31 December 2021 by Jennifer
Explanation of an ingenious thing: how you can use a CO2 monitor to check the air-freshness in a room, to guide your covid risk reduction.
I looked into this a while ago and I thought I’d do a write-up!
Wouldn’t it be useful if we could somehow measure “Has this room got any covid particles hanging in the air at the moment?”?
Unfortunately, we can’t yet do that in real-time. When scientists wanted to measure the covid levels in hospital ward air, they had to gather the samples first, then take them back to the lab and do some extra processes to find out what they’d actually gathered. They commented “The sampling and detection of airborne viruses poses several technological challenges” :-)
On the other hand, measuring carbon dioxide in the air is something people have been doing for ages. And that can give us a clue about what the covid risks might be in a particular space.
Carbon dioxide is one of the gases which we naturally breathe out in our breath. We breathe in oxygen, our bodies make use of it, and we breathe out carbon dioxide.
You’ll often hear it referred to as CO2, “see oh two”. The C means “carbon atom”. The O2 means “oxygen atoms, two of them”. They link up to make a molecule with the carbon atom in the middle. (The “di” in “dioxide” means two, as well.)
If CO2 builds up in a room, we sometimes sense that as the room becoming “stuffy”. If you ever felt sleepy in a stuffy room, then more refreshed after someone opened a window… part of that was probably the fresh oxygen coming in, and the breeze sweeping away some of the built-up CO2.
As you might know by now, someone currently infected with covid can breathe out covid virus on their breath. The teeny tiny particles of breath-liquid, with even tinier specks of virus in them, can hang in the air for some time. In still air, they might hang about for a few hours – though obviously they’d get blown away much quicker than that if you were outdoors.
Then the next person breathes in that same bit of air, and that’s probably the number one way that covid gets from one person to the next.
Fresh air will tend to sweep away any germs that are floating around – either from someone in the room, or from someone who already left the room a while ago.
(Even outdoors or in a well-ventilated room, you could still pick up germs from the breath of people close to you, same as you’d smell the smoke if someone was smoking right next to you. That’s why we still need good, face-fitting masks sometimes as well as fresh air. But the fresh air really helps.)
The ingenious idea was: OK, we can’t measure covid itself. But both covid virus and CO2 molecules float in human breath! So we can measure how much carbon dioxide is building up in a room, or bus, or train carriage, or hospital ward, as the people there are breathing. And that would give us a clue to how much covid virus would typically build up in that space, supposing an infectious person were there.
If a lot of CO2 is building up, it’s a clue that a lot of covid virus could build up in that room too, if anyone there was infected. And the other way round: if the CO2 level is low while people are there, then plenty of fresh air must be coming in to carry away what the humans are breathing out.
It’s what in science terms would be called a “proxy measure“: not exactly the same thing as what you’re really interested in, but related (at least some of the time) in a useful way.
It’s a dinky little thing – about 7cm by 7cm on its front face, similar to the palm of a smallish hand. It would easily fit in a pocket, and has a hole in the back where you can hang it over a screw-head on the wall.
Via Bluetooth, you can link one or more units to your phone, and view the results in an app which you can download for free. The app can memorise 7 days’ worth of how the levels went up or down, for you to look at afterwards.
Cost of that Aranet one is €199.00 plus taxes & delivery: expensive for the average family, but not enormous compared to running a school, a hospital or a business.
There are cheaper ones too. If you’re getting one, make sure it’s the “nondispersive infrared” type, also known as NDIR.
This kind of thing is what the UK government has promised to UK schools, although I don’t know if they’re buying the Aranet or a different model.
The CO2 level in air is measured in “parts per million”, often written “ppm”.
(Little devices like this won’t measure it to an exact number – it’ll be more like “give or take 50ppm”, which is good enough for ventilation monitoring.)
Outdoor air would typically read at about 400ppm to 450ppm, or higher if you’re near a busy road.
What’s a “good” level?
A study looking at a tuberculosis outbreak found the outbreak stopped after the ventilation was improved enough that the CO2 level went down to 600ppm. Getting it below 1,000ppm also made a significant difference to the chance of infection. (Summarised here.)
