[Factual] Does SARS-CoV-2 transmit via aerosol particles that are too small to be filtered by masks? If so, to what extent?

Continuing the discussion from Someone please help me refute what this doctor is saying about covid:

Please fight my ignorance here - although I do not dispute that masks are effective, I thought the disease could spread via aerosol particles too small to be filtered by cloth masks, if not respirators. Am I factually correct and if so to what extent?

The CDC’s scientific brief on SARS-CoV-2 transmission reads in part,

Exposure occurs in three principal ways: (1) inhalation of very fine respiratory droplets and aerosol particles, […]


The infectious dose of SARS-CoV-2 needed to transmit infection has not been established. […] animal studies and epidemiologic investigations (in addition to those described above) indicate that inhalation of virus can cause infection […]

Citations are available on their webpage. Just as an example, the CDC’s cite # 9 contains this:

Our finding that the proportion of small respiratory droplets [the majority of particles exhaled in all subjects, as has been previously observed (23)] increased at the peak of COVID-19 infection in NHPs (Fig. 4) confirms a previously published observation (17) from the exhaled aerosol profile of a single COVID-19 positive human subject, and suggests that, at peak infection, there may be an elevated risk of the airborne transmission of SARS-CoV-2 by way of the very small droplets that transmit through conventional masks and traverse distances far exceeding the conventional social distance of 2 m.


Also from the CDC on the efficacy of cloth masks versus aerosol particles,

Multi-layer cloth masks block release of exhaled respiratory particles into the environment, along with the microorganisms these particles carry. Cloth masks not only effectively block most large droplets (i.e., 20-30 microns and larger) but they can also block the exhalation of fine droplets and particles (also often referred to as aerosols) smaller than 10 microns ; which increase in number with the volume of speech and specific types of phonation. Multi-layer cloth masks can both block up to 50-70% of these fine droplets and particles and limit the forward spread of those that are not captured. Upwards of 80% blockage has been achieved in human experiments that have measured blocking of all respiratory droplets, with cloth masks in some studies performing on par with surgical masks as barriers for source control.


For cloth masks at least, it seems the quote in your second reply answers you question. However, how that is worded might be up for debate.

Some people might word the information in your post as “There aren’t any aerosols particles that are too small to be filtered by masks, because 50%-70% of them do get blocked.” But someone else might word it as “There are some particles too small to be filtered by masks, as only 50%-70% of aerosols get blocked.”

I would, however, argue that the first is more correct, as “filters” doesn’t usually mean “blocks entirely.” And the video being refuted seems to claim that no aerosols get blocked at all.

I would say you answered your own question, but I would love to see figures on other types of masks. You quote comparison with surgical masks, so it would be nice to see numbers on those. And then there are the respirators, like the KN95 and N95 and equivalents.

I suspect at least the latter would block aerosols more, but they might not. I’d also love data on how well they block them coming in, as that’s the main reason people wear the N95, it seems.

Well, keep in mind I’m only an amateur. The CDC link in post #2 also has numerous footnotes which I did not reproduce, but I think they are all pre-Delta variant.


Here’s a thread I started recently on masks and the Delta variant. @Retzbu_Tox posted a bunch of links in a July 20 post.

Awesome, thanks.


What I read 2 years ago, was that none of the medical staff in Wuhan who were working in the COVID isolation area, wearing full PPE and isolation masks, got sick. But that other medical staff wearing ‘surgical’ masks in the general hospital treating the general population did get sick. At the time, this was presented as evidence of the relative effectiveness of different kinds of mask.

Even if the masks don’t necessarily filter terrifically going or coming, they DO slow down the air, and by extension the aerosols. So if you socially distance and wear half-assed masks, that slowing, the dilution experienced by the aerosols over the six foot/2 meter distance, and whatever filtration effect both masks provide is going to make it considerably safer than standing right up on someone unmasked.

That’s the whole thing- both of you, especially the “sender” need to be wearing masks; relying on your own mask for filtration isn’t really the best tactic. And it’s why masks need to be mandated as well- saying it’s your choice is missing the entire boat in terms of how transmission actually works.

It should be understood that ‘surgical’ masks were never intended to protect the wearer from infection. They were designed to be used by people working in a sterile field (i.e. a surgical theater) to prevent the exhalation of droplets that could land on the patient and equipment. The primary concern in a sterile field is bacterial contamination and/or non-biological foreign particles because the body’s basic barrier to the environment (e.g. the skin) is open. Surgical masks made of impermeable polypropylene can limit the travel of exhaled droplets and provide protection to the mucosal surfaces in mouth and nose from the same but are in no way assured protection from aerosol contaminants; that is, particles that can remain suspended in stagnant air for many minutes. Common aerosols are cigarette smoke, perfumes, and the volatile organic compounds emitted by new plastics, all of which you can easily smell even wearing a tightly tied surgical mask because of the open gaps on the side.

