Do Houseplants Clean Your Air?
When I was younger, I scoffed at people who talk to their plants.
I scoffed at my hippy-dippy friend and her philodendron Fred. (Fred? Why not Phil?) I scoffed at new-age commentators on TV. Later. I scoffed at King Charles, who remarked in 2010, while still a mere prince, that "I happily talk to the plants and trees, and I listen to them."
Two discoveries have made me stop scoffing.
First, talking to plants does seem to help them grow. At least two studies demonstrate this. According to researchers, it's not the love and encouragement in what you say, or even the carbon dioxide you exhale, but rather the vibrations from your voice that help. So, yes, talk to your plants – or play music for them. As long as the volume isn't too high, they'll flourish.
Second, I discovered this week that Charles also spoke of how talking to plants helps keep him "relatively sane." If such a benign activity can promote mental health, I've got nothing bad to say about it. The monarchy needs as much help as it can get.
This newsletter follows the same trajectory as what I've written so far: First I'm going to scoff about a new study claiming that houseplants can reduce the toxicity of indoor air. I'm going to scoff about media treatment of the study as well as the study itself. Then I'm going to take much of it back.
In spite of numerous red flags, the study turns out to be credible and potentially important, and it's a good illustration of how statistics can play a supporting role in scientific progress. I do have concerns about the findings, but it would take more than a scoff to dismiss them.
Some initial scoffing
The new study didn't break the internet, but it was covered this week by Fox, Yahoo, The Weather Channel, and a number of less popular websites and Twitter feeds, which is to say that millions of people may have heard about it.
What the study purportedly shows is that houseplants can remove carcinogens from gasoline fumes. According to media reports, the study is "groundbreaking" and suggests that buying more houseplants can de-toxify the air and reduce your risk of cancer.
When I first read this, I was excited. Indoor air can trap gaseous pollutants and create pollution levels several times higher than the air outside. The air filters in your HVAC system remove particulate matter, but not toxic gases. Meanwhile, the WHO estimates that poor indoor air quality is responsible for more than 6 million premature deaths per year. The idea that simply adding a few plants to your home or work space could improve the air quality is quite appealing, so, I decided to look at the actual study.
What I found, at first, was a bunch of red flags.
Some red flags
The "study" was conducted by researchers at University of Technology Sydney in collaboration with Ambius, an Australian company that specializes in indoor plantscaping. It's not a published study. The only publicly available information about it is located on the Ambius website and in a statement to the press. On the Ambius website you can download a "report" that turns out to be a one-page graphic consisting mainly of background and hype. This one-pager actually tells you less about the study than the Ambius website does, though there isn't much there either. In the end, all you can get are a few blurbs.
At this point, I began to scoff. Here the media has reported on a "study" that we know almost nothing about except through the company that funded it and stands to profit from the findings. None of the media reports noted that this so-called study amounts to nothing more than a few blurbs on a company website, accompanied by glowing hyperbole from a manager at Ambius and a press release.
Statistical red flags
Without knowing more about the study, could it be credible? Maybe. Following a 1989 NASA report, a number of studies have shown that houseplants can reduce air pollution. Crudely speaking, toxins pass through the plant into the root system, where bacteria in the rhizosphere (the soil surrounding the roots) break the toxins down into non-toxic chemicals.
However, researchers have poo-poohed the idea that buying more plants for your living room could meaningfully impact air quality. Why not? The consensus seems to be that plants don't clean the air quickly enough. Prior studies have focused on plant performance in small, airtight chambers, which don't reflect real-life conditions in which air is continuously moving in and out of larger spaces (e.g., the room you're sitting in right now). If plants instantly sucked toxins out of the air, they would be great at cleaning the spaces where we live and work. Instead, they take many hours to clean a room. Meanwhile, more polluted air continues to seep in.
Here's where statistics and other kinds of math play a role. Researchers can calculate something called the clean air delivery rate (CADR), which is then used to estimate how much pollutant a given plant will remove from a given space over a given period of time. (I describe the math in the Appendix if you're interested.)
CADR estimates tell us that in order for houseplants to meaningfully improve the quality of air in a room, a ridiculous number of plants would be needed. One estimate determined that for a typical 1,500 square foot house, 680 potted plants would be needed to achieve a tiny but measurable increase in air quality. That was charitable. The co-author of a 2020 review estimated that a substantial improvement in air quality would require about 10 potted plants per square foot. So, if you live in a 1,000 square foot apartment, you'd need 10,000 plants. Welcome to the jungle.