Tuberculosis and covid both transmit via floating in the air, so we could make a guess that similar levels might be good for covid too. So it would make sense to say 600ppm is good to aim for. But if you can’t get it that low, 700ppm is better than 800ppm, 800ppm is better than 900ppm and so on.
The excellent FAQs on Protecting Yourself from Aerosol Transmission, by a group of top professors, summed up like this:
The exact level considered “safer” for CO2 varies and we have seen various recommendations from 500 to 950 ppm.
Choosing a general level like this is a compromise to make the method feasible and simple enough for many people. This is the same reason that a single distance (e.g. 1 m or 2 m) is quoted for social distancing (even though we know that 1.5 m is better than 1 m, and 2 is better than 1.5 m etc.).
A key goal is to make clear that the many shared spaces with 2000 or 3000 ppm CO2 are unsafe, so that people realize that they have to take action to improve the situation there.
Surveying classrooms, offices etc. with a CO2 monitor can be useful to determine which ones may have the worst ventilation, and prioritizing our actions there.
(paragraph breaks added by me for ease of reading)
They also explain:
- ~800 ppm, 1% of the air you are breathing has already been breathed by someone in the space. This can start to be risky.
- ~4400 ppm, 10% of the air you are breathing has already been breathed by someone else. This is a very dangerous situation. Levels this high are commonly observed in densely occupied spaces with low ventilation such as many schools.
Aside from infection risks in stale air, there’s been a lot of research into how CO2 itself affects human functioning. Levels over 1,000ppm are associated with things like headaches, asthma and not thinking as clearly. For example,
Thirty male active commercial airline pilots performed three 3-h flight simulations on the flight deck at 700, 1500, 2500 ppm CO2. … the pilots were more likely to successfully perform five of the seven most difficult maneuvers at the lower CO2 concentration.
In the experiment with the pilots, only the CO2 level itself was changed. In similar experiments, the dozy effect wasn’t always narrowed down to CO2 in particular: if you’re just looking at “stale air” overall, then the effects can also be from other molecules building up in a room over time, e.g. from paint or furniture. But either way, fresh air was better.
So this kind of research had already made pretty clear that fresh air helps with health and thinking. Places like schools should arguably aim for below 1,000ppm of CO2 for that reason, even aside from covid risk.
Of course, finding out the air is “too stale” doesn’t actually fix the air! It’s just a clue.
The meter can be useful for answering
- Would it be a good idea to open the window for a while?
- Opening the window in here is definitely enough when there’s a breeze, but what about on a day when the air is still?
- Does this classroom need an air filter as well, because opening the windows isn’t doing enough?
If you can’t get a monitor to use every day, you might at least try to borrow one for a trial sometimes, to get a sense of how well-aired your space typically is.
In some contexts, the CO2 information can enable people to choose whether to enter the space at all. For instance, this photo from @NOGjp shows a real-time display board in the entrance lobby of a cinema in Japan, in July 2021. Each segment of the chart represents the CO2 level in one of the cinema rooms.
Customers can be reassured about the air quality before they go in – or, if the level went higher and they didn’t think it was good enough, “vote with their feet” and skip it.
Some destinations are now showing their CO2 levels online, so you can check the air quality before you decide to even go there in the first place. The more shops & venues do it, the more it becomes a form of advertising, where the air quality is an attraction to some customers.
Supposing you’ve got a CO2 monitor, and you want to use it to assess the room you’re in. What practical factors might be useful to keep in mind?
- The reading might be different in different parts of the room, depending on the air flow, so you probably want to measure in a few different places. There can be “dead spots” for air movement at points you wouldn’t expect, like near a window.
- Keep the meter away from your face, so your breath doesn’t confuse it into giving a wrongly high reading. (50cm should be enough of a distance.)
- The freshness level is going to vary depending on how many people are using the space. What seems like “enough” ventilation when there’s three of you might not be enough for 10 or 30. If the room would normally hold 30 people, it probably makes sense to do the measuring while 30 people are in it.(People who do this kind of testing on a regular basis get around this limitation by bringing their own supply of carbon dioxide. They can artificially add CO2 to an empty room for the test – then monitor how quickly or slowly it gets aired away.)