Respirator masks like the N95-standard respirator are designed to make a good seal to the face, and provide reasonable protection for the wearer against aerosols for a certain period of time. The melt-blown polypropylene that N95 respirators are made out of stop small aerosols not by physically blocking them as that would also severely inhibit the flow of air but rather by drawing the air over electrostatically-charged fibers which grab the particles, and in the case of bacteria and virions, hold onto them until they desiccate and become inactivated. As the respirator becomes saturated with use, either from material inhaled through it or by moist exhalation, the effectiveness drops, hence why such items are considered to be ‘single use only’. A properly cared for N95 respirator that is not saturated with humidity can function effectively for many hours of use, and as long as it is allowed to dry can be reused for some period although with progressive degradation in filtering function.

Cloth masks are of marginal use in protecting users from aerosols or preventing their spread and I have yet to see a study that demonstrates otherwise with any kind of real control. My own casual experiment with improvised cloth barrier convinced me that they are of negligible utility for protection. There have been a number of studies purporting the utility of cloth masks in real world application based upon the difference in rates of infection between masked (cloth) and unmasked people but they do not account for differences in behavior that one might reasonably expect between someone who is taking the precaution of wearing a mask and someone who isn’t. I see the CDC guidance that someone cited above regarding woven cloth masks but I’ve frankly become somewhat dubious about the basis for that guidance and would personally not rely on a non-respirator mask indoors for protection.


You’re misremembering the details a little bit. Among the original Wuhan doctors, many of them got sick due to rapidly running out of adequate PPE, as happened amongst many countries during the first wave. Shortly after Wuhan was sealed, China sent in a “surge” of 44,000 medical staff from across China to help out in Wuhan and the amazing statistic was that none of the 44,000 staff got sick. One major innovation China made during the surge was that each medical staff was responsible for its own supply line for PPE from their home town hospital to Wuhan, ensuring that everyone could be adequately supplied in a distributed manner. This allowed a standard of protection that other countries would struggle to match from a logistics perspective.

For context, this was the PPE setup that Wuhan medical workers were expected to adopt:


China showed it was possible to adopt a protecting regime at scale that could provide absolute protection to HCW but it went far beyond just masks. Unfortunately, the rest of the world seemed intent on ignoring China’s success and you have many countries where COVID caseload is low and PPE is abundant but medical workers are still getting infected.

Just a quick note here about “aerosol particles that are too small to be filtered”: the very finest aerosol particles are easy to filter out, because they diffuse around so rapidly by Brownian motion. For any filter, there is a moderate particle size that is most difficult to trap, larger particles being easier to trap for several reasons, and smaller ones being easier to trap by diffusion. This is the “most penetrating particle size” or “MPPS”.

I’m guessing the MPPS for masks generally is smaller than infectious particles or droplets. But it isn’t quite right to speak of “too small” in this context.

Well that would be misleading, no? Because full PPE includes covering the eyes which is another potential route of infection.

The problem is, the one thing everyone knows about China is “don’t trust China”. Which is a good rule but unfortunately every news story goes through that same lens.
So this last year and a half, where China as a whole had very few cases, and cities like Beijing reported zero cases, people outside China have been happy to handwave that as propaganda, despite how easy it would be to disprove were it just a lie. Let alone something that could feasibly be lied about, like the number of medical workers that got sick.

I don’t see a detail that I got wrong.

The distinction wasn’t doctors working in COVID wards vs general hospital staff voluntarily wearing surgical masks. It was doctors coming in from other provinces that had the luxury of independent PPE supply lines vs Wuhan doctors in scrounged PPE. The difference wasn’t just the mask but the entire regiment of PPE.

As I wrote in the post you replied to.

But they weren’t wearing surgical masks, they were wearing whatever they could scrounge together. All the doctors were treating COVID patients, just some had inconsistent PPE and others had consistent PPE.

Pardon if this is a repeat:

Fig. 4, in particular, shows the volume size distribution of respiratory particles. Oversimplified, masks are good at cutting off pretty much everything above a few tens of micron, but poor* at reducing anything below a micron, with variable efficacy in between. And when sneezing, coughing, or speaking, we emit a lot of bigger particles that masks work great for. Less so when breathing normally.

I would grade “Covid is spread by aerosol particles that are too small to be filtered by masks” as partially true. It is spread, in part, by particles that are reduced but not eliminated by masks due to their size. It is also spread, in part, by particles that masks work great for. My takeaway from the paper is that masks work, but not perfectly, so distancing and air flow are still important. And more protection is good in saturated environments.

DOI: 10.1126/science.abg6296

*ETA the log scale makes them look worse than they are.