The claim comes with a NASA logo attached, which is most of the reason it has survived for more than three decades. Keep a few spider plants and a peace lily, the story goes, and they will pull the formaldehyde out of your living room.
The study behind the story is real. It was carefully done, by actual NASA researchers at the Stennis Space Center in Mississippi, and you can download the whole thing today for free.
It also has almost nothing to say about your living room. Not because the scientists got anything wrong, but because of some arithmetic that nobody did until 2020.
This article is about that gap. If you love your plants, keep reading anyway. Nothing here is a reason to get rid of them.
What the 1989 NASA report tested
The document is a NASA Technical Memorandum, number 101766, titled Interior Landscape Plants for Indoor Air Pollution Abatement. B.C. Wolverton, Anne Johnson and Keith Bounds published it on 15 September 1989.
Read the methods and the whole thing snaps into focus.
The experiments ran in four sealed Plexiglas chambers. Two measured 30 inches on each side, and two stood about 60 inches tall, which works out to somewhere between roughly 0.4 and 0.9 cubic metres of air. The lids were bolted down with wing nuts to make them airtight. One plant went in each box.
Then the researchers injected a single chemical: benzene, trichloroethylene or formaldehyde, tested one at a time rather than together. They sampled the air at the moment of injection, again at six hours, and again at 24 hours. Controls included empty sealed chambers to test for leaks, pots of soil with no foliage at all, and a pot whose soil surface was sealed under pea gravel.
Twelve species went through the main screening, including bamboo palm, Chinese evergreen, English ivy, peace lily and mother-in-law’s tongue. These are the plants that still appear on every NASA air-purifying plant list on the internet, nearly four decades later.
The concentrations are the first clue
The screening tests ran at roughly 15 to 20 parts per million, and one table in the report shows initial benzene as high as 65 parts per million. A part per million means one molecule of the chemical for every million molecules of air. A separate set of runs used under 1 part per million.
For scale, the Environmental Protection Agency reports that concentrations of many volatile organic compounds run up to ten times higher indoors than outdoors, and that one of its own exposure studies found about a dozen common organic pollutants at two to five times outdoor levels inside homes.
The report says so itself. It notes that the high-concentration results sat far above the levels commonly found in indoor air. The authors flagged their own limitation. Everyone who repeated the headline skipped that sentence.
The system was not a houseplant
Here is the detail that gets lost most often. The thing Wolverton’s team was proposing was a device, not a pot on a windowsill.
A fan pulls room air through a bed of activated carbon, the carbon traps the pollutant molecules, and the plant’s roots and the microorganisms living with them break down what the carbon has captured. The carbon does the catching. The biology does the digesting. The plant was the living half of a machine, not the machine itself.
And the application was specific. The report discusses energy-efficient sealed buildings, and it also discusses NASA’s interest in closed habitats such as a space station or a moon base. In a sealed habitat, the air is not exchanged with anything. Ever. Removing a molecule from that air is the only way it leaves.
That is not bad science. That is over-extrapolated science, and the distinction is the whole point of this article. The chamber and the spacecraft are the same kind of place. Your house is not.
Why the box matters more than the plant
In a sealed chamber, a benzene molecule has exactly one exit, which is the plant and its soil. Given 24 hours and nowhere else to go, a measurable fraction of it will leave by that route.
A house works nothing like that. Air leaves constantly, through gaps around windows, through vents, up the chimney, out the bathroom fan, and every time someone opens a door. That flow is described as air changes per hour, meaning how many times the full volume of air in a room is replaced in an hour. A figure of about one air change per hour is commonly used for typical buildings.
So the question is whether a plant removes anything fast enough to compete with a building that is already flushing itself out.
There is also a quieter finding worth knowing, because it tells you where the action is. A 2023 study in Environmental Science and Pollution Research exposed potted English ivy to gasoline vapors in a flowing chamber for 21 days and tracked seven target compounds. The plants cut concentrations of most of those compounds by 25 to 32 percent, and the researchers’ focus was the bacterial community in the soil, which shifted measurably in response.
In other words, most of the removal credited to houseplants is being done by microbes in the potting medium, not by the leaves. That is interesting biology. It is also why a sealed box with a pot in it behaves so differently from a room with a plant in the corner.
The arithmetic somebody finally did
In 2020, Bryan Cummings and Michael Waring of Drexel University published a review in the Journal of Exposure Science and Environmental Epidemiology with a title that leaves little room for interpretation: potted plants do not improve indoor air quality.