- For ventilation via windows, the weather on the day will make a difference. As well as how breezy it is outside, there’s the effect of warm air rising and cold air sinking. If your room is warm and outdoors is cold, then opening a window will mean that some warm air rises up-and-out, and some cold air comes in-and-down, creating a bit of natural airflow – whereas if the temperature outside is the same as inside, you don’t get that effect.
- Obviously, opening a window in cold weather loses some of the heat from your warmed-up air! But you may find you don’t need the window open very long to do a swoosh of air which sweeps away the staleness. So even if you don’t want to leave the window open, it’s still worth opening it from time to time, and experimenting with how much benefit you can get without getting too cold.
- If opening windows and doors in your room isn’t doing the trick, then perhaps opening a window in another nearby room will help to get some air moving through the building. (But this might not be a good idea if there are unmasked people elsewhere in the building, whose germs you don’t want flowing through your room!)
- In vehicles, you may find you get a very different read-out depending on whether the vehicle is moving, or whether the engine is on. Some airflow relies partly on the movement of the vehicle; air conditioning relies on power to run at all. In a parked car with the engine off and the windows shut, the air can become stale quite fast.
- The CO2 meter only knows about CO2. So there’s one more aspect to keep in mind – or, preferably, simplify out – which I talk about in the next section.
To use CO2 as our proxy measure, we’re relying on a link between “human breath” and “CO2 in the room”. If the CO2 went up or down for a reason which was nothing to do with humans breathing out, then obviously that’s not telling you about covid risks.
One example is that anything with an open flame will be adding to the CO2 in that room. An open fire, or a gas cooker, could contribute to the room perhaps feeling stuffy, and you feeling dozy – but wouldn’t be adding to any covid risk, because fires and cookers can’t be breathing out covid virus.
Another variable would be: Suppose you have a good air filter running in your room, so that any virus in the air would be partly caught by the filter. But the filter doesn’t catch CO2. Then the covid risk would be lower with the filter turned on – but that difference wouldn’t show up on the CO2 meter.
Or if you’ve got loads and loads of plants, then those might make the CO2 level a bit lower (because plants take in carbon dioxide and give out oxygen). But plants won’t be changing the level of covid virus. So they wouldn’t make the covid risk lower, even though you saw a lower number on the meter.
Each of these factors disrupt the neat connection between “human breath as a whole” and “CO2 in particular”, which we were using as our shortcut.
This means that to use the CO2 measurement as a clue to the covid risk, you have to take note of whether there are any other factors at play which affect either the CO2 level or the covid risk.
In some situations, it’s possible to intentionally do another measurement without any of the “extra” things happening as well, to keep it simple.
A very cool thing about sorting out air quality in our buildings is that it will also improve lots of other things besides reducing covid transmission: better thinking, fewer headaches, and even probably a reduction in other illnesses which transmit via the air.
The eleven professors who wrote the FAQs on Protecting Yourself from Aerosol Transmission recommend “Every public place should display CO2” – like the cinemas and shops in Japan. I approve of that idea. Real-time public visibility could go a long way to encourage better air quality!
- CO2 measurement page at Clean Air Crew, including a couple of videos, and links to lots of other pages: ventilation, filtration etc etc.
- How to use ventilation and air filtration to prevent the spread of coronavirus indoors, by Professor Shelly Miller.
- DIY Air Filters For Classrooms? A good short briefing on how and why to build a Corsi-Rosenthal box – a DIY portable air filter, typically cheaper than a pre-made one. (And see next couple of links if you’re in the UK and want to build one.)
- The DIY Air Filter movement comes to the UK. Adapts the Corsi-Rosenthal design to make use of filters which are easier to get hold of in the UK. By Stefan Stojanovic at Parents United.
- Materials suitable for UK Air filter design. Goes with the previous link.
- FAQs on Protecting Yourself from Aerosol Transmission. Very good explainer by scientists I consider reliable. (It’s the same document which I quoted above about recommended levels to aim for.)
- Indoor aerosol science aspects of SARS-CoV-2 transmission, by William W Nazaroff. Long, geeky scientific paper, explaining lots of the background science of how covid transmits through the air.