdoor air — the natural process our technology reproduces indoors.

Two points in it matter to how WellisAir works:

The hydroxyl takes the hydrogen. In the authors’ words, when a trace gas contains hydrogen atoms, “OH reacts by taking one hydrogen atom” — forming water and leaving an unstable fragment that reacts onward. They give methane as the example: CH₄ + OH → CH₃ + H₂O.

The pool is recycled, not spent. “In the atmosphere, OH radicals are also regenerated during the various reactions, so very little is used up overall.” That is why a molecule which survives only moments can keep working on a room.

Paul Crutzen, who won the Nobel Prize for his work on atmospheric ozone, called OH the “detergent of the atmosphere.”

Riedel, K. & Lassey, K. (2008). “Detergent of the atmosphere”. Water & Atmosphere 16(1), NIWA — read the article (PDF).

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.

Toluene comes off paint, adhesives and solvents; formaldehyde off pressed wood and new furnishings; ammonia off cleaning products and litter trays. These are laboratory tests on the WellisAir unit, not peer-reviewed studies.

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

Intertek measured the unit’s ozone output across its operating range:

0.002–0.006 ppm.

California limits an air cleaner used in an occupied space to 0.05 ppm, tested to the UL 867 method, so the top of that range is about one eighth of the limit. The unit is CARB certified.

Intertek, report 103502990CRT-001 — what was measured, and how.

Peer-reviewed clinical study, Medicina Oral Patología Oral y Cirugía Bucal, 2023. A WellisAir WADU-02 ran for five days in a working 10.9 m² dental clinic with normal cleaning continuing. Airborne molds and yeasts fell 66.7% (p = 0.03).

Read the study →

Every report above was produced by a laboratory or research group we do not own — Korea Conformity Laboratories, Intertek, and CRESCA at the Polytechnic University of Catalonia — and the dental-clinic result was peer reviewed and published independently of us. CARB certification and ETL listing are third-party as well.

What we claim and what we deliberately do not — including the absence of any bacteria or virus claim — is set out on Concerns and common questions.