At this point, with so much to scoff at, I was on the verge of moving on, but I decided instead to reach out to Dr. Fraser Torpy, the lead author of the new study, in case I was misunderstanding anything. It turns out I was. Thanks to several emails from Dr. Torpy, and some further digging, my scoffing has been replaced by appreciation for the new study – tempered though by lingering concerns about the methodology and findings.
The new study
According to Dr. Torpy, the new study is under review, but Ambius decided to promote the work anyway, because, as he noted, industry is always in a hurry.
Dr. Torpy, as it turns out, has been publishing studies on bioremediation of air pollution for over two decades. As far as I can tell, the science is reputable and the new study is an extension of prior work that he and others have conducted.
What's particularly groundbreaking about this study is the focus on gasoline exhaust, as opposed to other pollutants, and a consideration of more than just one or two of its toxic constituents.
Gasoline fumes contain many volatile organic compounds (VOCs), and many of those VOCs are toxic. Prior studies showing that plants can remove VOCs from the air focused on one or two VOCs at a time. The new study examined multiple VOCs in gasoline fumes simultaneously. Here's how Dr. Torpy described it in an email to me:
"We (and others in the field) have previously tested single VOCs, or once, pairs of VOCs. We discovered that there is an interaction effect: exposure to one VOC can enhance or delay the removal of another. So in real world spaces VOC phytoremediation could be very different from predictions made on single VOCs. Thus we tested the simultaneous removal of the VOC mixture of primary health concern: gasoline vapour."
The simultaneous focus on multiple VOCs is an important step, because, ultimately, your concern should not be that one or two VOCs from gasoline exhaust seep into your room, but rather that they all do.
Study methods and findings
Dr. Torpy and colleagues made use of the Ambius Live Picture Go, which is essentially a plant in a bamboo frame that can be hung on the wall, like a picture. The more Live Picture Gos you hang, the larger the "green wall" you create.
For this study, Dr. Torpy and his team introduced gasoline fumes to a sealed chamber containing a wall of Live Picture Go plants. (This "wall" was not the size of the one in your room; rather, it was one side of a chamber that could fit on the top of an office desk.)
To create gasoline vapor, 250 microliters of gasoline compounds were mixed together and placed in an 80 degree bath inside the chamber and allowed to volatilize. The amount of each VOC in the chamber's air was measured hourly by means of a gas chromatography/mass spectrometry (GC/MS) instrument, which, in one of his emails to me, Dr. Torpy described as an innovative move.
After an 8-hour period, almost all of the most toxic VOCs had been removed. For instance, the percentages removed for the most dangerous VOCs were as follows:
–Alkanes: 97.9%
–Cyclopentanes: 88.18%
–Benzenes: 85.96%
If you're into gardening, the plants used were pothos, syngonium, and spider plant – all readily found at your local nursery (and easy to grow).
Concerns
The findings of the new study sound quite impressive, but there there are still causes for concern. Once again, the plants were tested in a small, airtight chamber, and the effectiveness statistics describe what happened at the end of an an 8-hour period.
The fact that the plants removed over 85% of the most dangerous toxins found in gasoline fumes is great news. However, it took them 8 hours to do so. In an actual room, fumes would continue to seep in while the plants are doing their work, meaning that they won't be effective unless they can keep up.
As I mentioned, prior research suggests that plants work so slowly, you'd need to convert your room to a jungle in order to experience better air quality. On the Ambius website, nothing is stated about how rapidly the plants removed gasoline pollutants. We know they performed well after 8 hours, but assuming the fairly linear effects found in other studies, that's too slow to be of much use.
In Dr. Torpy's first email to me, he noted that prior research, including a 2019 study that he co-authored, predicted slow removal of benzene, but in the new study "we found quite the opposite." That's a tantalizing comment, so I asked whether he'd be willing to share his CADR data or at least explain what the benefits of that more rapid removal rate would be. [Update: In an email sent after this newsletter was published, Dr. Torpy stated that his team did not calculate CADR rates, and he did not explain in what sense the new data showed more rapid benzene removal.]
Yesterday I reached out to another expert in the field to get his views on possible removal rates. This expert, who asked to remain anonymous, gave me permission to paraphrase but not quote his response. He also noted that he was unfamiliar with the new study.
The gist of what the expert told me is that Dr. Torpy's work is good science, at that it's plausible that removal rates for benzene derivatives and other compounds were more rapid than in prior studies. However, the diffference is a relative one and may have limited practical significance. Plants still work slowly, even when supported by technologies ranging from additions to the soil to fans that direct airflow.