What they did was translation, and it was overdue. They collected 12 published chamber studies and converted 196 individual experimental results into a single common unit: clean air delivery rate, or CADR. That is the metric used to rate portable air cleaners, and it means the volume of air per hour that is effectively stripped of a pollutant.
Putting plants and air cleaners into the same unit is what made the comparison possible, and nobody had done it.
The result: the median CADR for a single plant came out at 0.023 cubic metres per hour. That is 23 litres of air an hour, per plant, which is roughly the volume of a small kitchen bin. Once an hour.
Their conclusion follows from the number. To match the pollutant removal that ordinary outdoor-to-indoor air exchange already provides in a typical building, you would need somewhere between 10 and 1,000 plants per square metre of floor space.
Run it on a real room
Numbers like that stay abstract until you put them in a room you can picture. So take an ordinary living room: 12 feet by 15 feet, which is 180 square feet, with an 8-foot ceiling.
That is about 16.7 square metres of floor and about 41 cubic metres of air.
At the low end of the published range, 10 plants per square metre, that room needs roughly 167 plants. At the high end, 1,000 per square metre, it needs about 16,700.
Work it from the other direction and you land in the same place. At one air change per hour, that room is already receiving about 41 cubic metres of outside air every hour. Divide that by 0.023 and you get roughly 1,800 plants to do the equivalent work. That calculation is simple arithmetic on the published median, and it is worth doing yourself with your own room’s dimensions.
One hundred and sixty-seven plants is a dense green wall covering most of a living room. Eighteen hundred is a greenhouse. Six pots on a windowsill is not a rounding error on that figure. It is not on the same chart.
The American Lung Association reached the same conclusion in a post first published in February 2024 and updated in October 2025, stating flatly that “the verdict is now in: houseplants do not improve air quality.” The EPA says much the same thing in its own guidance on improving indoor air quality, noting that there is no evidence a reasonable number of houseplants removes significant amounts of pollutants from a home.
If cleaner indoor air is the goal, the order that works is source control first, then ventilation, then filtration, and each of those has its own guide on this site. Plants are not a step on that list.
The other side of the ledger
A fair accounting does not stop at the air-cleaning claim. Plants also do a few things that nobody puts on the care tag. None of these is a reason to panic. All of them are a reason to water more carefully.
Overwatered pots grow mold. Mold needs moisture and almost nothing else. The CDC advises keeping indoor humidity no higher than 50 percent throughout the day and says mold grows wherever there is water. A saucer that stays full and soil that never dries out is a small permanently damp spot in an otherwise dry house. If you can see fuzz on the surface of the potting mix, or the pot smells like a basement, that is the problem announcing itself.
Potting media is a living reservoir. A 2020 study in Environment International sequenced settled dust from 82 homes in Beijing. The bacterial communities in that dust were shaped mainly by factors tied to the occupants and to houseplants. The fungal communities tracked environmental factors and outdoor plant diversity instead. The fungal portion was largely made up of potential allergens, a median of 61 percent of the fungal sequences, with the genus Alternaria most common.
That is not a finding of harm, and it was not a US sample. It is a finding that plants are part of the indoor microbial picture rather than separate from it.
Plants emit volatile compounds of their own. This is the part the plant-air-quality literature is least settled on. Cummings and Waring explicitly listed possible negative effects from other plant-sourced emissions as a question for future research, which is a scientist’s way of saying the net ledger has not been worked out. Flowering indoor plants also produce pollen, which matters to a small number of people and not at all to most.
When to involve a clinician
This is a small section because the situation is uncommon, not because it is unimportant.
If you have asthma, a mold allergy, chronic lung disease or a weakened immune system, the CDC names you among the people most affected by damp, moldy indoor environments. In that situation, a cough or wheeze that eases when you leave the house and returns when you come back is worth raising with your doctor, and worth mentioning alongside anything in the home that stays wet, plant pots included.
The CDC does not recommend testing your home for mold, because no agreed standard exists for an acceptable level. Find the moisture and fix it instead. For a plant, that usually means emptying the saucer, letting the top of the soil dry between waterings, and repotting if the medium has gone permanently soggy.
What plants seem to do for people
Here is where the Drexel researchers themselves pointed. Having dismantled the air-cleaning claim, Cummings and Waring called for future work on uptake mechanisms, alternative biofiltration, and “biophilic productivity and well-being benefits.”
That last phrase is the one worth following, and it deserves reporting at its real strength rather than at headline strength.
The study that bears most directly on this is also the most sobering. A 2023 field study in Frontiers in Psychology placed plants in nine Dutch organizations, each with a control workspace and an intervention workspace, measured before and after, with the final measurement at least four months after the plants arrived. The researchers tracked 18 outcome variables.