In the end, the expert's view was that you'd probably still need quite a few plants to reduce gasoline exhaust compounds in your indoor air.
Conclusion: What can you do?
As I've mentioned in other newsletters, studies on the health effects of variables such as diet are inherently messy, thanks to the variety of what we eat, the range of other activities that affect our health, and the difficulties in obtaining precise measurements of dietary intake.
The houseplant studies circumvent these problems by using small, airtight chambers and focusing on limited numbers of pollutants. This enables rigor, but the data may not generalize well to real-world conditions, which are vastly more complex.
What the researchers hope is that by studying key variables under tightly controlled conditions, we can learn enough about them to make predictions as to what will happen under more realistic circumstances.
For this reason, I wouldn't discount the results of studies that are rigorous to the point of artificiality. Houseplants clearly detoxify the air. Whether you benefit from the plants you have in your particular room is uncertain. Probably not, but maybe.
This raises the question of what you can do, besides purchasing more plants. I asked Dr. Torpy how to maximize the benefits of a "green wall". Here is his response:
"Healthy plants are essential to maintain the microbial community, so the right species for the light level, humidity, substrate watering method are important. Secondly, airstream exposure to the substrate has a very major effect on IAQ [indoor air quality] management: this is why the active green walls we have tested...are so much more effective than passive systems. Putting a GW [green wall] in a location where it receives maximum exposure to air flowing in a room will make a lot of difference!"
Based on a combination of what Dr. Torpy wrote, published research on the topic, and my own opinions, here's what I would conclude:
1. One or two potted plants probably won't produce a net improvement in the air quality of your room. (They may look nice though. And, if you talk to them, they'll grow more rapidly.)
2. A substantial number of houseplants may produce a net improvement in air quality. The plants should be healthy and receive as much air flow as possible. How to define "substantial" is unclear, but the more the better. Give uncertainties in the literature, I wouldn't discount the value of adding a few strategically-placed plants to a room where you spend a lot of time.
3. Commercially available "green walls" may be even more beneficial. How much more is difficult to say.
Here's one more thing you can do:
4. Be cautious about statistics in the news.
Apart from relying uncritically on blurbs posted by the funders of the new study, some journalists lauded the removal of 97% of gasoline toxins. This is misleading, because the removal rate was only that high for one class of toxins, the alkanes, and, as I've said, 8 hours was needed to reach that level, which may not be rapid enough to have practical benefits (unless you've made your room a jungle).
Even Science Daily, a fairly reputable source, cites that 97% figure out of context, and quotes an Ambius manager as saying that "we see over and over again the effects plants have in improving health, wellbeing, productivity and office attendance for the thousands of businesses we work with".
Ambius may believe that the plants they sell make their clients more healthy, but I wouldn't count that as objective evidence of anything other than the fact that people tend to see what they're motivated to see.
In the end, all we have so far with respect to meaningful reductions in gasoline fumes is a definite maybe. Perhaps that's reason enough to maintain houseplants.
Thanks for reading!
Appendix: CADR calculation
The clean air delivery rate (CADR) can be used to estimate how many plants are needed to improve the air quality of a particular room by a particular amount. At present, it's not something you can calculate at home.
The formula for determining CADR is
CADR = E x R
E refers to how much pollutant is removed by a filtration system (e.g., a plant) after one passage of air through the system.
(The technical term for E is single pass removal efficiency, or SPRE, which is (I - O)/I x 100, where I is the original concentration of pollutant, and O is the concentration after air has passed through the filtration system.)
R is the rate of airflow through the filtration system.
Let's imagine one particular houseplant, which I'll call Fred. A CADR value is a way of describing how efficiently Fred can can remove a pollutant from the air. CADR values can be calculated for any pollutant or pollutants that Fred is exposed to.
If you divide a particular CADR by the volume of a given room, you obtain a number called the biofilter refreshment capacity, or BRC. Thus, given the volume of the room you're sitting in, the BRC would describe how efficiently Fred removes any given pollutant from the air in the room.
Further refinements can be added if we know how large Fred is, how long the pollutants linger in the room, etc., but in the end, the estimates will still be crude. They're based on a single pass of polluted air over Fred, and they can only roughly approximate real-life fluctuations in outdoor pollution, the rate at which air enters and leaves the room, the effects of opening windows and doors, changes in temperature and humidity, and so on.
From the perspective of residents and indoor workers, what's important about CADR calculations is the hint that for a given room, some critical mass of houseplants might genuinely improve the air quality. It's still unclear how "jungly" that room would need to be.