Five of them moved. People reported fewer complaints about dry air, a greater sense of privacy, a more attractive workspace, more satisfaction with the workspace, and fewer health-related complaints specifically while at work. The other 13 did not budge. The outcomes at the far end of the causal chain, including mental health, absenteeism and job satisfaction, showed no measurable effect at all.
Smaller experimental work points the same direction. A 2025 study in Scientific Reports put 41 participants through a demanding working-memory task, once in a room with a medium-sized living wall of plants and once with a shelf of objects instead. Heart rate variability indicated more physiological relaxation with the plants present, and people rated the room more favorably. Positive emotions rose only in the real room, not in the virtual-reality version.
Forty-one people, two sessions each. That is a real finding and a very small one.
A 2024 literature review in Annali di Igiene surveyed the wider greenery-and-health literature and reports correlations between exposure to indoor and outdoor greenery and better mental health, lower stress, and improvements in measures including blood pressure and concentration. It is a narrative review of mostly correlational research, which is a weaker design than the air-quality question eventually received.
So label this honestly. It is preliminary research. The samples are small, the follow-up is short, most outcomes are self-reported judgments about a room, and none of these designs can separate the plant from the attention and novelty that arrived with it. What shows up most consistently is that plants change how a space feels and how people rate it.
That is not nothing. It is also a different tier of evidence entirely from the air-cleaning claim, which was tested directly and did not survive.
What to do with your pothos
Keep it. That is the short version, and it is not a consolation prize.
The specific things worth changing are small. Empty the saucers so nothing sits in water. Let the top of the potting mix dry between waterings, which is better for most houseplants anyway and gives mold less to work with. Repot anything whose soil has gone permanently wet and sour. If someone in the house has asthma or a mold allergy, treat a smelly pot as a thing to fix rather than a thing to live with.
And keep one piece of vocabulary. Clean air delivery rate is the unit that made the plant question answerable, and it is the standard rating for portable air cleaners. The next time something is marketed to you on an air-quality claim, whether it has leaves or a power cord, that is the number to ask for. If the seller cannot give you one, you have learned something.
Sources
- Journal of Exposure Science & Environmental Epidemiology (via PubMed). “Cummings BE, Waring MS. "Potted plants do not improve indoor air quality: a review and analysis of reported VOC removal efficiencies." Journal of Exposure Science & Environmental Epidemiology, 2020. PubMed. DOI: 10.1038/s41370-019-0175-9.” doi.org
- National Aeronautics and Space Administration. “Wolverton BC, Johnson A, Bounds K. Interior Landscape Plants for Indoor Air Pollution Abatement. NASA Technical Memorandum 101766, 15 September 1989.” ntrs.nasa.gov
- US Environmental Protection Agency. “Improving Indoor Air Quality.” epa.gov
- US Environmental Protection Agency. “Volatile Organic Compounds' Impact on Indoor Air Quality.” epa.gov
- American Lung Association. “Actually, Houseplants Don't Clean the Air.” lung.org
- Drexel University. “Potted Plants Do Not Improve Air Quality.” drexel.edu
- Centers for Disease Control and Prevention. “Mold.” cdc.gov
- Environmental Science and Pollution Research (via PubMed). “Dela Cruz M, Svenningsen NB, Nybroe O, Muller R, Christensen JH. "Removal of a complex VOC mixture by potted plants-effects on soil microorganisms." Environmental Science and Pollution Research International, 2023. PubMed. DOI: 10.1007/s11356-023-26137-8.” doi.org
- Frontiers in Psychology (via PubMed). “de Vries S, Hermans T, Langers F. "Effects of indoor plants on office workers: a field study in multiple Dutch organizations." Frontiers in Psychology, 2023. PubMed. DOI: 10.3389/fpsyg.2023.1196106.” doi.org
- Scientific Reports (via PubMed). “Serra H, Zavattaro C, Eid M, et al. "Biophilic interventions in real and virtual environments reduce stress during cognitively demanding tasks." Scientific Reports, 2025. PubMed. DOI: 10.1038/s41598-025-23224-3.” doi.org
- Environment International (via PubMed). “Ding LJ, Zhou XY, Zhu YG. "Microbiome and antibiotic resistome in household dust from Beijing, China." Environment International, 2020. PubMed. DOI: 10.1016/j.envint.2020.105702.” doi.org
- Annali di Igiene (via PubMed). “Paniccia M, Acito M, Grappasonni I. "How outdoor and indoor green spaces affect human health: a literature review." Annali di Igiene, 2024. PubMed. DOI: 10.7416/ai.2024.2654.” doi.org